Control of Air Pollution from Motor
Vehicles: Tier 3 Motor Vehicle Emission
and Fuel Standards
Summary and Analysis of Comments
&EPA
United States
Environmental Protection
Agency
-------
Control of Air Pollution from Motor
Vehicles: Tier 3 Motor Vehicle Emission
and Fuel Standards
Summary and Analysis of Comments
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.
&EPA
United States
Environmental Protection
Agency
EPA-420-R-14-004
March 2014
-------
Index of Tier 3 Proposal Commenters
Commenter
A 2nd Opinion Inc.
Advanced Biofuels Association
Advanced Biofuels USA
Advanced Engine Systems Institute
Afton Chemical Corporation
Algenol Biofuels, Inc.
Alliance of Automobile Manufacturers
American Academy of Pediatrics et al.
American Coalition for Ethanol
American Council on Renewable Energy
American Energy Alliance (coordinated
comments2)
American Fuel & Petrochemical
Manufacturers
American Gas Association
American Lung Association
American Lung Association (coordinated
comments)
American Lung Association, District of
Columbia
American Lung Association in Greater
Chicago
American Lung Association in
Massachusetts, et al.
American Lung Association Lehigh
Valley, PA
American Lung Association Maryland
Abbreviation
AFBA
AFBUSA
AESI
Alliance
ACE
ACORE
AFPM
AGA
ALA
ALA
Docket ID Number
EPA-HQ-OAR-2011-
0135-4684
EPA-HQ-OAR-2011-
0135-4318
EPA-HQ-OAR-2011-
0135-4357
EPA-HQ-OAR-2011-
0135-47311
EPA-HQ-OAR-2011-
0135-4359
EPA-HQ-OAR-2011-
0135-4325,4358
EPA-HQ-OAR-2011-
0135-4461
EPA-HQ-OAR-2011-
0135-4760
EPA-HQ-OAR-2011-
0135-4313
EPA-HQ-OAR-2011-
0135-4808, 4892
EPA-HQ-OAR-2011-
0135-4887
EPA-HQ-OAR-2011-
0135-1374, 4276, 4730,
4922, 4923
EPA-HQ-OAR-2011-
0135-4303,4306
EPA-HQ-OAR-2011-
0135-4765, 4920
EPA-HQ-OAR-2011-
0135-4885, 0668, 4620,
4810
EPA-HQ-OAR-2011-
0135-4175
EPA-HQ-OAR-2011-
0135-4278
EPA-HQ-OAR-2011-
0135-4732
EPA-HQ-OAR-2011-
0135-3194
EPA-HQ-OAR-2011-
0135-4174
-------
American Lung Association North
Carolina
American Lung Association South
Central Pennsylvania
American Lung Association Virginia
American Lung Association, Regional
Leadership Council in Northeast Ohio
American Motorcyclist Association
American Petroleum Institute
American Thoracic Society
American Trucking Association
Americas Petroleum Regional Technical
Committee
Appalachian Mountain Club
Association of Global Automakers
Aston Martin Lagonda Ltd.
Attorney General of Connecticut, et. al.
Attorney General of New York, et. al.
Automotive Aftermarket Industry
Association
Automotive Specialty Products Alliance
Biotechnology Industry Organization
BlueGreen Alliance
BMW of North America, LLC
Boulder County Board of County
Commissioners
Boulder County Board of Health
BP Products North America Inc.
Business for Innovative Climate and
Energy Policy
AMA
API
ATA
APTRC
AMC
Global
AAIA
ASPA
BIO
BP
BICEP
EPA-HQ-OAR-2011-
0135-4177
EPA-HQ-OAR-2011-
0135-3452, 3453, 3469
EPA-HQ-OAR-2011-
0135-4173
EPA-HQ-OAR-2011-
0135-4890
EPA-HQ-OAR-2011-
0135-4880
EPA-HQ-OAR-2011-
0135-4276, 4922, 4923
EPA-HQ-OAR-2011-
0135-47303
EPA-HQ-OAR-2011-
0135-4731
EPA-HQ-OAR-2011-
0135-4292, 4916
EPA-HQ-OAR-2011-
0135-4452
EPA-HQ-OAR-2011-
0135-4461
EPA-HQ-OAR-2011-
0135-4262
EPA-HQ-OAR-2011-
0135-4783
EPA-HQ-OAR-2011-
0135-4689
EPA-HQ-OAR-2011-
0135-4300
EPA-HQ-OAR-2011-
0135-4268
EPA-HQ-OAR-2011-
0135-4454
EPA-HQ-OAR-2011-
0135-4730
EPA-HQ-OAR-2011-
0135-4297
EPA-HQ-OAR-2011-
0135-4809
EPA-HQ-OAR-2011-
0135-4809
EPA-HQ-OAR-2011-
0135-4682
EPA-HQ-OAR-2011-
0135-3467
11
-------
California Air Resources Board
Care 2 (coordinated comments)
Ceres Investor Network on Climate Risk
Chesapeake Bay Foundation
Chevron Products Company
Chicago Metropolitan Agency for
Planning
Children's Environmental Health
Network
Chrysler Group LL
CHS, Inc.
City of Philadelphia Department of
Public Health Management Services
Clean Air Council
Clean Air Watch
Clean Fuels Development Coalition
Coordinated comments
Concerned Citizens Around Murphy
Consumer Specialty Products
Association
Consumers Union
Consumers Union (coordinated
comments)
CountryMark Cooperative
Cummins Inc.
Delaware Department of Natural
Resources
Department of Defense Clean Air Act
Services Steering Committee
Dresser Trap Rock, Inc.
CARB
CEHN
CFDC
CSPA
DNRC
EPA-HQ-OAR-2011-
0135-4261,4919
EPA-HQ-OAR-2011-
0135-2679
EPA-HQ-OAR-2011-
0135-4295
EPA-HQ-OAR-2011-
0135-1145
EPA-HQ-OAR-2011-
0135-4758
EPA-HQ-OAR-2011-
0135-4291
EPA-HQ-OAR-2011-
0135-4780
EPA-HQ-OAR-2011-
0135-4326
EPA-HQ-OAR-2011-
0135-4891
EPA-HQ-OAR-2011-
0135-4730
EPA-HQ-OAR-2011-
0135-4730
EPA-HQ-OAR-2011-
0135-4730
EPA-HQ-OAR-2011-
0135-4690
EPA-HQ-OAR-2011-
0135-2753, 2763, 2776,
3197,4254,4255
EPA-HQ-OAR-2011-
0135-3196
EPA-HQ-OAR-2011-
0135-4268
EPA-HQ-OAR-2011-
0135-4602
EPA-HQ-OAR-2011-
0135-4600
EPA-HQ-OAR-2011-
0135-4781
EPA-HQ-OAR-2011-
0135-4309, 4464
EPA-HQ-OAR-2011-
0135-4685
EPA-HQ-OAR-2011-
0135-4274
EPA-HQ-OAR-2011-
111
-------
Earthjustice (coordinated comments)
E.I. du Pont de Nemours and Company
Emissions Control Technology
Association
Energy Future Coalition and Urban Air
Initiative
Environment America (coordinated
comments)
Environmental Defense Fund
Environmental Justice Leadership Forum
on Climate Change
Environmental Law and Policy Center
Evangelical Environmental Network
(coordinated comments)
ExxonMobil
Ferrari
Flint Hills Resources, LP
Ford Motor Company
General Motors LLC
Governors' Biofuels Coalition
Growth Energy
HEAL Utah
HEAL Utah, Utah State Legislators
Hyundai Motor Group
Independent Fuel Terminal Operators
Association
INEOS Bio
Institute for Energy Research
International Council on Clean
Transportation
ECTA
EOF
ELPC
FHR
GM
IFTOA
IER
ICCT
0135-3591
EPA-HQ-OAR-2011-
0135-4301
EPA-HQ-OAR-2011-
0135-4263, 4267, 4293
EPA-HQ-OAR-2011-
0135-4353
EPA-HQ-OAR-2011-
0135-2685
EPA-HQ-OAR-2011-
0135-4355
EPA-HQ-OAR-2011-
0135-4730
EPA-HQ-OAR-2011-
0135-4731
EPA-HQ-OAR-2011-
0135-4683
EPA-HQ-OAR-2011-
0135-4307
EPA-HQ-OAR-2011-
0135-4277
EPA-HQ-OAR-2011-
0135-4287
EPA-HQ-OAR-2011-
0135-4349
EPA-HQ-OAR-2011-
0135-4288
EPA-HQ-OAR-2011-
0135-4252
EPA-HQ-OAR-2011-
0135-4681
EPA-HQ-OAR-2011-
0135-4453
EPA-HQ-OAR-2011-
0135-4455
EPA-HQ-OAR-2011-
0135-4731
EPA-HQ-OAR-2011-
0135-4310
EPA-HQ-OAR-2011-
0135-4691
EPA-HQ-OAR-2011-
0135-3195
EPA-HQ-OAR-2011-
0135-4304,4917
IV
-------
International Federation of Inspection
Agencies
Irving Oil Terminals Inc.
Johnson Matthey
Kentucky Division for Air Quality
Kinder Morgan Transmix Company,
LLC
League of Conservation Voters
(coordinated comments)
Lotus Cars Ltd.
Magellan Midstream Partners, L.P.
Manufacturers of Emission Controls
Association
Marathon Petroleum Company LP
Maryland Department of the
Environment
Maryland Department of Transportation
MathPro
McLaren Automotive Limited
Medical Advocates for Healthy Air
Mercatus Center at George Mason
University
Mercedes-Benz USA, LLC on behalf of
Daimler AG
Metropolitan Washington Air Quality
Committee
Michigan Department of Environmental
Quality
Mid-Atlantic Regional Air Management
Association Inc.
Mid-Continental Energy
Mitsubishi Motors R&D America, Inc
Monroe Energy, LLC
IFIA
MECA
MFC
MWAQC
MDEQ
MARAMA
MCE
MRDA
EPA-HQ-OAR-2011-
0135-4292, 4916
EPA-HQ-OAR-2011-
0135-4260
EPA-HQ-OAR-2011-
0135-4730
EPA-HQ-OAR-2011-
0135-4243
EPA-HQ-OAR-2011-
0135-4893
EPA-HQ-OAR-2011-
0135-4257
EPA-HQ-OAR-2011-
0135-4687
EPA-HQ-OAR-2011-
0135-4284, 4688, 4926
EPA-HQ-OAR-2011-
0135-4675, 4899
EPA-HQ-OAR-2011-
0135-4302
EPA-HQ-OAR-2011-
0135-4253
EPA-HQ-OAR-2011-
0135-4881
EPA-HQ-2001-0135-
4730
EPA-HQ-OAR-2011-
0135-4803
EPA-HQ-OAR-2011-
0135-4266
EPA-HQ-OAR-2011-
0135-4280
EPA-HQ-OAR-2011-
0135-4676
EPA-HQ-OAR-2011-
0135-4055
EPA-HQ-OAR-2011-
0135-4883
EPA-HQ-OAR-2011-
0135-4246
EPA-HQ-OAR-2011-
0135-4807
EPA-HQ-OAR-2011-
0135-4281
EPA-HQ-OAR-2011-
0135-4678
-------
Motor & Equipment Manufacturers
Association
National Association of Clean Air
Agencies
National Association of Convenience
Stores
National Automobile Dealers
Association
National Biodiesel Board
National Corn Growers Association
National Marine Manufacturers
Association
National Propane Gas Association
Natural Resources Defense Council
Natural Resources Defense Council
(coordinated comments)
Navigant Economics
New York State Department of
Environmental Conservation
NGV America
Northeast States for Coordinated Air Use
Management
Office of County Executive, Anne
Arundel County, Maryland
Open Joint-Stock Company
Outdoor Power Equipment Institute
Ozone Transport Commission
PBF Energy Inc.
PennFuture
Penn Environment (coordinated
comments)
Pennsylvania Department of
Environmental Protection
Peoples Republic of China
MEMA
NACAA
NACS
NADA
NBB
NCGA
NMMA
NPGA
NRDC
NESCAUM
OPEI
OTC
EPA-HQ-OAR-2011-
0135-4324
EPA-HQ-OAR-2011-
0135-4275
EPA-HQ-OAR-2011-
0135-4327
EPA-HQ-OAR-2011-
0135-4351
EPA-HQ-OAR-2011-
0135-4312
EPA-HQ-OAR-2011-
0135-4285
EPA-HQ-OAR-2011-
0135-4294
EPA-HQ-OAR-2011-
0135-3200, 4354, 4692
EPA-HQ-OAR-2011-
0135-4286
EPA-HQ-OAR-2011-
0135-2775
EPA-HQ-OAR-2011-
0135-4731
EPA-HQ-OAR-2011-
0135-4298
EPA-HQ-OAR-2011-
0135-4456
EPA-HQ-OAR-2011-
0135-4283
EPA-HQ-OAR-2011-
0135-4882
EPA-HQ-OAR-2011-
0135-1627
EPA-HQ-OAR-2011-
0135-4457
EPA-HQ-OAR-2011-
0135-4459
EPA-HQ-OAR-2011-
0135-4823
EPA-HQ-OAR-2011-
0135-4730
EPA-HQ-OAR-2011-
0135-4272
EPA-HQ-OAR-2011-
0135-4269
EPA-HQ-OAR-2011-
0135-4918
VI
-------
Petroleum Marketers Association of
America
Phillips 66 Company
Philadelphia Physicians for Social
Responsibility
Poet, LLC
Refinery Automation Institute, LLC
Renewable Fuels Association
Respiratory Health Association
Revecorp Inc.
Robert Bosch GmbH, Gasoline Systems,
Germany
Shell Oil Products for Shell and Motiva
Sierra Club
Sierra Club (coordinated comments)
Sierra Club Southeastern Pennsylvania
Group
Sierra Club, Clean Air Watch,
Respiratory Health Association
Small Business Refiners
Society of Independent Gasoline
Marketers of America
Southern California Association of
Governments
State of Utah
Sugar House Community Council
Sunoco Logistics Partners L.P.
Sutherland Asbill & Brennan LLP
Textile Industries, Inc.
Thomas Jefferson University Hospital
PMAA
RFA
RHA
SBR
SIGMA
SCAG
EPA-HQ-OAR-2011-
0135-4265
EPA-HQ-OAR-2011-
0135-4463
EPA-HQ-OAR-2011-
0135-4730
EPA-HQ-OAR-2011-
0135-4462
EPA-HQ-OAR-2011-
0135-3471
EPA-HQ-OAR-2011-
0135-4350
EPA-HQ-OAR-2011-
0135-4731
EPA-HQ-OAR-2011-
0135-4816
EPA-HQ-OAR-2011-
0135-3468
EPA-HQ-OAR-2011-
0135-4805
EPA-HQ-OAR-2011-
0135-4311,4723
EPA-HQ-OAR-2011-
0135-0671, 4242
EPA-HQ-OAR-2011-
0135-4730
EPA-HQ-OAR-2011-
0135-4308
EPA-HQ-OAR-2011-
0135-4804
EPA-HQ-OAR-2011-
0135-4327
EPA-HQ-OAR-2011-
0135-4352
EPA-HQ-OAR-2011-
0135-4879
EPA-HQ-OAR-2011-
0135-4176
EPA-HQ-OAR-2011-
0135-4290
EPA-HQ-OAR-2011-
0135-4884
EPA-HQ-OAR-2011-
0135-4731
EPA-HQ-OAR-2011-
0135-4730
Vll
-------
Truck and Engine Manufacturers
Association
Turner, Mason & Company
U.S. Coalition for Advanced Diesel Cars
Union of Concerned Scientists
Union of Concerned Scientists
(coordinated comments)
United Automobile Workers
United States Congress, House of
Representatives, Pennsylvania, 7th
District
United Steel workers Union
Utah Air Quality Board
Vehicle Services Consulting, Inc.
VNG.CO
Volkswagen Group of America, Inc.
Volvo Car Group
WE ACT for Environmental Justice
Weaver and Tidwell LLP
Wespath Investment Management
Wyoming Refining
EMA
UCS
UAW
USW
VSCI
EPA-HQ-OAR-2011-
0135-4314
EPA-HQ-OAR-2011-
0135-4757
EPA-HQ-OAR-2011-
0135-4759
EPA-HQ-OAR-2011-
0135-4317
EPA-HQ-OAR-2011-
0135-1625
EPA-HQ-OAR-2011-
0135-2184
EPA-HQ-OAR-2011-
0135-4279
EPA-HQ-OAR-2011-
0135-4305
EPA-HQ-OAR-2011-
0135-4197
EPA-HQ-OAR-2011-
0135-4806
EPA-HQ-OAR-2011-
0135-4348
EPA-HQ-OAR-2011-
0135-4299
EPA-HQ-OAR-2011-
0135-4296
EPA-HQ-OAR-2011-
0135-1626
EPA-HQ-OAR-2011-
0135-4458
EPA-HQ-OAR-2011-
0135-1849
EPA-HQ-OAR-2011-
0135-4460
1- Docket Number EPA-HQ-OAR-2011-0135-4731 is testimony from the Chicago, IL public hearing on
April 29, 2013.
2- Comments from the private citizens coordinated/facilitated by a non-governmental organization.
3- Docket Number EPA-HQ-OAR-2011-0135-4730 is testimony from the Philadelphia, PA public hearing
on April 24, 2013.
Vlll
-------
Acronyms and Abbreviations
AAM
ABT
AECD
AEO
A/F
AKI
AQCD
ARV
ASTM
bbl
BC
BOB
BTU
bpcd
BTEX
CAA
CaRFGS
CBOB
CAFE
CARS (or ARE)
CASAC
CBA
CBI
CCD
CCR
CFR
CG
CMAQ
CNG
CO
CO2
COA
CR
CRC
CREE
DALYs
DFE
DI
DoE, DOE
DOR
DPF
DRIA
Alliance of Automobile Manufacturers
Averaging, Banking, and Trading
Auxiliary Emission Control Device
Annual Energy Outlook
air/fuel ratio
Anti-Knock Index ((R+M)/2 octane)
Air Quality Criteria Document
Accepted Reference Value
ASTM International (formerly American Society for Testing
Materials)
Barrel
Black Carbon (commonly referred to as "soot")
Blendstock for Oxygenate Blending
British Thermal Unit
Barrels per Calendar Day
Benzene, Toluene, Ethylbenzene, and Xylene
Clean Air Act
California's Phase 3 Reformulated Gasoline (Program)
Conventional Blendstock for Oxygenate Blending
Corporate Average Fuel Economy
California Air Resources Board
Clean Air Science Advisory Committee
Cost Benefit Analysis
Confidential Business Information
Combustion Chamber Deposit
California Code of Regulations
Code of Federal Regulations
Conventional Gasoline
Community Multi-scale Air Quality Model
Compressed Natural Gasoline
Carbon Monoxide
Carbon Dioxide
Certificate of Analysis
Compression Ratio
Coordinating Research Council
Carbon Related Exhaust Emissions
Disability Adjusted Life Years
Denatured Fuel Ethanol
Direct Injection
U.S. Department of Energy
Direct Ozone Reduction
Diesel Particulate Filter
Draft Regulatory Impact Analysis
IX
-------
DTC
E10
E15
E30
E16-50
E51-83
E85
EFTA
EIA
EISA
EJ
EMTS
EO
EPA
EPAct
ETBE
ETW
EU
FCC
FCEV
FE
FID
FID
FFV
FR
FRM
FTP
FWP
GDI
GHG
GTAB
GVWR
H6C
HAP
HC
HDCC
HDGV
HDV
HEI
HHDGV
HIA
ICE
ICI
Diagnostic Trouble Codes
Gasoline containing 10 percent ethanol by volume
Gasoline containing 1 5 percent ethanol by volume
Gasoline-ethanol blend containing 30 percent ethanol by volume
Gasoline-ethanol blends containing 16 to 50 percent ethanol
volume
by
Gasoline-ethanol blends containing 51-83 percent ethanol by volume
Common name for gasoline-ethanol blends containing 51-83
ethanol by volume
percent
European Free Trade Association
Energy Information Administration
Energy Independence and Security Act of 2007
Environmental Justice
EPA Moderated Trading System
Executive Order
Environmental Protection Agency
Energy Policy Act of 2005
Ethyl Tertiary Butyl Ether
Emission Test Weight
European Union
Fluid Catalytic Cracker
Fuel Cell Electric Vehicle
Fuel Economy
Flame lonization Detector
Fuel Injector Deposit
Flex -Fuel Vehicle
Federal Register
Final Rulemaking
Federal Test Procedure
Fuel Warming Potential
Gasoline Direct Injection
Greenhouse Gas
Gasoline Treated as Blendstock
Gross Vehicle Weight Rating
Harvard Six Cities Study
Hazardous Air Pollutant
Hydrocarbon
High Density Close Coupled
Heavy-duty Gasoline Vehicle
Heavy-duty Vehicle
Health Effects Institute
Heavy Heavy-duty Gasoline Vehicle
Health Impact Assessment
Internal Combustion Engine
Independent Commercial Importer
-------
ILCP
IUVP
IRS
ISA
ISBL
ISO
IVD
LAC
LET
LDV
LOT
LEV
LHDGV
LML
LNG
LPG
LVW
MET
MDPV
mg/kg
MLEB
MMT
MPG
MSAT
MTBE
MY
NAAQS
NATA
NGL
NGO
NHTSA
NLEV
NMHC
NMMAPS
NMOG
NOx
Non-VCSB
NPRM
NRC
NSR
O3
O3 AQCD
OBD
OEM
OMB
Inter-laboratory Cross-check Program
In-Use Verification Program
Internal Revenue Service
Integrated Science Assessment
Inside Battery Limits
International Standards Organization
Intake Valve Deposit
Lowest Additive Concentration
Lean Best Torque
Light-duty Vehicle
Light-duty Truck
Low Emission Vehicle
Light Heavy-duty Gasoline Vehicle
Lowest Measured Level
Liquefied Natural Gas
Liquefied Propane Gas
Loaded Vehicle Weight
Minimum spark advance for Best Torque
Medium-duty Passenger Vehicle
Milligram per kilogram
Mid-level Ethanol Blend
Methylcyclopentadienyl Manganese Tricarbonyl
Miles Per Gallon
EPA's Mobile Source Air Toxics Rule
Methyl Tertiary Butyl Ether
Model Year
National Ambient Air Quality Standard
National Air Toxics Assessment
Natural Gas Liquids
Non-governmental Organization
National Highway Transportation Safety Administration
National Low Emission Vehicle
Non-methane Hydrocarbon
National Morbidity, Mortality, and Air Pollution Study
Non-methane Organic Gas
Oxides of Nitrogen; Nitrogen Oxides
Non- Voluntary consensus-based standards body
Notice of Proposed Rulemaking
National Research Council
New Source Review
Ozone
Ozone Air Quality Criteria Document
Onboard Diagnostics
Original Equipment Manufacturer
(White House) Office of Management and Budget
XI
-------
ON
OTAQ
ORVR
OSBL
PAHs
PHEV
PBMS
P.E.
PFI
PFID
PM
PN
POM
Ppm
PR
PSI
PTD
QALYs
R&D
RBOB
RFA
RFG
RFS
RIA
RICE
RIN
RSD
RVP
S&A
SAB
SAB-Council
SAB-EEAC
SAE
SBA
SB AR Panel
SBREFA
SCF
SCR
SFTP
SHED
SIP
SO2
SOA
Octane Number
Office of Transportation and Air Quality
Onboard Refueling Vapor Recovery
Outside Battery Limit
Polycyclic Aromatic Hydrocarbons
Plug-In Hybrid Electric Vehicle
Performance-Based Measurement System
Professional Engineer
Port Fuel Injected
Port Fuel Injector Deposit
Particulate Matter
Particle Number
Polycyclic Organic Matter
Parts Per Million
Precision Ratio
Pounds per Square Inch (pressure)
Product Transfer Document
Quality- Adjusted Life Years
Research and Development
Reformulated Blendstock for Oxygenate Blending
Regulatory Flexibility Analysis
Reformulated Gasoline
Renewable Fuel Standard
Regulatory Impact Analysis
Reciprocating Internal Combustion Engine
Renewable Identification Number
Remote Sensing Device
Reid Vapor Pressure
Summary and Analysis of Comments for Control of Air Pollution
from Motor Vehicles: Tier 3 Motor Vehicle Emission and Fuel
Standards (this document)
Science Advisory Board
SAB Advisory Council on Clean Air Compliance
Analysis
Environmental Economics Advisory Committee of the SAB
Society of Automotive Engineers
Small Business Administration
Small Business Advocacy Review Panel
Small Business Regulatory Enforcement Fairness
Act
Standard Cubic Feet
Selective Catalytic Reduction
Supplemental Federal Test Procedure
Sealed Housing for Evaporative Determination
State Implementation Plan
Sulfur Dioxide
Secondary Organic Aerosol
Xll
-------
SRM
SVM
THC
TPI
TTB
TZEV
ULSD
UFP
use
VCSB
VMT
VOC
VRR
VSL
VSLY
WOT
WTO
WTP
Standard Reference Material
Small Volume Manufacturer
Total Hydrocarbon
Test Performance Index
Tax and Trade Bureau
True Zero Emission Vehicle
Ultra-low Sulfur Diesel
Ultrafine Particulates
United States Code
Voluntary Consensus-based Standards Body
Vehicle Miles Traveled
Volatile Organic Compound
Value of a Risk Reduction
Value of Statistical Life
Value of Statistical Life Year
Wide Open Throttle
World Trade Organization
Willingness to Pay
Xlll
-------
Table of Contents
Chapter 1 General
Chapter 2 Authority and Need for the Tier 3 Standards
Chapter 3 Emissions and Air Quality Emissions Impacts
Chapter 4 Tier 3 Vehicle Program
Chapter 5 Tier 3 Gasoline Sulfur Program
Chapter 6 Regulatory Streamlining and Technical Amendments
Chapter 7 Costs
Chapter 8 Estimated Benefits of the Proposed Rule
Chapter 9 Economic Impact Analysis
Chapter 10 Statutory Process
Chapter 11 Comments on Proposed Regulatory Text
Chapter 12 Other
xiv
-------
Tier 3 Summary and Analysis of Comments
1. General
The following comments relate in general to the Notice of Proposed Rulemaking
(NPRM). The comments in this chapter are not on any specific aspect of the proposed rule;
rather, they are directed to the general substance of the proposal. More detailed comments on
specific provisions of the proposal can be found in later chapters of this Summary and Analysis
of Comments.
For more information on the proposed rule, see the Federal Register at 78 FR 29816,
published on May 21, 2013. The public comments submitted on this rule can be viewed online
at www.regulations.gov (the public docket for this rulemaking is docket number EPA-HQ-
OAR-2011-0135).
1.1. General Support for the Proposed Standards
What Commenters Said:
We received many comments supporting the proposed rule, which generally stated that
they support the rule itself and/or efforts to reduce pollution from motor vehicles and reduce
sulfur in gasoline. Commenters additionally expressed support for various aspects of the rule,
including: the systems approach of regulating vehicles and fuel together, harmonization with
California low emission vehicle (LEV) III standards, coordination with EPA's light-duty
Greenhouse Gas (GHG) standards, low sulfur fuels enabling advanced vehicle technologies,
help for states in meeting National Ambient Air Quality Standards (NAAQS), and finalizing
the rule by the end of 2013 (in order to retain a January 1, 2017 program start date).
However, many of these commenters also stated that, although they support the rule,
they believe that additional work should be done before the rule is finalized. Commenters
offered various suggestions on how they believe that the rule could be improved, and those
specific comments can be found throughout this Summary and Analysis of Comments
document.
Our Response:
We appreciate the support we have received from these commenters and many other
parties during the development of the Tier 3 final rule. The Tier 3 program will establish more
stringent vehicle emissions standards and will reduce the sulfur content of gasoline beginning
in 2017, as part of a systems approach to addressing the impacts of motor vehicles and fuels on
air quality and public health. As described in the preamble to the final rule, we continue to
believe that the Tier 3 program is necessary, and is technologically and economically feasible
in the time frame allowed.
For responses to specific issues raised in public comments, please see the separate
chapters of this Summary and Analysis of Comments document.
1-1
-------
Chapter 1: General
1.2. General Opposition to the Proposed Standards
What Commenters Said:
We also received comments expressing opposition to the proposed rule. These
comments listed various concerns such as flawed analyses or processes, and cost/supply
impacts. These comments are generally summarized below, and are discussed in more detail in
later chapters of this Summary and Analysis of Comments Document.
Some commenters stated that they believe the rule is unnecessary because EPA failed to
provide an adequate scientific justification, technical need, or cost effectiveness for the rule.
These commenters further stated that the rule would impact fuel domestic supplies and affect
energy security.
Several commenters raised the concern that they believe the rule will result in higher
costs to the refining and/or vehicle manufacturing industry, and thus raise consumer costs.
Some of these commenters noted concerns about impacts on jobs in these industries, and a
number of these commenters further stated that the costs would result in little environmental
benefit.
Individual commenters also stated that they do not agree with EPA's assessments of the
magnitude to which vehicle emissions cause air pollution, or the health benefits of the rule. We
also received a number of comments raising concerns that the rulemaking is an example of
government intrusion.
Commenters:
American Energy Alliance
American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM)
Dresser Trap Rock, Inc.
ExxonMobil
Flint Hills Resources, LP (FHR)
Marathon
Monroe Energy, LLC
Private Citizens
Refinery Automation Institute, LLC
Sunoco Logistics Partners L.P.
United States Congress, House of Representatives, Pennsylvania, 7th District
Our Response:
We have in fact taken the comments we received on the proposal into account and made
changes to some of the programmatic requirements, where appropriate, which should ease
implementation (e.g., gasoline sulfur ABT program design) without jeopardizing the benefits of
the program. We believe that the final rule implements the Tier 3 program in a manner
1-2
-------
Tier 3 Summary and Analysis of Comments
consistent with our legal obligations, with sound science, and with sound environmental,
energy, and economic policy.
For responses to specific issues raised in public comments, please see the separate
chapters of this Summary and Analysis of Comments document.
1-3
-------
Tier 3 Summary and Analysis of Comments
2. Authority Under the Clean Air Act and Need for the Tier 3 Standards
2.1. Authority
2.1.1. Use of Authority
What Commenters Said:
Commenter: Mercatus Center at George Mason University
Additionally, the EPA regulates both PM and Ozone under the National Ambient Air Quality
Standards (NAAQS), which makes the proposed regulation an indirect, and perhaps impractical,
way to achieve the EPA's objectives.
There are also several unusual aspects to this rule worth mentioning. First, a revision of the
ozone NAAQS standards was recently returned to the EPA by the Administrator of the Office of
Information and Regulatory Affairs (13). The EPA was urged to reconsider its ozone proposal
until after a new review of the scientific literature has been conducted by the Clean Air Science
Advisory Committee (CASAC). The EPA may want to consider following the same advice for
this regulation since it is also related to ozone.
Another unusual aspect of the regulation is that one outcome the agency identifies as a basis for
regulating is to help states and localities comply with other EPA regulations. The Notice of
Proposed Rulemaking (NPRM) states, "these reductions would help state and local agencies in
their effort to attain and maintain health-based National Ambient Air Quality Standards
(NAAQS)" (14). The EPA should allow states flexibility to achieve standards in ways they think
are best, rather than mandating how to comply through further regulations.
It seems odd to use Tier 3 gasoline standards to reduce PM levels when the EPA has the
authority to reduce PM directly through the NAAQS.
The EPA has issued this regulation as a result of the authority granted it by the Clean Air Act.
Given that this regulation is not required by statute, the EPA would be well advised to consider
holding off on issuing such a regulation until the benefits are more certain.
Our Response:
The commenter states that "the EPA regulates both PM and Ozone under the National
Ambient Air Quality Standards (NAAQS), which makes the proposed regulation an indirect, and
perhaps impractical, way to achieve the EPA's objectives." The commenter also states that
EPA's Tier 3 rulemaking is "a basis for regulating is to help states and localities comply with
other EPA regulations." These statements appear to confuse the setting of the NAAQS with
EPA's authority to finalize national rulemakings under Clean Air Act Title II. Section 109 (42
U.S.C. 7409) of the Clean Air Act directs the Administrator to propose and promulgate
"primary" and "secondary" NAAQS for criteria pollutants (identified under section 108). The
Act defines primary standards as "ambient air quality standards the attainment and maintenance
2-1
-------
Chapter 2: Authority Under the Clean Air Act and Need for the Tier 3 Standards
of which in the judgment of the Administrator, based on [air quality] criteria and allowing an
adequate margin of safety, are requisite to protect the public health." In setting primary ambient
air quality standards, the EPA's responsibility under the law is to establish standards that protect
public health, without regard to the costs of implementing those new standards. How States
choose to achieve the NAAQS is not prescribed when the NAAQS is promulgated - the NAAQS
do not reduce criteria pollutant levels by themselves.
The Clean Air Act also authorizes EPA to establish emissions standards for motor
vehicles to address air pollution that may reasonably be anticipated to endanger public health or
welfare (section 202). EPA also has authority to establish fuel controls to address such air
pollution as well as fuels emissions products that may impair vehicle emissions controls (section
211). The EPA's exercise of these authorities in promulgating the Tier 3 rule does not conflict
with States' authority and flexibility in choosing control measures for their plans to attain and
maintain the NAAQS. Indeed, as noted by the commenter, emission reductions resulting from
the Tier 3 rule will assist state and local agencies in their efforts to attain and maintain health-
based NAAQS "as expeditiously as practicable" (Section 172(a)2). We received several
comments from states supporting the Tier 3 program, as described elsewhere in this document.
The final Tier 3 rulemaking will provide the public with very significant health benefits that are
achieved at a reasonable cost (refer to Chapter 8 of this Summary and Analysis of Comments
document for specific responses to comments regarding the size and certainty associated with the
Tier 3 rule's estimated benefits).
We also disagree with the commenter's suggestion that EPA should reconsider its current
action on the Tier 3 rulemaking based on the delay of most recent ozone NAAQS. This advice is
irrelevant in the current circumstances due to the very different nature of the two actions - the
setting of a NAAQS and the promulgation of a national mobile-source air quality rule. The
overwhelming consensus of the scientific literature finds human exposure to ozone to be
harmful. The final Tier 3 rule will reduce emissions related to ozone formation and will lead to a
significant public health benefit.
2.1.2 Authority for Gasoline Sulfur Standards
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM)
API opposes the requirement to further reduce average gasoline sulfur with this Tier 3
regulation. This rulemaking is discretionary, and API has serious doubts as to the Agency's
justification for it.
Commenter: Monroe Energy, LLC
The Tier 3 sulfur standards are not required by law.
2-2
-------
Tier 3 Summary and Analysis of Comments
Our Response:
Consistent with our proposal, we are adopting gasoline sulfur controls under our
authority in section 21 l(c)(l) of the Clean Air Act. This section gives us the authority to
"control or prohibit the manufacture, introduction into commerce, offering for sale, or sale" of
any fuel or fuel additive for use in a motor vehicle, motor vehicle engine, or nonroad engine or
nonroad vehicle (1) whose emission products, in the judgment of the Administrator, cause or
contribute to air pollution which may reasonably be anticipated to endanger the public health or
welfare [section 21 l(c)(l)(A)] or (2) whose emission products will impair to a significant degree
the performance of any emission control device or system which is in general use, or which the
Administrator finds has been developed to a point where in a reasonable time it would be in
general use were the fuel control or prohibition adopted [section 21 l(c)(l)(B)]. Consistent with
our proposal, we are finalizing controls on gasoline sulfur levels based on both of the Clean Air
Act criteria.
We believe that the final rule implements the Tier 3 program in a manner consistent with
our legal obligations, with sound science, and with sound environmental, energy, and economic
policy, and we disagree with comments suggesting that the Tier 3 rule is not justified. The Tier 3
program, including the gasoline sulfur standards, will reduce ambient levels of air pollution that
endanger public health and welfare and will provide important benefits to the public such as
preventing PM- and ozone-related premature deaths. The Tier 3 Preamble very clearly explains
the need for the Tier 3 standards in Section II, our technical justification for the vehicle emission
controls in Section IV, and technical justification for the fuel standards in Section V. A vast
body of underlying technical analyses supporting the Tier 3 standards can be found in our
Regulatory Impacts Analysis, which reflects the best methods, data and assumptions available at
the time of the rulemaking analysis.
2.1.3 General Support
What Commenters Said:
Commenter: American Lung Association
The Clean Air Act grants EPA the authority to set standards for vehicles and fuels and to reduce
air pollution that threatens public health under Section 211. The Clean Air Act grants the EPA
administrator the authority to limit the sulfur and gasoline which reduces the efficiency of
emission control technologies, and leads to greater tailpipe pollution. When pollution from
motor vehicles endangers public health, Section 202 of the Act requires the EPA administrator to
take necessary action.
Commenter: Mid-Atlantic Regional Air Management Association Inc. (MARAMA)
State and local air pollution control agencies have very limited authority to control motor vehicle
emissions and fuels. Section 177 of the Clean Air Act authorizes states to opt into California's
2-3
-------
Chapter 2: Authority Under the Clean Air Act and Need for the Tier 3 Standards
Low Emissions Vehicle tailpipe standards, but Section 211 of the Clean Air Act limits states'
ability to establish clean gasoline standards.
Commenter: Sierra Club
Under both provisions, the statutory threshold for determining endangerment is precautionary in
nature, allowing for regulatory action before the threatened harm occurs.13 "Indeed, the very
existence of such precautionary legislation would seem to demand that regulatory action precede,
and, optimally, prevent, the perceived threat." Moreover, the Administrator's endangerment
determination is entitled to great deference, so long as the Administrator provides adequate
public endangerment justification.15 The statute "allows for a somewhat attenuated chain of
causation ... Regulation may be premised on a determination that an air pollutant emitted from a
new automobile is likely to contribute to air pollution which endangers the public health.16 The
Administrator's charge of protecting the public from danger, dictates that the Administrator "be
accorded flexibility, a flexibility that recognizes the special judicial interest in favor of protection
of the health and welfare of people, even in areas where certainty does not exist."17
Upon making its threshold endangerment determination, as EPA has appropriately done here, the
agency has an explicit duty to promulgate standards under Section 202. The statute provides
further direction on the stringency of such standards.
[Regulations ... applicable to emissions of hydrocarbons, carbon monoxide, oxides of nitrogen,
and paniculate matter from classes or categories of heavy-duty vehicles or engines ... shall
contain standards which reflect the greatest degree of emission reduction achievable through the
application of technology which the Administrator determines will be available for the model
year to which such standards apply, giving appropriate consideration to cost, energy, and safety
factors associated with the application of such technology.
And further: The Administrator shall promulgate ... regulations applicable to evaporative
emissions of hydrocarbons from all gasoline-fueled motor vehicles ... shall take effect as
expeditiously as possible and shall require the greatest degree of emission reduction achievable
by means reasonably expected to be available for production during any model year to which the
regulations apply.
Section 202's mandatory duty requires EPA to assess "the greatest degree of emission reduction"
that is "achievable"; that assessment should be informed by the authority conferred to it under
Section 211 of the Act. Section 211 authorizes the Administrator to "regulat[e], control or
prohibit" motor vehicle fuels and fuel additives if the subject fuel or fuel additive or any
emission product from it may contribute to the endangerment of public health or welfare, or if
the emission products of such fuel or fuel additive will impair to a significant degree the
performance of any emission control device or system which is in general use or has been
developed to a point where in a reasonable time it would be in general use if such regulation
were adopted. As described above, EPA has met its initial threshold consideration by providing
strong evidence of public health and welfare endangerment.38 EPA also proposes the integrative
Tier 3 standard based on evidence that gasoline sulfur impairs the emissions control systems of
vehicles.39 Accordingly, EPA should exercise its authority to regulate pursuant to Section 211,
2-4
-------
Tier 3 Summary and Analysis of Comments
and should use this authority to inform the stringency of pollution control limits that are
"achievable" under Section 202.
EPA Demonstrates that the Proposed Rule is Technologically and Economically Feasible:
Moreover, in assessing "achievability," EPA fulfills its obligation to appropriately consider
technological feasibility, economic feasibility and energy impacts. EPA is entitled to significant
deference in demonstrating feasibility in light of the complex scientific and technical analysis it
must undertake. "EPA need only show that technology will be available in the lead time
provided, not that the technology is currently available." "EPA will have demonstrated the
reasonableness of its basis for prediction [that standards are feasible] if it answers any theoretical
objections ..., identifies the major steps necessary in refinement of the [technology], and offers
plausible reasons for believing that each of those steps can be completed in the time available."46
In addition, cost is a consideration but should not be an obstacle for promulgating technology
forcing regulation.47[EPA-HQ-°AR-2011-°135-4311-A1p9]
In sum, EPA's record amply demonstrates the pollutant reduction levels and fuel quality controls
comprising the proposed Tier 3 Rule are technologically and economically feasible. Indeed, EPA
properly uses its authority to protect public health from motor vehicle air pollution by proposing
an all-system, integrative Tier 3 standard, and Sierra Club urges EPA to finalize the rule without
delay.
Sierra Club applauds the U.S. Environmental Protection Agency for exercising its Clean Air Act
authority to promulgate the all-system integrative Tier 3 motor vehicle pollution and fuel content
standards. The proposed Tier 3 standards will achieve significant air pollution reductions that
will provide very real public health benefits, that cannot be achieved without the Rule's proposed
low sulfur gasoline standard. On behalf of Sierra Club's millions of members and supporters, we
urge EPA to finalize these standards before December 31, 2013.
1242U.S.C. §7521(c).
13 Ethyl Corp. v. Envtl. Prot. Agency, 541 F.2d 1. 13 (D.C. Cir. 1976).
15 Train v. Natural Resources Defense Council, Inc., 421 U.S. 60, 75, 95 (1975); Nat'I Petrochemical &
Refiners Ass'n v. EPA, 287 f.3d 1130 (D.C. Cir. 2002).
16 Ethyl Corp. at 16.
17 Id. (referencing Environmental Defense Fund, Inc. v. Ruckelshaus, 439 F.2d 584, 598 (1971).).
38 Ethyl Corps .v. Envtl. Prot. Agency, 541 F.2d 1, 23-23 (D.C. Cir. 1976) (endangerment threshold must
be based on assessment of risk and proof of facts).
39 Id. (threshold determination must be factually based). EPA's research demonstrates significant
reductions in NOx, CO, and total HC for Tier 2 vehicles as a result of sulfur content 10 ppm and lower.
EPA makes the case that the proposed standard is an inseparable and indispensable component of
achieving meaningful air quality benefits. Lowering the sulfur content of fuels both "enablefs] vehicles
designed to the proposed Tier 3 tailpipe exhaust standards to meet these standards for the duration of their
useful life, and [] facilitatefs] immediate emission reductions from all the vehicles on the road at the time
the sulfur controls are implemented." Control of Air Pollution from Motor Vehicles: Tier 3 Motor Vehicle
Emission and Fuel Standards. U.S. Environmental Protection Agency, Proposed Rule, RIN 2060-AQ86
at 67 (Mar. 29, 2013)(hereinafter "Proposed Rule"). Without the proposed sulfur content standards, the
Tier 3 pollution reductions cannot be achieved. Id.
46NRDCv. EPA, 655 F.2d 318, 328, 333 (D.C. Cir. 1981); See alsoHusqvarnav. EPA, 254 R.3d 195,
201 (EPA not obliged to provide detailed solutions to every engineering problem, but only need to
identify the major steps for improvement and give plausible reasons for its belief that the industry will be
2-5
-------
Chapter 2: Authority Under the Clean Air Act and Need for the Tier 3 Standards
able to solve those problems in the time remaining); NRDC v. Thomas, 805 F.2d 410, 429-430 (D.C. Cir.
1986).
47 Husqvarna AB v. E.P.A., 254 F.3d 195, 201 (D.C. Cir. 2001)("In construing similar language included
in CAA section 202, we explained in NRDC v. Thomas that the mere fact that the provisions 'seek to
promote technological advances while also accounting for cost does not detract from their categorization
as technology-forcing standards.' 805 F.2d at 428 n. 30. The 'Congress intended the agency to project
future advances in pollution control capability. It was 'expected to press for development and application
of improved technology rather than be limited by that which exists today.' NRDC v. EPA, 655 F.2d 318,
328 (D.C.Cir.1981) (quoting S.Rep. No. 91-1196, at 24 (1970))."
Commenter: Sierra Club
Similarly, Section 211 authorizes the Administrator to regulate and/or prohibit fuels or fuel
additives if the fuel or fuel additive, or emission product of the fuel causes or contributes to air
pollution that may reasonably be anticipated to endanger the public health or welfare. The statute
states:
The Administrator may, on the basis of information ... by regulation, control or prohibit the
manufacture, introduction into commerce, offering for sale, or sale of any fuel or fuel additive
for use in a motor vehicle, motor vehicle engine ... if, in the judgment of the Administrator, any
fuel or fuel additive or any emission product of such fuel or fuel additive causes, or contributes,
to air pollution or water pollution ... that may reasonably be anticipated to endanger the public
health or welfare, or (B) if emission products of such fuel or fuel additive will impair to a
significant degree the performance of any emission control device or system which is in general
use, or which the Administrator finds has been developed to a point where in a reasonable time it
1 9
would be in general use were such regulation to be promulgated.
12. 42U.S.C. §7521(c).
Commenter: Union of Concerned Scientists (UCS)
Additionally, Section 211 of the Clean Air Act allows EPA to establish a fuel standard if
emissions from fuel combustion cause or contribute to pollution that endangers public health or
the fuel impairs the performance of the emissions control device or system. EPA is using both
authorities in promulgating this rule.
Commenter: Sierra Club
EPA Makes an Appropriate Threshold Determination that Motor Vehicle Pollution Endangers
Public Health
Clean Air Act Sections 202 and 212 Require Regulation Where Pollution Endangers Public
Health or Welfare
Clean Air Act Sections 202 and 212 provide broad authority to the Environmental Protection
Agency to regulate motor vehicle pollution and fuel quality to reduce pollution that may
reasonably be anticipated to endanger public health or welfare. Specifically, Section 202 requires
2-6
-------
Tier 3 Summary and Analysis of Comments
the Administrator to regulate motor vehicle pollutant emissions upon making a public
endangerment determination. The statute states: [EPA-HQ-OAR-2011-0135-4311-A1 p. 3]
The Administrator shall by regulation prescribe (and from time to time revise) in accordance
with the provisions of this section, standards applicable to the emission of any air pollutant from
any class or classes of new motor vehicles or new motor vehicle engines, which in his judgment
cause, or contribute to, air pollution which may reasonably be anticipated to endanger public
health or welfare ...regulations ... shall contain standards which reflect the greatest degree of
emission reduction achievable through the application of technology which the Administrator
determines will be available for the model year to which such standards apply, giving
appropriate consideration to cost, energy, and safety factors associated with the application of
such technology.11
11 42U.S.C. §7521 (a)(l),(3)at2511-29.
Commenter: Union of Concerned Scientists (UCS)
Section 202 of the Clean Air Act requires that the Administrator of the Environmental Protection
Agency (EPA) set emission standards from motor vehicles if reductions from vehicles are
needed and cost-effective. President Obama directed the EPA to promulgate the Tier 3 standards
at the same time that he asked for the greenhouse gas standards for passenger vehicles in a
Presidential memorandum released May 21, 2010 (3). [EPA-HQ-OAR-2011-0135-4317-A1, p.
2^_
3 - The White House. 2010. Presidential Memorandum Regarding Fuel Efficiency Standards. Online at
http://www.whitehouse.gov/the-press-office/presidential-memorandum-regarding-fuel-efficiency-
standards accessed on June 28, 2013
Our Response:
We acknowledge the comments and their support of our authority for the Tier 3
rulemaking.
2.2. Need for the Tier 3 Standards
What Commenters Said:
EPA received numerous comments that affirm the need for the emissions reductions, air quality
improvements, and health benefits that will result from the Tier 3 program. These comments are
from a broad range of stakeholders, including state and local governments, emissions control
suppliers, environmental organizations, health organizations, consumer groups, labor groups,
private citizens, and others. The following list illustrates the breadth and variety of commenters
that have expressed the need for the Tier 3 standards:
Advanced Engine Systems Institute (AESI)
American Academy of Pediatrics
American Lung Association
American Lung Association District of Columbia
2-7
-------
Chapter 2: Authority Under the Clean Air Act and Need for the Tier 3 Standards
American Lung Association Illinois
American Lung Association in Greater Chicago
American Lung Association in Massachusetts et al.
American Lung Association Lehigh Valley, Pennsylvania
American Lung Association Maryland
American Lung Association North Carolina
American Lung Association of the Mid-Atlantic
American Lung Association South Central Pennsylvania
American Lung Association State and Local Chapters
American Lung Association Virginia
American Lung Association, Regional Leadership Council in Northeast Ohio
American Public Health Association
American Thoracic Society
Appalachian Mountain Club
Asthma and Allergy Foundation of America
Attorneys General of New York, Connecticut, Delaware, Maryland, Massachusetts, Maine,
New Hampshire, North Carolina, Oregon, Rhode Island, Vermont, Washington,
Washington D.C., New York City, and Chicago
BlueGreen Alliance
Boulder County Board of Commissioners and Boulder County Board of Health
Business for Innovative Climate and Energy Policy (BICEP)
Ceres Investor Network on Climate Risk
Chesapeake Bay Foundation
Chrysler Group LLC
Chicago Metropolitan Agency for Planning
Children's Environmental Health Network
City of Philadelphia Department of Public Health Management Services (AMS)
Clean Air Council
Consumers Union (CU)
Delaware Department of Natural Resources
Emissions Control Technology Association (ECTA)
Environmental Defense Fund
Environmental Law and Policy Center (ELPC)
HEAL Utah on behalf of members of the Utah Legislature
Health Care without Harm
Kentucky Division for Air Quality
Maryland Department of the Environment
Maryland Department of Transportation
Medical Advocates for Healthy Air and Clean Air Carolina
Metropolitan Washington Air Quality Committee
Mid-Atlantic Regional Air Management Association (MARAMA)
Mom's Clean Air Force
National Association of Clean Air Agencies (NAC AA)
National Association of City and County Health Officials
Natural Resources Defense Council (NRDC)
Navigant Economics
2-8
-------
Tier 3 Summary and Analysis of Comments
New York State Department of Environmental Conservation (NY DEP)
Northeast States for Coordinated Air Use Management (NESCAUM)
Office of County Executive, Anne Arundel County, Maryland
Ozone Transport Commission (OTC)
PennFuture
Pennsylvania Department of Environmental Protection (PA DEP)
Philadelphia Physicians for Social Responsibility
Private Citizens (over 170,000 comments)
Sierra Club
Sierra Club Southeastern Pennsylvania Group
State of Utah
Sugar House Community Council
Trust for America's Health
U.S. Coalition for Advance Diesel Cars
Union of Concerned Scientists (UCS)
United Steelworkers Union (USW)
Utah Air Quality Board
WE ACT for Environmental Justice
Wespath Investment Management
Many comments submitted by the organizations listed above include extensive discussion of why
the Tier 3 standards are needed both nationally and for the particular locales represented by the
commenters. In general, commenters made the following points regarding the need for the Tier
3 standards:
- Tier 3 will significantly reduce tailpipe emissions of harmful pollutants such as
particulate matter and gaseous pollution, including nitrogen oxides and volatile organic
compounds which are ozone precursors. For example:
o In addition to citing EPA's analyses of emission reductions from the proposed
Tier 3 standards, many commenters cite a 2011 study by the National Association
of Clean Air Agencies (NAC AA) that estimated the costs and air quality benefits
of a Tier 3 program modeled on California's Low Emissions Vehicle (LEV III)
program, including tighter vehicle emissions standards and an average gasoline
sulfur standard of 10 ppm. According to the NACAA study, the proposed Tier 3
standards would to cut emissions of NOx, CO and VOCs by 29%, 38% and 26%
respectively by 2030.
Emission reductions from the Tier 3 gasoline sulfur standards will be immediate for the
existing fleet. For example:
o Commenters again cite the 2011 NACAA study showing that reducing the sulfur
content of gasoline would have an immediate effect in 2017, with an expected
260,000 ton reduction of NOx. The NACAA study estimates that is the equivalent
of taking 33 million cars off the nation's roads for a year.
The Tier 3 standards are needed to improve air quality, both nationally and locally. For
example:
2-9
-------
Chapter 2: Authority Under the Clean Air Act and Need for the Tier 3 Standards
o Commenters cited EPA's proposed Tier 3 analyses and the American Lung
Association's State of the Air 2013 report, which found that nearly 132 million
people, or 42 percent of the nation's population, live in areas where ozone and
particulate matter are at levels that are "unhealthful."
• Commenters expressed concern for the general public and especially for
sensitive populations such as children, older adults, those with existing
conditions such as asthma, reduced lung function, chronic obstructive
pulmonary disease, heart disease and diabetes.
o Commenters from specific locales often mentioned specific examples of poor air
quality in their areas, or reported failing grades for various measurements of
ozone and particle pollution in the air according to ALA's State of the Air 2013
report.
The Tier 3 standards are critical to helping areas to attain the National Ambient Air
Quality Standards (NAAQS), as well as assisting areas in staying in attainment. For
example:
o Numerous state commenters report on their difficulty in achieving and
maintaining the current NAAQS and express concerns over meeting potentially
tighter future standards. They call for national Tier 3 standards to help bring
needed emission reductions to their states, especially in those areas affected by
pollutant transport.
o The Ozone Transport Commission (OTC) concluded based on its own modeling
that "attainment of the 2008 health-based ozone standard will be impossible in the
OTR without additional emission reductions from highway vehicles and other
mobile sources."
o Some commenters point out that their states have already implemented many
stringent controls, and in the absence of national Tier 3 program, areas needing
additional emission reductions may have to implement more expensive, less cost-
effective measures, including additional controls on stationary sources (including
already controlled sources) to meet their statutory clean air obligations.
• Achieving equivalent emission reductions of the magnitude that will result
from Tier 3 could be extremely difficult, if not impossible, in areas where
stationary sources are already highly controlled and there are not enough
other sources to implement controls.
Tier 3 will provide substantial health benefits, including preventing instances of
premature deaths, respiratory related ER visits and hospitalizations, acute respiratory
symptom days, including asthma attacks, and missed days of work or school. The Tier 3
standards will also save health care costs. For example:
o Many commenters cited the American Lung Association's report, "A Penny for
Prevention: The Case for Cleaner Gasoline and Vehicle Standards," which
estimated the full implementation of cleaner gasoline and vehicles in 2030. For
the eastern half of the United States only, the report projected that Tier 3 would
prevent more than 2,500 premature deaths each year, avoid more than 15,000
asthma attacks each year, and avert more than 3.1 million missed work and school
2-10
-------
Tier 3 Summary and Analysis of Comments
days each year. The monetized health and economic benefits would range
between $8.5 billion and $22 billion annually.
o Some commenters also cite a study by Navigant Economics showing that the
health benefits of the Tier 3 program have an estimated value of $5 to $6 billion
annually by 2020, and $10 to $11 billion annually by 2030.
The Tier 3 program is worth the cost. For example:
o Many commenters make two related points: the Tier 3 program can be achieved at
a reasonable cost, and the benefits of the Tier 3 program far outweigh the costs.
Commenters cite analyses such as EPA's Regulatory Impact Analysis of the Tier
3 (noting that the costs analyses are corroborated by independent experts), the
American Lung Association's report, "A Penny for Prevention: The Case for
Cleaner Gasoline and Vehicle Standards," and the Navigant Economics study.
Tier 3 is important for reducing near-roadway concentrations of vehicle emissions
o Commenters pointed to studies linking roadway-related air pollution with
increased risks and rates of asthma and other respiratory illnesses, such as the
2010 report by the Health Effects Institute (HEI), which showed that near-
roadway exposure to traffic pollution is high and affects a larger population than
previously thought.
o Commenters also expressed that traffic-related air pollution and the health
impacts associated with such pollution raise significant environmental justice
concerns.
The Tier 3 standards have important environmental benefits. For example:
o Commenters note that pollutants from motor vehicle emissions (NOx, particulate
matter, ozone) contribute to the acidification of lakes and streams, the loss of
native species, poor air quality, and poor visibility. Emission reductions due to the
Tier 3 standards will reduce damage caused by acid deposition, reduce
eutrophication, and reduce visibility impairment
o Some commenters cite specific examples of areas where such environmental
benefits are especially important, e.g., national parks, the Chesapeake Bay.
Tier 3 provides economic and employment benefits. For example:
o Some commenters point to the high cost of health care, especially emergency
room visits and hospital admissions, and applaud the expected cost savings
projected by EPA's analyses, the American Lung Association's report, "A Penny
for Prevention: The Case for Cleaner Gasoline and Vehicle Standards," and the
Navigant Economics study, as illustrated above.
o Some commenters note that Tier 3 will create jobs at refineries and at high-tech
companies developing and deploying state of the art emission control equipment
for vehicles, according to the same Navigant Economics study
o Some commenters also express concerns about the economic burden that would
be placed on local industries and businesses in areas that cannot meet their
statutory clean air obligations without the Tier 3 standards
2-11
-------
Chapter 2: Authority Under the Clean Air Act and Need for the Tier 3 Standards
The Tier 3 rule should be implemented as soon as possible. For example:
o Many commenters expressed their strong desire for the EPA to issue the final Tier
3 standards by the end of 2013 and implement the standards starting in 2017 as
proposed, in order to realize the emission reductions, air quality improvements,
and health benefits of the Tier 3 standards as soon as possible.
o Some commenters specifically expressed concerns about lost benefits due to any
delay in the standards.
EPA also received comments from oil industry commenters questioning the need for the Tier 3
standards. The commenters include:
American Petroleum Institute (API)
Association of Fuel & Petrochemical Manufacturers (AFPM)
Chevron
Flint Hills Resources, LP (FHR)
Marathon Petroleum Company LP (MFC)
Monroe Energy LLC
Commenters claim that EPA has not demonstrated the need for the Tier 3 gasoline sulfur
standards and make the following general points. These points are expounded upon in the
commenter's detailed comments and thus are also addressed throughout this Summary and
Analysis of Comments document:
The environmental benefits of the Tier 3 are negligible with respect to both reductions in
emissions inventories and improvements in air quality. For example:
o Commenters cite two-API sponsored studies conducted by ENVIRON which
showed that in 2022, the summertime ozone precursor emissions of volatile
organic compounds (VOCs), oxides of nitrogen (NOx), and carbon monoxide
(CO) from gasoline-fueled light-duty vehicles are projected to be reduced by 8%,
11% and 7%, respectively, due to Tier 3. ENVIRON also found that, in 2022,
Tier 3 would yield a maximum ozone benefit of about less than 1 ppb and mean
monthly summer 2022 PM2.5 concentrations of no more than 0.1 ug/m3. The
commenters point out that EPA's modeling calculates Tier 3 reductions in ozone
of 0.5 - 1.35 ppb and in PM2.5 of 0 - 0.05 ug/m3 in years 2017 - 2030. The
commenters conclude that reductions of this magnitude are "negligible."
The benefits of the Tier 3 program are not worth the cost. For example:
o The commenters claim that EPA's analyses of the impacts of the Tier 3 as
described in the Draft Regulatory Impact Analysis are flawed and conclude that
"the issues below, when taken collectively, demonstrate that the costs of the Tier
3 proposed rule are greater than the benefits."
• A flawed baseline
• Underestimated costs
• Overestimated emissions benefits
2-12
-------
Tier 3 Summary and Analysis of Comments
• Implausible health benefits due to highly conservative, unrealistic
assumptions
o The commenters further suggest that EPA should withdraw and resubmit a new
DRIA consistent with OMB guidelines.
Implementation of the gasoline sulfur standards in 2017 is not necessary. For example:
o Some commenters argue that a January 1, 2017 start date for implementation of
the gasoline sulfur standards is not necessary to reduce emissions from the in-use
vehicle fleet. The commenters:
• Question that the Tier 3 standards will have immediate emission
reductions for the existing vehicle fleet
• Point to the ENVIRON studies of Tier 3 impacts mentioned above to
argue that the emissions impacts of reducing gasoline sulfur will have only
a de minimis impact on air quality.
• Further claim that the implementing the rule on January 1, 2017 will not
help nonattainment areas reach attainment of the NAAQS.
References:
American Lung Association (2013). State of the Air 2013. Available at:
http://www.stateoftheair.org/2013/assets/ala-sota-2013 .pdf.
• American Lung Association (2013). A Penny for Prevention: The Case for Cleaner Gasoline and
Vehicle Standards. Available at: http://www.lung.org/healthy-air/outdoor/defending-the-clean-
air-act/interactive-presentations/cleaner-gasoline-and-vehicles-report-2013 .pdf
• ENVIRON International Corporation (2013). Effects of Light-duty Vehicle Emissions Standards
and Gasoline Sulfur Level on Ambient Ozone. Available at:
http://www.api.org/~/media/Files/News/2013/13-April/ENVIRON-Sep2012-Effects-of-LDV-
Emiss-Stds-Gasoline-Sulfur-level-on-Ozone.pdf.
ENVIRON International Corporation (2013). Effects of Light-duty Vehicle Emissions Standards
and Gasoline Sulfur Level on Ambient Particulate Matter. Available at:
http://www.api.Org/~/media/Files/Policy/Alternatives/Environ-API-report-vehicle-emissions.pdf.
• Health Effects Institute Panel on the Health Effects of Traffic-Related Air Pollution (2010).
Traffic-related air pollution: A critical review of the literature on emissions, exposure, and health
effects. HEI Special Report 17. Available at: http://pubs.healtheffects.org/view.php?id=334
• National Association of Clean Air Agencies (2011). Cleaner Cars, Cleaner Fuel, Cleaner Air: The
Need for and Benefits of Tier 3 Vehicle and Fuel Regulations. Available at:
http://www.4cleanair.org/documents/NACAATier3VehandFuelReport-EMBARGOED-
Oct2011.pdf
• Navigant Economics (2012). Economic Analysis of the Implications of Implementing EPA's
Tier 3 Rules. Available at: http://www.ectausa.com/061212-Economic-Analysis-of-the-
Implications-of-Tier-3-Sulfur-Reduction-Final_embargoed.pdf
Our Response:
We agree with the majority of commenters that there is a significant need for the
emissions reductions, air quality improvements, and health benefits provided by the Tier 3
standards, and we disagree with comments from the fuel industry arguing that we have not
2-13
-------
Chapter 2: Authority Under the Clean Air Act and Need for the Tier 3 Standards
clearly demonstrated that need. As documented in our Regulatory Impacts Analysis (RIA) of the
Tier 3 rule (Chapters 7 and 8) and described in the Preamble (Sections II, III, and VIII), the Tier
3 program will significantly reduce motor vehicle emissions, leading to significant air quality
improvements and helping state and local areas attain and maintain the existing health-based air
quality standards in a cost-effective and timely way. The breadth and variety of comments we
received describe the importance of Tier 3's expected impacts from the perspective of states,
environmental and health organizations, and the public further demonstrates that the air quality
improvements and health benefits that will result from the Tier 3 standards are needed.
With respect to the commenters' points about the expected emission impacts of the
standards, we agree with the majority of commenters that Tier 3 standards will significantly
reduce tailpipe emissions of nitrogen oxides (NOx), Volatile Organic Compounds (VOC),
particulate matter (PM^.s), and air toxics. For example, our emissions and air quality analyses of
the final Tier 3 standards (Section III of the Preamble, Chapter 7 of the RIA) show that in 2030,
NOx and VOC emissions from on-highway vehicles will be reduced by about 330,000 tons and
170,000 tons, respectively, or about 25% and 16% of emissions from on-highway vehicles.
These reductions will continue beyond 2030 as more of the fleet turns over; by 2050, when Tier
3 vehicles would make up almost the entire fleet, NOx and VOC will be reduced by nearly 31%
for on-highway vehicles. Moreover, immediate reductions are expected in 2017 when the
gasoline sulfur standards take effect. The gasoline sulfur standards, which take effect in 2017,
will provide large immediate reductions in NOx emissions from existing gasoline vehicles and
engines on the road today, e.g., NOx emissions will be reduced by about 260,000 tons, or about
10% of emissions from on-highway vehicles, in 2018 alone. Given these important emission
reductions, we disagree with fuels industry commenters that the emission reductions due to Tier
3 are "negligible." We respond to more specific comments on our emissions inventory modeling
and estimated emissions reductions in Chapter 3.1 of this Summary and Analysis of Comments
document.
Similarly, we agree with the majority of commenters that the Tier 3 standards will
significantly improve air quality and that they are critical to helping areas attain and maintain the
National Ambient Air Quality Standards (NAAQS). As demonstrated in Sections II and III of
the Preamble, and Chapter 7 of the RIA, the emissions reductions from the Tier 3 standards are
projected to lead to significant decreases in ambient concentrations of ozone, PM2.5 and air toxics
(including notable nationwide reductions in benzene concentrations) by 2030, and will
immediately reduce ozone when the sulfur controls take effect in 2017. Over 149 million people
currently live in areas designated nonattainment for one or more of the current NAAQS. In the
absence of additional controls such as Tier 3 standards, many areas will continue to have ambient
ozone and PM2.5 concentrations exceeding the NAAQS in the future. Our air quality modeling
indicates this action will meaningfully decrease ozone concentrations in many areas of the
country. Furthermore, numerous state air quality agencies and organizations representing
geographic groups of these state agencies have affirmed in their comments that air quality
improvements from Tier 3 will be a critical part of areas' strategies to attain and maintain the
NAAQS. Our analysis of the air quality impacts of the Tier 3 standards clearly shows that these
improvements are significant, and the need for these improvements is corroborated by the
majority of comments themselves as well as other studies of Tier 3's impacts presented in these
comments. Thus, we disagree with fuel industry comments that the air quality improvements
2-14
-------
Tier 3 Summary and Analysis of Comments
due to Tier 3 are "negligible," Our responses to specific claims about our air quality analysis are
found in Chapter 3.2 of this document.
We agree with the many commenters who stated that the final Tier 3 rulemaking will provide the
public with very significant health benefits that are achieved in a cost-effective manner. Our
analysis of the final rulemaking (Preamble Section VIII and RIA Chapter 8) estimates that by
2030, the annual emission reductions of the Tier 3 vehicle and fuel standards will annually
prevent between 660 and 1,500 PM-related premature deaths, between 110 and 500 ozone-
related premature deaths, 2,200 hospital admissions and asthma-related emergency room visits,
19,000 asthma exacerbations, 30,000 upper and lower respiratory symptoms in children, and 1.3
million lost school days, work days, and minor restricted activity days. Given the expected
health benefits stemming from air quality improvements due to Tier 3's expected emissions
reductions, we again conclude that Tier 3's emission and air quality impacts are not "negligible"
as the fuel industry commenters claim. The estimated annual monetized health benefits of the
Tier 3 standards in 2030 (2011$) will be between $7.4 and $19 billion, assuming a 3-percent
discount rate (or between $6.7 billion and $18 billion assuming a 7-percent discount rate). We
project the final fuel standards to cost on average 0.65 cent (i.e., less than a penny) per gallon of
gasoline, and the final vehicle standards to have an average cost that increases in proportion to
the increase in stringency during the phase-in period, from $28 per vehicle in 2017 to $72 per
vehicle in 2025, when the standards are fully phased in (our cost analyses can be found in
Preamble Section VII and RIA Chapters 2, 5 and 8). Costs in 2030 are estimated to be
approximately $1.5 billion. Using the more conservative benefits estimate, the 2030 benefits
outweigh the costs by a factor of 4.5. Using the upper end of the benefits range, the benefits
outweigh the costs by a factor of 13. Thus, even taking the most conservative benefits
assumptions, benefits of the final standards clearly outweigh the costs.
We therefore disagree with comments suggesting that the benefits of the Tier 3 program
are not worth the cost. We further disagree with each of the fuel industry commenters' specific
points that our baseline is flawed, that we underestimate costs and overestimate benefits, and that
estimated health benefits are implausible due to unrealistic assumptions. Our methods have been
thoroughly peer reviewed and reflect the best methods, data and assumptions available at the
time of the rulemaking analysis. Furthermore, these methods are consistent with OMB and
internal EPA guidelines for analyzing the impacts of a national-level rulemaking, and we reject
the claim made by fuel industry commenters that EPA should withdraw and resubmit a new
Regulatory Impacts Analysis. Chapters 3, 4, 5, 7, 8 and 10 of this document contain more
detailed responses to the commenters' specific claims.
With respect to comments on other impacts of the Tier 3 standards, we agree with the
many commenters who stated that Tier 3 is important for reducing near-roadway concentrations
of vehicle emissions and that Tier 3 has significant environmental benefits. Tier 3 will reduce
exposure to vehicle pollution for the millions of people living, working, and going to school near
major roads, and thus delivering significant emissions benefits to communities across the U.S.
beginning in 2017 when the gasoline sulfur standards take effect. The reduction in air pollutants
resulting from the Tier 3 program will also have environmental, or "welfare," co-benefits in
addition to human health benefits, including changes in visibility, materials damage, ecological
effects from PM deposition, ecological effects from nitrogen and sulfur emissions, vegetation
2-15
-------
Chapter 2: Authority Under the Clean Air Act and Need for the Tier 3 Standards
effects from ozone exposure, and climate effects. For example, our analysis shows that the Tier
3 standards will provide improvements in visibility and sulfur deposition, as well as substantial
decreases in nitrogen deposition as a result of the standards (Preamble Section III and RIA
Chapter 7). We note that despite our goal to quantify and monetize as many of the benefits as
possible for the final rulemaking, the welfare co-benefits of the Tier 3 standards remain
unquantified and non-monetized in this RIA due to data, methodology, and resource limitations.
As a result, the benefits quantified in this analysis are likely underestimates of the total benefits
attributable to the final program.
While some commenters point to employment as another benefit of the Tier 3 standards,
as discussed in Preamble Section X.D and RIA Chapters 9.2.2 and 9.3.2, EPA expects the
employment effects of these standards to be small. Some commenters argue that the standards
may increase, rather than decrease, employment (see Chapter 9.3 of this Summary and Analysis
of Comments document); EPA does not quantify most of the employment impacts of the
standards and thus does not evaluate these estimates.
Given the important emissions reductions, air quality improvements, and health benefits
that Tier 3 will deliver, we agree with the majority of commenters that the Tier 3 standards
should be implemented as soon as possible. We are finalizing a Tier 3 program that, as
proposed, begins with model year 2017 for vehicle emission controls and with calendar year
2017 for the gasoline sulfur standards. As described briefly above and in Section III of the
Preamble and Chapter 7 of the RIA, the Tier 3 standards will provide immediate and meaningful
emission reductions and air quality improvements starting in 2017 due to the sulfur standards.
We reject the fuel industry commenters' claims to the contrary, and address their specific points
(e.g., that 2017 implementation will not help ozone NAAQS attainment) in Chapters 3 and 5 of
this document.
2-16
-------
Tier 3 Summary and Analysis of Comments
3. Emissions and Air Quality Emissions Impacts
The comments in this chapter correspond to Section III of the preamble to the proposed
rule and address the emissions and air quality impacts of the program. The comments received
and our responses to those comments are located below.
3.1 Emission Impacts of the Proposed Program
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM); Marathon
Overestimated emissions benefits. We recommend that EPA uses EIA's AEO 2013 as the
baseline to account for declining gasoline demand. EPA relied on EIA's Annual Energy Outlook
in 2011 (AEO 2011), which projected 4% higher gasoline demand in 2030 vs. 2012.
The analysis in Figure 10 [EPA-HQ-OAR-2011-0135-4276-A2] compares gasoline demand and
vehicle miles travelled in AEO 2011 with AEO 2013. Using as baseline AEO 2013 instead of
AEO 2011, in 2030 gasoline demand is projected to be 26% lower, vehicle miles travelled 11%
lower, and gasoline consumption in gallons per mile 16% lower.
In line with this analysis, baseline emissions are expected to be lower than EPA's assumed
baseline. As a result, the emissions benefits from Tier 3 are overstated.
Our Response:
For purposes of this final rule, we estimated emissions reductions compared to a
reference case that assumed renewable fuel volumes and ethanol blends based on the U.S.
Energy Information Administration's Annual Energy Outlook 2013 (AEO2013). Furthermore,
the future year projections of vehicle population and vehicle miles travelled were updated to
reflect the latest estimates from AEO2013 as well. Additional details can be found in Section III
of the preamble for the final rule.
3.1.1. Emissions Inventory Modeling Methodology
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM); Marathon Petroleum Company LP (MFC)
EPA should correct a number of deficiencies in the methods incorporated into the MOVES
model which lead to an overstatement of the emissions inventory benefits of Tier 3.
A recent assessment of the version of MOVES used by EPA to estimate the emissions benefits of
the Tier 3 proposal identified several key issues with this model, including the following:
3-1
-------
Chapter 3: Emissions and Air Quality Emissions Impacts
The FTP driving cycle used by EPA in its sulfur test program to develop fuel sulfur correction
factors to implement in the MOVES model for adjusting the exhaust emissions of Tier 2 and
newer light-duty vehicles is a mild cycle that does not cover the full range of accelerations in-use
and does not include high-speed operation. In fact, the FTP test represents only a small fraction
of the running exhaust emissions, particularly for Tier 2 vehicles. Consequently, the fuel sulfur
impacts for running exhaust emissions which EPA developed based on the FTP cycle are not
representative of the vehicle operating modes that produce the bulk of the exhaust emissions, and
this renders the adjustments for 2004 and later model years in the version of MOVES used for
the NPRM highly questionable.
The 2003 and older model year methods also rely on FTP-based measurements and the
concerns about the representativeness of the FTP cycle for sulfur effects modeling are already
described above.
Our Response:
The FTP driving cycle used in the in-use sulfur program was chosen to represent the type
of driving that manufacturers must cover in their certification testing to evaluate the in-use
emissions performance. The commenter raises a valid argument that the contributions of
emissions are expected to vary by vehicle operating modes and that the effect of sulfur on
exhaust emissions for the high-speed operations, not covered by FTP cycle, may not perform
similarly to the FTP cycle. However, the commenter failed to acknowledge that the CRC E-60
study cited by the commenter suggests that the magnitude of the fuel sulfur effects over the
US06 cycle for NMHC and NOx was found to be larger than that found for the FTP cycle. In
fact, the effects of sulfur going from 30 ppm to 5 ppm observed in the study on US06 cycle
(statistically significant reduction of 40% and 44% for NMHC and NOx, respectively) are larger
than the effects on FTP composite observed from the in-use sulfur study (statistically significant
reduction of 11% and 23% for NMHC and NOx, respectively). When using predictive models to
support rules, EPA has to make judgments regarding what modeling information is available and
whether the information is of adequate quality to support the conclusion being reached. Thus, in
this instance, although it may be difficult to compare the results from two different test programs
due to methodological differences, the finding does suggest that the application of the results
from FTP cycle to broader ranges of vehicle operation is a reasonable approach for modeling the
effects of sulfur on Tier 2 and newer vehicles.
What Commenters Said:
The review described here focused on the integration of the EPAct data into the NPRM MOVES
model. The EPAct-based equations form the basis for all
2004 and newer model year exhaust corrections (for both Tier 2 and Tier 3 certified vehicles).
Two items addressed in this discussion are (1) inconsistencies in equation coefficients between
EPA references and (2) the representativeness of the EPAct Program test fleet.
3-2
-------
Tier 3 Summary and Analysis of Comments
There are numerous inconsistencies between the EPA evaluation of the EPAct test data as
documented in EPA-420-R-13-002 and the methods incorporated into MOVES for estimating
non-sulfur fuel effects for 2004-and-later model years. It is unclear if errors were made in
reporting or in the incorporation of methods into MOVES.
The first discrepancy found was that the EPAct analysis report developed equations for two
forms of reduced models, as described in Section 5 of this report: the 11-term reduced models
and the 16-term reduced models. The report goes into detail as to why the 11-term reduced
models are statistically preferable, and the 11-term reduced models are the only form of the
equations provided in the Executive Summary of the report. This becomes a discrepancy
because it is the 16-term reduced models that were programmed into NPRM MOVES for all
pollutants and process combinations except one (NOx start exhaust equations in NPRM MOVES
are based on the 11-term reduced model). EPA needs to provide the technical justification for
the selection of the NPRM MOVES model equations and why those differ from the
recommendations of the EPAct Program analysis report.
Our Response:
The coefficients used in the NPRM application of MOVES, and reported in the docket
memorandum1, reflect the state of analysis at the time it needed to be completed for the NPRM.
However, following the initiation of inventory generation and air-quality modeling supporting
the NPRM, analysis of the EPAct dataset continued through October, 2012. The sets of
coefficients used in the FRM application of MOVES reflect the subsequent developments and
analysis described in the final report.2 The sets of coefficients used in the NPRM could be
considered as "draft" versions, and those used in the FRM as "updated" or final versions based
on the analyses described in the final report.
What Commenters Said:
The second discrepancy is that the THC running exhaust equation coefficients shown in Table 6-
6 do not appear anywhere in the EPA Program analysis report. These THC equation coefficients
used by NPRM MOVES are undocumented. It is presumed that these are some variation of the
16-term reduced model of the EPAct Program analysis report based on which parameters have a
non-zero coefficient.
Our Response:
These coefficients are documented in the docket memorandum as are the other
coefficients1. The commenter correctly notes that this set of coefficients is a version of the 16-
term model. However, they can be considered a "draft" version, reflecting model fitting using all
15 vehicles. Those in the final report reflect additional quality assurance, as described in
1 U.S. EPA, 2013. "Memorandum to Docket: Updates to MOVES for the Tier 3 NPRM"
2 U.S. EPA, 2013. Assessing the Effect of Five Gasoline Properties on Exhaust Emissions from Light-Duty Vehicles
Certified to Tier-2 Standards: Analysis of Data from EPAct Phase 3 (EPAct/V2/E-89). Final Report. EPA-420-R-
13-002. Assessment and Standards Division, Office of Transportation and Air Quality. Ann Arbor, MI.
3-3
-------
Chapter 3: Emissions and Air Quality Emissions Impacts
Chapter 6 of the Report. As a result, two vehicles were dropped before final model fitting,
leading to small changes in the coefficients. However, the qualitative relationships among the
terms in both the draft and final models are similar.
In addition, the coefficients used in the NPRM are documented in the final report in Appendix E
"Preliminary Reduced Models," as noted in 5.4 (page K
used to identify influential vehicles, as described in 5.5.
9 r-M-,
"Preliminary Reduced Models," as noted in 5.4 (page 100). These preliminary models were
What Commenters Said:
The third discrepancy is that three instances were found in which coefficients were inconsistently
assigned to a given fuel parameter. These are shown in Table 6-7. It is not clear if the EPAct
reporting is in error or if the MOVES modeling equations are in error.
Our Response:
The commenter has correctly noted an editorial error in the report.
In the report tables for the 16-term models, the effects are not listed in the correct order,
as in the Appendices. For THC Start emissions, the commenter has noted an editorial error in
Table 57. In this table, the labels for the model terms etOHxTSO, etOHxT90 are not listed in the
correct order, leading to incorrect associations of coefficients with model terms. The correct
order should be etOHxRVP, etOHxTSO, etOHxT90.
The same editorial error applies to the CO start term listed in Table 6-7 above. In report
table 65, it is incorrectly associated with the etOHxT90 effect, rather than the etOHxTSO effect.
However, in both these cases, the editorial errors are not germane to the FRM application
of the model, as the terms used in MOVES for the FRM analysis are those for the 11-term
models listed in the previous tables (Table 56 for THC, Table 64 for CO). The terms are listed
in correct order in these report tables.
What Commenters Said:
The fourth discrepancy, albeit a minor one, was that there were a number of instances where the
equation coefficients differed between NPRM MOVES and the EPAct Program analysis report
and the difference was not due to round off error.79 For example, the CO running exhaust T50
term is 0.024856 in NPRM MOVES and 0.02484 in the EPAct Program analysis report.
Our Response:
The commenter correctly lists the term used in the NPRM model, as well as that from the
16-term CO model, listed in Table 65. Again, the small difference highlighted is not germane to
3-4
-------
Tier 3 Summary and Analysis of Comments
the model applied in the FRM, which is that listed in Table 64. In this model, the T50 term is
0.0261, slightly larger than that in the 16-term model. However, numeric values of coefficients
nearly always have differences between models fit starting with 11 vs. 16 terms. This outcome is
routine and expected.
What Commenters Said:
The second issue concerns the representativeness of the EPAct Program test fleet, which was
also discussed in Section 5 [EPA-HQ-OAR-2011-0135-4276-A13].
Our Response:
The comment is premised on an assumption that fuel effects would or should differ
between vehicles or groups of vehicles with differing emission levels, whether due to technology
or to age. In fact, evidence suggests that fuel effects among vehicles with different emissions
levels will differ in absolute terms (i.e., g/mi) but tend to be similar in relative terms (i.e., ratios,
percents). That is to say, fuel effects (as well as other effects) are multiplicative rather than
additive.
One example that illustrates the point is available in the EPAct dataset itself. The first
plot (Figure 3-1) shows the data for cold-start NOx, averaged by aromatics level and by vehicle.
If we look at the trend for the Altima, we can see that its NOx increases from about 0.13 to 0.178
g/mi as aromatics increases from 15% to 35%, giving an absolute difference of nearly 0.05 g/mi.
In contrast, the corresponding increase for the cleanest vehicle, the Sienna, is about 0.002 g/mi
(0.0081 - 0.0059 g/mi). Thus, start emissions for the Altima are 21-22 times higher than those
for the Sienna at both aromatics levels.
The second plot (Figure 3-2) shows the same data plotted on a natural-log scale. What
immediately strikes the viewer is the similarity in logarithmic trends for all the vehicles across
the entire range of emissions, allowing for statistical variability. Again focusing on the Altima
and Sienna, examination shows that the slopes of the trends are nearly identical for these two
vehicles; that is to say, the logarithmic differences are extremely close. As differences in
logarithms can be interpreted as ratios, the implication is that the effect of aromatics for these
two vehicles, expressed as a ratio, i.e., multiplicatively, is nearly identical. Examination of plots
for the various emissions, as shown in the body of the report, as well as in the Appendices,
shows that this pattern is typical of emissions behavior in relation to changing fuel effects.
As the models developed from these data are fitted to logarithmic transforms of the
emission results, the various coefficients can be viewed as representing multiplicative effects of
the changing fuel properties. Similarly, the models are applied in MOVES so as to generate
multiplicative effects. On this basis, we find it reasonable to apply the multiplicative effects
across groups of vehicles of differing standards and different ages, with the realization that
projected differences in emissions related to changes in fuel properties will be proportional to the
projected emission levels. As previously explained, EPA has to make judgments regarding what
modeling information is available and whether the information is of adequate quality to support
the conclusion being reached. Thus, based on available evidence, it is reasonable to assume that
the fact that the EPAct sample comprised relatively young and clean (Tier 2) vehicles does not
3-5
-------
Chapter 3: Emissions and Air Quality Emissions Impacts
impair the applicability of the models to the Tier 2 fleet as these vehicles age and acquire
mileage.
Figure 3-1 Linear effects plot for cold-start NOx vs. aromatics, with data averaged by two aromatics levels and by
vehicle (Source: EPAct Phase 3)
0.16
0.17
0.16
0.15
0.14
0.12
0.11
1 aC
38 0.09
g 0.06
0.07
0.06
0.05
0.04
0.03-
0.02
0.01
0.00
•D
vehicle -
20
Aromatics (%)
'ALT1MA
COHOLLACE
'ODYSSEY
' CAUBER «™ CAMRY
EXPLGRERXLT F150
' OITTLDOKXE — SIENNA
30
•CMC
' IMFftLALS
• SIIVEFWDO
40
'COBALT
'LIBERTY
3-6
-------
Tier 3 Summary and Analysis of Comments
Figure 3-2 Linear effects plot for cold-start In(NOx) vs. aromatics, with data averaged by two aromatics levels
and by vehicle (Source: EPAct Phase 3)
-1-
-2
-3:
-4-
-5:
-61
vehicle
* COROLLACE
Aramaics (%)
' CALIBER •"" CAMRY «-" CMC
EXFLCflERXLT F150 «•*•" IMFftLALS
'OITTLOOKXE «™ SIENNA — SIli/EFftDO
'COBALT
'LIBERTY
What Commenters Said:
The extrapolation of Tier 0 Non-sulfur fuel effects to Tier 1 through NLEV technologies has not
been properly validated.
Our Response:
The commenter is correct that the Complex and Predictive models were developed using
data from Tier 0 vehicles and that the resulting models are applied to Tier 1 vehicles in MOVES.
Similarly, it is correct that the EPAct models were developed using a set of low-mileage Tier 2
vehicles, and that in MOVES, the models are applied to Tier 2 vehicles as they age and acquire
mileage. Nonetheless, we argue that these extrapolations are reasonable and appropriate given
the ways in which the models were developed and the fuel adjustments applied.
The comment is premised on an assumption that fuel effects should differ in some way
between vehicles of differing technologies ages or "high-emitter" status. As mentioned above,
fuel effects are represented in MOVES as multiplicative effects that are proportional to base
emission levels. We agree with the commenter that when fuel effects are expressed as absolute
changes in mass (e.g., g, g/mi, g/kg, etc.), vehicles of different technologies or ages will differ
strongly. We do, however, assume that proportional, or relative fuel effects (expressed as
3-7
-------
Chapter 3: Emissions and Air Quality Emissions Impacts
fractions, ratios or logarithmic differences), can be seen as similar, and thus transferrable on
average across technology groups and ages.
The data available to directly evaluate this assumption are limited. Nonetheless, a
limited volume of data can be compiled from the results of the EPAct program, as described
below.
The models applied in MOVES were developed from the results of EPAct Phase 3, in
which 15 Tier 2 vehicles were measured on 27 fuels. In addition, in EPAct Phase 5, three
vehicles manufactured in the 1990's were measured on three fuels used in Phase 3. Thus, using
these results, it is possible to make a direct comparison of emissions for sets of pre-Tier 2 and
Tier 2 vehicles measured on the same fuels (at 75°F).
The three 1990's vehicles are briefly described in Table 3-1. Note that these vehicles
range from 10-17 years in age and that all had presumably accumulated over 150,000 mi.
In terms of fuels, we limited the comparison to 2 fuels with 0% and 10% ethanol content,
respectively. These two fuels are closely matched in terms of RVP and aromatics levels, and
differ slightly in T50. However, they differ widely in T90, which is not expected to strongly
affect NOx emissions, but is expected to influence HC, CO and PM emissions. The properties of
these fuels are summarized in Table 3-2. Note that emissions on both fuels were acquired for
only two of the three vehicles.
Table 3-1 Characteristics of three "high-mileage" pre-Tier 2 vehicles measured in EPAct Phase 5
Make/Model
Chevrolet Tahoe
Ford Taurus
Dodge Dakota
Engine
V8-5.7L
V6-3.0L
Model Year
1997
1990
1993
Odometer (mi)
221,000
X90,4003
229,000
Table 3-2 Selected properties of two fuels measured in EPAct (Phase 5)
Fuel No.
6
7
Ethanol (vol. %)
10.56
<0.10
Aromatics (vol. %)
15.0
17.0
RVP (psi)
7.24
7.15
T50 (°F)
188.5
193.1
T90 (°F)
340.4
298.4
At the outset, we averaged the results by vehicle and fuel, and plotted the results for both
cold-start and hot-running phases of the LA92 cycle. Results for NOx, THC and PM are shown
3 The odometer for this vehicle is assumed to have been rolled over.
3-8
-------
Tier 3 Summary and Analysis of Comments
below in Figures Figure 3-3Figure 3-5. Note that the results are shown on a logarithmic scale
(base 10). This view of the data facilitates showing the results for all the vehicles in one plot. In
addition, differences in logarithms can be interpreted as proportional or relative differences
between the fuels for the various vehicles.
Aside from the fact that their emissions are higher, the differences in logarithms between
the two fuels are not obvious for the older "high-mileage" vehicles. Although the sample of
older vehicles is small, a qualitative view of the plots suggests that no clear and obvious
differences between Tier 0, Tier 1 and Tier 2 vehicles are evident.
It is helpful to follow up with a closer examination of this subset of results. As
mentioned, differences in logarithms represent ratios, e.g., log a - log b = log(a/b). However, for
purposes of summary, it is more intuitive to express the results as percent differences between
fuels 6 and 7 (relative to fuel 7). Accordingly, mean percent differences for the set of vehicles
are presented below in Figure 3-3 through Figure 3-5 for NOx, THC and PM, respectively. In
each plot, the differences are ranked from smallest to largest, meaning that the ordering of the
vehicles differs in each chart.
In reviewing the charts, it is clear that when the differences in emissions between the two
fuels are viewed as fractions, there are no clear or obvious differences between the 1990's
vintage high-mileage vehicles and the MY 2008 Tier 2 compliant low-mileage vehicles.
Generally, the two high-mileage vehicles differ in the signs of their effects, with one vehicle
showing a positive and the other a negative change. Cold-start PM is the only case in which
both vehicles have negative effects and have low rankings. Hot-running THC also stands out as
the only case in which the two older vehicles have the largest and smallest fractional effects. In
the remaining cases the older vehicles are distributed evenly across the rank order.
Overall, the available evidence suggests that when fuel effects are expressed as relative
multiplicative factors, as they are in both the EPAct analyses, and in their applications in
MOVES, it is reasonable to assume that the proportional effects are transferrable across different
vehicle technologies, as well as across other factors such as age, mileage or "high-emitter"
status. Furthermore, "That a model is limited or imperfect is not, in itself a reason to remand
agency decisions based upon it." Appalachian Power Company v. EPA, 251 F.3d 1026, 1051
(D.C. Cir. 2001)(internal citations omitted). "It is only when the model bears no rational
relationship to the characteristics of the data to which it is applied that [courts] will hold that the
use of the model is arbitrary and capricious." Appalachian Power Company v. EPA, 135 F.3d
791, 802 (D.C. Cir. 1998)(internal citations omitted).
3-9
-------
Chapter 3: Emissions and Air Quality Emissions Impacts
Figure 3-3 Mean percent difference in NOx cold-Start (top) and hot-running (bottom) emissions from
14 Tier 2 and 2 pre-Tier 2 vehicles measured on two fuels
(
E)
SIENNA ^^H
)UTLOOKXE
Taurus(90)
CAMRY
ALTIMA
PLORERXLT
SILVERADO
CALIBER
IMPALALS
F150
Tahoe(9?P
COBALT
COROLLACE
LIBERTY
^^
^H
^H
^m
•
•
•
-100
-50
0 50 100
Mean Difference (%)
150
200
(
E)
CIVIC
FOCUS
COROLLACE
ODYSSEY
Taurus(90)
SILVERADO
SIENNA
F150
CAMRY
ALTIMA
Tahoe(97)
•
IMPALALS
LIBERTY
)UTLOOKXE
PLORERXLT
CALIBER
^™
^
•
•
•
1
-100
-50
0 50 100
Mean Difference (%)
150
200
3-10
-------
Tier 3 Summary and Analysis of Comments
3-11
-------
Chapter 3: Emissions and Air Quality Emissions Impacts
Figure 3-4 Mean percent difference in THC cold-start (top) and hot-running (bottom) emissions from
14 Tier 2 and 2pre-Tier 2 vehicles measured on two fuels
-50
-50
A 1 Tl l\ /I A
ALII MA
C AMPV
l«MIVII\Y
LIBERTY
Tahoe(97)
FOCUS
ODYSSEY
COBALT
IMPAL/BBB
CALIBER
SIENNA
EXPLORERXLT
OUTLOOKXE
SILVERADO
^m
•
•
50 100
Mean Difference (%)
150
COfl 1C
rULU j
C 1\IIC
LI vIL.
1 1 DCDTV
LI btKI Y
CAMRY
F150 ]
OUTLOOKXE
COROLLACE
IMPALALS
SILVERADO
COBALT
EXPLORERXLT
CALIBER
ALTIMA
Taurus(90)
^m
•
M
•
i
50 100
Mean Difference (%)
150
3-12
-------
Tier 3 Summary and Analysis of Comments
3-13
-------
Chapter 3: Emissions and Air Quality Emissions Impacts
Figure 3-5 Mean percent difference in PM cold-start (top) and hot-running (bottom) emissions from
14 Tier 2 and 2pre-Tier 2 vehicles measured on two fuels
UL
EXP
1
SI
ALTIMA
ODYSSEY ^^^^H
CIVIC
F150
LIBERTY
.ORERXLT
FOCUS
MPALALS
LVERADO
Fahoe(97)
SIENNA
Taurus(90)
•
^H
^m
•
•
•
—
-200 -100 0 100 200 300
Mean Difference (%)
400
500
SI
EXP
1
1
ou
cc
1 1 RPRTV
LI DCI\ 1 T
r— U— Q/Q7\
anoe^y /)
LVERADO
F150
ODYSSEY
.ORERXLT
MPALALS
SIENNA
CIVIC 1
CAMRY
aurus(90) •
TLOOKXEB
ALTIMA
CALIBER
IROLMM
-200
-100
100 200 300
Mean Difference (%)
400
500
3-14
-------
Tier 3 Summary and Analysis of Comments
What Commenters Said:
Tier 0 exhaust impact equations, for modeling fuel effects, are extrapolated to El5, which is not
supported by data.
The model does not restrict El5 usage to the fleet legally allowed to use the fuel (2001-and-later
model year light-duty vehicles). Rather all gasoline vehicles were modeled as operating on El 5
at a uniform market share.
Our Response:
The MOVES model is not currently capable of partitioning a fuel supply market share
differently based on vehicle model year. This indeed leads to modeling pre-2001 vehicles with
an E10/E15 fuel supply, using the Complex model to apply fuel effects on these vehicles. We do
not mean to imply that pre-2001 vehicles would be fueling with E15 in the future; this outcome
is rather an artifact of how fuel supplies are partitioned inherently in the MOVES model. In the
interest of completeness, we performed further analysis using a fuel supply that does not contain
E15. As shown in Figure 3-6, the application of adjustments based on the Complex model results
in only minor fuel effect differences between E10 and El 5, as this model is not as sensitive to
ethanol level as the fuel effects model used in later model years (based on EPAct Program). We
do not believe that this difference contributes significantly to the overall emissions inventories
used in the Tier 3 modeling.
3-15
-------
Chapter 3: Emissions and Air Quality Emissions Impacts
Figure 3-6 NOx start emissions based on the Complex model for mix of E10/E15 and E10 only
NOx START EMISSIONS FOR TYPICAL MARKET SHARE V. NO E15
400000
200000
— Reference (27% E15)
— No E15
1990
1995
2000 2005
VEHICLE MODEL YEAR
2010
2015
What Commenters Said:
The NPRM MOVES has two severe flaws in the manner by which RVP impacts are modeled:
(1) predictive equations are extrapolated far outside the range of RVP measured in the underlying
test data; and (2) temperature interactions with RVP are not addressed.
RVP impacts on exhaust have critical flaws, primarily in winter season modeling, due to
excessive RVP extrapolation and a failure to evaluate temperature interactions.
Our Response:
The commenter is correct in pointing out that the fuel effects applied in the MOVES
modeling do extrapolate beyond the RVP range spanned by the data underlying the models used
to estimate fuel-effect adjustments i.e., 7-10 psi.
Despite the implication that this approach would result in large errors, we believe that it is
reasonable to assume that the relationships between log(Emissions) and RVP can be extrapolated
linearly from the range from 7-13 psi. This extrapolation, while not necessarily exact, can be
presumed to yield reasonable approximations over both summer and winter RVP ranges.
3-16
-------
Tier 3 Summary and Analysis of Comments
Evidence supporting this assumption is available from the same study cited by the commenter,
namely, CRC E-74b.
This study measured emissions of the three gaseous pollutants on 15 vehicles (MY 1994-
2004), spanning various vehicle technologies and emission standards. Measurements were
conducted on the FTP cycle at two temperatures (50 and 75 °F). Selection of study fuels
attempted to focus on varying ethanol and RVP while maintaining other important properties
constant (e.g., aromatics, T50 and T90), with the goal of assessing interactions between
temperature and RVP effects over the temperature range applicable to the southwestern U.S.
We performed a simple re-analysis of the study results, focusing on a subset of the data
allowing assessment of the linearity of relationships between log-transformed emissions and
RVP. We thus examined results for a subset of five fuels spanning ethanol levels on the range
from 0-10 vol. %, and RVP on a range from 7-13 psi. This subset was also restricted to a
temperature of 75 °F, as the fuels with 7 psi RVP were not measured at the lower temperature. In
addition, we examined results for Bags 1 and 2 of the FTP, as the calculations in MOVES handle
start and running emissions separately.
Emissions vs. RVP are presented on logarithmic scales for Cold-start and Hot-running
FTP emissions in Figures Figure 3-7Figure 3-9 below. In this presentation, the data are averaged
by RVP and vehicle. For each vehicle, a linear trend line is shown. An overview of the results
suggests that while some uncertainty exists as to whether some of the trends with RVP have a
small degree of curvature, linear extrapolation over the measured range can be expected to give
reasonable results.
The graphic results were followed up with statistical modeling. We modeling the natural-
log transform of the emissions versus ethanol, RVP, a quadratic term for RVP (to test for
curvature) and an ethanol RVP interaction. For modeling purposes, the fuel properties were
standardized to neutralize collinearity between the 1st and 2n order terms. "Mixed modeling"
techniques were used with random intercepts fit for each vehicle to isolate the effects of fuel
properties from the substantial variability contributed by the vehicles.
The results are briefly summarized in Table 3-3 below. A statistically significant
RVPxRVP interaction was evident only for CO Cold-start emissions (Bag 1), suggesting some
degree of curvature in the relationship over the range of 7-13 psi. In addition, significant linear
trends for RVP were apparent only for CO and NOx running emissions. In the remaining cases,
no significant relationships between exhaust emissions and RVP were apparent, either linear or
quadratic.
In the case of CO start emissions, models were fit with and without the quadratic term,
and the difference in projected emissions estimated at 7 and 13 psi for E10 gasoline. The
differences between the two models are approximately 3% at 7 psi and 7% at 13 psi. Obviously,
there is no suggestion of curvilinearity in the relationship between emissions and RVP for any
the remaining emissions or test phases.
3-17
-------
Chapter 3: Emissions and Air Quality Emissions Impacts
On the whole, we conclude that the available evidence gives no suggestion of strong
curvilinearity in the relationship between RVP and logarithms of emissions. Accordingly,
pending further evidence, we consider it reasonable to extrapolate these trends linearly from the
"summer RVP range" (7-10 psi) to the "winter range" (10-13 psi).
Table 3-3 Summary of statistical modeling of ethanol and RVP effects on FTP emissions (Source: CRC E-74b,
results measured at 75 F on EO and E10 fuels)
Emission
CO
HC
NOx
Phase
1
2
1
2
1
2
Model term
Ethanol
•
•
•
•
—
—
RVP
—
•
—
—
—
•
RVPxRVP
•
—
—
—
—
—
EthanolxRVP
—
—
—
—
—
—
3-18
-------
Tier 3 Summary and Analysis of Comments
Figure 3-7 CO emissions vs. RVP at 75°F, measured on the FTP cycle (Source E-74b)
100.0-
£ mo-
il
D
re
m
0 1'°;
0,1-
(a) Cold-Start (Bag 1)
±____
* * " *
* * ^nulp;
10
11
RVP (psi)
IN
0>
m
o
u
(b) Hot-Running (Bag 2)
10
11
RVP (psi)
12
13
12
13
92
For example, winter season RVP caps for controlling CO exhaust.
3-19
-------
Chapter 3: Emissions and Air Quality Emissions Impacts
This may be due to the limited temperatures evaluated (50 and 75 degrees Fahrenheit).
Figure 3-8 THC emissions vs. RVP at 75°F, measured on the FTP cycle (Source E-74b)
*E 1.00:
B
to
0.01-
(a) Cold-Start (Bag 1)
10
11
RVP (psi)
1,000:1
£ 0.100:
"6
ri
o>
w
(b) Hot-Running (Bag 2)
10
11
RVP Cpsi)
12
12
13
13
3-20
-------
Tier 3 Summary and Analysis of Comments
Figure 3-9 NOx emissions vs. RVP at 75°F, measured on the FTP cycle (Source E-74b)
10.00:
E 1.00
CT
fO
CQ
X
O 0.10
0.01
10
11
12
RVP (psi)
vehicle
6 3 7 E g
14 iAA 15
10.000
9 10
RVP (psi)
11
12
>M»f 9 &AA 1 o -H-+ 11 eee 12 Bea 13 >w»i< 14 AAA 15
13
13
3-21
-------
Chapter 3: Emissions and Air Quality Emissions Impacts
With respect to the interaction between temperature and RVP, the commenter notes
correctly that the datasets used to estimate fuel effects did not incorporate the effect of
temperature.
At the outset, we can point out that MOVES does not apply temperature adjustments to
hot-running CO, THC or NOx emissions, thus obviating the need to consider interactions
between temperature and fuel effects. It is widely accepted that the effect of temperature on hot-
running gaseous emissions is negligible.
For start emissions of CO, THC and NOx, however, MOVES applies both temperature
and fuel effects, under an assumption that they can be applied multiplicatively and
independently. However, when multiplicative effects are jointly applied to mass emissions (in
"linear" as opposed to logarithmic space), the results appear "interactive" in that the different
effects either reinforce or damp each other. For example, start emissions increase substantially
as temperature declines, implying that fuel effects are amplified at lower temperature, whether
positive or negative. The net result can be either increased or decreased emissions, depending on
the nature of the fuel effects.
We can focus on start CO as an example. For CO, fuel effects are calculated using the
Complex model for 2000 and earlier model years4, and using the EPAct models for 2001 and
later model years. In both cases, the results applied are broadly consistent with those of the study
cited by the commenter (CRC E-74b)5. As the two studies applied differing approaches to
statistical analysis, the respective model coefficients cannot be compared in terms of magnitude.
However, they may be compared qualitatively in terms of sign.
For cold-start CO emissions (Bag 1), the Complex Model has a small but positive linear-
effect coefficient for RVP. This result is directionally similar to the E-74b Composite CO model,
which also has positive RVP coefficients, as well as a positive interaction between RVP and
temperature. This result implies that for "cold" temperatures below 50°F, increasing RVP should
increase CO, with the effect amplified by decreasing temperature. The application of the
Complex Model in MOVES gives qualitatively similar results with a positive RVP coefficient
amplified by the multiplicative temperature adjustment.
For MY2001 and later, the EPAct models can be applied to start emissions specifically.
The CO start model has a negative linear coefficient for RVP, (meaning that emissions decline as
RVP increases). In E-74b, a piece-wise fit was used, giving negative and positive coefficients for
RVP < 9 and > 9 psi, respectively. In addition, a positive interaction term was included in the
reduced model (Table 5-14, page 64). As expected, the temperature coefficient is negative,
suggesting that an "interference" interaction applies, i.e., that the combined effects of RVP and
temperature would have a mutual "damping" effect. The net results of the E-74b model are
shown in Figure 5-1 (page 76) of the CRC report. At temperatures below 50 °F, the trends
portrayed show an "interference" effect, i.e., that increasing RVP decreases CO start emissions,
4 For THC and NOx, analogous effects are calculated using applications of the EPA Predictive Model.
5 Coordinating Research Council. Effects of Vapor Pressure, Oxygen Content, and Temperature on CO Exhaust
Emissions. CRC Report E-74b. Alpharetta, Georgia, (Prepared by Sierra Research, Inc. Sacramento, CA), May,
2009.
3-22
-------
Tier 3 Summary and Analysis of Comments
with the absolute margin (in g/mi) increasing with declining temperature. As mentioned, despite
the differences in underlying data and modeling approaches, the application of independent RVP
and temperature effects in MOVES gives similar results. This outcome results from the
multiplicative combination of a negative temperature effect (CO increases and T decreases) with
a negative RVP effect (CO decreases as RVP increases). The net result is that the temperature
effect is reduced by increasing RVP at lower temperatures, which is direct!onally similar to the
result obtained in E-74b, and suggested by the commenter as a correct representation of CO
behavior in relation to RVP and "cold" temperature.
Results for THC and NOx are similar. The models applied in MOVES do not contain
temperature effects, but multiplicative combinations of the fuel and temperature effects results in
interaction effects in the projected emission volumes. Thus, on the whole, we conclude that the
multiplicative combination of temperature and fuel effects as applied in MOVES does allow for
interactions between these effects.
What Commenters Said:
EPA assumes that the proposed evaporative emission standards reduce permeation emissions by
75%, emissions from vapor leaks by 70%, and emissions from liquid leaks by 30-45%. These
assumptions are based on simple engineering estimates and are uncertain.
Our Response:
The evaporative emissions modeling for the FRM has been significantly updated based
on completion of additional test programs and peer review comments. The NPRM modeling for
the Tier 3 control case was based on an engineering analysis; the revised modeling for the final
rule incorporates the latest data and has been improved to reflect more real-world physical
processes.
What Commenters Said:
EPA estimates that the proposed evaporative emission standards will provide 34% of the total
VOC emission reductions expected from the Tier 3 proposal by 2030; however, the MOVES
model does not assume any reduction in permeation emissions from Tier 2 vehicles relative to
earlier vehicle technologies. In contrast, data from the CRC Project E-77 studies (17, 18, 19) on
evaporative permeation show that some Tier 2 vehicles have much lower permeation emissions
than pre-Tier 3 vehicles (e.g., vehicles certified to California's "Zero Evaporative Emission
Standards"). Therefore, it is likely that a portion of the assumed permeation benefit for Tier 3
already is occurring in Tier 2 vehicles. Thus, the benefits of the proposed Tier 3 evaporative
emission standards are overstated.
17 - Haskew, H. and Liberty, T, Enhanced Evaporative Emission Vehicles, CRC Project E-77-2, March
2010
18 - Haskew, H. and Liberty, T, Vehicle Evaporative Emission Mechanism: A Pilot Study, CRC Project
E-77, June 24, 2008
3-23
-------
Chapter 3: Emissions and Air Quality Emissions Impacts
19 - Haskew, H. and Liberty, T, CRC Project E-77-2c, Study to Determine Evaporative Emission
Breakdown, Including Permeation Effects and Diurnal Emissions Using E20 Fuels on Aging Enhance
Evaporative Emissions Certified Vehicles, December, 2010
Our Response:
Data from the CRC E-77 programs show that pre-Tier 2 technologies have higher
permeation emission rates than Tier 2 vehicles. The CRC E-77 test data indicate that the Tier 2
permeation rates in the version of MOVES used for the Tier 3 FRM analysis properly reflect the
data, but pre-Tier 2 rates are underestimated. Our Tier 3 analysis is not ascribing emissions
benefits to Tier 3 that are already occurring, and Tier 3 benefits are not being overstated.
What Commenters Said:
The method by which exhaust basic emission rates were developed for Tier 3 vehicles, based on
the ratio of exhaust standards, failed to account for the different certification fuels inherent in
those standards.
One key concern noted was that within EPA's ratioing method for NPRM MOVES, the
difference in certification fuels between Tier 1 and Tier 3 vehicles was not accounted for in
the calculation of exhaust basic emission rates. This is problematic as the differences in
certification fuel in terms of sulfur content and oxygenate level are significant for the Tier 3
case. Because of certification fuel differences, the application of ratios based on strict
numeric exhaust standards is flawed if fuel differences are not explicitly factored in.
Our Response:
We agree that this question was relevant to the NPRM MOVES analysis. However, for
the FRM analysis, we addressed the question of changing certification fuels by incorporating fuel
adjustments in the "GeneralFuelRatioExpression" calculations to account for the adoption of
Tier 3 certification fuel starting in MY2017. One such adjustment accounts for reduction in fuel
sulfur to 10 ppm and is calculated by re-centering the preexisting sulfur model around a new
base sulfur calculation at 10 ppm. A second adjustment accounts for the adoption of E10
certification fuel and is calculated modifying the existing adjustments using the EPAct models
by assigning the Tier 3 certification fuel as the "base" fuel for MY2017 and later. With these
adjustments in place, Tier 3 vehicles operating on in-use fuel with 10 ppm sulfur would show no
sulfur effect, as the in-use and certification levels match. Similarly, Tier 3 vehicles operating on
an in-use fuel with the same non-sulfur properties as the Tier 3 certification fuel would have no
fuel adjustment for the non-sulfur fuel properties. Conversely, emissions for Tier 3 vehicles
operating on fuels with properties differing from the E10 certification fuel are adjusted
appropriately relative to the certification fuel.
What Commenters Said:
3-24
-------
Tier 3 Summary and Analysis of Comments
In order to capture the potential emissions effects of reducing gasoline sulfur levels below 30
ppm, a separate equation was developed for use in the EPA mobile source emissions inventory
model, MOVES. This equation was based solely on the EPA study of sulfur effects on exhaust
emissions of in-use Tier 2 light-duty vehicles. The coefficients applied to this equation were
from mixed model results with inherent assumptions which were acknowledged within the in-use
sulfur report (see comments below). The sulfur effects were then applied multiplicatively to
other gasoline fuel effects in MOVES. EPA did not provide a comparative analysis of expected
emissions impacts due to the sulfur equation change in MOVES.
Our Response:
The estimated emissions impacts from the updated sulfur model are presented in a
separate memorandum to the docket.6
3.1.1.1. Additional Comments from Sierra Research Submitted as a Part of
API/AFPM Comments
What Commenters Said:
An assessment of the emissions benefits of the Tier 3 proposal was prepared by Sierra Research
and attached to API/AFPM's comments (Attachment No. 13).
The linear correction factor implemented in NPRM MOVES conflicts with the EPA analysis of
the in-use sulfur program to develop the percent changes.
Our Response:
As described in a separate memo to the docket7, since the results from the in-use sulfur
program based on log-transformation were reverse-transformed to linear space prior to being
incorporated into MOVES, we believe that the percent changes from the mixed model were
correctly applied. Furthermore, the relationship between changes in gasoline sulfur content and
NOx, HC, NMHC and NMOG emissions is typically linear. The linearity of sulfur impacts on
NOx, NMHC and NMOG emissions is supported by past studies with multiple fuel sulfur levels
all of which compare gasoline with differing sulfur levels that are below approximately 100 ppm
(e.g., CRC E-60 and 2001 AAM/AIAM programs as well as comments submitted to this
rulemaking by MECA cited within Preamble IV.A.6). As stated within Preamble IV.A.6 of the
Tier 3 final rule, the relative linearity of the effect of gasoline sulfur level on NMOG and NOx
emissions allows exhaust emissions results generated within EPA and other studies of gasoline
sulfur at levels immediately above or below either 10 ppm or 30 ppm to be normalized to either
10 ppm sulfur or to 30 ppm sulfur. This allowed EPA to evaluate vehicle emission control
6 U.S. EPA. 2014. Memorandum to Docket: Updates to MOVES for the Tier 3 FRM Analysis
7 U.S. EPA. 2013. "Memorandum to Docket: Updates to MOVES for the Tier 3 NPRM"
3-25
-------
Chapter 3: Emissions and Air Quality Emissions Impacts
system response to changes in gasoline sulfur content either between or within the range defined
by the proposed level of 10 ppm S and the current standard of 30 ppm S.
What Commenters Said:
The NPRM MOVES sulfur correction update for light-duty Tier 2 vehicles does not account
for the difference between normal and high-emitting vehicles, whereas distinct sulfur impacts
for normal and high emitters are accounted for in the preexisting MOVES sulfur corrections
for 2003 and older model year vehicles.68 The average odometer of the in-use test program on
which this update is based is about 30,000 miles compared to an approximate odometer level
of 120,000 miles for the average on-road vehicle.69 Therefore, this new method is biased high
due to absence of a high emitter adjustment in NPRM MOVES given that the test fleet on
which it is based is newer than the average in-use fleet. This issue of high emitters should
have been apparent to EPA, as previous sulfur studies have evaluated both new and aged
catalysts given the clear importance of the representativeness of the vehicle catalyst.70'71
68 High emitters are generally less sensitive to fuel sulfur content.
69 As per Footnote 43, the average vehicle in the on-road light-duty fleet is 10.8 years old (calendar year 2011
estimate) with an approximate odometer reading of just under 120,000 miles.
70 T. D. Durbin, J. W. Miller, J. T. Pisano, T. Y. Younglove, C. G. Sauer, S. H. Rhee, T. Huai, and G. I. MacKay,
"The Effect of Fuel Sulfur on NH3 and Other Emissions from 2000-2001 Model Year Vehicles, CRC Project E-
60," May 2003, available at www.crcao.com.
71 "Summary: CRC Sulfur/LEV Program," Coordinating Research Council Report, CRC Project No. E-42, December
22, 1997.
Our Response:
The commenter claimed that the absence of high emitter adjustment for 2004-and-later
vehicles in MOVES would result in overestimation of the effects of sulfur on emissions based
on two studies. However, we do not believe that these studies provide support for the
commenter's claim. The first study by Durbin et al. concluded that for both NMHC and NOx,
the effects of catalyst age on FTP composite emissions were not statistically significant for the
tested fleet. The CRC Sulfur/LEV study showed that when sulfur was reduced from 600 to 40
ppm, aged catalysts demonstrated slightly lower but similar magnitudes of sulfur effect on
FTP composite emissions as compared to the new catalysts.
What Commenters Said:
The correction factors listed in Table 6-2 for "all other types" represent those applied by the
model for heavy-duty gasoline vehicle types. The documentation wholly omits any discussion
of the derivation of these values. EPA needs to document how these values were derived
before they can be reviewed.
Our Response:
3-26
-------
Tier 3 Summary and Analysis of Comments
The MOVES docket memo for the Tier 3 FRM analysis presents the derivation of the
sulfur coefficients for heavy-duty gasoline vehicles.
What Commenters Said:
For motorcycles, the method assumes no impact from fuel sulfur (see Table 6-2). EPA has no
data from motorcycles on fuel sulfur effects; however, for model years 2003 and older, the model
maps motorcycle fuel corrections to those of light-duty gasoline vehicles. So in this instance,
the new method (to have no sulfur impact for motorcycle exhaust) is inconsistent with the
method applied to all other model year motorcycles. This is a minor point since the contribution
of motorcycles to the on-road inventory is minor.
Our Response:
In modeling the emissions impact from the proposed rule, we did not attempt to update
the fuel sulfur algorithm in MOVES for the legacy vehicles (pre-2004 model year vehicles)
including motorcycles, since (as the commenter pointed out) the emissions contribution from
them is very small. The contribution of pre-2004 model year motorcycles to the on-road
inventory for all criteria pollutants is less than 0.5% in 2018 and less than 0.1% in 2030. In
terms of the total Tier 3 emission reductions, pre-2004 model year motorcycles account for 0.1%
and 0.0% of the Tier 3 emission reductions in 2018 and 2030, respectively, for all criteria
pollutants. For 2004-and-later model year vehicles, the fuel sulfur algorithm was updated based
on the latest data from the in-use sulfur program only for cars and trucks. We believe that the
changes in assumption of no sulfur impact for 2004-and-later model year motorcycles is justified
due to lack of data and because both methodologies result in consistent outcome of almost no
impact of fuel sulfur from motorcycles on national emissions inventories.
What Commenters Said:
The model inappropriately extrapolates log-log sulfur corrections towards the zero asymptote for
key sectors of the 2003-and-older model year fleet in order to evaluate proposed sulfur
requirements. The result is large changes in exhaust emissions that are not confirmed by actual
data.
NPRM MOVES incorporated no new data for estimating 2003 and older model year vehicle
fuel sulfur impacts. In order to evaluate fuel sulfur levels at or below 10 ppm as part of the
regulatory evaluation, the model simply extrapolates preexisting sulfur adjustment equations
present in the public version of MOVES. Thereby, the NPRM MOVES model methods are
that of MOVES2010a with one minor change. Because this sulfur correction is an
extrapolation (most of the underlying methods developed did not contain any fuels below 30
ppm S, see Table 6-5), EPA programmed in a cap at the maximum allowable exhaust reduction
due to sulfur below 30 ppm. In MOVES2010a that cap is 15 percent (defined relative to 30
' U.S. EPA. 2014. Memorandum to Docket: Updates to MOVES for the Tier 3 FRM Analysis, Chapter II.A.
3-27
-------
Chapter 3: Emissions and Air Quality Emissions Impacts
ppm fuel); in NPRM MOVES that cap is 60 percent (defined relative to 30 ppm fuel).72
EPA's justification is that the in-use sulfur test program with Tier 2 vehicles showed reductions
up to 60 percent; therefore, greater reductions should be permissible when extrapolating
previous test programs beyond the lower sulfur limit tested.
Figure 6-3 presents an example of how the linear extrapolation compares to that used by
NPRM MOVES for NOx emissions from normal emitting NLEVs (passenger cars).
Figure 6-3
NPRM MOVES NLEV (Passenger Car) Sulfur Impact
Running Exhaust NOx, Normal Emitter
Q.
Q.
O
0%
-10%
-20%
0)
'
-30%
-40%
-50%
u
-60%
-70%
• Linear Extrapolation
Tier 3 NPRM Extrapolation
(Including-60% Cap)
10 15 20
Gasoline Sulfur Content (ppm)
25
30
72 The 15 percent cap in MOVES2010a was present, because the model already did permit the extrapolation of fuel
sulfur effects below 30 ppm as part of emission inventory development calculations.
Our Response:
In modeling the emissions impact from the proposed rule, we did not attempt to update
the fuel sulfur algorithm in MOVES for the pre-2004 model year vehicles because we were not
aware of any additional data that provided information on fuel sulfur below 30 ppm for the
legacy vehicles. Therefore, the commenter is correct that the effects of sulfur for pre-2004
model year vehicles are based on extrapolation.
3-28
-------
Tier 3 Summary and Analysis of Comments
However, the commenter did not accurately characterize the effect of extrapolation in the
plot they presented above. Figure 3-10 through Figure 3-13 show the effect of sulfur on running
exhaust emissions from pre-2004 model year vehicles for sulfur levels below 30 ppm, based on
MOVES. The red lines represent the old sulfur floor of 15 percent, previously in MOVES2010a,
and the blue lines represent the new sulfur floor of 60 percent in Tier 3 NPRM version of
MOVES. For all pollutants, the lines lie on top of each other for sulfur level at or above 10
ppm, the average in-use sulfur level modeled in the Tier 3 rule. The two lines start to diverge
only below 10 ppm. These trends are also true for the sulfur effects on starts. The plots indicate
that the modification made to the sulfur floor in the Tier 3 NPRM version of MOVES did not
affect the sulfur adjustment and thus, did not impact the emissions inventories estimated in the
rule.
Lastly, the effects of lowering sulfur level on emission for pre-2004 model year vehicles
are less than 2% for all criteria pollutants in both 2018 and 2030, contrary to the commenter's
claim that the effects based on extrapolation would result in large changes. In terms of the
emissions reductions estimated from the Tier 3 standards, the contribution from pre-2004 model
year vehicles represent 4% and 8% of the total reductions in 2018 for NOx and VOC,
respectively, and 0.1% and 0% in 2030 for NOx and VOC, respectively.
3-29
-------
Chapter 3: Emissions and Air Quality Emissions Impacts
Figure 3-10 Effect of sulfur on running exhaust NOx emissions from pre-2004 model year
vehicles in MOVES
1.0
0.9
°-8
0.7
e
i
0.6
1 0.5]
1J
| 0.4
H51
|°.3
^
w 0.2
0.1
0.0
10 20
Sulfur Level
PLOT Old Sulfur Floor New Sulfur Floor
30
3-30
-------
Tier 3 Summary and Analysis of Comments
Figure 3-11 Effect of sulfur on running exhaust THC emissions from pre-2004 model year
vehicles in MOVES
1.0
0.9
0.8
a
° 0.7
g
I
°
8
a
OJ
0.6
0.4
tj 0.3
%
w 0.2
0.1
0.0
10 20
Sulfur Level
PLOT Old Sulfur Floor New Sulfur Floor
30
3-31
-------
Chapter 3: Emissions and Air Quality Emissions Impacts
Figure 3-12 Effect of sulfur on running exhaust CO emissions from pre-2004 model year
vehicles in MOVES
1.0
0.9
0.8
0.7
I
> 0.6
a °-5
1
0.4
cfl
0.2
0.1
10 20
Sulfur Lev el
PLOT Old Sulfur Floor New Sulfur Floor
30
3-32
-------
Tier 3 Summary and Analysis of Comments
Figure 3-13 Effect of sulfur on running exhaust PM2.5 emissions from pre-2004 model year
vehicles in MOVES
1.0
.y
.B
en °-'
o
•^
OJ
.> 0.6
rt
f2 n c
" U.D
| 0.4
W
^ m
fcj U.J
|
^ . -
0.2
.1
0.0
10 20
Sulfur Level
FLOT Old Sulfur Floor New Sulfur Floor
30
What Commenters Said:
The 2003 and older model year methods assign light-duty corrections directly to motorcycles
and heavy-duty vehicles. The method assumes all sulfur impacts are equivalent across vehicle
classes mapped by model year. Therefore, the sulfur correction for a 2003 model year light-duty
vehicle (meeting the NLEV standard) is that applied by the model to a 2003 motorcycle and a
2003 heavy-duty gasoline vehicle. The lack of vehicle class specific impacts is problematic for
motorcycles and heavy trucks as the underlying data are based on light-duty gasoline vehicles
only (and the catalyst technology employed is not constant across vehicle classes for any given
model year). This lack of vehicle class specificity is inconsistent with the 2004 and later model
year approach that has distinct corrections by vehicle class as shown in Table 6-2.
Our Response:
We believe that the modeling approach that applies distinct sulfur effects by vehicle type
for 2004-and-later model year vehicles is an improvement over what was done previously. We
acknowledge that the fuel sulfur effect for pre-2004 model year motorcycles and heavy-duty
gasoline vehicles were based on the effect from light-duty vehicles, but we do not expect the
3-33
-------
Chapter 3: Emissions and Air Quality Emissions Impacts
effect of sulfur on heavy-duty gasoline vehicles to be significantly different than that of light-
duty since the combustion and exhaust aftertreatment systems are similar in principle. The
impact on emissions is minimal; the contributions to the total inventory from pre-2004 model
year motorcycles and heavy-duty gasoline vehicles are less than 1.8% in 2018 and 0.1% in 2030
for VOC and NOx. Again as previously explained, when using predictive models to support
rules, EPA has to make judgments regarding what modeling information is available and whether
the information is of adequate quality to support the conclusion being reached.
What Commenters Said:
EPA revised the estimated proportion of fuel sulfur emitted as exhaust sulfate PM for NPRM
MOVES. The exhaust sulfate PM is the only exhaust PM component impacted by fuel sulfur.
The revised approach basically halved the amount of fuel sulfur coming out as exhaust sulfate.
The sulfate correction to PM exhaust was not developed using data below 30 ppm and was a
linear extrapolation assuming no lube oil contribution as the method presumes no sulfate
exhaust at zero ppm S fuel. The method should address the sulfate component due to lube oil,
especially at ultra-low sulfur levels assumed. However, this methodological issue is a minor
point as the sulfate portion of PM exhaust is minor.
Our Response:
Due to limited data on lubricating oil contributions to fuel sulfur for gasoline vehicles,
and the small contribution of sulfate to PM from gasoline vehicles (as the commenter pointed
out), no changes were made to the methodology for modeling sulfate emissions in the final rule.
What Commenters Said:
Finally, EPA's analysis of the In-Use Study data suggests that there is no effect of gasoline
sulfur content on PM emissions. However, the MOVES model used by EPA to compute the
benefits of the proposed Tier 3 regulation assumes that reducing sulfur content will reduce PM
emissions.
Our Response:
No sulfur effect on PM emissions was modeled in MOVES for Tier 2 and later vehicles,
consistent with the results from the In-Use Sulfur Program.
What Commenters Said:
All NLEV through Tier 3 exhaust basic emission rates are ratioed from Tier 1 vehicles in
MOVES. Ratios are separate for low-to-mid power range (ratioed to FTP) and high power range
(ratioed to US06). These power ranges were illustrated previously in Figure 6-1. Tier 1 was the
latest technology for which EPA felt it had a full enough record of data across the vehicle useful
3-34
-------
Tier 3 Summary and Analysis of Comments
life to estimate deterioration in the original release of MOVES2010. For NPRM MOVES, no
attempt was made to update NLEV and Tier 2 basic emission rates with new data available since
the original MOVES2010 development, nor was an attempt made to validate the extrapolation
assumptions such as the proportions of emissions by exhaust mode and deterioration rates.96
94 J. Warila, "Developing Emission Rates Representing the Effect of Proposed Tier 3 Standards on Gaseous
Emissions from Gasoline-Powered Light-duty Vehicles (HC/CO/NOx)," EPA OTAQ memorandum to the
Docket Number EP A-HQ-O AR-2011-0135, December 2011.
95 "Development of Emission Rates for Light-Duty Vehicles in the Motor Vehicle Emissions Simulator
(MOVES2010)," EPA-420-R-11-011, August 2011.
Our Response:
As the MOVES documentation9 and docket memorandum10 make clear, emission rates
for NLEV, Tier 2 and Tier 3 vehicles are scaled relative to rates for Tier 1 vehicles, with scaling
factors assigned using FTP and US06 results from the In-use Verification Program. As the
documentation also describes, we estimate deterioration for these vehicles using proportional
scaling applied by a log-linear trend (through nine years of age), in which slopes in
In(Emissions) are assumed to be similar. Beyond nine years, deterioration is assigned through a
set of proportional factors. Based on examination of large volumes of data for Tier 1 and earlier
vehicles, we believe that this assumption is reasonable. Note that under these multiplicative or
"logarithmic" assumptions, late-model year (e.g., 2013) vehicles show much lower absolute
deterioration than previous model year (e.g., 2001) vehicles; that is, the late-model year vehicles
have much less absolute increase in the average emission rate over a given period of time (in
g/hr, g/mi, etc.). In addition, the multiplicative deterioration model, which we believe to be
technically appropriate, assigns far lower deterioration to Tier 2 or Tier 3 vehicles than some
alternative models, such as assuming that deterioration in start emissions could be estimated as a
fixed additive increment. Again, the consistent patterns in multiplicative scaling seen in data for
Tier 1 and earlier vehicles give a reasonable expectation that similar patterns would apply to
NLEV and Tier 2 vehicles.
With regard to assumptions regarding the proportions of emissions in various operating
modes, we disagree that we "extrapolate" these assumptions in developing emission rates. In
fact, the questions of time fraction in operating modes and the emission rates for differing
technologies do not intersect. As the operating modes for running operation are defined in terms
of speed and acceleration, rather than vehicle technology or emissions standards, they are
influenced by vehicle weight and driving patterns. We would not expect changes in rates,
however they are assigned, to give associated changes in operating-mode fractions.
What Commenters Said:
9 U.S. EPA. 2011. Development of Emission Rates for Light-Duty Vehicles in the Motor Vehicle Emissions
Simulator (MOVES2010). EPA-420-R-11-011. Assessment and Standards Division, Office of Transportation and
Air Quality, Ann Arbor, MI.
10 U.S. EPA. 2013. "Memorandum to Docket: Updates to MOVES for the Tier 3 NPRM"
3-35
-------
Chapter 3: Emissions and Air Quality Emissions Impacts
EPA updated estimates of the impact of fuel system leaks on evaporative emissions using data
from a Colorado field program. This program provided data on the incidence of vehicles with
leaks in the evaporative emission control system and in emissions associated with these vehicles.
While this database is suitable for use in updating emission inventories, there remains
considerable uncertainty in projecting emissions associated with "leakers" due to uncertainty in
incidence and emission impact of leakers. This leads to uncertainty in projected evaporative
emissions and benefits of the Tier 3 proposal.
Our Response:
The analysis and report determining the leak prevalence rates underwent significant
revisions between the NPRM and FRM and was therefore sent out for a second peer review. The
second review did not result in revisions to the analysis, but did have some minor updates to the
report.11 The analysis revisions included uncertainties for prevalence rates and are shown in the
report.
What Commenters Said:
Vapor venting algorithms used to estimate evaporative emissions for multi-day diurnals and for
vehicles with leaks were updated for the Tier 3 analysis with the DELTA model. Although the
model is an improvement over the previous method of estimating vapor growth by accounting for
canister capacity, back purge, and multiple day diurnals, a peer review found weaknesses in the
model, most importantly in assumptions about evaporative canister back-purge,100 which leads to
additional uncertainties regarding EPA's estimates of evaporative emission benefits for the Tier 3
proposal.
Our Response:
For the NPRM, EPA used an estimate for back-purge based on a marine study. The
FRM modeling, using the DELTA model for diurnal emissions, now includes a revised back-
purge estimate based on the completed multi-day diurnal test program of nine vehicles
representing top sales of the national fleet, tested over fourteen days of diurnals in a SHED.12
What Commenters Said:
As part of the NPRM MOVES development, EPA reviewed the permeation impact of ethanol-
containing gasoline. EPA's evaluation found, based on several CRC studies103 that the
permeation impact for E15 is not statistically different from E10. The conclusion that ethanol-
related permeation impact does not vary by ethanol content (for the range of
11 DeFries, T., Palacios, C., Weatherby, M, Stanard, A., Kishan, S. (2013) Estimated Summer Hot-Soak
Distributions for Denver's Ken Caryl I/M Station Fleet.
12 Lindner, J, Sabisch, M., Glinsky, G., Stewart, I, St. Denis, M., Roeschen, J. (2013) Multi-Day Diurnal Testing
3-36
-------
Tier 3 Summary and Analysis of Comments
6 to 20 percent by volume) is supported by the CRC studies and the review completed as part of
CRC Project A-73-1.104 Given this finding, NPRM MOVES used the previously estimated E10
permeation impact for E15 gasoline in the RIA methods.105
101 "Mobile Source Hydrocarbon Speciation Profiles for the Tier 3 Rule NPRM and Anti-backsliding Study
AQ Modeling," U.S. EPA memorandum, Docket EPA-HQ-OAR-2011-0135, January 2013.
102 "Hydrocarbon Composition of Gasoline Vapor Emissions from Enclosed Fuel Tanks - Draft,", EPA-
420-R-l 1-018,2011.
103 M. Beardsley, "Updates to MOVES for the Tier 3 NPRM Analysis," EPA OTAQ memorandum to the
Docket Number EPA-HQ-OAR-2011-013 5, March 2013.
104
"Development of Inventory and Speciation Inputs for Ethanol Blends," CRCReportNo. A-73-1,Report
No. SR2012-04-01, prepared by Sierra Research for Coordinating Research Council, May 2012.
105 It was noted that the NPRM MOVES memorandum contained a typo for the discussion on this subject matter.
The memo reports the ethanol permeation increase for vehicles meeting enhanced evaporative
standards (and newer) as "213 percent increase for newer technology vehicles." The correct value used by
the model is an increase of 113.8 percent (or a multiplicative adjustment of 213.8).
There are two aspects, outlined below, in which the model's permeation impacts due to
ethanol could be improved upon.
1. If the permeation impacts of E6 to E20 are not statistically different, then the
modeled permeation increase could be based on the data collected from all ethanol
blends. The NPRM MOVES impact is based only on E6 to E10 fuels.
Our Response:
Analysis of the E-77 program using this approach (including E6-E20 fuels) affects the
ethanol effect on permeation by less than 10%. This will have very small effects on modeling
results.
2. An examination of the available data shows that vehicles meeting Tier 2 or LEV II
evaporative standards are statistically distinct from vehicles meeting enhanced
evaporative standards. The NPRM MOVES treats all vehicles meeting enhanced
evaporative and subsequent standards as a single group.
Our Response:
Performing a mixed-model regression on all data available from E-77 test programs
shows no significant interaction effect for vehicle standard and ethanol.13 The version of
MOVES used for the Tier 3 final rule's analysis continues to treat all vehicles meeting enhanced
evaporative and subsequent standards as a single group with regard to the effect of ethanol in
fuel.
What Commenters Said:
13 U.S. EPA, 2014, "Development of Evaporative Emissions Calculations for Tier 3 FRM" memorandum to the Tier
3 docket.
3-37
-------
Chapter 3: Emissions and Air Quality Emissions Impacts
CRC A-73-1 completed the analyses to address (1) and (2) above. The analyses showed that the
presence of ethanol in gasoline increased permeation by 116 and 75 percent, respectively, for
vehicles certified to enhanced evaporative and Tier 2/LEVII standards. This compares to the 114
percent increase modeled by NPRM MOVES for both sets of standards. As such, the NPRM
MOVES ethanol permeation impact is overstated for the newest technology vehicles.
Our Response:
Our analysis differs from A-73-113. We did not find a statistical significance in ethanol
effects for different vehicle standards in the E-77 data sets whether we use the E20 tests or not in
the analysis.
What Commenters Said:
Commenter: Governors' Biofuels Coalition
EPA must ensure that its fuel models are updated to reflect new science and the realistic
emissions from the combustion of commercial gasoline (as opposed to unrealistic certification
fuels).
Our Response:
EPA agrees with the commenter. Accordingly, the emissions model used to estimate the
emissions reduction from the proposed rule included updated fuel-effects model incorporating
the latest researches on the effects of fuel properties on on-road emissions.14'15'16
What Commenters Said:
Commenter: Pennsylvania Department of Environmental Protection (DEP)
Nevertheless, DEP has concerns about... potential gaps or flaws in the emissions modeling.
Our Response:
EPA disagrees with the assertion that there are potential gaps or flaws in the emissions
modeling, especially since the commenter did not provide any detail about the areas of concern.
14
U.S. EPA, 2013. Assessing the Effect of Five Gasoline Properties on Exhaust Emissions from Light-Duty
Vehicles certified to Tier 2 Standards: Analysis of Data from EPAct Phase 3 (EPAct/V2/E-89). Final Report. EPA-
420-R-13-002.
15 The Effects of Ultra-Low Sulfur Gasoline on Emissions from Tier 2 Vehicles in the In-Use Fleet, EPA-420-R-14-
002.
16 U.S. EPA. 2014. Memorandum to Docket: Updates to MOVES for the Tier 3 FRM Analysis
3-38
-------
Tier 3 Summary and Analysis of Comments
Again as previously explained, EPA recognizes that any modeling analysis, and any projection
of future conditions, inherently has uncertainties. But "[i]t is only when the model bears no
rational relationship to the characteristics of the data to which it is applied that [courts] will hold
that the use of the model is arbitrary and capricious." Appalachian Power Company v. EPA, 135
F.3d 791, 802 (D.C. Cir. 1998)(internal citations omitted).
What Commenters Said:
The EPA should explain in greater detail the methodology for obtaining the emission reduction
estimates that would immediately occur when 10 ppm sulfur average levels at the refinery gate
become available to the nation's fleet.
Our Response:
We disagree with the commenter that our methodology for estimating emissions
reductions from 10 ppm sulfur was not adequately explained. Our modeling methodology is
clearly documented in Chapter 7 of the RIA, and the Emissions Modeling Technical Support
Document. The underlying peer-reviewed study of the effects of sulfur on Tier 2 vehicles are
described in Section IV of the Preamble and in Chapter 1 of the Regulatory Impact Analysis.
Furthermore, the memorandum to the docket on updates to MOVES describes how the results
from the in-use sulfur study were incorporated into the MOVES model used for the air quality
and inventory analyses.
What Commenters Said:
Aviation fuel; gasoline subject to the averaging, banking, and trading program; diesel fuel; and
home heating oil in the liquid fuels transportation network all can lead to contamination of Tier 3
gasoline. As EPA states in the proposed rule in reference to a 10 ppm cap, 'The U.S. gasoline
distribution system poses contamination challenges that make it difficult to set and enforce such
a tight downstream per-gallon sulfur standard.' (78 FR 29927). It is quite possible during the first
three years of this program that the 10 ppm refinery gate average could be much higher
downstream of the refinery when the fuel reaches the Northeastern United States. The EPA
should explain what will be the average downstream level of sulfur in gasoline available to the
motoring public and indicate that the emissions modeling is based off this realistic sulfur level.
As indicated in the previous comment, even levels of sulfur near 15 ppm will have deleterious
effects on gasoline. Further, fuels available in terminals in Pennsylvania today exceed the current
30 ppm Tier 2 sulfur fuel standard. Some gasoline sulfur levels in Pennsylvania approach 50
ppm. The modeled emission reductions and monetized health benefits should be based on what
gasoline sulfur level will actually reach the consumer.
The EPA should verify the accuracy of the emissions modeling results if intra-company sulfur
credit trading is allowed. If refineries can use intra-company sulfur credit trading, how does the
varying average sulfur content of gas affect modeling assumptions? It seems likely that gasoline
sulfur levels will vary largely by producer, region, and even batch. If the average sulfur content
varies widely, emission reductions could vary widely by region which means that some areas
will not achieve the reductions promised by this program while other achieve higher reductions
3-39
-------
Chapter 3: Emissions and Air Quality Emissions Impacts
than promised. This is important to quantify to allow an accurate representation of the monetized
benefits.
Our Response:
The emissions modeling assumed 10 ppm sulfur everywhere nationwide, because there is
too much uncertainty associated with the individual refinery compliance decisions and the
movement of fuel through the fuel distribution system to predict in-use sulfur levels in specific
geographic areas. This is consistent with our expectation of what the national average in-use
sulfur levels will be at retail. Potential contamination downstream of the refinery, or increases in
sulfur due to small volume additives are expected to be minor, and offset by minor downstream
decreases from downstream blending of lower sulfur blendstocks. As the commenter notes, the
averaging, banking, and trading program means that some gasoline could be above 10 ppm sulfur
and some gasoline could be below 10 ppm sulfur. However, because refineries generating
credits and using credits are interspersed across the country, and because most areas receive a
considerable portion of their fuel by pipeline, barge, rail or truck from refineries in other areas,
we expect the variation in average sulfur levels across the country to be limited.
EPA is committed to monitor and further evaluate in-use sulfur levels and their impact on
vehicle emissions. Such ongoing evaluation will include analyses of: in-use fuel surveys; batch
data that refineries are required to submit; and the sulfur credit market. It will also include
evaluation of any issues or concerns that might arise during implementation of the program.
Finally, we will also carry out an ongoing evaluation of data submitted by the vehicle
manufacturers on the performance of their Tier 3 vehicles in-use.
3.1.2. Criteria and Toxic Pollutant Emission Impacts
What Commenters Said:
Commenter: Chevron Products Company
Impact on Emissions and Air Quality: The NPRM indicates that by 2050, the proposed Tier 3
program would reduce NOx and VOC emissions by nearly 40% from the level of emissions
projected without Tier 3 controls. Missing from EPA's assessment is a reference point that
allows the comparison of Tier 3 to other similar control measures. Although a 40% reduction is
large on a percentage basis, this reduction is off of a relatively low Tier 2 baseline, and it gives
the reader an incomplete picture of the absolute emissions reductions from the proposal.
Our Response:
In Chapter 7 of the Regulatory Impact Analysis for the proposal, in addition to providing
the reduction in percent space, the estimates of absolute emissions reductions (in U.S. Short
Tons) were also presented.
3-40
-------
Tier 3 Summary and Analysis of Comments
What Commenters Said:
Commenter: Refinery Automation Institute, LLC
1) Using EPA's own Complex Model Phase 2 tool, the decrease in going from the current 30ppm
S to 10 ppm S in the levels of VOC, Toxics, and NOX is very small, less than 1% (Attachment
I)
Our Response:
We disagree with the comment. To estimate the emissions reduction from the Tier 3
standard, substantial updates have been made in MOVES, including replacing the Complex
Model with a new fuel-effects model for 2001-and-later model year vehicles, based on recent
research on the effects of fuel properties on on-road emissions.17'18'19 However, for pre-2001
model year vehicles, the Complex Model is still used. Based on the model results, we expect
substantial emissions reductions in many harmful pollutants with the Tier 3 rule (see Section III
of the preamble to the final rule).
3.1.3. Greenhouse Gas Emission Impacts
What Commenters Said:
Commenter: Environmental Defense Fund (EDF)
Tier 3 Rule Will Result in a Net Decrease in GHG Emissions:
EPA has estimated that the proposed Tier 3 standards will result in net GHG reductions. The
proposed sulfur and tailpipe standards are projected to reduce nitrous oxide and methane
emissions, which are both potent greenhouse gases. Higher energy use will be required to
remove sulfur at refineries, which will marginally increase CO2 refinery emissions, but this
modest increase is expected to be offset by reductions in vehicle emissions of more potent
greenhouse gases, including nitrous oxide and methane.
Our Response:
EPA agrees that Tier 3 will result in a net decrease in GHG emissions.
U.S. EPA, 2013. Assessing the Effect of Five Gasoline Properties on Exhaust Emissions from Light-Duty
Vehicles certified to Tier 2 Standards: Analysis of Data from EPAct Phase 3 (EPAct/V2/E-89). Final Report. EPA-
420-R-13-002.
18 The Effects of Ultra-Low Sulfur Gasoline on Emissions from Tier 2 Vehicles in the In-Use Fleet, EPA-420-R-14-
002.
19 U.S. EPA. 2014. Memorandum to Docket: Updates to MOVES for the Tier 3 FRM Analysis
3-41
-------
Chapter 3: Emissions and Air Quality Emissions Impacts
What Commenters Said:
Commenter: Marathon Petroleum Company LP (MFC)
In section III B.7., EPA states that 'We do not expect the Tier 3 vehicle standards to result in any
discernible changes in vehicle CO2 emissions or fuel economy.' This is because EPA assumes
that all the increased refinery GHG emissions will be offset by N2O and CH4 reductions in
tailpipe emissions. EPA estimates refinery GHG emissions to be 4.6 MMTCO2e for 2017. EPA
attempts to downplay this estimate but if their refinery by refinery model is as accurate as they
claim, there can't be much downside to this estimate.
Our Response:
The EPA statement that is quoted in this comment is referring to per-vehicle CC>2
emissions and vehicle fuel efficiency. We are not aware of any technological reasons that vehicle
emission controls responding to the Tier 3 vehicle standards would affect vehicle CO2 emissions
or fuel economy in any significant way.
As the commenter notes, we also considered the CC>2 emissions from refineries, and the
non- CO2 greenhouse gas emissions (CH4 and N2O) from vehicles. We believe that the potential
increase in GHG emissions from the Tier 3 standards would primarily come from refinery
emissions. However, we expect the combined impact of CH4 and N2O emission reductions from
the vehicles and CC>2 emission increases from the refineries would result in a slight net decrease
on a C (^-equivalent basis in 2018, and by 2030 a larger net decrease of between 2.5 to 2.7
MMTCO26 (see Section III of the preamble for additional detail).
What Commenters Said:
Commenter: WE ACT for Environmental Justice
And last but certainly not least, the importance of this rule on climate change. While the rule
documentation specifically states that an increase or decrease in greenhouse gas emissions
cannot be ascertained with certainty at this point because of the expected emissions increases in
the refinery process to reduce the amount of sulfur in the fuel, what you are proposing here will
ultimately, whether now or in the near future, reduce GHG emissions. As an engineer that
worked in refineries and many other chemical facilities, I am confident that technology will
advance accordingly so we can say - with confidence - that one day, there will be a net reduction
in CO2 emissions and subsequently a net reduction in negative health impacts. That is why it is
especially important that we take every step to improve our air quality, especially for those that
suffer respiratory and cardiovascular challenges during extremely hot weather or periods of 'heat
waves' that we will continue to experience more frequently due to our changing climate.
Our Response:
We appreciate the comment. We expect the combined impact of CH4 and N2O emission
reductions from the vehicles and CC>2 emission increases from the refineries to show a slight net
3-42
-------
Tier 3 Summary and Analysis of Comments
decrease in 2018 on a CO2 equivalent basis. While still small, this net decrease grows to a range
between 2.5 to 2.7 MMTCO2e by 2030.
3.2. Air Quality Impacts of the Proposed Program
3.2.1. January 1, 2017 Implementation is Not Necessary To Help Areas Attain the Ozone
NAAQS
We received comments reflecting two views: 1) that the standards, in the proposed timeframe,
are not needed by states and others to attain the NAAQS; and 2) that the standards are important
to the states for attaining the NAAQS and that the proposed timing is relevant for them.
What Commenters Said:
American Fuel & Petrochemical Manufacturers (AFPM), American Petroleum Institute
(API) and Marathon Petroleum Company LP (MFC):
EPA asserts that reducing gasoline sulfur to 10 ppm on January 1, 2017 will have immediate
benefits for the existing vehicle fleet by reducing emissions from Tier 2 vehicles. Even if that is
true, that does not justify January 1, 2017 implementation of the 10 ppm gasoline sulfur standard.
Furthermore, implementing the rule on January 1, 2017 will not help nonattainment areas reach
attainment. For the current ozone NAAQS (promulgated in 2008), the Agency defined the
classifications of nonattainment designations as: marginal, moderate, serious, severe, or extreme,
and set deadlines for these areas to come within attainment for each category.
The only way to attain by the end of 2015 is to have the 2013-2015 summers be clean. Similarly,
the Moderate areas need the 2016 summer to be clean to be reclassified as attainment. Tier 3 is
too late for the Marginal and Moderate areas. In contrast, taking EPA's claims of emissions
reductions benefits as true, Tier 3 could benefit the Serious, Severe, and Extreme areas, but
January 1, 2017 implementation is not necessary to help these areas reach attainment by
December 31, 2021. We suggest, consistent with the need to provide 5 years of lead time as
described below, and the requirements for the Serious areas to demonstrate attainment, that EPA
implement this rule on January 1, 2019.
The Agency claims that this proposed Tier 3 standard is necessary for States to attain the existing
ozone NAAQS. Last year, EPA promulgated nonattainment areas for the 2008 ozone NAAQS.
Most — 36 are marginal out of 46 total nonattainment areas - must be in compliance by 2015
based on monitoring data for 2013, 2014, and 2015. Tier 3 will not help these 36 marginal
nonattainment areas because it will not be effective before 2017.
Commenter: American Lung Association of the Mid-Atlantic
Indeed, as the October 2011 report on this subject by the National Association of Clean Air
Agencies very plainly points out, the ability of States and localities to attain the 2008 ozone
standard is tied directly to timely promulgation and implementation of Tier 3 standards.
3-43
-------
Chapter 3: Emissions and Air Quality Emissions Impacts
Commenter: American Lung Association Illinois
Recently the National Association of Clean Air Agencies also pointed out that the ability of
States and localities to attain these standards is tied directly to the timely promulgation and
implementation of Tier 3.
Commenter: Appalachian Mountain Club (AMC)
Many states, especially here in the Mid -Atlantic, may have difficulties attaining or maintaining
the National Ambient Air Quality Standards for ozone in coming years, without Tier 3.
Commenter: Attorney General of Connecticut, et.al; Attorney General of New York, et.al
By setting more stringent emission standards for new passenger cars and trucks and cutting the
allowable sulfur content in gasoline, the Tier 3 standards would reduce nitrogen oxides and
volatile organic compounds by 80 percent and particulate matter by 70 percent. These reductions
would go a long way toward limiting the production of smog. As EPA recognized, "[i]n the
absence of additional controls, many areas will continue to have ambient ozone concentrations
exceeding the NAAQS in the future." 78 Fed. Reg. at 29,819.
As EPA aptly noted in the preamble to the proposed rule, "few other national strategies exist that
would deliver the same magnitude of multi-pollutant reductions projected to result from the
proposed Tier 3 standards." 78 Fed. Reg. at 29,819. Therefore, we strongly support the adoption
of the proposed Tier 3 standards and urge that they be finalized by the end of the year so that our
States and cities may realize their benefits as soon as possible. Thank you for your consideration
of our comments on this important matter.
Commenter: Chicago Metropolitan Agency for Planning
However, CMAP's actions alone will not bring the region into attainment of air quality
standards. National action is required to bring overall mobile source emission rates down. This
will have a significant effect both on emissions within the region and pollutants transported from
other parts of the country. Transport in particular is a serious issue - the background monitors in
the region are almost at the nonattainment level, indicating that out-of-region sources are a
dominant contributor to the region's air quality problems.
Previous motor vehicle emission and fuel standards have had a major impact on air quality.
Standards are becoming tighter as scientific understanding of air pollution's impact on people
and the environment improves, and rightly so. In order to meet the new standards, tighter motor
vehicle emission and fuel standards are required; the proposed Tier 3 standards should be
adopted.
Commenter: City of Philadelphia Department of Public Health Management Services (AMS)
The Tier 3 would reduce nitrogen oxide, volatile organic compound, emission by eight and three
percent by 2017, by 2030 by 28 and 23 percent, respectively. The approximate emission of
3-44
-------
Tier 3 Summary and Analysis of Comments
benzene and total air emissions would be reduced by four and three percent by 2017, and 36 and
23 percent by 2030. These reductions will significantly improve to help benefit Philadelphians,
and as well help us to meet the National Ambient Air Quality Standards.
Commenter: Delaware Department of Natural Resources (DNRC)
The proposed Tier 3 standards will improve the effectiveness of motor vehicle emissions
controls and help Delaware to meet and maintain compliance with the current ozone standard
and any potentially tighter future standard.
Delaware has been in non-attainment for the pollutant ozone since the standard was first
established in 1971. Over the past 40 years, Delaware has reduced the impacts of ozone on our
citizens through the adoption of numerous control measures. Delaware control measures adopted
since 1990 have reduced ozone precursor pollutants, volatile organic compounds (VOC) and
oxides of nitrogen (NOx), by 70%; yet Delaware's air quality remains unhealthy. EPA has
designated portions of the State of Delaware as non-attainment for the health based 2008 75 ppb
ozone standard, and in 2012 Delaware's air monitors recorded 39 exceedances of the standard,
with one exceedance being 40% above the standard. In Delaware, mobile sources emitted 64% of
state-wide NOx emissions for 2008.
Delaware has very limited authority to control motor vehicle emissions and fuels. As a result,
Delaware has chosen to exercise the authority provided in Section 177 of the Clean Air and has
adopted California's Low Emissions Vehicle tailpipe standards, but Section 211 of the Clean Air
Act limits Delaware's ability to establish clean gasoline standards.
Ozone remains a major challenge, and motor vehicles are a major source of air pollution in
Delaware, and in areas upwind of Delaware. We urge the EPA to finalize this proposal.
Commenter: Environmental Defense Fund (EDF)
More than 158 million people in the U.S. currently live in areas that have been designated as
nonattainment for one or more of the NAAQS, which are health-based national pollution
standards. Achieving and maintaining the NAAQS has been particularly challenging for high-
growth areas that are experiencing a significant increase in vehicle use. Tier 3 will yield
significant emission reductions in fine particulates and ozone, improving air quality across the
nation and assisting states in meeting their obligations under the NAAQS program.
Over 138 million people currently live in ozone nonattainment areas, which include many major
population centers. EPA's analysis found that decreases in ozone, likely due to tailpipe
reductions in NOx and VOCs as a result of Tier 3 exhaust and fuel standards, will be dramatic
enough in some areas to lower ozone levels from above the 2008 8-hour standard to below it.
This has important implications for the 46 areas (representing 27 full or partial counties) that
were designated as nonattainment areas in 2012 for the 2008 ozone standard. Specifically, EPA
projects that, in 2017, ozone design values in Bucks County, Pennsylvania, Arlington County,
Virginia, and St. Louis County, Missouri - with a total population of almost 2 million people -
will move from being above the standard to below. And by 2030, EPA has modeled that Hudson
3-45
-------
Chapter 3: Emissions and Air Quality Emissions Impacts
County, New Jersey and Brazoria County, Texas - with a projected population in 2030 of over 1
million people - will have ozone design values reduced from above the standard to below as a
result of the proposed rule. And, in 2030, more than 200 counties are projected to have ozone
design value (DV) decreases of greater than 1.5 ppb.
Additionally, EPA estimates that in 2030, 60 counties in the U.S. will move from above 60 ppb
to at or below 60 ppb under the Tier 3 scenario, compared to the reference case. This indicates
that the reductions made by Tier 3 could also be vital in helping states meet a strengthened ozone
standard.
EPA also projects that the emissions reductions resulting from the Tier 3 program would help
states attain and maintain the PM2.5 NAAQS. In 2030, over 100 counties are projected to have
fine particulate design value decreases of greater than 0.05 micrograms per cubic meter.
Commenter: HEAL Utah on behalf of Jeff Miller et al:, and on behalf of Utah State Legislators
As members of the Utah Legislature, we have regularly heard from our state environmental
regulators about the challenges they face in meeting EPA air quality standards, such as the
National Ambient Air Quality Standards for PM 2.5. Our regulators have passed dozens of new
rules, instituted public education programs and tightened controls on industry, yet still struggle to
meet those standards.
One of the main culprits, particularly for our wintertime inversion episodes, is our transportation
sector. Our regulators estimate that 57 percent of the pollutants that plague us during those poor
air quality winter days come from tailpipe emissions.
Along Utah's Wasatch Front, we have boosted investment in transit and put in place programs to
encourage alternative fuel vehicles, such as CNG cars and trucks. Yet tailpipe emissions remain
a significant polluter.
Thus, it was with great interest, that we have learned of your agency's proposed Tier 3 Vehicle
Emission and Fuel Standards Program, which would require lower—sulfur gasoline and more
advance vehicle pollution control systems. We believe these safeguards would have a major
impact on Utah's poor air quality.
Commenter: Kentucky Division for Air Quality
Currently, Kentucky has a number of areas that need to improve air quality to meet the 2008 8-
hour ozone National Ambient Air Quality Standard (NAAQS). Tier 3 is a vital program that will
assist urban areas in Kentucky and in other states in meeting and attaining the ozone NAAQS.
Specifically, by 2030, the proposed Tier 3 program would result in reductions in on-road mobile
source emissions of nitrogen oxides (NOx), volatile organic compounds (VOC) and carbon
monoxide of at least 25 percent from current levels. Further, the Tier 3 program would reduce
on-highway emissions of NOx and VOC nearly 40 percent by 2050, when Tier 3 vehicles would
comprise almost the entire fleet, greatly assisting our heavily populated urban areas in realizing
clean air. Additionally, the program would reduce sulfur dioxide (SO2) by 16,621 tons in 2017
3-46
-------
Tier 3 Summary and Analysis of Comments
and by 17,267 tons in 2030. Direct particulate matter (PM2.5) would be reduced by 121 tons in
2017 and 7,458 tons in 2030.
More than 158 million people across the nation currently live in areas where the air they breathe
violates at least one of the health-based NAAQS. Although interstate transport has in the past
been a major contributor to air quality problems in some areas, programs and controls have been
put in place to address this concern. Mobile source emissions, including those from the passenger
cars and light trucks that are the focus of this proposal, are a primary contributor to these
violations, playing an especially central role in elevated levels of ozone and fine particulate
matter (PM2.5). The vehicles to be affected by Tier 3 standards contribute to other air quality
problems as well, including toxic air pollution, regional haze and the eutrophication of water
bodies.
Commenter: Metropolitan Washington Air Quality Committee (MWAQC)
The Washington region has made great strides in cleaning the air thereby, reducing the health
impacts of poor air quality, which cause a number of respiratory problems (e.g., asthma) and
premature death. It has done so by lowering emissions of pollutants that produce Smog and fine
particles in the atmosphere. As a result, the region has met the annual fine particle standards,
which-were published in 1997 (15 |ig/m3) and 2013 (12 |ig/m3) as well as daily fine particle
standards, which were published in 1997 (65 |ig/m3) and 2006 (35 |ig/m3). Besides this, the
region has also met the previous 1-hour ozone standard (0.12 ppm) published in 1979. However,
there are tougher challenges ahead for the Washington region. The region needs to meet the
current ozone standard (0.075 ppm) by 2015 and also needs to make sure that the region remains
in attainment for the current fine particle standards (annual: 12 |ig/m3, daily: 35 |ig/m3). EPA is
also planning to propose a possibly tougher ozone standard this year. EPA may also further
tighten the fine particle standard in the future.
In view of the challenges ahead for meeting and/or maintaining the tougher federal ozone and
fine particle standards, the Washington region needs to reduce the emissions of pollutants
producing ozone and fine particles (VOC, NOx, PM2.5, and SO2) significantly. The Tier 3 rule
will help the Washington region immensely as it will provide cleaner gasoline and vehicles. This
will also help the three jurisdictions in this region as they implement strategies to meet and/or
maintain the above mentioned federal standards.
Passenger vehicles are the largest emitters of nitrogen oxides (NOx) and one of the largest
emitters of volatile organic compounds (VOCs), carbon monoxide (CO), and PM2.5 in the
Washington region. A study by the National Association of Clean Air Agencies (NAC AA)
estimated that the Tier 3 program can reduce gasoline vehicle emissions of nitrogen oxides,
carbon monoxide, and volatile organic compounds by 29, 38 and 26 percent respectively.
Reductions in emissions of the above pollutants will cause reduction in the levels of ozone, fine
particles, and carbon monoxide. Clearly, implementation of the Tier 3 rule will lead to
significantly cleaner air in the Washington region, providing important health benefits to
millions of people.
3-47
-------
Chapter 3: Emissions and Air Quality Emissions Impacts
Commenter: Mid-Atlantic Regional Air Management Association Inc. (MARAMA)
Northeast States for Coordinated Air Use Management released their Assessment of Clean
Gasoline in the Northeast and Mid-Atlantic States.
Over the first decade of the 21st century, the Mid-Atlantic region achieved significant
improvements in air quality. Ozone remains a major challenge, and motor vehicles are a major
source of air pollution in our region. We encourage actions to cost-effectively reduce air
pollution from motor vehicles as soon as reasonable, enabling the public to breathe healthy air.
In the absence of a national Tier 3 program, areas needing additional emission reductions may
have no choice but to turn to other, more expensive, less cost-effective measures, including
additional controls on stationary sources (including already controlled sources) to meet their
statutory clean air obligations. This would place an unfair economic burden on local industries
and businesses when a more cost-effective national program which directly affects the
contributing sources is available. Moreover, achieving equivalent emission reductions of the
magnitude that will result from Tier 3 could be extremely difficult, if not impossible, in areas
where stationary sources are already highly controlled and there aren't enough other sources to
implement controls.
Commenter: National Association of Clean Air Agencies (NACAA)
State and local air pollution control agencies are relying on EPA to adopt the Tier 3 rule. Section
177 of the Clean Air Act authorizes states to opt into California's LEV III tailpipe standards, but
not all states are able to take advantage of this opportunity. Moreover, the Clean Air Act
precludes all states except California from adopting low-sulfur gasoline standards. Therefore, it
is imperative that the federal government take action this year to adopt Tier 3. If the rule is not
promulgated by December 31, 2013, Tier 3 may not apply to the 2017 model year of vehicles
and an entire year of benefits will be lost. This delay will have a serious and adverse impact on
human health and welfare.
Commenter: Natural Resources Defense Council (NRDC)
In the proposed rulemaking, EPA notes that over 158 million people, or half of the U.S.
population, live in areas that fail to meet one or more National Ambient Air Quality Standards
(NAAQS). Emissions from motor vehicles are a particularly important contributor to NAAQS
non-attainment. EPA projects that by 2014 "in many nonattainment areas, cars and light trucks
will contribute 30-45 percent of total nitrogen oxides (NOX) emissions, 20-25 percent of total
volatile organic compound (VOC) emissions, and 5-10 percent of total direct particulate matter
(PM2.5) emissions". By reducing these compounds through the Tier 3 rules, meaningful and
necessary reductions in ozone, PM and other air pollution can be achieved. EPA projects that the
Tier 3 proposal will bring many counties from above to below the ozone NAAQS.
Commenter: Northeast States for Coordinated Air Use Management (NESCAUM)
Even with the projected benefits associated with programs currently in effect, many of our most
populous areas are predicted to be nonattainment for the current 0.075 ppm ozone NAAQS in
3-48
-------
Tier 3 Summary and Analysis of Comments
2015. Current nonattainment designations of the ozone NAAQS fail to capture the extent of the
ozone pollution problem in the eastern United States. When EPA determined the attainment
status of areas in the U.S., it based its determinations on ozone monitoring data collected during
either 2008-2010 or 2009-2011. In the eastern states (Texas to North Dakota and to the east),
there were 99 monitors measuring ozone levels in violation of the 0.075 ppm NAAQS during
these years. This, however, presents a misleading picture of the extent of the air pollution
problem facing the states. In contrast to the time periods used by EPA to designate ozone
nonattainment areas, during 2010-2012 there were 252 monitors measuring violations of the
ozone NAAQS in these same states, an increase of over 150% in the number of violating
monitors. Many of these monitors are in cities, towns, and counties that EPA did not originally
identify as having ozone pollution problems.
Attaining the standard in these areas will require additional NOx reductions within our region as
well as in upwind areas that contribute to the region's pollution burden. Tier 3 is the most
significant strategy that the federal government could implement to help states attain and
maintain the NAAQS for ozone. The combined near-term benefits of the low sulfur gasoline
provisions and the increasing benefits of the tailpipe standards would help areas that need
additional reductions to attain, and assist other areas to stay in attainment.
The NESCAUM region, home to over 42 million people, is subject to episodes of poor air
quality resulting from ground-level ozone and fine particle pollution. During severe events, the
scale of the problem can extend beyond NESCAUM's borders and include over 200,000 square
miles across the eastern United States. Local and regional sources as well as air pollution
transported hundreds of miles from distant sources outside the region contribute to elevated
ozone and fine particle concentrations in the region.
NOx emissions contribute to a number of adverse public health and environmental outcomes.
NOx is the most important contributor to regional ozone concentrations and an important
precursor to fine particulate matter formation. These two pollutants are responsible for tens of
thousands of premature deaths, hospital admissions, and lost work and school days in the U.S.
annually. NOx is also a key factor in a number of environmental problems that affect the
Northeast.
Ozone remains a persistent pollution problem in parts of the NESCAUM region during warm
weather months. The evolution of severe ozone episodes often begins with the passage of a large
high pressure area from the Midwest to the middle or southern Atlantic states. Three primary
pollution transport pathways affect air quality in the region: long-range, mid-level, and near-
surface. During severe ozone episodes associated with high-pressure systems, these pathways
converge on the Mid-Atlantic area, where sea and bay breezes act as a barrier and funnel ozone
and other air pollutants up the Northeast Corridor.
Collectively, NOx emissions and ambient ozone concentrations in the region have dropped
significantly since 1997, along with the frequency and magnitude of exceedances of the health-
based ozone NAAQS. Despite this demonstrated progress, many of the most populous areas of
the region continue to violate the current 0.075 ppm ozone NAAQS. Attaining the standard in
these areas will require significant additional NOx reductions within the Northeast and in upwind
3-49
-------
Chapter 3: Emissions and Air Quality Emissions Impacts
areas. Federal measures such as the Tier 3/low sulfur gasoline program will significantly reduce
NOx emissions and help states achieve the requisite reductions.
Looking toward the future, additional NOx reductions will be critical to ozone attainment in a
broader swath of the region if EPA were to adopt a more health protective ozone NAAQS in the
range of 0.060 - 0.070 ppm as recommended by EPA's Clean Air Scientific Advisory
Committee (CASAC).
Commenter: Sierra Club
Further, the proposed rule will improve air quality across the country, especially those regions in
non-attainment with the health-based National Ambient Air Quality Standards ("NAAQS"). As
the RIA demonstrates, without Tier 3, several counties will continue to violate the 8-hour ozone
standard. Specifically, without the benefits of the Tier 3 controls, in 2017, 40 counties, with a
projected population of almost 50 million people, would experience levels of ozone that exceed
the 2008 8-hour ozone standard of 75 ppb. Tier 3 will assist those areas with attainment in 2017
and beyond, including some counties that would reduce ozone design values from above the
level of the standard to below it as a result of pollution reductions achieved solely by the
proposed rule. Moreover, the benefits of reduced ozone resulting from the proposed rule will
help fill the regulatory gap created by EPA's overdue review of the 2008 ozone standards.
Similarly, the proposed Tier 3 standard will assist many regions of the country that are currently
designated nonattainment for PM2.5 to come into compliance with the NAAQS. At present,
there are 50 PM2.5 nonattainment areas (for the 2006 NAAQS) with a population totaling over
105 million people. EPA provides evidence that 24-hour PM2.5 will decrease for many counties
by 2030 as a result of the proposed rule's projected emissions reductions of PM2.5, NOx SOx,
and VOCs.
The proposed rule also will decrease ambient levels of NO2. Reductions in NO2 resulting from
Tier 3 will "help any potential nonattainment areas to attain and maintain the standard," thereby
facilitating compliance with the health-based NAAQS. Moreover, reductions in NO2 will be
greatest in urban areas, helping to prevent the adverse health effects of NO2 exposure.
Commenter: State of Utah
Utah currently has three non-attainment areas for the PM2.5 NAAQS. In those areas, Nitrogen
Oxides (NOx) and Volatile Organic Compounds (VOCs) from on-road mobile sources are a
primary contributor to the PM2.5 in our air. Preempted by federal statute, Utah is unable to set its
own vehicle emission standards. The proposed Tier 3 Program, therefore, presents a promising
strategy to reduce NOx, VOCs, and direct PM2.5 from on-road mobile sources. If EPA's
projections are accurate, the proposed Tier 3 Program, in conjunction with new mileage
efficiency standards for vehicles, will reduce combined NO, and non-methane organic gas
emissions by 80 percent on a fleet average basis and paniculate emissions by 70 percent on a
per-vehicle basis by the year 2030.
Commenter: Utah Air Quality Board
3-50
-------
Tier 3 Summary and Analysis of Comments
Utah faces recurrent episodic air quality challenges for ozone and PM2.5. The state currently has
three non-attainment areas for the PM2.5 NAAQS. In those areas, on-road mobile sources are
currently one of the largest source categories for emissions of NOx and Volatile Organic
Compounds (VOCs). However, by federal statute, the state is preempted from setting vehicle
emission standards, leaving us with limited options for reducing on-road mobile source
emissions. The proposed Tier 3 Program fills this gap by providing a means of greatly reducing
NOx, VOCs, and direct from onroad mobile sources. Specifically, the proposed Tier 3 Program
will reduce combined NOx and nonmethane organic gas (NMOG, a subset of VOC) emission
standards by 80 percent on a fleet average basis and particulate emission standards by 70 percent
on a per-vehicle basis.
Commenter: Respiratory Health Association (RHA)
Anyone who has trouble breathing or has limited lung capacity will benefit from cleaner, less
polluted air. The Tier 3 rule will help achieve the drastic ozone reductions still needed to
achieve the current inadequate ozone standard. A strong Tier 3 standard will also be extremely
helpful in meeting an ozone standard based on science that the EPA should establish later this
year.
As the ozone standard still isn't correctly set, the reality is that the number is certainly higher
than the 158 million people EPA says are exposed to unhealthy levels of air pollution. Far too
many people do not even know that their air is not meeting standards that will protect their
health.
Commenter: Ozone Transport Commission (OTC)
Motor vehicles are the Ozone Transport Region's largest source of NOX, which is the most
important contributor to elevated regional ozone concentrations. EPA's Tier 3 proposal would
reduce NOX.
EPA is required under the Clean Air Act to set NAAQS that are protective of human health and
welfare. EPA lowered the health-based 8-hour ozone NAAQS to 75 parts per billion in 2008 and
is anticipated to promulgate a more stringent standard in 2014. The OTC's modeling efforts
demonstrate that gasoline-powered vehicles remain a significant contributor to ground level
ozone. Based on this modeling demonstration, attainment of the 2008 health-based ozone
standard will be impossible in the OTR without additional emission reductions from highway
vehicles and other mobile sources. Ozone precursor emissions from mobile sources are the
largest contributor to ozone levels within the OTR. As stated in the Preamble for EPA's
proposed rule 'Control of Air Pollution From Motor Vehicles: Tier 3 Motor Vehicle Emission
and Fuel Standards,' the vehicle emission standards, combined with the proposed reduction of
gasoline sulfur content from the current 30 parts per million (ppm) average down to a 10-ppm
average, is expected to result in a dramatic emission reduction of NOR, VOC, direct PM2.5,
carbon monoxide and air toxics. (78 Fed. Reg. 29816). Cleaner vehicles under the Tier 3
program will significantly reduce ozone precursor emissions and other pollutants as these
vehicles replace the existing vehicle fleet. Cleaner fuels will have the significant added
advantage of reducing emissions from the in-use fleet by enabling catalytic converters to reduce
pollution from all gasoline-powered vehicles by limiting 'NOR creep' associated with sulfur
3-51
-------
Chapter 3: Emissions and Air Quality Emissions Impacts
build up in the catalyst. Without clean gasoline, existing and improved vehicle emission
standards will not be as effective.
The Ozone Transport Commission (OTC) member states call on the U.S. Environmental
Protection Agency (EPA) to significantly reduce pollution from gasoline-powered motor
vehicles by promulgating stringent vehicle emission standards and lower sulfur content standards
for gasoline. Adoption of federal 'Control of Air Pollution from Motor Vehicles: Tier 3 Motor
Vehicle Emission and Fuel Standards' as proposed by EPA on March 29, 2013 will reduce ozone
and ozone precursors in the Northeast and Mid-Atlantic states, as well as in upwind states, that
significantly contribute to nonattainment of the ozone National Ambient Air Quality Standards
(NAAQS) in the Ozone Transport Region (OTR).
Commenter: Maryland Department of Transportation
Many counties in Maryland have been designated nonattainment for ozone and/or particulate
matter, which concerns us greatly. Maryland has already implemented many stringent controls,
costly transit improvements, and mobile source emission reduction strategies. However, we need
more effective programs that provide additional reductions of harmful emissions. We believe
that the Tier 3 Program is a positive step forward in efforts to reduce emissions from motor
vehicles and a logical progression of the work initiated in the Tier I Program and enhanced by
the Tier 2 Program. We strongly support the immediate benefits that will accrue to our region as
soon as the near term reduction of sulfur in fuels takes place and the longer term benefits that
will be realized as new technology (Tier 3) vehicles are introduced.
Commenter: New York State Department of Environmental Conservation
Air pollutant concentrations continue to exceed health-based National Ambient Air Quality
Standards in much of the United States, including portions of New York. Mobile sources,
particularly light duty vehicles, are often the dominant emission source for ozone precursors.
EPA's analysis suggests that in 2030, on-road mobile source emissions of oxides of nitrogen
(NOx) and volatile organic compounds (VOCs) will be reduced by 28% and 23% respectively.
These ozone precursor emission reductions will help reduce the incidence of adverse health
impacts from cardiovascular and respiratory illnesses and conditions known to be associated with
exposure to elevated levels of ozone.
The benefits of reducing average gasoline sulfur content to 10 ppm are not limited to future Tier
3 and LEV III vehicles. EPA's emissions modeling suggests that roughly half of the NOx
reductions associated with full implementation of Tier 3 occurs immediately in 2017 due to
emission reductions from the existing vehicle fleet, attributable to enhanced operation of existing
emission control equipment.
If the reductions associated with Tier 3 are not realized, alternative reductions will be necessary
across much of the country. There may not be sufficient feasible alternatives, and even if
feasible, any such alternatives will likely be more expensive and more disruptive. This is the
lowest hanging fruit remaining on the tree.
3-52
-------
Tier 3 Summary and Analysis of Comments
The sulfur standards and the vehicle standards they facilitate are vital to providing clean, healthy,
air to millions of New Yorkers.
These reductions are critical because there are two nonattainment areas in New York State for
the 2008 ozone national ambient air quality standard (New York-N. New Jersey-Long Island,
NY-NJ-CT and Jamestown, NY) that are required pursuant to EPA's designation to attain by
December 31,2015.
Specifically, the United States Environmental Protection Agency (EPA) designated the New
York — Northern New Jersey — Long Island, NY-NJ-CT area 'nonattainment' of the 2008
ozone national ambient air quality standard (NAAQS) of 0.075 parts per million (ppm) on May
21, 2012 with a 'marginal' classification. The DEC submitted an official reclassification request
to the EPA on June 20, 2012. The DEC requested to be reclassified pursuant to Clean Air Act
(CAA) section 181(a)(4) because the New York —Northern New Jersey — Long Island, NY-NJ-
CT ozone nonattainment area's design value of 0.084 ppm is within five percent of the
'moderate' classification threshold of 0.085 ppm. In the alternative, the DEC requested a
'voluntary' reclassification to 'moderate' pursuant to CAA section 181(b)(3). A voluntary
reclassification under section 181(b)(3) of the CAA is a non-discretionary EPA action; the
statute provides that '[t]he Administrator shall grant the request of any State to reclassify a
nonattainment area in that State... to a higher classification.' EPA, however, has not taken any
action because it questions its authority to grant a voluntary reclassification of the multi-state
region without the support of Connecticut and New Jersey. It should be noted that the
Administrator seemingly has the authority under CAA section 107(d)(3) to revise the designation
of an area if 'available information indicates that the designation of any area or portion of an area
within the State or interstate area should be revised.' In that regard, monitoring data for 2012
indicates the New York City metropolitan areas design value of 0.086 ppm is above the threshold
for a 'moderate' classification. DEC screening modeling also predicts non-attaining ozone design
values of 0.082 ppm in 2015 and 0.076 ppm in 2018.
Therefore, the reductions associated with the Tier 3 gasoline sulfur control and vehicle emissions
standards, estimated below, will be critical in meeting the emission reduction requirements of a
moderate nonattainment area.
EPA provides national emission reductions (from MOVES), but does not break them down by
state. Using a VMT ratio (4.3%, from data on the FHWA web site (1)) yields an estimate of
11,000 tons per year NOx in 2017 from the existing fleet and 22,000 tons of NOx in 2030. VOC
emission reductions are estimated at 1700 tons in 2017 and 9700 tons in 2030. PM2.5 is 5 and
320 and benzene is 70 and 370 respectively. These additional reductions, on top of what is
already being achieved, are significant relative to programs that the State has adopted and plans
to adopt as it deals with ozone nonattainment. NOx reductions in the range of 30 and 60 tons per
day in 2017 and 2030 respectively, coupled with VOC reductions in the same timeframe of 4.6
and 26 tons per day would contribute to the State's continued effort to control ozone precursor
emissions.
New York, through its adoption of California's LEV III vehicle standards, will receive some of
the emission reductions described above.
3-53
-------
Chapter 3: Emissions and Air Quality Emissions Impacts
Alternatives are less feasible, more expensive, and more disruptive.
The air quality impacts of the emissions from motor vehicles can't be placed in a compartment
separate from other emission sources. If the necessary emissions reductions from this proposed
rule are not realized, then they will have to be obtained in some other manner from other sectors
as the Department develops State Implementation Plans to address nonattainment issues within
the state.
Each year the Department works to identify strategies that result in meaningful State
Implementation Plan reductions. While these efforts have identified areas for significant
reduction in the past (NOx RACT, NOx SIP Call, Consumer Products, AIM Coatings, etc.) work
continues on identifying what, if any, additional reductions are available from these program
categories. If the reductions from Tier 3 are not realized, the Department is concerned that there
are not enough source categories left to regulate to make up the difference.
And just to point out that metropolitan New York is in non-attainment for the 2008 ozone
standard, and we have, in fact, based on actual monitored data requested a bump up of that non-
attainment status to moderate.
Well, again, Tier 3 for New York by itself- Tier 3 by itself doesn't do all that much for us
because we have the California program, and, in fact, New Jersey has the California program as
well. But we will certainly see the transport benefits associated with it as we get improvement in
air quality upwind from Tier 3.
Our issues are we're - we are non-attainment based on current monitoring data, based on 2012
data. We're still non-attainment. In fact, 2012 made us worse because it got rid of a cool
summer. So when we look at our current non-attainment status and the fact that we've requested
a bump up to moderate. We need to get the reductions to get into attainment in the 2018 time
frame, which would be required. 2015 looks virtually impossible.
While the emissions benefits associated with the Tier 3 standards won't fully accrue until the
fleet turnover, the environmental benefits from gasoline sulfur reductions will also improve the
effectiveness of millions of catalytic converters already in the field. We know that some of the
effects of sulfur on catalysts is reversible, so as soon as it goes into effect, we get benefits. Your
documents suggest 284,000 tons of NOx nationally in 2017. That's probably in the order of
10,000 tons of NOx for New York.
And to illustrate some of that impact, I looked at data - I/M data for 2012 in New York, and we
had 17,000 vehicles fail initial I/M tests for catalytic converter. And we don't know how many
might've been repaired before they were tested, so we know that sulfur improvements would
reduce that catalyst impact, and, therefore, not only save some money for the owners for catalyst
repairs, but also give us environmental benefits.
The real big bite occurs with the fuel sulfur benefit that allows us to get immediate improvement
and, you know, just doing a little pro rata of EPA's numbers, it looks like 10,000 tons of NOx for
New York. So, yeah, we can use that.
3-54
-------
Tier 3 Summary and Analysis of Comments
Commenter: American Lung Association
On March 29, 2013, EPA proposed an implementation rule for the 2008 ozone National
Ambient Air Quality Standards (NAAQS). EPA assumed a Tier 3 program with strong low
sulfur gasoline standards in its baseline analysis for attainment of the ozone NAAQS adopted in
2008. State and local governments are also preparing to meet NAAQS for paniculate matter,
nitrogen dioxide, and sulfur dioxide. Tier 3 will be a critical tool for local and state governments
to meet these clean air goals. In absence of Federal Tier 3 standards, state and local governments
will have to turn to other measures. In most areas, mobile sources comprise a large percentage of
the emission inventory across the nation. Finding pollution reductions equal to those that would
have come from a Federal Tier 3 program will be difficult.
Commenter: Environmental Defense Fund (EDF)
Tier 3 will make significant and immediate progress to reduce the pollution that forms harmful
ozone. The emission reductions made under this program are a crucial piece of what must be a
multipronged strategy to reduce ozone. Tier 3 will help states meet the 2008 Ozone NAAQS and
prepare to meet the expected 2013 Ozone NAAQS.
Our Response:
Chapter 7 of the RIA and Section III of the preamble to the final rule include EPA
projections of emission reductions and air quality improvements from the Tier 3 standards. EPA
believes that the emissions reductions and air quality improvements from the Tier 3 standards
will help states to attain and maintain the NAAQS "as expeditiously as practicable." (Sections
172(a)(2) and 181 of the CAA). The Tier 3 emissions reductions and air quality improvements
occur in a timeframe that is relevant for 2008 ozone nonattainment areas classified as moderate,
serious, severe and extreme. By beginning the program in 2017, it will aid attainment of the
NAAQS in moderate areas by December 2018. Furthermore, the Tier 3 standards will help any
areas that attain the NAAQS (including current marginal nonattainment areas) to maintain that
NAAQS in the future. EPA received numerous comments from states and other groups
indicating that the emission reductions and air quality improvements need to occur as soon as
possible. Timely implementation of Tier 3 will both provide public health benefits as soon as
possible, and also assist states as they develop attainment and maintenance plans (to avoid the
need for other, more costly state/local measures).
3.2.2. Air Quality Impacts of the Tier 3 Standards are Negligible
Three commenters, API, AFPM and Marathon, reference two studies done by Environ to assert
r-*-, 90 91 -^^
that the air quality benefits of the Tier 3 rule are negligible. ' Another commenter, EDF, noted
20 ENVIRON, Effects of Light-duty Vehicle Emissions Standards and Gasoline Sulfur Level on Ambient Ozone,
Final Report, prepared for the American Petroleum Institute, September 2012
21 ENVIRON, Effects of Light-duty Vehicle Emissions Standards and Gasoline Sulfur Level on Ambient Fine
Paniculate Matter, Draft Final Report, prepared for the American Petroleum Institute, June 2013
3-55
-------
Chapter 3: Emissions and Air Quality Emissions Impacts
that the Environ studies underestimate the benefits of the Tier 3 program because they analyze
2022, which is an interim year when the benefits of Tier 3 are not fully realized. Another
commenter, ECTA, hired Navigant Economics to compare the Environ studies and the EPA
analysis of Tier 3 impacts and concluded that the results, although not directly comparable, are
largely consistent and that the projected incremental benefits of Tier 3 vastly exceed the
incremental costs. Another commenter, Alliance and Global, noted that both the June 2013
Environ study and a 2013 Bloomberg Government ("BGOV") "Regulatory Analysis of EPA's
Tier 3 rule" focus on one co-benefit of sulfur reduction without addressing the totality of other
co-benefits.
What Commenters Said:
Commenter: American Fuel & Petrochemical Manufacturers (AFPM); American Petroleum
Institute (API); Marathon Petroleum Company LP (MFC)
Emissions benefits of the Tier 2 program continue to be realized as the vehicle fleet turns over.
In modeling the environmental impacts out to year 2022, recent studies by ENVIRON showed
that Tier 3 would yield incremental reductions in mean monthly summer 2022 PM2.5
concentrations of no more than 0.1 ug/m3 in contrast to a maximum incremental reduction of 2.7
ug/m3 in mean monthly PM2.5 ambient levels under the federal Tier 2 program. Similarly for
ozone, ENVIRON found the maximum ozone benefit expected from Tier 3 to be less than 1 ppb,
relative to a maximum ozone benefit of 12 ppb anticipated from the federal Tier 2 program.
EPA's modeling calculates Tier 3 reductions in ozone of 0.5 - 1.35 ppb and in PM2.5 of 0 - 0.05
ug/m3 in years 2017 - 2030. It should be mentioned that the current ozone and PM2.5 NAAQS
are 75 ppb and 12 ug/m3, respectively. The ENVIRON studies support the conclusion that
EPA's Tier 3 standards for new vehicle emissions and gasoline sulfur will provide negligible
reductions in emissions inventories, and negligible improvements in air quality.
Negligible environmental benefits:
The emissions inventory and air quality impacts of the Tier 3 Proposal are negligible. Even if
one accepts an assertion that EPA's air quality modeling analysis is accurately assessing the
impacts of the emissions reductions claimed for the Tier 3 proposal, these are likely to be
negligible. As the data analysis below shows, the proposed Tier 3 standards for new vehicle
emissions and gasoline sulfur will provide negligible environmental benefit, with respect to both
(a) reductions in emissions inventories, and (b) improvements in air quality. API recently
sponsored an assessment of the incremental nationwide emissions inventory reductions and air
quality benefits associated with the adoption of progressively more stringent light duty vehicle
emissions standards and gasoline sulfur limits over time (11). The studies, conducted by
ENVIRON, (12) (13) and provided as Attachments No. 9 and No. 10, showed that in 2022, the
summertime ozone precursor emissions of volatile organic compounds (VOCs), oxides of
nitrogen (NOx), and carbon monoxide (CO) from gasoline-fueled light-duty vehicles are
projected to be reduced by 62%, 80% and 51% respectively as a result of the implementation of
the federal Tier 2 program. In contrast, implementation of a federal Tier 3 program would further
reduce VOC, NOx and CO emissions by only 8%, 11% and 7%, respectively. Similarly, Tier 3 is
expected to yield lower reductions in PM precursors in comparison to those achieved by the Tier
3-56
-------
Tier 3 Summary and Analysis of Comments
2 program. The study showed incremental reductions in SO2 and direct PM2.5 of 92% and 19%,
respectively, attributable to the adoption of Tier 2, compared with additional reductions in SO2
and direct PM2.5 of 64% and 5%, respectively, as a result of Tier 3. It should be noted that the
absolute level of SO2 emissions in Tier 2 and 3 is two orders of magnitude lower than NOx
levels (e.g., 48.4 vs. 2879 Mg/day) and thus a 64% improvement of a very small number is
insignificant.
Following the trend in emissions inventories, ENVIRON found that Tier 3 is expected to yield
negligible improvement in ambient ozone levels relative to the large reductions that have
occurred (and are expected to continue) as a result of Tier 2. The maximum ozone benefit
expected from Tier 3 is less than 1 ppb, relative to a maximum ozone benefit of 12 ppb
anticipated from the federal Tier 2 program.
Similarly, ENVIRON concluded that incremental reductions in the monthly mean of ambient
PM2.5 concentrations attributable to Tier 3 will be negligible in comparison to those expected
from Tier 2. The study showed that Tier 3 would yield incremental reductions in mean monthly
summer 2022 PM2.5 concentrations of no more than 0.1 ug/m3 in contrast to a maximum
incremental reduction of 2.7 ug/m3 in mean monthly PM2.5 ambient levels under the federal
Tier 2 program.
Commenter: Emissions Control Technology Association (ECTA)
Most recently, on June 6, 2013, the American Petroleum Institute (API) released a study
prepared by ENVIRON International Corporation which concluded that relative to Tier 2, Tier 3
would only deliver small benefits in terms of PM2.5 reduction. The implication is that Tier 3 is
not justified because it will deliver small marginal environmental benefits. We strongly disagree
with this conclusion.
At ECTA's request, Dr. Hal Singer of Navigant Economics performed a review of ENVIRON's
study to assess its creditability. Dr. Singer compared the ENVIRON analysis with EPA's
analysis presented in the proposed rule and concluded that the results are largely consistent. He
also concluded that the point estimate of 2022 used in the ENVIRON study is misleading
because it does not reflect the full environmental benefit of PM2.5 reductions achieved under the
rule. Only a portion of the fleet will be upgraded to tighter Tier 3 standards by 2022. EPA's
analysis shows that a significant environmental benefit from Tier 3, in terms of PM2.5
reductions, will not be achieved until 2030 when 80% of the light-duty fleet is Tier 3 compliant.
ENVIRON was silent on the projected benefits from Tier 3 in 2030.
At your request, I have reviewed ENVIRON's study of the projected benefits from Tier 3, and I
compared its major results against those of the EPA. ENVIRON estimates that relative to
emissions reduction attributable to Tier 2, by 2022Tier 3 would reduce emissions from PM2.5 by
only 3 percent, and would reduce the monthly mean 24-hour PM2.5 concentrations in the Eastern
portion of the United States by less than 0.1 micrograms per cubic meter (|ig/m3). Because these
incremental reductions in PM2.5 are relatively smaller than the reductions achieved by Tier 2,
the ENVIRON study gives the impression that Tier 3 might not be justified.
3-57
-------
Chapter 3: Emissions and Air Quality Emissions Impacts
To survive a cost-benefit test, however, there is no reason why Tier 3 must deliver the same
benefits as Tier 2; rather, Tier 3 need only deliver incremental benefits in excess of incremental
costs. EPA's analysis shows that the proposed benefits of Tier 3 by 2030 would exceed the
proposed costs by between $4 and $20 billion. Thus, opponents of Tier 3 need to show that
EPA's benefits by 2030 are overstated by between $4 and $20 billion. Yet the ENVIRON study
is silent about the projected benefits from Tier 3 in 2030.
Moreover, the ENVIRON study does not even imply that Tier 3-related benefits by 2030 would
be significantly less than EPA's estimated benefits. As demonstrated below, there appears to be
no material difference in the projected incremental benefits of Tier 3 (relative to Tier 2) between
the ENVIRON study and the EPA study. Because ENVIRON's estimated reductions by 2022
are larger than those of EPA by 2017 but smaller than those of EPA by 2030 (with the exception
of NOx), it is conceivable that the implied social benefits from the ENVIRON study in 2030
exceed those of EPA.
For whatever reasons, the ENVIRON study focused on environmental benefits achieved by
2022, just five years after the implementation of the proposed rules in 2017. As EPA correctly
explained, however, it is mistake to look for PM2.5-related benefits before Tier 3-compliant
vehicles represents a majority of the fleet: "Reductions in direct emissions of PM2.5are projected
to result solely from the proposed vehicle tailpipe standards, so meaningful reductions are
realized mainly as the fleet turns over". Indeed, the ENVIRON study expressly allows for the
possibility of further reductions in PM2.5 after 2022, yielding larger social benefits: "The main
limitation of this study is introduced by the lack of complete phase-in of the LEV III standard by
2022, the basis year for comparing emission standards. Some additional improvements in PM2.5
beyond 2022 are expected as the LEV III standard fully matures". By providing benefits for 2022
only, the ENVIRON study makes it difficult to compare its results with those of EPA, which
presented benefits in 2017 (the first year of implementation) and in 2030 (when 80 percent of the
light-duty fleet are Tier 3-compliant).
With this caveat in mind, it is possible to compare the projected benefits across the two studies.
The ENVIRON study estimates incremental benefits by 2022 with partial LEV III technology
penetration of 8%, 7%, 11%, 64% and 5% in VOC, CO, NOx, SO2, and PM2.5 emissions,
respectively, from CONUS emissions of light-duty vehicles. The projected reductions in
emissions decline to 6%, 7%, 5%, 50%, and 3% in VOC, CO, NOx, SO2, and PM2.5,
respectively, when expressed as a share of emissions from all "on-road vehicles". By
comparison, the EPA projects reductions by 2017 of 3%, 4%, 8%, 51%, and 0.1% in VOC, CO,
NOx, SO2, and PM2.5 emissions, respectively; by 2030, EPA projects reductions of 23%, 30%,
28%, 51% and 10% in VOC, CO, NOx, SO2, and PM2.5 emissions, respectively.
Accordingly, ENVIRON's projected 3% decline in PM2.5 emissions by 2022 is consistent with
EPA's estimates of 0.1% decline by 2017 and 10% decline by 2030.The same is true of
ENVIRON's projected declines in VOC and CO emissions, which also are larger than EPA's
estimates for 2017 but smaller than EPA's estimates for 2030. There is effectively no difference
in the two estimates for SO2. Accordingly, the only so-called inconsistency across the two
studies is for NOx; but even here the projected differences are small (5% for ENVIRON in 2022
versus 8% for EPA in 2017).
3-58
-------
Tier 3 Summary and Analysis of Comments
In summary, unless and until ENVIRON produces emission-reduction estimates for 2030
attributable to Tier 3, it is difficult to compare the projections of the two models. As the record
currently stands, the two models are largely consistent, and the projected incremental benefits of
Tier 3 still vastly exceed the incremental costs.
Commenter: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
In the past few weeks, two analyses have been published that critique EPA's Proposed Tier 3 and
Proposed Gasoline Sulfur Standards. Both of these analyses are flawed because they "cherry
pick" one co-benefit of sulfur reduction and over-focus on it, without addressing the totality of
other co-benefits.
Automaker Response to ENVIRON Modeling Study on EPA Tier 3 Impact on PM2.5 Air
Quality - On June 20, 2013, API announced a new modeling study by ENVIRON International
Corporation which concludes that pending EPA Tier 3 regulations, including the proposed
reduction of one part of the current EPA market gasoline sulfur standard (the "refinery annual
average" limit) from 30 to 10 ppm, would do little to reduce fine particulates (PM2.5) by 2022,
at least in summertime in the U.S. eastern states:
"Overall, the modeling results suggest that large improvements in summertime ambient ground-
level PM2.5 concentrations occur in the eastern US as a result of the switch from Tier 1 to Tier 2
standards. However, relatively small additional reductions in 2022 PM2.5 concentrations are
predicted to result from the transition to a Federal standard similar to the California LEV III
standard, even when considering emissions reductions due to a lower gasoline sulfur content in
the LEV III scenario. These results are due to one or both of the following factors depending on
the pollutant: (1) the change in emissions between the Tier 2 and LEV III scenarios is relatively
small compared to the change between the Tier 1 and Tier 2 scenarios, and (2) the Tier 2 LDV
emissions of PM2.5 precursors constitute a relatively small fraction of the total inventory. In
particular, in the case of the PM2.5 precursor that experiences the largest relative reduction from
the Tier 2 to LEV III scenarios, SO2, on-road LDV emissions constitute only 0.2% of the total
anthropogenic inventory and hence SO2 LDV emission reductions result in small ambient PM2.5
benefits.
The main limitation of this study is introduced by the lack of complete phase-in of the LEV III
standard by 2022, the basis year for comparing emission standards. Some additional
improvements in PM 2.5 beyond 2022 are expected as the LEV III standard fully matures. In
addition, this study does not address wintertime PM2.5 benefits."
(Emphasis added), Sec. 3.2, p. 19, ENVIRON, "Effects of Light-duty Vehicle Emissions
Standards and Gasoline Sulfur Level on Ambient fine Particulate Matter," Final Report Prepared
for the American Petroleum Institute , June 2013, Project No. 06-31891 A.
ENVTRON's Conclusion Fails to Address Full Co-benefits of Tier 3 and Reduced Gasoline
Sulfur: In addition to the very important limitations of the ENVIRON study acknowledged at the
end of the report's conclusion, it is misleading to focus on only one co-benefit of market gasoline
3-59
-------
Chapter 3: Emissions and Air Quality Emissions Impacts
sulfur reduction (in this case reduction of S02 as a precursor to PM2.5 vehicle emissions)
without acknowledging the entire suite of significant additional co-benefits resulting from EPA's
proposed Tier 3 standard and more stringent market gasoline sulfur standard. These include
reducing various other vehicle emissions of concern, and enabling more fuel efficient vehicle
technologies. Use of ENVIRON's findings requires appropriate additional context. Likewise,
discussions of regulatory costs must be viewed in terms of the totality of the benefits.
ENVTRON's modeling analysis compares the effects of EPA emission standards and reductions
in market gasoline sulfur on PM2.5 precursor exhaust emissions in Tier 1 vs. Tier 2 standards,
compared with Tier 2 vs. proposed Tier 3 (represented by California LEV III) scenarios for one
season and one geographic area. Unfortunately, the conclusion focuses on only one among many
co-benefits.
SO2, as one precursor to PM2.5, will in fact increase year round, if the sulfur content of the fuel
is enough to "poison" the catalyst. Sulfur coats the surface and also chemically impairs the
catalyst action on the exhaust, preventing it from capturing three other key exhaust emissions,
NOX (Nitrous Oxides), CO (Carbon Monoxide), and VOCs (Volatile Organic Compounds —
precursors for Ozone and smog formation in the atmosphere) and converting them into nitrogen,
carbon dioxide and water. Reliable catalytic converter action is critical to meeting tough new
vehicle tailpipe emission limits. EPA's own new study data confirm the benefits of reducing
gasoline sulfur for in use Tier 2 vehicles, even below the new reductions EPA has proposed (See
EPA Regulatory Impact Analysis, Draft EPA 420-D-13-003). Tier 3 would bring national
gasoline sulfur closer to California, Japan, and EU levels.
Finally, the proportion of SO2 as a PM2.5 precursor from vehicle emissions was reduced under
the study's modeling scenarios by 92% between Tier 1 and 2, and by 64% between Tier 2 and 3,
both considerable achievements for air quality across the nation. In any event, under both sets of
scenarios, the vehicle emission contributions to PM2.5 have always been overshadowed by other
anthropomorphic sources, so there is nothing really new here from an EPA policy standpoint. In
addition, when you also consider the cumulative outcome adding ENVIRON's own Tier 3
projections of other reductions in vehicle emissions, 11% for NOX, 7% for CO, and 8% for
VOCs, and 5% for PM2.5 (and only to 2022, not even later years when additional reductions
would occur) plus the vehicle technology advancements promoted by reduction of gasoline
sulfur from Tier 3, the error in highlighting just SO2 or PM2.5 outcomes becomes obvious. That
said, we agree that NAAQS and State Implementation Plans should continue to include
reductions from stationary sources where the cost/benefit is justified.
Automaker Response to BGOV Regulatory Analysis of EPA's Tier 3 Rule: On June 13, 2013, a
Bloomberg Government ("BGOV") "Regulatory Analysis of EPA's Tier 3 rule" © Bloomberg
2013, Bloomberg Finance LP, was published that, among other things, found "...SO2 reductions
achieved by Tier 3 will be relatively expensive, with a cost of nearly $76,500 a ton of S02
reduced" and ends saying "Bloomberg Government's analysis finds that Tier 3 will have a costly
but minimal impact on certain pollutants targeted by the rule and that the health benefits are
possibly overstated based on recent academic research. Also, viewing sulfur independently,
would yield more cost-effective ways of meeting Tier 3's proposed requirements."
3-60
-------
Tier 3 Summary and Analysis of Comments
As with the ENVIRON report conclusion, you cannot disaggregate one co-benefit of gasoline
sulfur reduction for analysis (e.g., its reduction of SO2 vehicle emissions) from the entirety of all
the emission benefits from reducing sulfur in marketplace gasoline. Since you cannot logically
isolate the benefits for only one component of the exhaust without counting the benefits for the
other components, it is erroneous to assign the entire annualized cost of the sulfur reduction to
just one component benefit (i.e., taking $1.3B \17,000 tons of SO2 reduction to reach $76,500)
which strongly overstates the proportionate cost. The Tier 3 benefits of reducing sulfur would
not only enable prospective reductions for future vehicles, the reduction in market fuel would
also reduce emissions in the existing fleet and in other types of engines.
Commenter: Environmental Defense Fund (EDF)
In addition to overstating the costs of Tier 3, API erroneously claims that Tier 3 will not generate
any health benefits. Despite numerous independent, credible analyses, API has said that there
would be no benefits in reduced ozone or PM2.5 from Tier 3 versus Tier 2. However, their
ozone and PM assessments, conducted by ENVIRON International Corp., are seriously flawed
and misleading, and both underestimate the significant benefits of the Tier 3 program. Both
analyses estimate the benefits for the year 2022, instead of 2030 when the benefits of Tier 3 will
be more fully realized.
Commenter: National Association of Clean Air Agencies (NACAA)
NACAA strongly supports EPA's Tier 3 proposal to further strengthen the federal program to
regulate emissions from passenger cars and light trucks and lower sulfur levels in gasoline. We
are so supportive because we know of no other strategy that can achieve such substantial,
immediate and cost-effective reductions in air pollution as Tier 3.
Our Response:
The air quality impacts predicted to result from the Tier 3 rule are significant. As
NACAA pointed out, these air quality improvements are greater than could be achieved by any
other known, practical measure in the same timeframe. As described in Chapter 8 of the RIA,
these air quality improvements will provide the public with very significant health benefits and
the benefits of the final standards clearly outweigh the costs. In addition, as pointed out by the
Alliance and Global, the benefits of the rule come from the entire suite of emissions reductions
due to reducing sulfur in fuel, not just the reduction in SO2 from on-road vehicles.
3.2.3. Inadequate Technical Justification - Air Quality Impacts
What Commenters Said:
Commenter: American Fuel & Petrochemical Manufacturers (AFPM), American Petroleum
Institute (API) and Marathon Petroleum Company LP (MFC)
3-61
-------
Chapter 3: Emissions and Air Quality Emissions Impacts
API and AFPM submitted a report from Sierra Research in support for their comments.22 The
Sierra report included two comments on Air Quality Impacts (see Section 7 of Sierra report).
The first comment pertains to EPA's model performance evaluation and the second comment
pertains to the grid size for EPA's air quality analysis.
Based on the data presented by EPA, the agency's evaluation of model performance shows
that the model generated acceptable predictions of overall ozone and PM concentrations.
However, EPA failed to evaluate how well the model predicts VOC to NOx ratios or model
responsiveness to reductions in VOC, NOx, and/or PM emissions. This is an important issue
because uncertainty and compensating errors in the emission inventories and other inputs to
the air quality model make it possible for model performance in predicting ozone and PM
concentrations to be acceptable even if the model cannot accurately predict responses to
emission reductions. Given that the point of the modeling analysis is to assess the air quality
impacts of the emission reductions estimated to result from the Tier 3 proposal, EPA's failure
to perform this evaluation is a major flaw.
That there are large uncertainties in projected mobile source emission inventories is
highlighted by the large changes in those inventories whenever emission modeling software is
updated. Most recently, the release of the MOVES model increased motor vehicle emissions
considerably and altered VOC/NOx emission ratios. Uncertainty in emissions and in model
response to emission reductions is largest in urban areas where motor vehicle emissions are
concentrated, so the ability of the Tier 3 modeling to adequately respond to Tier 3 emission
reductions is also very uncertain in the absence of a meaningful evaluation of this issue by
EPA. In addition to failing to evaluate the performance of the model in response to changes in
emissions, EPA also failed to assess the role and relative importance of NOx or VOC
emission reductions in reducing ambient ozone concentrations or the role of PM, VOC, and
NOx emission reductions in reducing ambient concentrations of PM2.5. As is well known,
depending on the VOC/NOx ratio, NOx emission reductions can either increase or decrease
ambient ozone concentrations. In addition, NOx emission reductions can significantly impact
PM2.5 formation. In the absence of any assessment of the role and importance of NOx
emission reductions, it is impossible to determine the air quality benefits of the Tier 3
proposal in general and the proposed reductions in gasoline sulfur levels in particular.
The motor vehicle NOx emissions targeted by the Tier 3 proposal are concentrated in urban
centers where NOx reductions are less effective in reducing ozone; therefore, understanding
the model's response to NOx is particularly important for this rulemaking. An example of how
the complicated relationship between NOx and ozone, ozone trends, and sensitivity to
hydrocarbon emission control could have been evaluated by EPA can be seen in an evaluation
conducted in Atlanta, Georgia. That analysis found a local source of ozone produced by the
Atlanta urban area superimposed on an elevated regional background. Effects of NOx
reductions were complicated and dependent on meteorology and location. NOx control was
judged effective in reducing the regional background, but VOC control was more effective in
controlling ozone produced locally in the urban plume. Again, EPA's failure to assess this
issue makes it impossible to determine the air quality benefits of the Tier 3 proposal.
22 Sierra Research, June 20137. Assessment of the Emission Benefits of U.S. EPA's Proposed Tier 3 Motor Vehicle
Emissions and Fuel Standards. Prepared for: American Petroleum Institute.
3-62
-------
Tier 3 Summary and Analysis of Comments
Another approach to evaluating the performance of the model in responding to changes in
emissions is to focus on the variation of emissions from weekdays to weekends, which
provides a real-world test of emission inventories and modeling. Emissions of ozone
precursors are lower on weekends, primarily due to changes in the patterns of motor vehicle
use. Observations of weekend and weekday ozone provide very relevant real-world data with
which to evaluate the impact of motor vehicle emissions on air quality, yet the Tier 3 RIA
does not even include any discussion of weekend/weekday ozone, much less an assessment.
The potential impact of this approach can be seen in a recent study of Detroit and Atlanta
motor vehicle emissions with the MOVES model, which found that motor vehicle emissions
of NOx were 23.7% lower on Saturdays and 49.9% lower on Sundays in compared to
weekdays, while VOC decreased by 7% on Saturdays and 20% on Sundays. These weekend
motor vehicle NOx emission reductions are of the same order as those projected by the Tier 3
regulations. Another recent modeling study of the weekend/weekday phenomena in the
Midwest found that weekend emissions generally resulted in lower ozone over most of the
modeling domain, but higher ozone locally in two urban areas, and that NOx reductions on the
weekend were responsible for the weekend/weekday ozone differences.
Weekend ozone levels have decreased relative to weekday levels recently, and there is now
little difference in weekend and weekday ozone in most areas of the country as noted in a
recent study completed by Environ. The study authors note that precursors of ozone have
decreased significantly and that NOx emission reductions have been larger, leading to an
increase in VOC/NOx ratios. The faster decline in NOx emissions is a result of regulations
that targeted the largest stationary NOx sources in the eastern half of the U.S. (acid rain rules,
NOx SIP call, CAIR rules). Currently, most urban areas have similar ozone on weekends and
weekdays even though NOx emissions from motor vehicles are much lower on weekends.
This indicates that mobile source NOx emission reductions do not have a large effect on ozone
near urban areas, which is a critical issue with respect to the Tier 3 proposal that EPA has
failed to address.
Our Response:
The CMAQ model evaluation performed for the Tier 3 rule is consistent with prior
agency rulemakings, such as the 2017-2025 Light-Duty Vehicle Greenhouse Gas Emission
Standards Final Rule (77 FR 62624, October 15, 2012).
There is no independent way to evaluate the impact of VOC and NOx emissions
sensitivities on CMAQ projections since they cannot be isolated in the real world. However,
there have been several studies which evaluate the sensitivity of the system as a whole to
emissions reductions (called dynamic evaluation). Numerous dynamic evaluations of CMAQ's
ability to simulate the change in air quality resulting from emissions reductions have been
conducted and summarized in the peer-reviewed literature. For instance, Napelenok et al (2011)
concluded that the CMAQ model "is able to reproduce the observed change in daily maximum 8-
hour ozone levels" at the majority of locations when emissions uncertainty is considered. Other
dynamic evaluations (Zhou et al., 2013, Godowitch et al., 2010, Gilliland et al., 2008, Godowitch
et al., 2007) have suggested that CMAQ may be a conservative estimate of the air quality
improvements resulting from emissions reductions. Overall, the ozone, PM2.5, air toxics
3-63
-------
Chapter 3: Emissions and Air Quality Emissions Impacts
concentrations and nitrate and sulfate deposition model performance results for the 2005 CMAQ
simulations performed for the Tier 3 proposed rule are within the range or close to that found in
other recent applications. The model performance results, as described in this report, give us
confidence that our applications of CMAQ using this 2005 modeling platform provide a
scientifically credible approach for assessing ozone and PM2.5 concentrations for the purposes
of the Tier 3 rule.
EPA and other community modeling efforts have long-established that NOx controls are
the primary mechanism for reducing ozone levels regionally and that VOC reductions can lead to
more local reductions within specific urban areas (NRC 1991, US EPA 2010, and US EPA
2013). This finding points to the need for a combined NOx and VOC reduction strategy to
reduce ozone levels over the U.S. The Tier 3 proposal has been shown by EPA's fully-evaluated
modeling analyses to reduce ozone in a significant way over large parts of the U.S.
In addition, the recently released second draft of the ozone Risk and Exposure
Assessment (available at: http://www.epa.gov/ttn/naaqs/standards/ozone/s_o3_2008_rea.html)
evaluated ozone changes that would occur with NOx reductions alone versus what would occur
with combined NOx and VOC reductions.
References:
Gilliland, A.B., Hogrefe, C., Finder, R.W., Godowitch, J.M., Foley, K.L., Rao, S.T. (2008). Dynamic evaluation of
regional air quality models: assessing changes in O3 stemming from changes in emission and meteorology.
Atmospheric Environment, 42: 5110-5123.
Godowitch, J.M., Pouliot, G.A., Rao, S.T. (2010). Assessing multi-year changes in modeled and observed urban
NOx concentrations from a dynamic model evaluation perspective. Atmospheric Environment, 44: 2894-2901.
Godowitch, J.M., Hogrefe, C., Rao, S.T. (2008). Diagnostic analysis of a regional air quality model: changes in
modeled processes affecting ozone and chemical-transport indicators form NOx point source emissions reductions.
Journal of Geophysical Research-Atmospheres, 113(D19): D19303, DOI: 10.1029/2007JD009537.
Napelenok, S.L., Foley, K.M, Kang. D.W., Mathur, R., Pierce, T., Rao, S.T. (2011). Dynamic evaluation of regional
air quality model's response to emission reduction in the presence of uncertain emission inventories. Atmospheric
Environment, 45: 4091-4098.
National Research Council (1991): Rethinking the Ozone Problem in Urban and Regional Air Pollution, National
Academy Press, Washington D.C., 524pp.
U.S. Environmental Protection Agency (2010): Regulatory Impact Analysis for the Reconsideration of the 2008
Ozone National Ambient Air Quality Standard, RTF NC, 89pp.
U.S. Environmental Protection Agency (2013): Integrated Science Assessment for Ozone and Related
Photochemical Oxidants, EPA 600/R-10/076F, RTF NC, 1251pp.
Zhou, W., Cohan, D.S., Napelenok, S.L. (2013). Reconciling NOx emissions reductions and ozone trends in the
U.S., 2002-2006, Atmospheric Environment, 70: 236-244.
What Commenters Said:
Commenter: American Fuel & Petrochemical Manufacturers (AFPM), American Petroleum
Institute (API) and Marathon Petroleum Company LP (MFC)
Many studies have shown that ozone formation and sensitivity to NOx control depend on the
size of the modeling grid. In the Tier 3 NPRM modeling analysis, a uniform 12 km modeling
3-64
-------
Tier 3 Summary and Analysis of Comments
grid is used to cover the entire United States. The effect of modeling grid size on model
predictions was not evaluated by EPA.
The 12 km grid is likely to be adequate outside of urban areas and large point sources where
pollutant gradients are weak. However, the 12 km grid is inadequate to resolve pollutant
concentrations in urban areas where spatial gradients are large. This leads to an overestimation
of mixing in urban centers and an underestimation of NOx inhibition effects. Higher grid
resolution is particularly important for the Tier 3 NPRM because strong motor vehicle
emission sources—such as major highways and urban centers—are poorly resolved by a 12
km grid and the benefits of NOx control near motor vehicle emission sources is likely to be
overestimated. This overestimation will be carried over into the population exposure modeling
and amplified due to the large population density in the poorly resolved urban areas.
Our Response:
The use of a 12km grid resolution is appropriate when evaluating regional- to national-
scale emission reduction impacts (http://www.epa.gov/ttn/scram/guidance/guide/final-03-pm-rh-
guidance.pdf) as was done in the Tier 3 analysis. Use of a 12 km grid resolution is also
consistent with prior agency rulemakings, such as the NOx SIP Call rulemaking. (63 FR 57356,
(October 27, 1998).(See also, Michigan v. EPA, 663 F.3d 663 (D.C. Cir., 2000). The credible
model performance reported in the operation model evaluation
(http://www.epa.gov/otaq/documents/tier3/454rl3006.pdf) confirms that 12km resolution
modeling provides a good representation of the physical and chemical processes important to
simulating ozone concentrations. Recent work analyzing the accuracy of 12km resolution 2007
CMAQ simulations compared to equivalent 4km resolution simulations showed no systematic
improvement in model performance at the finer 4km resolution compared to the 12km resolution
(Dolwick et al, 2013). An additional study analyzing ozone response to NOx emissions
reductions at 12km and 4km resolutions (Simon et al, 2013), showed similar regional patterns in
ozone changes at the two resolutions and also did not show any systematic difference in ozone
response to NOx cuts at 12 vs. 4 km resolutions when looking across all sites in the Northeastern
US over the entire 2007 ozone season.
References:
Dolwick, P., Baker, K., Kelly, I, Misenis, C., Phillips, S., Possiel, N., Simon, H., Timin, B. Comparison of CMAQ
model performance over the Northeast United States as a function of grid resolution (12km vs 4km) for a 2007
annual model simulation, Community Modeling and Analysis Annual Conference, Chapel Hill, NC, October 2013:
http://cmascenter.org/conference/2013/agenda.cfm
Simon, H., Baker, K., Possiel, N., Dolwick, P., Timin, B. Model Resolution and Ozone Sensitivity to NOx
Emissions Changes in the Northeastern US, Community Modeling and Analysis Annual Conference, Chapel Hill,
NC, October 2013: http://cmascenter.org/conference/2013/agenda.cfm
What Commenters Said:
Commenter: American Fuel & Petrochemical Manufacturers (AFPM), American Petroleum
Institute (API) and Marathon Petroleum Company LP (MFC)
3-65
-------
Chapter 3: Emissions and Air Quality Emissions Impacts
API and AFPM submitted a report from Sierra Research in support for their comments.23 The
Sierra report included the following comment on speciation profiles, see Section 6.3.2 of
Sierra report, which we've addressed here.
Another area of importance are the speciation profiles for headspace vapors of EO, E10, and
El5 that were used by EPA in conjunction with the air quality modeling analysis. EPA
developed new speciation profiles for the Tier 3 analysis101'1 2 using headspace measurements
from a subset of EPAct test program fuels. No attempt was made, however, to determine if
these research fuels are representative of commercial fuels.
The EO, E10, and El 5 headspace profiles were each created by averaging headspace
measurements from two of the EPAct fuels. One fuel in each pair was low in aromatics (15
vol.% nominal) and the other high (35% vol. nominal). The E10 and E15 headspace profiles
were also used to create the E15 evaporative emissions profile by taking the ratio of E15/E10 for
each hydrocarbon species and applying the ratio to the existing E10 profile.
Unfortunately, there are differences in the detailed composition of the EPAct fuels that cause
speciation differences that are not related to ethanol content. These differences could create a
bias when comparing air quality impacts of EO, E10, and E15. The distribution of individual
aromatics differs between fuels in an unexpected manner. For instance, the El5 fuel has
higher benzene and toluene headspace concentrations compared to the E10 fuel, but lower
concentrations of other aromatics like propyl- benzene. It would be preferable to generate the
profiles from composited commercial fuels or to perform a simple correction based on ethanol
content.
Our Response:
This rule does not compare air quality impacts of different ethanol blends so this
comment is beyond the scope of this rule.
3.2.4. Tier 3 Proposal Analysis Overstates the Air Quality Benefits of Reducing Vehicular
NOx Emissions in Urban Areas
What Commenters Said:
Commenter: American Fuel & Petrochemical Manufacturers (AFPM), American Petroleum
Institute (API) and Marathon Petroleum Company LP (MFC)
In the DRIA, EPA did not provide information on how the Tier 3 rule and changes in ozone
levels may result in a disproportionate impact on populations in urban centers. In their recent
report on the Benefits and Costs of the Clean Air Act (EPA, 2011), EPA noted that in many
urban areas, the ozone levels were 15 to 20 ppb higher with the Clean Air Act Amendments
23 Sierra Research, June 20137. Assessment of the Emission Benefits of U.S. EPA's Proposed Tier 3 Motor Vehicle
Emissions and Fuel Standards. Prepared for: American Petroleum Institute.
3-66
-------
Tier 3 Summary and Analysis of Comments
(CAAA) than without. EPA attributed this to NOx scavenging whereby nitrogen oxides, while
participating as an ozone precursor, can also serve to scavenge or reduce ozone, particularly
during the peak ozone season and in urban centers where ozone levels might otherwise be quite
high. Thus, the effect of the CAAA controls was to suppress NOx scavenging in the city centers,
where "dis-benefits" of the CAAA are the largest. A similar phenomenon may occur when the
Tier 3 rule is implemented. Since OMB's guidelines call for regulatory agencies to assess
distributional effects, EPA's DRIA is deficient in this area (OMB Circular A-4, page 14).
Our Response:
The CMAQ model used to project changes in ozone from the Tier 3 standards accounts
for interactions between photochemistry, background concentrations of ozone, VOC and NOx,
local emissions and meteorology. As described in Chapter 7.2.4.1.1 of the RIA, there is one
county in 2018 that is projected to have an increase in modeled ozone design value concentration
(Cuyahoga County, OH, where Cleveland is located). When NOx levels are relatively high and
VOC levels relatively low, NOx forms inorganic nitrates (i.e., particles) but relatively little
ozone. In addition, NOx can react directly with ozone resulting in suppressed ozone
concentrations near NOx emissions sources. Such conditions are called "NOx-saturated." Under
these conditions, VOC reductions are effective in reducing ozone, but NOx reductions can
actually increase local ozone under certain circumstances. We believe that this is the case in
Cuyahoga County in 2018. In 2030, when the fleet would be composed of vehicles meeting the
new standards and the NOx and VOC emissions reductions are larger, this ozone disbenefit is
eliminated, and the design values for all the modeled counties are decreasing.
Our analysis is consistent with OMB and internal guidelines for conducting national-level
regulatory impact analyses.
3.2.5. Tier 3 Rule will Increase PM2.5 Levels in 10 Nonattainment Counties
What Commenters Said:
Commenter: Chevron
By 2030, the Tier 3 rulemaking will actually increase PM2.5 levels in 10 nonattainment counties.
This 'disbenefit' effect is due to reductions in reactive precursor species (like NOx) which can
often result in increases in the secondary pollutants ozone and PM.
Our Response:
The increases in PM levels that were reported in the proposed rule were due to a series of
conservative assumptions and uncertainties related to fuel parameters in 2017, and also an
emissions processing issue which erroneously increased direct PM emissions in about one third
of modeled counties, see Chapters 7.2.4.2.3 and 7.1.3.2.2 of the DRIA for the proposed rule.
EPA noted that we did not believe these increases would actually occur. This was corrected for
the final rule and as noted in Chapter 7.2.4.2.3 of the RIA, we do not expect that any increases in
PM2.5 will occur.
3-67
-------
Chapter 3: Emissions and Air Quality Emissions Impacts
3.2.6. Mobile Source Air Toxics Reductions are Based on Antiquated Data
What Commenters Said:
Commenter: American Fuel & Petrochemical Manufacturers (AFPM), American Petroleum
Institute (API) and Marathon Petroleum Company LP (MFC); Truck and
Engine Manufacturers Association (EMA):
Mobile Source Air Toxics: EPA claims reductions in Mobile Source Air Toxics (MSAT) that are
based on the use of antiquated databases. In the case of air toxics, EPA makes reference to a
2005 National Air Toxic Assessment (2005 NATA) database. Since EPA issued the Tier 2
vehicle and fuel standards and the MSAT2 regulations subsequent to releasing the 2005 NATA,
deriving the proposed Tier 3 mobile source air toxics benefits from the 2005 NATA database
yields estimates that are highly inflated, not real-world, and seriously suspect. EPA needs to
develop a current toxics emissions database before it can make any claims about MSAT benefits.
The second error of note is EPA's allocation of the percentage of outdoor air toxics that can be
ascribed to mobile sources. EPA cites the 2005 NATA in claiming that mobile sources account
for "43 percent of outdoor toxic emissions and over 50 percent of the cancer risk and non-cancer
hazard associated with primary emissions." (See 78 Fed. Reg. at 29837.) As in the case of EPA's
erroneous premise pertaining to UFPs, EPA's assertion regarding the percentage of air toxics
associated with mobile sources is premised on outdated and overstated data.
More specifically, in attempting to justify a contemporary rulemaking using 2005 data, EPA is,
in effect, ignoring the remarkable advancements that have been made in reducing motor vehicle
emissions, especially heavy-duty vehicle emissions, subsequent to 2005. As evidenced by the
findings of Phases 1 and 2 of the Advanced Collaborative Emissions Study (ACES), of which
EPA is a co-funder, emissions of criteria pollutants and air toxics have been reduced across the
board by up to 99% from new technology diesel engines ("NTDEs"). (See, e.g., Presentation of
Imad Khalek (SwRI) to the CRC Real World Emissions Workshop, April 2013.)
Moreover, of the more than 14 million on-road diesel vehicles in operation today, more than
40% are NTDEs, and more than 60% of the miles driven by diesel vehicles are driven by
NTDEs. (See, e.g., Presentation of Dan Greenbaum at HEI Annual Conference, April 2013.)
That remarkable rate of penetration of NTDEs since 2007 (the date of introduction for most
NTDEs) has had a very significant effect on the aggregate emissions of air toxics that can be
ascribed to mobile sources. EPA should account for these developments as they significantly
undercut (as the Agency should have hoped) the premise for the current rulemaking.
Our Response:
Contrary to the commenters' assertions, the reductions in mobile source air toxics
estimated to result from the Tier 3 standards are not derived from the 2005 NATA database.
They were modeled for this rule using the MOVES model, taking into account impacts of Tier 2
vehicle and fuel standards, MSAT2 regulations, and all other mobile source programs currently
3-68
-------
Tier 3 Summary and Analysis of Comments
in place. See Chapter 7.1.2 of the draft RIA for a list of rules included in the inventory analysis
for the proposed rule and Memorandum to Docket: Updates to MOVES for the Tier 3 NPRM
Analysis (March 11, 2013, Docket EPA-HQ-OAR-2011-0135) for information on the air toxics
updates included in MOVES for the proposed rule. Furthermore, the version of the MOVES
model used in the Tier 3 analysis relies on emissions data from the ACES program to model
toxic emissions from 2007 and later diesel engines.
3.2.7. Mobile Source Air Toxics Tables are Confusing
What Commenters Said:
Commenter: American Fuel & Petrochemical Manufacturers (AFPM), American Petroleum
Institute (API) and Marathon Petroleum Company LP (MFC)
A table in the Preamble of the proposed Tier 3 rule adds further doubt to EPA's toxics
conclusions. Table 111-12, titled: "Percent of Total Population Experiencing Changes in Annual
Ambient concentrations of Toxic Pollutants in 2030...." presents information that is confusing
and inconsistent. Table III-l claims a 36% reduction in benzene on road inventories by 2030 but
the Preamble text associated with table III-12 claims that over 80% of the population will see a
decrease of at least 2.5% and in fact the 25-50% reduction value for Benzene shows that no
percent of the population will see this level of reduction. There are similar problems for all of the
air toxics for which EPA is claiming emission reduction benefits.
Our Response:
Table III-l presents reductions in onroad inventories and Table III-12 presents changes in
ambient concentrations of pollutants. Table 111-12 is presenting the percentage of the population
that is projected to experience various percent changes in ambient concentrations of each of the
air toxics due to the Tier 3 standards. Adding the rows in the benzene column that are associated
with a percent change of at least 2.5 % gives an answer of 81%.
3.2.8. EPA Should Use a Well-to-wheels, Lifecycle Basis for Gasoline and Ethanol Air
Quality Impacts
What Commenters Said:
Commenter: American Fuel & Petrochemical Manufacturers (AFPM), American Petroleum
Institute (API) and Marathon Petroleum Company LP (MFC)
We recommend that EPA uses a well-to-wheels, lifecycle basis for gasoline and ethanol air
quality impacts. EPA focused only on the vehicle/fuel system (use emissions) and ignored the
fuel production emissions. According to the National Academy of Sciences (NAS) report
("Renewable Fuel Standard: Potential Economic and Environmental Effects of U.S. Biofuel
3-69
-------
Chapter 3: Emissions and Air Quality Emissions Impacts
Policy", October 2011), ethanol has higher overall emissions than gasoline (see graphs below).
This is a result of significantly higher production emissions for ethanol arising from production,
which includes agriculture and the biorefmery. Note the relatively small differences between
gasoline and ethanol in tailpipe emissions (use). Figure 11 [EPA-HQ-OAR-2011-0135-4276-A2]
summarizes the NAS report findings.
As the volume of ethanol increases significantly in EPA's baseline between 2017 and 2025, EPA
does not make adjustments to the air quality impact of ethanol emissions. As a result the air
quality benefits of the proposed Tier 3 rule are overstated.
Our Response:
EPA did use a well-to-wheels approach, and accounted for impacts of increased use of
ethanol on emissions associated with fuel production and distribution. Details of the approach
can be found in a memorandum to the docket (Development of Air Quality Reference Case
Upstream and Portable Fuel Container Inventories for Tier 3 Proposal, Memorandum from Rich
Cook, December 12, 2012). The commenter also recommends EPA use AEO 2013. EPA is
using AEO 2013 for its final rule analysis. EPA disagrees with the assertion that it has
overstated the Tier 3 benefits.
3.2.9. State of Hawaii Should be Exempted from the Standards
What Commenter Said:
Commenter: Chevron:
Additionally, Tier 3 requirements are not justified by expected air quality improvements in
certain unique climates like that of Hawaii. We propose that, because the Tier 3 requirements are
costly and are highly unlikely to have a beneficial impact to the air quality, the state of Hawaii
should be exempted from the program.
Our Response:
The Tier 3 rule will reduce emissions of NOx, SOx, PM and VOCs in Hawaii. These
emission reductions are expected to reduce ambient concentrations of ozone, PM, NOx and SOx
which will positively impact human health as well as visibility, deposition and ozone-related
harm to vegetation. Air quality impacts were not able to be modeled in HI due to the size of the
air quality modeling domain, not because we thought that air quality improvements were
unlikely.
3.3. Health and Environmental Effects of Criteria and Air Toxics Pollutants
3.3.1. Tailpipe Reductions Will be Offset by Increases in Emissions from Refineries
What Commenters Said:
3-70
-------
Tier 3 Summary and Analysis of Comments
The commenter is concerned about increases in emissions near refineries, particularly in
environmental justice communities. This commenter is concerned that tailpipe reductions will
be offset by emissions increases from refineries. The commenter is also specifically concerned
that the sulfur removed from gasoline will be emitted to the air as SO2.
Commenter: Concerned Citizens Around Murphy
Our Response:
Overall, the Tier 3 rule will deliver significant emissions and air quality benefits to
communities across the U.S., including environmental justice communities, especially as a result
of the near-roadway emissions reductions that will begin occurring in 2017.
While Tier 3 standards will result in very large emission reductions from both new and
existing vehicles, the additional gasoline hydrotreating will also cause a relatively small increase
in emissions at the refinery. EPA analyzed these impacts in detail using our refinery-by-refinery
analysis and concluded that even in the worst case, the emissions would be sufficient to cause
only a small number of refineries to trigger the need for new permits (see Section V.K of the
preamble and Chapter 5.4 of the RIA). Concerns that the sulfur removed from the gasoline might
increase emissions of SO2 are unfounded, as the process of desulfurizing the gasoline results in
elemental (solid) sulfur being removed, which is then sold as a byproduct by the refinery for
other purposes.
In all cases both minor and major NSR permit applications are subject to public
comment. For any federal NSR permit applications submitted to EPA Regional Offices, we are
committed to assisting members of the local EJ communities in understanding the applications
and our proposed permits, in offering comment, and participating in our decisions, consistent
with our EJ2014 implementation plan, Considering Environmental Justice in Permitting.
See http://www.epa.gov/environmentaljustice/resources/policv/plan-ej -2014/plan-ej-
permitting-201 l-09.pdf
3.3.2. Benefits of this Rule that are Based on Reduction in UFP are Erroneous
What Commenter Said:
Commenter: Truck and Engine Manufacturers Association (EMA)
As with most of its significant rulemakings impacting mobile sources, the Preamble to the Tier 3
Rule provides an overview of EPA's perspective on the potential health impacts of air pollution,
including the air pollution constituents that are emitted from motor vehicles. (78 Fed. Reg. at
29829-29850.) While a detailed response to all of EPA's "overview" is beyond the scope of
these comments, there are two clear errors in EPA's health impacts assessment that should be
corrected and accounted for in the Agency's final rulemaking documents.
3-71
-------
Chapter 3: Emissions and Air Quality Emissions Impacts
The first error relates to EPA's assertions regarding the purportedly unique health impacts that
can be ascribed to ultrafine particles ("UFPs"). EPA cites its own Integrated Science Assessment
("ISA") published in 2009 as support for its assertions. (78 Fed. Reg. at 29831.) However, and
contrary to EPA's claims, more recent work published by the Health Effects Institute ("HEI") in
January 2013 - HEI Perspectives 3, "Understanding the Health Effects of Ambient Ultrafine
Particles" - has shown that the available data do not support a conclusion that UFPs play any
unique or significant role in engendering potential adverse health effects. In that regard, the
overall conclusions of HEI's expert panel bear repeating:
While selected [animal exposure] studies show evidence for UFP effects, the current evidence,
when considered together, is not sufficiently strong to conclude that short-term exposures to
UFPs have effects that are dramatically different from those of larger particles.... There are no
long-term animal exposure studies of UFP health effects.
One explanation that must be considered for the [epidemiologic] results to date is weakness in
the true underlying relationship between UFP exposures and adverse effects - that the null
hypothesis being tested by these studies is true.
The available observational study designs have also not been able to clearly demonstrate whether
UFPs have effects independent of those for related pollutants... .No epidemiologic studies of
long-term exposures to ambient UFPs have been conducted.
Toxicological studies in animals, controlled human exposure studies, and epidemiologic studies
to date have not provided consistent findings on the effects of exposures to ambient levels of
UFPs, particularly in human populations. The current evidence does not support a conclusion
that exposures to UFPs alone can account in substantial ways for the adverse effects that have
been associated with other ambient pollutants such as PM2.5. (HEI Perspectives 3, p. 5.)
Thus, to the extent that the proposed Tier 3 Rule is premised on any projected amelioration of the
health effects ascribed to ambient exposures to UFP, that purported justification for the
rulemaking is unfounded and in error, as evidenced by the scientific findings developed
subsequent to EPA's 2009 ISA. As a result, EPA should correct that error in the Agency's final
rulemaking record.
Our Response:
EMA comments that any projected benefits of this rule that are based on reduction in
UFP are erroneous. The quantified PM-related benefits associated with this rule are based on
reduction in emissions that affect ambient concentrations of PM2.5 mass, not UFP. Reduction in
UFP concentrations is not a significant premise for the Tier 3 standards. However, we
qualitatively summarize EPA's most current assessment of UFP health effects, reported in the
2009 ISA, as part of our broader summary of evidence related to PM health effects in Section
II.B.2 of the preamble to the final rule.
3-72
-------
Tier 3 Summary and Analysis of Comments
EPA will consider the HEI publication as part of its review of the larger body of evidence
for PM health effects when developing the next PM ISA during the next review of the PM
NAAQS. The HEI study does not substantively change our assessment of the benefits associated
with the Tier 3 rule.
3-73
-------
Tier 3 Summary and Analysis of Comments
4. Proposed Vehicle Emissions Program
4.1. LD Exhaust Standards
4.1.1. General
4.1.1.1. Comments Generally Supportive of the Proposed Standards and Program
What Commenters Said:
Consumers Union
New car buyers will also benefit. Starting in 2017, new cars will have tighter limits on tailpipe
emissions, including carbon monoxide and benzene, which can linger in garages and even
attached residential living space (4). The proposed rule also offers automakers an incentive to go
beyond the minimum 8-year/ 80,000-mile warranty currently required for emissions control
systems, and extend it to 15-years/150,000 miles for new vehicles. This move could improve
reliability and lower costs to maintain emissions control systems.
Pennsylvania Department of Environmental Protection (DEP)
The proposed vehicle standards would reduce tailpipe and evaporative emissions from passenger
cars, light-duty trucks, medium-duty passenger vehicles, and some heavy-duty vehicles. These
proposed vehicle standards are intended to harmonize with California's Low Emission Vehicle
program, creating a federal vehicle emissions program that would allow automakers to sell the
same vehicles in all 50 states.
DEP supports EPA's proposed Tier 3 emission standards for vehicles. The proposed standards
would be met by vehicle manufacturers beginning in model year 2017 and phasing in through
later model years. Vehicles sold in states such as Pennsylvania that have adopted light-duty
vehicle emission standards promulgated by California Air Resources Board (CARB) will be
meeting the same standards for model year 2015 and later vehicles. States that adopted CARB
light-duty vehicle standards comprise nearly 50 percent of the national market for these vehicles.
Technological developments in the automotive field that have already occurred or that are in the
developmental pipeline as a result of CARB and EPA efforts will allow automobile
manufacturers to effectively meet these more stringent standards for NOx, VOC, and PM2.5 by
the year 2021.
Southern California Association of Governments (SCAG)
SCAG supports efforts to minimize emissions from vehicles, and has included a regional clean
freight corridor system in the recently adopted 2012-2035 Regional Transportation
Plan/Sustainable Communities Strategy. SCAG has also been active with our partner agencies in
Plug-in Electric Vehicle readiness planning.
Our Response:
4-1
-------
Chapter 4: Proposed Vehicle Emissions Program
EPA acknowledges the general comments in support of the proposed standards.
4.1.1.2. Harmonization with CA LEVIII Standards
Alliance of Automobile Manufacturers (Alliance) and Association of Global Automakers
(Global)
And I also want to thank the California Air Resources Board for its role in coordinating with
EPA to help ensure harmonization of the LEV III and Tier 3 regulations.
So why do auto makers support this proposal? Well, Tier 3 brings a harmonized approach, as
you've heard today. It builds upon the successes we've had in the national greenhouse gas and
fuel economy programs. It stays true to the simple principle of providing the cleanest vehicles to
everyone nationally. And it provides the fuels that we need.
American Honda Motor Co., Inc.
Much has been said during the past few years about the importance of regulatory harmonization.
With near unanimity, the auto industry supported the 2012-2016 and 2017-2025 vehicle
greenhouse gas standards, in large part because it represented a national, unified solution to a
complex set of competing federal and state standards. Honda produces a wide range of vehicles
for an equally wide demographic of buyers that requires multidimensional considerations about
what to produce, how many to produce, the cost to produce our products, and where those
models will most likely be purchased. Having to deal with competing federal and state regulatory
requirements only complicates those decisions further.
As the EPA considers setting Tier 3 vehicle emissions and fuel requirements, we urge the agency
to harmonize - to the greatest extent possible - with California's LEV III regulations, including
certification fuel characteristics and market fuel characteristics. Doing so would enable one fleet
of vehicles to meet all U.S. regulations, substantially easing both the regulatory burden and cost
of complying - letting us provide our customers greater affordability - and allowing us to design
our vehicles to maintain consistent performance and emissions durability across the nation.
One of the more important regulatory achievements of this administration has been its actions to
harmonize the federal fuel economy and vehicle greenhouse gas regulations with those of
California. A harmonized set of tailpipe emissions and fuel regulations would complement a key
objective laid out by President Obama in the May 21, 2010 White House memorandum. It states:
"The national program should seek to produce joint Federal standards that are harmonized with
applicable State standards, with the goal of ensuring that automobile manufacturers will be able
to build a single, light-duty national fleet." We couldn't agree more. Honda urges the agency to
finalize a set of standards that offers the greatest degree of harmonization with state standards, as
well as to ensure that the fuel operated in these vehicles will accommodate the advanced
emissions control systems needed to meet the proposed standards.
4-2
-------
Tier 3 Summary and Analysis of Comments
Honda urges the Agency to finalize a set of standards that offers the greatest degree of
harmonization with State standards, as well as to ensure that the fuel operated in these vehicles
will accommodate the advanced emissions control systems needed to meet the proposed
standards.
BMW of North America, LLC
BMW commends EPA for their collaboration with ARB and with the automakers toward a single
national program for criteria pollutants in developing the complex regulation proposal which will
substantially impact emissions for vehicles beyond a decade to 2025.
In addition to improving air quality across the country, a single national program of harmonized
standards would allow for wise financial and resource investments by the auto industry, as well
as increased energy security for the nation.
Through a variety of ground breaking engine technologies, BMW has substantially decreased the
criteria emissions of its fleet since the adoption of Tier 2 Motor Vehicle Emission Standards.
BMW has maintained a leading role in deploying innovative engineering solutions to meet the
challenge of stricter emission standards: BMW engines have won numerous engine awards
among others for High Precision Direct Injection, twin-scroll turbocharger, and BMW
Valvetronic fully variable valve control.
To that end, compliance flexibility and adequate lead time are two key factors which can create
the required boundary conditions for development of innovative and creative engineering
solutions and achievement of product maturity aimed to meet ambitious emissions standards.
50-State Certification and Harmonization Tier 3 / LEV III: BMW strongly supports one of the
primary goals in the Tier 3 regulation - harmonization of the federal and the California criteria
emission programs. Any continuation of two different standards needs to allow the
manufacturers the possibility to choose between the standards and mutual recognition.
Harmonization is still needed for ARB and U.S. EPA in the areas of vehicle standards, test
procedures, and certification processes in order to establish a common set of vehicle criteria
emission standards nationwide.
California Air Resources Board (CARB)
The California Air Resources Board (CARB) is fully supportive of the proposed Tier 3 federal
test procedure (FTP) and supplemental federal test procedure (SFTP) requirements. With a few
exceptions, the proposed requirements mirror the Low- Emission Vehicle (LEV III) requirements
effectively addressing manufacturers' concerns regarding compliance with both national and
California emission programs. In order to further harmonize program requirements of both
agencies, CARB intends to align the LEV III rule with a number of the Tier 3 FTP and SFTP
requirements after Tier 3 is finalized as noted below and offers the following comments on the
proposal.
4-3
-------
Chapter 4: Proposed Vehicle Emissions Program
CARB shares U.S. EPA's goal of reducing vehicle emissions to improve air quality. Despite
significant progress in reducing smog-forming and particulate matter emissions from the
passenger vehicle fleet, California needs further reductions in order to meet state and federal
ambient air quality standards. To help achieve these standards, CARB adopted the Low-
Emission Vehicle (LEV III) regulations last year as part of our Advanced Clean Cars program.
The Advanced Clean Cars program combines the control of smog-causing pollutants and
greenhouse gas emissions into a single coordinated package of requirements for model years
2015 through 2025 and assures the development of environmentally superior cars that will
continue to deliver the performance, utility, and safety vehicle owners have come to expect.
CARB is mindful of the cooperative effort between CARB and U.S. EPA in the development of
the new vehicle exhaust: emission standards for California's LEV III program and the federal
Tier 3 program. It is our intent to evaluate the Tier 3 rule once it is finalized to determine where
it is appropriate to further align the LEV III regulation with the federal rule, without sacrificing
the stringency and emission benefits of the LEV III program. Since any delay in finalizing Tier 3
would potentially impact the implementation date of this important program, CARB urges U.S.
EPA to proceed, expeditiously to finalize the program before the end of this year. We also look
forward to continue working with U.S. EPA in an attempt to resolve any remaining differences
between the two programs after the Tier 3 regulations are finalized.
Our primary goal continues to be improving air quality in California. To this end, California has
a separate and unique Zero-Emission Vehicle Program, which we will, continue to pursue in
order to ensure the prompt and successful deployment of advanced technology zero-emission
vehicles. Accordingly, any determination of the extent to which it would be appropriate for
California to incorporate elements of the Tier 3 program into our light-duty vehicle program
would need to take into consideration our ability to sustain progress towards California's long
term plans for transforming the vehicle fleet for reduction of criteria pollutants. In addition, any
modifications to the LEV III program must be structured to assure that the emission reductions
provided by the California program will be maintained.
Chrysler Group LLC
Chrysler strongly supports harmonization of the federal Tier 3 regulations with California LEV
III regulations. As such, we urge EPA to make the necessary adjustments as suggested below and
as suggested in the Alliance's comments to truly achieve the goal of One National Criteria
Emissions Program that includes one certification fuel with harmonized test procedures and
certification processes.
Cummins Inc.
In this rulemaking, EPA proposes to coordinate its Tier 3 program for reduction of tailpipe and
evaporative emissions closely with California's LEV III standards and with EPA's and
California's greenhouse gas requirements for light-duty vehicles. Cummins strongly supports the
goal of harmonization with California and other requirements. Consistency between EPA and
California programs would allow manufacturers to design products for 50 states and help avoid
the additional costs of parallel design, development, calibration, and manufacturing.
4-4
-------
Tier 3 Summary and Analysis of Comments
Ford Motor Company (Ford)
Harmonization: Consistent with the approach taken in the One National Program (ONP) for
Corporate Average Fuel Economy (CAFE) and Greenhouse Gas (GHG), the proposed Tier 3 rule
is closely aligned with California's Low Emission Vehicle (LEV III) emissions program. This
alignment will allow automakers to design and build a single model that can be tested once and
sold nationally. We appreciate the significant efforts made by both EPA and California Air
Resource Board (CARB) to harmonize the Tier 3 and LEV III requirements, and we encourage
the Agencies to continue to work with Industry to minimize any remaining differences in the two
programs, while retaining the interim provisions finalized in LEV III.
I want to thank EPA and the California Air Resources Board for their efforts to harmonize
regulations to improve air quality. Ford supports this effort, and will continue working to
improve air quality while harmonizing standards.
Tier 3, working in concert with LEV III, helps provide clean vehicles to all 50 States. It allows
auto makers to design and build a single model rather than two versions meaning different
requirements, test and certify a national vehicle one time. And importantly, it provides a step
towards the fuels needed to meet these standards.
We appreciate the significant efforts made by EPA, both EPA and CARB, to harmonize Tier 3
and LEV III rules, and we encourage the Agency to continue to work with industry to minimize
remaining differences in the two programs.
General Motors LLC (GM)
GM appreciates the efforts of EPA and the California Air Resources Board (CARB) staffs
working together and with industry toward a harmonized national program for criteria emissions,
with the goal of enabling manufacturers to design, develop, certify, build, and sell the same clean
vehicles throughout all fifty states. This will allow our engineers to focus on a single emissions
control system for criteria emissions, and builds upon the single national program for greenhouse
gas (GHG) emissions and fuel economy (FE). A national approach is especially important given
the scope of the challenges we are facing today to meet increasingly stringent GHG/FE standards
and other regulatory requirements while simultaneously needing to meet all customer demands in
an intensely competitive market.
One of the biggest challenges in integrating the California LEV III program into a national
program occurs during the first couple of years of the program. Transit!oning from a program
covering about a third of nationwide volume to a national program covering 100% of nationwide
volume presents challenges in ramping up both the exhaust and evaporative requirements to
national levels.
These approaches will move us to a national program as soon as practicable and facilitate
compliance based on a single 50-state fleet rather than separate compliance determinations for
California/177 States versus the remaining U.S. states. GM supports such a 50-state compliance
approach, not just for the FTP NMOG+NOx fleet average and the evaporative phase-in, but for
4-5
-------
Chapter 4: Proposed Vehicle Emissions Program
other fleet averages and phase-ins as well including the SFTP fleet average, evaporative fleet
average and 3 mg/mi PM phase-in.
That brings me to the key reason we are here today, working closely with the EPA and the other
stakeholders to achieve a harmonized and appropriate structure Tier 3 program. The
establishment of a single national standard for fuel economy and CO2 regulation was quite
simply a much needed breakthrough that will enable more timely and efficient introduction of
technologies customers, manufacturers, and regulators want to succeed.
We know that EPA and California are committed to further reducing tailpipe criteria emissions,
and in order to achieve the level of reduction proposed, our vehicles will be so clean there will be
no reason to have competing regulatory requirements.
Our precious engineering resources are already stretched thin, and requiring them to design,
develop, and certify two of everything I think we can all agree is counterproductive. We need to
be able to focus on developing one of everything, which we can then sell throughout the U.S. to
provide everyone here the cleanest vehicles in the world. A harmonized Tier 3 program properly
structured can achieve this.
Hyundai Motor Group
As you know, this harmonization is very important for us and the auto industry as it allows us to
design to one national standard. Additionally, we have found it very beneficial that EPA was
willing to provide many opportunities to meet with industry and other stakeholders to discuss the
provisions of the rule while they were under development.
As I mentioned already, Hyundai Motor Group is pleased with EPA's efforts to align the Tier 3
rule with CARB's LEV III program. It's challenging to meet different requirements for various
regulatory agencies, particularly when it comes to testing. Having one consistent national
standard will increase laboratory throughput in addition to design and development cost savings.
We realize that there are challenges in adopting the same provisions as CARB, but we hope that
EPA and CARB will continue to work together to try to resolve as many of the remaining
differences as possible.
Manufacturers of Emission Controls Association (MECA)
MECA applauds EPA for developing a Tier 3 proposal that will establish a national set of
exhaust and evaporative emission standards for light-duty and medium-duty vehicles by largely
harmonizing their proposal with California's LEV III requirements.
Johnson Matthey
Furthermore, California has already set in motion its own tightened emission limits known as
LEV III, which are very similar to the Tier 3 proposal. This means our industry already has no
choice but to implement technologies for LEV III that will clearly also meet the Tier 3 proposed
4-6
-------
Tier 3 Summary and Analysis of Comments
limits. Thus, the adoption of the Tier 3 proposals would bring efficiencies to the industry by
having one common set of emissions to achieve for the U.S. market.
Mid-Atlantic Regional Air Management Association Inc. (MARAMA)
By harmonizing vehicle emission standards across the country, Tier 3 would facilitate
compliance by automobile manufacturers. The automobile manufacturers support Tier 3 because
it enables them to harness economies of scale by deploying advanced emission control
technologies in all new vehicles sold nationwide.
Mitsubishi Motors R&D America, Inc (MRDA)
We believe that harmonization with the California LEV 111 Program (LEV 111) must be EPA's top
priority when finalizing this rulemaking. As EPA notes numerous times in the proposed
rulemaking, it is extremely important to automotive manufacturers that a single fleet of vehicles
can be designed and produced for all 50 states, and be in compliance with the Tier 3, LEV III,
and Federal Greenhouse (GHG) and Corporate Average Fuel Economy (CAFE) requirements.
Although EPA indicates that harmonization with LEV III is a guiding principle in developing
these proposed regulations, there are still significant gaps that remain in achieving this necessary
goal. A harmonized program would allow manufacturers to develop cost-effective processes and
to deploy advanced technology vehicles in meeting these standards. Smaller companies, like
Mitsubishi Motors, have less financial resources and fewer vehicle lines to implement major
changes in a cost-effective manner, which makes the proposed compliance flexibilities (phase-
ins, early credits, interim in-use standards, etc.) outlined in the NPRM extremely important.
Mitsubishi Motors supports the intent to harmonize the proposed Tier 3 program with
California's LEV III program.
Motor & Equipment Manufacturers Association (MEMA)
The EPA's proposal is mindful of other regulatory frameworks and compliance regimes. Efforts
to harmonize Tier 3 vehicle emission requirements nationwide and to choreograph the timing of
Tier 3 implementation with the final rules for greenhouse gas (GHG) emissions and corporate
average fuel economy (CAFE) standards for light-duty vehicles are critical for interconnected
and highly complex vehicle manufacturing supply chains. Also, EPA's integrated systems
approach to vehicles and fuels, combined with the alignment to other parallel regulations, is
practical and cost-effective. Companies and government entities alike will benefit from the
resulting outcomes, such as streamlined costs for R&D and production and reduced burden for
multiple, overlapping testing and compliance protocols. Since the timing of this rule is critical, it
is important that EPA not delay its completion. If delayed, the benefits of synchronizing the
timing of the GHG emissions, CAFE standards, and other programs will be lost and may
negatively impact the states in achieving their respective air quality goals.
MEMA urges EPA to fully consider the public comments, particularly from the vehicle
manufacturers. In addition, EPA must continue to collaborate with states, like California, and
other stakeholders to avoid divergent policy pathways and competing regulatory regimes.
4-7
-------
Chapter 4: Proposed Vehicle Emissions Program
MEMA recommends that the Agency promulgate a final rule by end of 2013 in order to match
with other regulatory requirements standards affecting MY 2017 vehicles.
National Automobile Dealers Association (NADA)
Harmonized Emissions Mandates, Lead Time, Durability and Other Technical Standards: Last
year, the California Air Resources Board (CARB) finalized its Low Emission Vehicle (LEV) III
regulations. In order to minimize compliance costs and maximize compliance flexibilities, EPA's
final Tier 3 regulations should encourage if not mandate that CARB harmonize with the federal
scheme. Among other things, this means that fleet average emissions compliance should be
based on a manufacturer's nationwide sales. (See also 4.1.5.8 re: nationwide compliance
demonstration.)
New York State Department of Environmental Conservation
New York has a long history of adopting California's motor vehicle emissions control programs
to achieve its air quality objectives. New York most recently adopted the California Advanced
Clean Cars emissions program, which included the LEV III standards, in 2012 to achieve and
maintain reductions of criteria and greenhouse gas pollutant emissions.
Tier 3 should be harmonized with LEV III as quickly as possible.
The Department strongly supports this effort to harmonize federal emissions standards with
California's technology forcing LEV III standards. In order to maximize the air quality benefits
of harmonization, as well as reduce the industry compliance burden, Tier 3 should reach LEV III
stringency as quickly as possible.
Northeast States for Coordinated Air Use Management (NESCAUM)
By harmonizing vehicle emission standards with those in the California program, Tier 3 would
facilitate compliance by automobile manufacturers, enabling them to harness economies of scale
by deploying advanced emission control technologies in all new vehicles sold nationwide.
Pennsylvania Department of Environmental Protection (DEP)
DEP also supports EPA's efforts to harmonize a national emission standard program with
CARB's emission standards. This harmonization will allow automobile manufacturers to design,
produce and test vehicles that meet the same standard by the year 2021. The automobile
manufacturers also appear to support this effort because they recognize that some cost-savings
benefits will result from producing vehicles meeting the same national standard. Clearly,
harmonized programs would be significantly beneficial for both the environment and consumers.
The EPA should harmonize the phase-in of light-duty vehicles in the Tier 3 program with
CARB's LEV III Program. While EPA's proposal would result in the same emission standards
for model year 2025, the phase-in schedule lags slightly behind the schedule finalized by
California. Given the issues with fleet turnover limiting emission benefits and the fact that
4-8
-------
Tier 3 Summary and Analysis of Comments
manufacturers would already have developed the technology for light-duty vehicles to meet the
schedule that California is setting, EPA should finalize an emission standard phase-in that is
harmonized nationally.
Nevertheless, DEP has concerns about... some elements of the proposed Tier 3 program that are
not harmonized with California's Low Emission Vehicle III Program. DEP requests that EPA
examine and address these issues before finalizing the Tier 3 rule.
Respiratory Health Association (RHA)
Harmonizing Federal standards with California vehicle emission standards will bring deserved
and equitable relief from local pollution beginning in 2017. Late is better than never, and the
proposed Tier 3 rules uses a smart systems approach that treats fuel and the engine as
intertwined. It is the most cost-effective way to tackle the problem posed by the emissions for
millions of vehicles, and it will improve the lives and health of people living with or who face
increased risk of lung disease. We need EPA to finalize the strongest Tier 3 standards possible.
Southern California Association of Governments (SCAG)
SCAG appreciates EPA's efforts to harmonize the national vehicle tailpipe and evaporative
emission and gasoline fuel sulfur content standards with California standards.
SCAG also applauds EPA's close coordination with California's programs for greenhouse gas
emissions from light-duty vehicles.
There appears to be some differences between the proposed Tier 3 standards and California's
LEV III Regulations and California gasoline sulfur content standards. While acknowledging the
complexities and nuances of these programs, harmonization of the proposed Tier 3 with the
California program is important from a business and manufacturer continuity perspective. We
believe that such harmonization may yield savings to California residents through the
efficiencies of scale in the production of vehicles that meet one national standard.
Truck and Engine Manufacturers Association (EMA)
In this rulemaking, EPA proposes to coordinate its Tier 3 program for reduction of tailpipe and
evaporative emissions closely with California's LEV III standards and with EPA's and
California's greenhouse gas ("GHG") requirements for light-duty vehicles. EMA fully supports
the goal of harmonization with California's standards and other emissions-related requirements.
Indeed, EMA has long supported harmonized, nationwide programs for regulating emissions
from heavy-duty engines and vehicles. For example, most recently, EMA supported - and
continues to support - nationwide harmonization of the OBD program for heavy-duty vehicles as
well as a single national program for reducing GHG emissions.
In the context of the Proposed Tier 3 Standards, EMA supports harmonization where such
standards are technologically feasible and reasonable in light of the myriad requirements facing
manufacturers now and in the near future. At the same time, EPA must recognize the inherent
4-9
-------
Chapter 4: Proposed Vehicle Emissions Program
need for adequate leadtime and so must provide sufficient time for the implementation of any
new, aligned standards as well as an adequate period of stability between standard changes to
provide manufacturers an opportunity to recoup the investment required to meet the new
standards. Consequently, EPA must continue to work with California to assure the coordinated
and timely implementation of changes in California's regulations where needed to achieve
desired 50-state harmonization.
- Assure full alignment between EPA and ARB standards where reasonable and feasible, while
meeting EPA's obligations to finalize regulations that are both technologically feasible and
provide the necessary and required leadtime and period of stability
As you know, EMA and its members have been long-time proponents of regulatory alignment in
the United States and beyond. We urge ARB and CARB to work together to assure that LEV III
and Tier 3 are aligned in both intent and practice.
U.S. Coalition for Advanced Diesel Cars
However, it is vitally important that the Tier 3 rule and its implementation schedule be
harmonized with the California LEV 3 program so vehicle manufacturers and their suppliers can
focus their innovation, investments and related efforts on delivering clean advanced technology
cars and trucks to a national market.
Volvo Car Group
VCG supports the effort of the EPA to work with the California Air Resources Board (CARB)
towards a possible harmonization of the Tier 3 and LEVIII program.
It is therefore extremely important that the EPA, CARB and the industry continue to work
together to achieve harmonization.
Our Response:
EPA received numerous comments favoring close harmonization between the existing
California LEV III program and Tier 3 from a wide range of commenters, including vehicle
manufacturers, suppliers, auto dealers, states, NGOs, and private citizens. The Tier 3 standards
we are finalizing are closely coordinated with the LEV III program to create a vehicle emissions
program that will allow automakers to sell the same vehicles in all 50 states.1 We have worked
closely with individual vehicle manufacturers and their trade associations, who have emphasized
the importance of a harmonized national program. Together, the Tier 3, 2017 LD GHG, and
LEV III standards will maximize reductions in criteria pollutants, GHGs, and air toxics from
motor vehicles while streamlining programs and enabling manufacturers to design a single
vehicle for nationwide sales, thus reducing their costs of compliance. In this way, the Tier 3
1 In December 2012 EPA approved a waiver of Clean Air Act preemption for the California Air Resources Board's
(CARB's) LEV III program with compliance beginning in 2015. Twelve states adopted the LEV III program under
Section 177 of the Clean Air Act. These states include Connecticut, Delaware, Maryland, Maine, Massachusetts,
New Jersey, New York, Oregon, Pennsylvania, Rhode Island, Washington, and Vermont.
4-10
-------
Tier 3 Summary and Analysis of Comments
program responds to the May 21, 2010 Presidential Memorandum that requested that EPA
develop a comprehensive approach toward regulating motor vehicles, including consideration of
non-GHG emissions standards.2
We worked closely with CARB and the vehicle manufacturers, the latter both
individually and through their trade associations, to align the two programs. The Tier 3 program
is identical to LEV III in most major respects for light-duty vehicles (and heavy-duty vehicles, as
described in the preamble and in this Summary and Analysis of Comments document). The
levels and timing of the declining fleet-average NMOG+NOx standards are identical to those in
LEV III. Also, the final Tier 3 emissions bins to which manufacturers will certify individual
vehicle models in order to comply with the fleet-average standards are also identical to those in
LEV III. Similarly, the light-duty Tier 3 FTP PM standards and percent phase-in match those for
LEV III through MY 2024.
Some commenters observed that there are a few light-duty Tier 3 and LEV III provisions
that are different. For example, the LEV III program and the Tier 3 program have different light-
duty PM requirements late in the program (i.e., after MY 2024 (see Section IV.A.3.b. of the
preamble)), and the two programs have different final NMOG+NOx standards for small volume
manufacturers (Section IV.G.l). We are finalizing a revised SFTP (US06) PM standard, and
CARB has indicated in their comments that it plans to take similar action in near future. Also,
LEV III currently does not include an evaporative emissions leak test and standard; CARB has
indicated in their comments that they plan to adopt such requirements after Tier 3 is finalized.
CARB also indicated in their comments that they intend to consider several additional actions to
further align several minor aspects of LEV III with the Tier 3 program once Tier 3 is finalized.
Beyond the provisions mentioned above, the differences between the programs are not
major and most will exist only in the transitional years of the Tier 3 program. These additional
differences result from the fact that the LEV III requirements begin slightly earlier and that a
limited phase-in of some provisions is necessary for a smooth transition to overall aligned
programs. These temporary differences include the process for how early compliance credits are
generated and used (e.g., preamble Section IV.A.T.a); how quickly manufacturers will need
move toward certifying all of their vehicle models to longer useful-life values (e.g., Section
IV.A.T.c) and on the new test fuel (e.g., Section IV.A.T.d); and transitional emissions bins to
facilitate the transition from Tier 2 to Tier 3 (Section IV A.7.n).
Because of these temporary differences between Tier 3 and LEV III, some commenters
encouraged EPA to consider lead time and flexibility provisions to facilitate compliance as they
relate to the period before the transitional provisions are harmonized. As discussed in the
preamble and elsewhere in this document, we believe the proposed program design, as revised in
light of such comments, address these concerns of the commenters.
4.1.1.3. Treating Vehicles and Fuels as a System
2 The Presidential Memorandum is found at: http://www.whitehouse.gov/the-press-office/presidential-
memorandum-regarding-fuel-efficiency-standards.
4-11
-------
Chapter 4: Proposed Vehicle Emissions Program
What Commenters Said:
Ford Motor Company (Ford)
Once again, I appreciate the opportunity to provide testimony today here. Ford will continue
working to improve air quality while harmonizing standards in order to meet these challenging
emissions control requirements. Treating fuels and vehicles as a system continues to be essential
to achieving our collective air improvement goals.
Motor & Equipment Manufacturers Association (MEMA)
Also, EPA's integrated systems approach to vehicles and fuels, combined with the alignment to
other parallel regulations, is practical and cost-effective. Companies and government entities
alike will benefit from the resulting outcomes, such as streamlined costs for R&D and production
and reduced burden for multiple, overlapping testing and compliance protocols. (See also
4.1.1.2).
Mercedes-Benz USA, LLC on behalf of Daimler AG
Treating the fuel as part of the system in total criteria emissions and Green House Gas (GHG)
reduction is an important acknowledgement by the EPA that improvements proposed by the Tier
3 rulemaking are possible when all contributing factors are examined and optimized.
Our Response:
We agree with the commenters that treating vehicles and fuels as an integrated system is
an important aspect of the Tier 3 program design. In fact, we believe that this "systems
approach" is critical for the program to achieve its expected benefits. Section IV.A.6 of the
preamble discusses in more detail the relationship between fuel sulfur levels and the ability of
vehicles to meet the Tier 3 exhaust emission standards.
4.1.1.4. Interaction of Tier 3 Program with GHG Program
What Commenters Said:
Alliance of Automobile Manufacturers (Alliance) and Association of Global Automakers
(Global)
In many cases, GHG emissions reductions directly compete with criteria emissions reductions in
the vehicle design. Recognizing the goal of aggressive, simultaneous reductions in GHGs and
criteria emissions, automakers have worked with agency staff to develop a program that allows a
technically feasible and cost-effective introduction of advanced technology vehicles. However,
notwithstanding the research and dialogue that has taken place, it is impossible to accurately
predict the pace of invention and innovation, the future fuel supply and pricing, or, most
importantly, consumer purchasing behavior. Consequently, we suggest that the mid-term review
4-12
-------
Tier 3 Summary and Analysis of Comments
for the 2017-25 GHG/CAFE rule consider how the implementation of the GHG/CAFE rule is
impacting automakers' ability to achieve the Tier 3 requirements and vice-versa.
Ford Motor Company (Ford)
We also encourage the agencies to evaluate the effects on new test fuel and new procedures on
fuel economy and greenhouse gas emissions. This evaluation should consider both near-term
impacts associated with the phase-in period of the new fuel and procedures, and the long-term
implications of 2017 through 2025, One National Program.
Looking forward, there are other significant elements regarding future fuels and ultimately
engine design which can have a large impact on greenhouse gas compliance. The unprecedented
increase in CAFE standards and the associated greenhouse gas requirements are driving a
dramatic reduction or downsizing in engine size and number of cylinders.
Pennsylvania Department of Environmental Protection (DEP)
The EPA should consider whether reducing sulfur levels in gasoline reduces or increases back
pressure in pre-Tier 3 vehicles' exhaust systems, and then determine whether fuel economy is
affected positively or negatively.
If using 10 ppm sulfur gasoline in vehicles that were previously using 30 ppm gasoline
eliminates sulfur deposition on a catalyst substrate, a small change in exhaust system
backpressure should result. Depending on whether there is an increase or decrease in
backpressure, the vehicle's fuel economy would either be enhanced or harmed. EPA should
consider this in this rulemaking.
Our Response:
As discussed throughout the final rule, EPA has designed the final Tier 3 program in full
recognition of the parallel implementation of the GHG standards in the same 2017-2025 time
frame. By aligning the implementation schedules for both sets of standards, we are facilitating
the ability of manufacturers to meet one of their stated goals, the ability to develop product plans
that simultaneously account for the technological challenges of both programs well into the
future. We considered the feasibility of the Tier 3 standards in the context of the established
GHG requirements.
As part of the 2017-2025 GHG standards rulemaking, EPA committed to a midterm
evaluation of the GHG standards for model years 2022-2025, in coordination with the California
Air Resources Board and the National Highway Traffic Safety Administration. EPA will be
making a determination as to whether the 2022-2025 GHG standards remain appropriate under
section 202(a) of the Clean Air Act. In making this determination, EPA will consider a number
of factors relevant to setting GHG standards, such as technology effectiveness, costs, leadtime,
feasibility, and other relevant factors (see 40 CFR 86.1818-12(h)). It is important to note that the
midterm evaluation is focused on the GHG standards for MY 2022-2025; there is not a midterm
evaluation of the Tier 3 standards.
4-13
-------
Chapter 4: Proposed Vehicle Emissions Program
Regarding the consideration of the potential impact on GHG emissions and fuel economy
of sulfur deposition on catalysts, we are aware of no information to indicate a relationship
between fuel economy/GHG emissions and sulfur deposition related to back pressure increases.
4.1.1.5. Form of the Standards
What Commenters Said:
Children's Environmental Health Network (CEHN)
The phase-in goals for NMOG+NOx should be calculated on a per-vehicle basis (as is proposed
for the Tier 3 PM FTP standards) as opposed to the proposed fleet-wide basis (which would
allow manufacturers to offset higher-emitting vehicles with extra-clean models). All vehicles
should be held to a per-vehicle emissions standard in order to best protect children's health.
Our Response:
We proposed the NMOG+NOx standards in terms of corporate fleet averages. We
believe that this fleet-average approach affords manufacturers some flexibility to choose which
technologies to implement and on what timeline, limited by the phase-in schedule and the
increasingly stringent declining standards in this rule. We believe that this is an efficient way to
achieve even greater AQ and health benefits than a per-vehicle approach. We believe the fleet
average standards will provide greater AQ benefits because, when combined with the ability to
generate credits, the fleet average incentivizes manufacturers to achieve additional reductions in
the early years of the program when the standards are the least stringent.
In contrast, the per-vehicle standards we are finalizing for PM are not intended to force
increasing technological improvements over time, but rather to bring all new vehicles to the
levels being achieved by many vehicles today. In that context, fleet averaging is not appropriate,
since our intent is for all vehicles individually to meet the standard.
We note that the Tier 3 rule will result in very significant health benefits, including
specific benefits to children, as described in Section VIII.B of the preamble. We are finalizing
the form of the standards as proposed.
4.1.2. Levels of the Standards
4.1.2.1. General Comments on Level of the Standards
4.1.2.2. 120,000 Useful Life Standards
What Commenters Said:
Alliance of Automobile Manufacturers (Alliance) and Association of Global Automakers
(Global):
4-14
-------
Tier 3 Summary and Analysis of Comments
The CAA is equally clear in prohibiting EPA from setting durability requirements beyond
120,000 miles for vehicles less than 6,000 pounds GVWR. CAA § 207(c)(5)(B) states that the
full useful life for vehicles below 6,000 pounds GVWR is 120,000 miles. In the proposed rule,
EPA recognizes that the statute does not allow it to set durability requirements beyond 120,000
miles for vehicles less than 6,000 pounds GVWR. To avoid the statutory requirement, EPA again
proposes two compliance options for these vehicles. One of these options is viable for
manufacturers, and the other one complies with the CAA durability limit:
a. 150,000 mile durability: Manufacturers can choose to certify all of their vehicles below 6,000
pounds GVWR to the 150,000 mile fleet average standard that is harmonized with LEV III fleet
average.
b. 120,000 mile durability: A manufacturer can choose to certify a test group to the CAA
authorized 120,000 mile durability. However, in doing so, the manufacturer must meet a fleet
average for its entire fleet below 6,000 pounds GVWR that is 15 percent lower than the 150,000
mile fleet average.
Again, EPA has created an alternative compliance option that is so punitive that it effectively
forces manufacturers to comply with the option that violates the CAA. As discussed above
regarding the NMOG+NOx phase-in, EPA cannot avoid compliance with the CAA by offering
an alternative option that meets the statutory requirements, but is so burdensome that, in practice,
it will restrict manufacturer's choices and require compliance with the option that violates the
statute. As explained above, several courts have held that when an agency provides a regulatory
scheme where only one option is truly viable, the remaining option becomes a mandate.
In this case, option (b) above is unduly onerous, because a manufacturer's decision to certify
even one test group to 120,000 mile durability would trigger the requirement that the
manufacturer's entire light-duty fleet is held to a more stringent durability standard. The
structure of this alternative makes it clear that EPA's intent is to drive all manufacturers to
certify to 150,000 mile durability.
The California LEV III requirements are based on 150,000-mile durability standards, and
manufacturers are seeking harmonization between the Tier 3 and LEV III requirements. Even
though the California durability requirement is not consistent with the durability limits set forth
in the CAA, this is still no reason for EPA to either 1) violate the requirements of the CAA, or 2)
impose another Hobson's choice on manufacturers.
We agree that a vehicle certifying to 120,000-mile durability should meet a more stringent
standard than a vehicle certifying to 150,000-mile durability. Therefore, it is appropriate for
EPA's rules to provide that EPA will accept certification of vehicles to 150,000-mile durability
at the CARB standards. The rules should also provide that manufacturers may certify vehicles to
120,000-mile durability at a standard that is 85% of the proposed 150,000-mile bin standard. For
such vehicles, the fleet average would be calculated using the corresponding 150,000-mile bin
standard. This approach complies with the CAA without confronting manufacturers with an
onerous choice.
4-15
-------
Chapter 4: Proposed Vehicle Emissions Program
Recommendation: We recommend that vehicles certifying to 120,000 mile durability be required
to meet a "bin" standard that is 85% of the 150,000 mile bin. For calculation of the fleet average,
the bin emissions would be the corresponding 150,000 mile bin value. For example, a vehicle
that certified to 120,000 mile Bin 30 would need to meet a 26 mg/mile standard, both at
certification and in use. In calculating the fleet average, the vehicle would be considered a 30
mg/mile vehicle.
Chrysler Group LLC:
The CAA precludes EPA from requiring manufacturers to certify their vehicles that are less than
3,750 pounds loaded vehicle weight (LVW) and 6,000 pounds GVWR beyond a useful life of
120,000 miles (35). EPA acknowledges this in the preamble to the proposed rule.
Notwithstanding this clear statutory requirement, EPA proposes a regulatory structure that will
effectively force manufacturers to certify all of their vehicles to a useful life of 150,000 miles.
Specifically, EPA again thwarts the CAA by proposing two compliance options — one that does
not comply with the CAA's 120,000 useful life limitation but offers a reasonable compliance
standard, and another that provides the required statutory useful life protection but for which
compliance is so stringent that no reasonable company would ever choose it.
Under the more favorable 150,000-mile useful life compliance option, manufacturers may
choose to certify all of their vehicles below 3,750 pounds LVW and 6,000 pounds GVWR to the
150,000 mile fleet average tailpipe emissions standard (in contravention of the CAA) that is set
at a reasonable stringency and is harmonized with the LEV III fleet average standard. Under the
unfavorable 120,000 mile useful life option, a manufacturer may choose to certify a test group to
the CAA-authorized 120,000 mile useful life standard. However, if a manufacturer opts to certify
even a single test group to the 120,000 mile useful life standard (the sole useful life authorized
by the CAA), it must meet an unreasonably much more stringent fleet average standard for all of
its vehicles that are less than 3,750 pounds LVW and 6,000 pounds GVWR that is 15 percent
lower than the 150,000 mile fleet average standard.
As with the NMOG+NOx phase-in requirements for heavy-duty vehicles, EPA has created an
alternative compliance option that is so punitive that it effectively forces manufacturers to
comply with the option that violates the CAA. As detailed in footnote 35, the CAA prohibits
EPA from establishing a 150,000 mile useful life emissions standard for light-duty vehicles and
light-duty trucks of less than 3,750 pounds LVW. And, as discussed above, EPA cannot simply
circumvent the clear statutory limitation on useful life to 120,000 miles for vehicles less than
3,750 pounds LVW and 6,000 pounds GVWR by offering an alternative option that meets the
statutory requirements but is so burdensome that, in practice, it will restrict manufacturers'
choices and require compliance with the option that does not comply with the plain terms of the
statute. Again, EPA's dual-path approach impermissibly eviscerates the statutory limitations in
the CAA. Providing a significantly more burdensome alternative compliance path that meets the
statutory requirements does not, and cannot, remedy the fact that the primary compliance path
violates the statute.
As a general matter, Chrysler supports EPA's attempts to harmonize standards in the Tier 3 rules
with the California LEV III standards, which, as relevant here, require all vehicles to comply
4-16
-------
Tier 3 Summary and Analysis of Comments
with a 150,000 mile useful life standard. However, Congress has restricted EPA in what it can do
under the CAA, and, absent an act of Congress, again, EPA cannot thwart the statutory
protections contained therein. Thus, although EPA can incentivize manufacturers to certify to a
150,000 mile useful life standard to be consistent with the LEV III program, EPA cannot
penalize manufacturers for refusing to do so (37). Moreover, the regulatory scheme EPA
proposed cannot be characterized as an "incentive" to manufacturers to accept a 150,000 mile
useful life standard. Rather, for the reasons discussed below, the alternative compliance option
EPA proposed is a penalty on manufacturers that wish to adhere to the statutory limitation to a
120,000 mile useful life standard for vehicles less than 3,750 pounds LVW and 6,000 pounds
GVWR. This penalty has two aspects:
First, EPA proposes to adopt a 120,000 mile useful life NMOG+NOx emission standard that is
85 percent of the respective NMOG+NOx 150,000 mile standard. There is no sound scientific
basis for reducing the standard by 15 percent simply because the useful life is reduced from
150,000 to 120,000 miles. To the contrary, the available data suggest that vehicle degradation
flattens out considerably after 100,000 miles and, consequently, the difference in degradation-
related emissions between a vehicle that has accumulated 120,000 miles and a vehicle that has
accumulated 150,000 miles is less than 15 percent. Catalyst degradation data obtained in
connection with an evaluation of Chrysler aftertreatment systems indicate that the difference in
degradation is instead approximately 10 percent. In any case, EPA may not arbitrarily establish a
more stringent standard for the 120,000 mile useful life, without undertaking a quantitative
analysis of emissions degradation from 120,000 to 150,000 miles that correlates to the magnitude
of the more stringent standard for vehicles certified at 120,000 miles useful life. Accordingly,
imposing a 150,000 mile useful life standard and a 15 percent more stringent 120,000 mile useful
life standard are by no means equivalent from an emissions control standpoint and do not
provide manufacturers a realistic choice. Instead, it appears that EPA is creating a 120,000 mile
useful life standard that is artificially and unreasonably stringent—by requiring emissions
reductions that are far disproportionate to the benefit that would result from accepting a 150,000
mile useful life—in order to force manufacturers to select the illegal 150,000 mile useful life
option.
Second, EPA proposes to require that where manufacturers select the 120,000 mile useful life
option for even a single vehicle model—and even if that specific vehicle model is unique in
some respect or is produced in limited quantities—they must certify all of their light-duty
vehicles and light-duty trucks to the 15 percent more stringent 120,000 mile useful life
NMOG+NOx emissions standards (See footnote). For example, if Chrysler determines that it
needs to retain a 120,000 mile useful life for its Dodge Viper specialty car, which constitutes
approximately 0.1% of its total U.S. sales volume, it must meet 15 percent more stringent
standards for its entire fleet. As with the NMOG+NOx phase-in standards for heavy-duty
vehicles discussed above, this regulatory scheme, which sweeps in a manufacturer's entire fleet
if the manufacturer opts to pursue the 120,000 mile useful life option for a single vehicle model,
is a penalty imposed on manufacturers that refuse to forego their statutory protections. And, as
discussed above with respect to heavy-duty lead time, there is no rational basis for penalizing a
manufacturer's entire vehicle fleet simply because the manufacturer opts for the 120,000 mile
useful life for a single model or test group, which it may need to do out of necessity. There is no
rational relationship between the insignificant increase in emissions resulting from the one model
4-17
-------
Chapter 4: Proposed Vehicle Emissions Program
subject to the shorter useful life and the unreasonable and onerous emissions limitation that
would be imposed across the entire fleet. Such a poison pill or hammer is so unreasonable that it
can only be understood as a penalty for retaining the shorter useful life for even one model.
Again, that approach is unreasonable, unfair, and unlawful.
Chrysler nevertheless supports harmonization with the California LEV III requirements,
especially considering that the ARB has indicated that it may refuse to accept vehicles certified
to a 120,000 mile useful life standard as certified to the California requirements. However, EPA
can achieve this goal without unduly penalizing manufacturers that choose to certify one or more
test groups to a 120,000 mile useful life standard. As such, we believe EPA should, for purposes
of these Tier 3 rules, certify the 120,000 mile Tier 3 bin standards at 90 percent of the respective
California LEV III 150,000 mile useful life NMOG+NOx standards; but, in calculating the fleet
averages, any such vehicle would be considered at the California LEV III 150,000 mile standard.
Recommendation: Chrysler agrees that it is reasonable to adopt 120,000 mile Tier 3 useful life
standards that are lower than the corresponding California 150,000 mile LEV III useful life
standards. Chrysler recommends that EPA retain the 120,000 mile useful life NMOG+NOx bin
standards, especially where EPA has no statutory authority to impose 150,000 mile standards.
Chrysler also recommends that such 120,000 mile standards be set at 90 percent of the respective
California LEV III 150,000 mile useful life NMOG+NOx standards, with fleet average
calculations computed using the California LEV III 150,000 mile NMOG+NOx standards.
Chrysler Footnote: EPA states in the preamble to the proposed rule that it is proposing to adopt 120,000
mile useful life NMOG+NOx emissions standards that are 85 percent of the respective 150,000 mile
useful life NMOG+NOx standards, see 78 Fed. Reg. at 29,868 ("Numerically, we are proposing 120,000
mile useful life NMOG+NOx standards that are 85 percent of the respective NMOG+NOx 150,000 mile
standards."). The proposed rule language itself, however, provides only 85 percent fleet average
standards, and fails to provide 85 percent bin standards. Given the description in the preamble to the
proposed rule, we assume that this omission was simply an oversight and not intentional. However, if
EPA decides to adopt the useful life requirements as proposed, it is critical that EPA provide in the final
rule modified bin standards reflecting the 15 percent reduction for the 120,000 mile useful life
NMOG+NOx standards (i.e., a 26 mg/mi bin reflecting a 15 percent reduction from the 30 mg/mi bin).
Otherwise, EPA's proposed rule actually would require manufacturers that choose the 120,000 mile
useful life option to reduce emissions by more than 15 percent as compared to the 150,000 mile useful life
option, which does not appear to be EPA's intent.
New York State Department of Environmental Conservation
Most vehicles, excluding casualty losses, remain in service well beyond current useful life
requirements (10 years, 120,000 miles). Consequently, New York encourages EPA to adopt
more stringent full useful life requirements to prevent significant erosion of emission benefits as
Tier 3 vehicles age. Specifically, EPA should adopt California's LEV III useful life of 15 years,
150,000 miles.
Pennsylvania Department of Environmental Protection (DEP)
4-18
-------
Tier 3 Summary and Analysis of Comments
The EPA is correct to extend the regulatory useful life of vehicles to 15 years or 150,000 miles
(whichever occurs first). Motor vehicles built today can easily travel over 150,000 miles before
reaching the end of their useful life. Therefore, raising the useful life value to 150,000 miles is
appropriate as an option for vehicles restricted by the Clean Air Act to a 120,000 mile useful life
and as a requirement for vehicles not so restricted. The 'useful life' of a vehicle is the period of
time, in terms of years and miles, during which a manufacturer is responsible for the vehicle's
emissions performance. If the manufacturer certifies to the lower 120,000 mile useful life, it
would seem that the manufacturer should multiply the proposed 150,000 mile standard to obtain
the numerical fleet average by 0.80 not 0.85 as given in Table IV-2 due to the fact that 120,000
miles divided by 150,000 miles equals 0.80. By allowing manufacturers to use the 0.85 factor,
extra credit would be given to manufacturers for certifying to a lower useful life. CARB does not
allow manufacturers to receive credit for certifying vehicles to a 120,000 mile useful life. It
would be more appropriate for EPA to harmonize the Tier 3 requirement with the CARB would
be more appropriate for EPA to harmonize the Tier 3 requirement with the CARB standards
when it concerns useful life to provide the appropriate incentive to automobile manufacturers to
produce vehicles with a longer useful life.
Our Response:
The auto manufacturing industry has uniformly expressed the desire to produce and sell a
single national vehicle fleet, including a general ability and willingness of the industry to certify
their vehicles to a 150,000 mile, 15 year full useful life, as required by the LEV III program.
However, the CAA, written at a time when vehicles did not last as long as they do today,
precludes EPA from requiring a useful life value longer than 120,000 miles (and 10 or 11 years,
depending on vehicle category and weight) for lighter light-duty vehicles (LDVs and LDTs up to
3,750 Ibs loaded vehicle weight (LVW) and up to 6,000 Ibs GVWR (i.e., LDTls)). For heavier
light-duty vehicles (i.e., LDT2s, 3s, 4s, as well as MDPVs, representing a large fraction of the
light-duty fleet), this statutory restriction does not apply, and we are finalizing a 150,000 mile,
15 year useful life value, as proposed.
For the lighter vehicles, we are continuing to apply the 120,000 mile (and 10 or 11 year,
as applicable) useful life requirement from the Tier 2 program, also as proposed. (LEV III does
not allow 120,000 mile useful life certification beyond the phase-in period and therefore vehicles
certified as 50-state offerings under a harmonized approach will also need to comply with the
150,000 mile useful life for LEVIII regardless of the Tier 3 120,000 mile useful life option.) For
these lighter vehicles, manufacturers are allowed to certify to either useful life value. In order
for the Tier 3 NMOG+NOx standards to represent the same level of stringency regardless of
which useful life value manufacturers choose, we proposed and are finalizing proportionally
lower numerical values (85 percent of the NMOG+NOx 150,000 mile standards based on a data
analysis provided in Chapter 1 of the RIA) for the declining fleet average FTP NMOG+NOx
standards when a manufacturer chooses the 120,000 mile useful life for eligible vehicles. The
RIA analysis, also provided in the draft RIA for the proposed rule, provides robust support for
the 85-percent value for the adjustment to the NMOG+NOx standard. Commenters did not make
any specific comments about our analysis, including providing supporting information to support
a different percentage adjustment value.
4-19
-------
Chapter 4: Proposed Vehicle Emissions Program
We proposed that a manufacturer that certifies any vehicle model under the 120,000 mile
provision be required to certify all their LDVs and LDTls to the 120,000 mile useful life and
associated numerically lower FTP NMOG+NOx fleet average standard. Comments from the
auto industry expressed a concern that this approach would be inflexible to manufacturers' needs
and unnecessarily burdensome. We disagree that the requirements proposed for the 120,000
mile provision were punitive or create a mandate, but we have considered the concerns expressed
by commenters and believe that the emission benefits of Tier 3 program will be maintained if
certification of these lighter vehicles to the 120,000 mile useful life standards is allowed on a test
group basis. Therefore we are finalizing this revised approach.
We have also considered the comments that recommended that EPA set specific "bin"
standards at proportionally reduced levels (factor of 0.85) for manufacturers that choose the
120,000 mile useful life for their lighter vehicles and then include those vehicles in a single
150,000 mile useful life fleet average as proof of compliance. We do not believe it is appropriate
to allow a vehicle certified to (and only held to) a 120,000 mile useful life to have their projected
emission results included with vehicles that meet the higher 150,000 mile useful life
requirement. Manufacturers that choose the 120,000 mile useful life option likely would be
doing so specifically to avoid any durability liability beyond the 120,000 mile point, and
therefore their performance at the level of the standards is not guaranteed or legally required
beyond the 120,000 mile useful life. As the analysis that we performed and that we discuss in
Chapter 1 of the RIA was done on a fleet average basis, we believe that it is appropriate for
manufacturers to certify each test group to the same Tier 3 'bin' levels and then to average all the
120,000 mile test groups against the "85 percent" adjusted FTP NMOG+NOX standard for that
model year and separately average all of the 150,000 mile test groups to the unadjusted standard.
Manufacturers are required to maintain separate 120,000 mile and 150,000 mile fleet averages if
they choose to certify some vehicles to the lower useful life standards. With this continued focus
on respective fleet average standards for 120,000 mile and 150,000 mile useful life test groups,
we do not see an additional value to creating different sets of bins given the existing variety of
certification bins available above and below the two separate fleet average standards throughout
the phase-in and into the final program.
4.1.2.3. Specific Comments on FTP NMOG+NOx Standards
What Commenters Said:
Manufacturers of Emission Controls Association (MECA)
Two recent research programs have discussed pathways to reaching Tier 3, Bin 30 exhaust
standards with a light-duty diesel vehicle. The first of these is the Cummins Advanced
Technology Light-Duty Diesel Aftertreatment System (ATLAS) program, sponsored by the U.S.
Department of Energy (DOE). Details of this project were presented at the 2012 DOE DEER
Conference held in Dearborn, MI (see:
wwl.eere.energy.gov/vehiclesandfuels/resources/proceedings/2012_deer_presentations.html,
presentation by Mr. Gary Henry, Cummins, Inc.) and included in SAE paper no. 2013-01-0282
that was presented at the April 2013 SAE International Congress in Detroit, MI. Advanced diesel
4-20
-------
Tier 3 Summary and Analysis of Comments
emission control technologies, including a passive lean NOx adsorber and an SCR-coated DPF
that utilizes direct ammonia reductant injection, have been combined in this program with
advanced combustion controls on a 2.8 liter, 4 cylinder diesel engine to demonstrate a pathway
to Tier 3, Bin 30 exhaust emissions compliance on a full-size, light-duty pick-up truck. A second
approach to Tier 3, Bin 30 exhaust levels on a light-duty diesel vehicle was discussed by
Southwest Research Institute in SAE paper no. 2013-01-1301 (presented at the April 2013 SAE
International Congress in Detroit, MI). This project combines advanced diesel combustion
technologies, including high temperature glow plugs, with a close-coupled lean NOx adsorber
catalyst + catalyzed DPF emission system to significantly reduce cold-start emissions and
provide a pathway to Tier 3, Bin 30 exhaust emission compliance on a 2 liter diesel-equipped
sport-utility vehicle.
Our Response:
EPA acknowledges these comments supporting the feasibility of the 30 mg/mi
NMOG+NOx standard.
4.1.2.4. Specific Comments on SFTP NMOG+NOx Standards
What Commenters Said:
Alliance of Automobile Manufacturers (Alliance) and Association of Global Automakers
(Global)
SFTP Fleet Average and Family Emission Limits: Unlike Tier 3, Tier 2 (and LEV II) vehicles
certify to a different SFTP standard scheme. Tier 2 vehicles certify to an SFTP standard based on
the FTP bin to which the vehicle is certified, while in Tier 3 the FTP and SFTP standards are
independent. The Tier 3 and LEV III programs both include the previous generation vehicles in
the fleet average. Under the LEV III program, for a LEV II vehicle the certification data is used
in the calculation of the fleet average, projected out to 120,000 miles using a deterioration factor
or actual test results if testing was conducted using aged components. For compliance purposes
(confirmatory and in-use), the vehicle is held to the standard to which it was certified. However,
it appears that under the Tier 3 program, EPA will not allow projected certification data to be
used in the fleet average calculation for Tier 2 or interim Tier 3 vehicles. Instead, a
manufacturer-specified SFTP FEL, which a manufacturer would be held to under confirmatory
and in-use testing, would be the value used in the fleet average calculation and would become
the new certification standard for that vehicle.
Under EPA's approach of treating values used in the fleet average calculation as emission
standards, subject to confirmatory and in-use testing, manufacturers will be forced to add a
compliance margin when choosing an FEL under the Tier 3 program for these Tier 2/interim Tier
3 vehicles. For example, a MY 2018 Tier 2 (a.k.a., "interim Tier 3") vehicle with SFTP
certification data showing emissions of 95 mg/mile projected to 120,000 miles would use a 95
mg/mi value in the ARB's fleet average calculation. However, under EPA's Tier 3 program the
manufacturer might choose an FEL of 145 mg/mile (a 35% compliance margin, which is not
4-21
-------
Chapter 4: Proposed Vehicle Emissions Program
atypical) to allow sufficient compliance margin given the various sources of variability (vehicle-
to-vehicle, test-to-test, etc.) that a manufacturer must consider relative to confirmatory and in-use
compliance testing.
The net effect of this difference is to unnecessarily increase the stringency of the Federal SFTP
program over the California program. To address this difference, harmonization with LEV III
would only be applicable to "interim Tier 3" vehicles (Bin 125, Bin 160, and the Bin 20 or Bin
30 PZEV vehicles discussed above) and thus would only apply to a small sub-set of vehicles, and
only then through the MY 2019.
Recommendation: We recommend EPA adopt the following new paragraph (or changes to this
effect) in the interim provisions of section §86.181 l-17(b)(7)(iii)(C):
"(C) For vehicles certified to bins higher than Bin 70, and Bin 30 or Bin 20 vehicles which
received partial zero emission vehicle (PZEV) credit in California that certified to interim Tier 3
using carryover emissions data, compliance (confirmatory and in-use) will be based on the Tier 2
certification values contained in §86.1811-04."
SFTP 4,000-Mile Standard Prior to MY 2020: EPA has proposed to retain the 4K SFTP
standards from the Tier 2 program to prevent "backsliding" for vehicles certified in the earlier
portions of the Tier 3 program when the composite SFTP fleet average standards might allow
room for less robust SFTP calibrations. Under the proposal, these standards would apply to all
"Interim Tier 3 vehicles." The agency concluded the interim SFTP standards would not need to
apply to "Final Tier 3 Vehicles," since their emission system designs should be sufficiently
robust to mitigate the agency's concern about SFTP backsliding. "Final Tier 3 Vehicles" would
include those certified to the new useful life (i.e., 150K), Tier 3 certification fuel, and new PM
standards. All others would be considered "Interim Tier 3 Vehicles."
Since EPA's definition of Interim Tier 3 Vehicles would then include all vehicles certified using
the option to test with the California E10 certification fuel, all such vehicles would be subject to
the interim 4K SFTP standards - even if these vehicles were certified to some of the cleanest
bins and to 150K SFTP standards on E10. For example, a vehicle might be certified to Bin 70 or
below, perhaps even as low as Bin 20, plus the 150K useful life (for FTP and SFTP) and the new
PM standard, but it would still be considered "interim" because it was tested on the California
LEV III E10 certification fuel. Such vehicles should be expected to have calibrations as robust as
any Final Tier 3 Vehicles, including the new SFTP standards at full useful life, and should not be
subject to the interim 4K SFTP standards.
While we do not think there is a need for the interim 4K SFTP standards at all; if the agency is
going to insist on retaining them it would make more sense to link these standards to the
transitional Tier 3 bins, i.e., those above Bin 70, rather than linking them to the more complex
definition of Interim Tier 3 Vehicles. Additionally, having this simplistic link to the bin structure
rather than to multiple variables of the more complex Interim Tier 3 Vehicle definition would
simplify both reporting requirements for the manufacturers and data acquisition and tracking
burdens for the agency, and would do so without any loss of stringency or backsliding.
4-22
-------
Tier 3 Summary and Analysis of Comments
Recommendation: We recommend that EPA change the applicability of the 4K SFTP standard
such that it would apply to only those bins above Bin 70 rather than to all vehicles that have been
defined as Interim Tier 3 Vehicles through MY 2019 only.
Interim In-Use Standards: Similarly, interim in-use standards are appropriate for SFTP for both
the MDV and LDV classes. For the first time ever, the SFTP standards will apply at a 150,000
mile useful life. And the required levels will be far lower than current standard levels. These two
factors, coupled with the rapid introduction of new technologies driven by the GHG and fuel
economy requirements, justify the need for interim in-use SFTP standards to help manufacturers
manage their in-use compliance risk. [EPA-HQ-OAR-2011-0135-4461-A1, p. 24]
Recommendation: We recommend EPA continue the historic practice of allowing interim in-use
standards by harmonizing with the LEV III interim in-use standards for the HDV FTP and SFTP
requirements and the LDV SFTP requirements.
American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM)
In discussing the SFTP NMOG+NOx feasibility, EPA notes "The proposed new emission
requirements include stringent NMOG+NOx composite standards over the SFTP that would
generally only require additional focus on fuel control of the engines and diligent
implementation of new technologies like gasoline direct injection (GDI) and turbocharged
engines." There are no supporting data in the DRIA to substantiate this comment and in fact the
following section notes that "A range of technology options exist to reduce NMOG and NOx
emissions from both gasoline fueled spark ignition and diesel engines below the current Tier 2
standards. Available options include modifications to the engine calibration, engine design,
exhaust system and after treatment systems." and further add that "To achieve the NMOG+NOx
Tier 3 SFTP standard manufacturers will need to develop and implement technologies to manage
catalyst temperatures during high-load operation without using fuel enrichment."
California Air Resources Board (CARB)
CARB supports U.S. EPA's proposal to not include relaxed interim emission standards for non-
methane organic gas (NMOG) + oxides of nitrogen (NOx) and (PM) in the light-duty SFTP
program. CARB agrees that the technologies required for the proposed SFTP emission standards
are well-established and interim emission standards, typically reserved for new technologies, are
not needed. Accordingly, CARB intends to propose alignment with this proposal once the Tier 3
program is finalized.
International Council on Clean Transportation (ICCT)
The SFTP standards are too lenient and, as proposed, will not be effective. Current vehicles
certified to Tier 2 bin 2 or LEVII-SULEV have average NMHC+NOx emissions of less than 10
mg/mi, more than 80% below the proposed limit of 50 mg/mi in 2025. Similarly, the proposed
SFTP particulate standards are 3.3 times higher for vehicles 6000 GVWR than the proposed FTP
standard. Setting the SFTP standards properly is especially important for diesel engines, as diesel
4-23
-------
Chapter 4: Proposed Vehicle Emissions Program
emission control hardware requirements are largely set by the high load conditions on the SFTP.
SFTP NMHC+NOx standards should be set at no more than 20 mg/mi and SFTP particulate
standards at no more than 6 mg/mi.
SFTP Standards: For the most part, the overall stringency of the proposed rules is adequate and
the provisions, including leadtime and credit provisions, are appropriate. However, ICCT is
extremely concerned that the SFTP requirements are far less stringent than the FTP
requirements. While the proposed Tier 3 SFTP standards are a major improvement over the
SFTP standards for Tier 2, the Tier 2 SFTP standards were unchanged in stringency from the
Tier 1 SFTP standards and, thus, completely ineffective. Thus, despite the large reduction in the
SFTP standard levels, the proposed SFTP standards are still far too lenient and will not achieve
the objectives of the SFTP standards to reduce in-use emissions.
To demonstrate our concerns, we have analyzed the stringency of the proposed SFTP standards
in two different ways. The first method compares the proposed SFTP standards against current
SFTP emission levels. The second method compares the proposed SFTP standards against the
proposed FTP standards.
Current SFTP emission levels: Figure 1-5 [of number EPA-HQ-OAR-2011-0135-4304] in the
draft RIA (page 1-18), reproduced below, demonstrates that the average SFTP NMHC+NOx
emissions for current vehicles certified to Tier 2 bin 2 or LEVII-SULEV emission standards (the
orange bars) is less than 10 mg/mi, and the highest emissions seen is about 42.5 mg/mi. The
proposed SFTP NMHC+NOx standard drops from 103 mg/mi in 2017 to 50 mg/mi in 2025. So,
the proposed standard for 2017 is more than 10 times the average emissions of current vehicles
and the proposed 2025 standard is more than 5 times the average emissions of current vehicles.
Proposed SFTP versus proposed FTP standards: The original SFTP standards, adopted in 1996
and applied to Tier 1 vehicles, found that the incremental emissions on the SCO3 and USO6
cycles was similar in magnitude to the incremental emissions from the cold start on the FTP.
Thus, SFTP standards for Tier 1 vehicles were set at the same numeric level as the FTP
standards.
As the SFTP standards are hot, running emissions only, it is appropriate to separate the FTP
requirements into cold start emissions and hot, running emissions. The proposed NMHC+NOx
FTP standards are 30 mg/mile. The draft RIA states (page 1-6): [EPA-HQ-OAR-2011-0135-
4304-A1, pp. 13-14] 'Based on modal analysis of a gasoline powered vehicle being operated on
the FTP cycle, approximately 90 percent of the NMOG emissions occur during the first 50
seconds after a cold start. In addition, about 60 percent of the NOX emissions occur in these
early seconds.'
The Tier 2 bin 2 standards were 10 mg/mi for NMHC and 20 mg/mi for NOx. Using this ratio
and applying it to the cold start emission ratios from the draft RIA, 70% of NMHC+NOx
emissions on the FTP are from the cold start (90% x 1/3 + 60% x 2/3). This means that about
30% of NMHC+NOx FTP emissions are from hot, running operation and, thus, account for
about 9 mg/mi of the proposed FTP standards.
4-24
-------
Tier 3 Summary and Analysis of Comments
The proposed NMHC+NOx SFTP standards are 50 mg/mi, or 5.5 times higher than the hot,
running emissions portion of the FTP standards. This is so lenient as to be essentially no off-
cycle control. And this is with respect to current vehicles, much less for 2025 with several
additional generations of emission control development.
After application of the standard 50 percent compliance margin, hot, running NMHC+NOx
emissions on the FTP are about 4.5 mg/mi. Current SFTP NMHC+NOx emissions from Tier 2
bin 2 and LEVII-SULEV vehicles are a bit less than twice this amount. This is a reasonable ratio
between SFTP and hot, running FTP emissions. This, in turn, indicates that the hot, running FTP
comparison supports the analysis of SFTP emissions from current vehicles.
Both the current SFTP data and the proposed hot, running emissions on the FTP support actual
SFTP NMHC+NOx emissions of less than 10 mg/mi. After adding the standard x2 in-use
compliance margin, the SFTP NMHC+NOx standard should be set at no more than 20 mg/mi.
Setting appropriate SFTP standards is especially important for diesel vehicles. The cold start in
the FTP largely determines the emission control system design for gasoline vehicles. The
primary need for SFTP gasoline engine standards is to ensure that proper calibrations are used
off-cycle and that emissions remain reasonable in-use. However, for diesel engines, the emission
control system design is largely determined by high load operation. Thus, while the proposed
SFTP requirements would likely not impact gasoline hardware design, ineffective SFTP
standards could lead to selection of diesel emission control systems that are less effective in-use.
In fact, this has already been seen in Europe, where Euro IV and Euro V heavy-duty vehicles
equipped with selective catalytic reduction (SCR) systems have significantly elevated emissions
of nitrogen oxides (NOx) during in-use driving, particularly when operating in urban traffic. In
some cases, actual in-use urban emission levels may be as high as or higher than those from
much older vehicles with engines certified to more lenient emission standards (9). This illustrates
the importance of setting emission standards using representative test procedures and appropriate
standards.
ICCT does have concerns in two areas. First, while the proposed supplemental FTP standards
are a major improvement over the SFTP standards with Tier 2, they are still not stringent
enough. The Tier 1 SFTP standards were set at the same numeric level as the Tier 1 FTP
standards. While the proposed SFTP NMOG plus NOx standards are 67 percent higher than the
proposed FTP standards.
Mitsubishi Motors R&D America, Inc (MRDA)
We completely agree with the Alliance and Global Automakers comments regarding concerns
with the SFTP test. The SFTP test, which is used to calculate the NMOG + NOx fleet average, is
a composite of the SCO3 and US06 tests. The SCO3 test is expensive and time-intensive to run,
and the US06 is highly driver-dependent relative to any other vehicle emissions test. As the
overall emission standards for the FTP become more stringent in the proposed Tier 3 program,
automobile manufacturers will make significant changes to comply, e.g. changes to catalysts,
4-25
-------
Chapter 4: Proposed Vehicle Emissions Program
engine calibrations, and transmission gear ratios, etc. It is unknown how these changes will
affect the SFTP emissions and therefore compliance with the new standards.
The proposed Tier 3 program does not include early credits for SFTP compliance. Early credits
are a necessary compliance flexibility for automotive manufacturers, since they cannot control
for market uncertainty, i.e. which vehicles customers will purchase. It will be an unnecessary
(and extremely challenging) compliance burden for OEMs to comply with a federal regulation
that will not allow early credits, and with the LEVIII program that will allow early credits. This
discrepancy between the Federal and California programs can create significant differences
between the two sets of standards.
Additionally, the proposed Tier 3 program does not allow for interim in-use SFTP standards,
whereas they are included as part of the LEV III program. Without additional allowances for in-
use standards during the interim years, it will be difficult for manufacturers to learn about the
new technologies and how well they work over time. New technologies will be strongly
penalized under the Tier 3 program in comparison to the LEV III program. For example, if an
OEM starts basic development work in Spring/Summer 2014 in order to certify the vehicle in
2016, calibrations can be made to meet the required standards. However, the performance of the
vehicle over time is unknown. Therefore, interim in-use standards would help alleviate this
uncertainty.
Another area which demonstrates a lack of harmonization between the proposed Tier 3 program
and the LEV III program is SFTP fleet average compliance. EPA is proposing that non-Tier 3
(i.e. Tier 2) test groups use the Tier 3 120K Family Emission Limits (FEL) for fleet average
certification and for confirmatory and in-use standards. EPA is effectively penalizing the Tier 2
carryover vehicles by making it harder to meet their fleet average SFTP compliance level. This
requirement would retroactively change fleet average compliance standards for non-Tier 3
vehicles. Additionally, this requirement means that non-Tier 3 test groups would be counted
differently for fleet average compliance under LEVIII and Tier 3. A lack of harmonization in this
area will introduce a significant reporting burden for manufacturers. Duplicate certification and
confirmatory compliance strategies will be necessary and provide no air quality benefit. Similar
to LEV III, EPA should allow non-Tier 3 vehicles to use Tier 2 emissions certification data for
confirmatory and in-use requirements.
NPRM Comments: For the reasons stated above, we request the following:
-Allow early credits for SFTP compliance.
-Provide interim in-use standards for SFTP compliance.
-Harmonize with the LEVIII program for SFTP fleet average compliance by allowing non-Tier 3
vehicles to use Tier 2 emissions certification data for confirmatory and in-use requirements.
Supplemental Federal Test Procedure (SFTP): Allow early credits and provide interim in-use
standards for SFTP compliance. EPA should harmonize with the LEVIII program by allowing
non-Tier 3 vehicles to use Tier 2 emissions certification data for confirmatory and in-use
requirements.
Natural Resources Defense Council (NRDC)
4-26
-------
Tier 3 Summary and Analysis of Comments
Based on EPA's own analysis of current vehicle performance, the Tier 3 standards for the
Supplemental Federal Test Procedure (SFTP) should be more stringent. As presented in the Draft
Regulatory Impact Analysis (DRIA), Figure 1-4, model year 2010 and 2011 vehicles certified to
Bin 5 had average HC+NOx levels at or below EPA's proposed NMOG+NOx standard of 50
mg/mi for model year 2025. Further, Bin 2 vehicles, shown in DRIA Figure 1-5, had average
HC+NOx emissions below 15 mg/mi and maximum levels below 50 mg/mi. Under Tier 3, it is
expected that existing and new technologies enabling vehicles to reach Bin 2 and 3 levels will be
much more widespread in the fleet. Clearly, a standard even lower than 50 mg/m is technically
and economically feasible.
Union of Concerned Scientists (UCS)
Millions of Americans breathe cleaner air as a result of our nation's clean air laws but serious
challenges remain. More than 1 in 3 Americans still live in areas where air pollutant levels
exceed at least one of the health-based National Ambient Air Quality Standards. Passenger
vehicles remain the second largest emitters of nitrogen oxides (NOx) and volatile organic
compounds (VOCs) in the U.S. - the primary pollutants that form smog. These vehicles also
emit more than half of all carbon monoxide pollution and contribute to particulate matter
emissions. It is therefore essential that the Tier 3 standards for both PM and NMOG+ NOx be
sufficiently stringent as to provide quantifiable and real reductions in harmful emissions.
We urge EPA to consider levels for the SFTP for both PM and NMOG+ NOx that are
sufficiently stringent to ensure that they are suitably protective of human health and the
environment.
Supplemental Federal Test Procedure: EPA should revisit the non-methane organic gas plus
nitrogen oxides (NMOG+ NOx) and particulate matter (PM) exhaust emission requirements to
ensure that the emission requirements of this test procedure are suitably protective of human
health and the environment.
We are concerned that the proposed Supplemental Federal Test Procedure (SFTP) is not
sufficiently stringent. In 2025, the Federal Test Procedure (FTP) requires a NMOG+ NOx limit
of 30 mg/mile, regardless of class size, while the SFTP level is 50 mg/mile. The SFTP is 67%
higher than the FTP. Additionally, the Regulatory Impact Analysis (8) shows that the average
SFTP NMOG+ NOx emissions for current light duty vehicles certified to Tier 2 Bin 2 are less
than 10 mg/mile, and the highest emissions are just over 40 mg/mile. The RIA also shows that
Tier 2 Bin 5 vehicles currently average approximately 30mg/mile for light duty vehicles and up
to approximately 50 mg/mile for the largest light duty trucks. This means that the highest current
Tier 2 Bin 2 and the average Tier 2 Bin 5 emitters are already compliant with the proposed 2025
standard, and that the 2025 standard is five times the emissions of the average current Tier 2 Bin
2 vehicle.
Our Response:
4-27
-------
Chapter 4: Proposed Vehicle Emissions Program
We received a variety of comments regarding the proposed SFTP standards. On one
hand, several commenters stated that the proposed standards are too lenient, based on their
evaluation of vehicle emission test data we presented in the NPRM. We have considered these
comments and have reviewed the data from the NPRM. Our conclusion from that data continues
to be that the SFTP NMOG+NOx emission levels that we are finalizing are sufficiently low to
ensure that manufacturers will accomplish the purpose we had set out for these standards - to
largely eliminate the use of fuel enrichment events (and their emission consequences) and
discourage "backsliding" on SFTP emissions performance as new vehicles are developed,
including diesels whose NOx emission control hardware requirements have historically been set
by the high load conditions found on the SFTP.
As a result, we designed the standards to essentially eliminate fuel enrichment events and
their emissions consequences, thereby resulting in important emissions reductions. See Chapter
1 of the RIA for an analysis of this data. Based on our reassessment, we continue to believe that
significant additional real-world emission reductions will not result through SFTP NMOG+NOX
standards lower than the 50 mg/mi fully phased-in level we are finalizing. We considered the
analyses of commenters arguing that a lower standard is shown to be achievable by some
vehicles today, in part based on an assumed relationship between FTP and SFTP performance.
We do not see a strong enough relationship between these emissions to draw conclusions about
the standard levels. As we discuss here and in the preamble, we find it more compelling to
observe the close relationship between SFTP operation and fuel enrichment events.
The 50 mg/mi final level of the standard ensures that over the SFTP cycles, vehicles will
have little opportunity to deviate from the tight emission controls established for the FTP. Any
lack of attention to the SFTP cycles will result in exceedance of the standards due to the highly
non-linear emission performance of the emission control systems when fuel enrichment events in
gasoline vehicles occur or when diesels do not properly control NOx aftertreatment. (To achieve
additional meaningful reductions for SFTP "off-cycle" like operation, EPA would need to pursue
a "not to exceed" requirement similar to what is required for heavy-duty engines; such an
approach would require significant new research and we did not propose this for Tier 3.) Again,
we believe that the 50 mg/mi NMOG+NOx standard will ensure that the SFTP performance of
future vehicles with future technologies continues to be comparable to that attained by the
current Tier 2 fleet. Finally, the fuel enrichment limitation provisions we are finalizing will
further support the goals of SFTP standards.
On the other hand, a few commenters stated that as compared to the LEV III program,
compliance with the Tier 3 SFTP NMOG+NOx standards, including manufacturer-selected
FELs, is unnecessarily stringent. The LEV III program has the challenge of determining a fleet
average for SFTP that includes LEV II vehicles previously certified only to the 4,000 mile (4K)
standards and LEVIII certified to more stringent standards that apply to full useful life. LEV III
allows the LEV II results to be deteriorated to full useful life for purposes of fleet average
calculation. However these LEV II SFTP requirements remain only 4,000-mile standards.
Unlike LEV II, the Tier 2 SFTP standards have always been full useful life (not only
4,000-mile) standards, and we continue to require all Tier 3 vehicles to also meet full useful life
standards. When transitioning allowable "carryover" vehicles into "interim" Tier 3,
4-28
-------
Tier 3 Summary and Analysis of Comments
manufacturers will need to determine the self selected SFTP FEL that they believe represents the
full useful life performance of that vehicle. This self-selected standard will be what the
manufacturer will be held to for in-use testing in addition to determining the fleet average level
at certification. It is possible that the Tier 3 program may require a manufacturer to select an
FEL that is numerically higher than the carryover LEV II 4,000-mile result with deterioration
applied, but it is at the discretion of the manufacturer to select an FEL that represents expected
full useful life performance. This situation, if it occurs, is not inconsistent with our overall intent
for the SFTP standards, and we are finalizing these provisions as proposed.
Comments regarding the 4K SFTP standards for Interim Tier 3 vehicles suggested that if
we insist on retaining this requirement, it should only apply to Bins higher than Bin 70. The
purpose of retaining this requirement for all Interim Tier 3 vehicles is that by definition, Interim
Tier 3 vehicles are not meeting the full requirements of Tier 3, which include the final E10 test
fuel and the stringent PM standards for both FTP and SFTP. The 4K SFTP requirement is a very
effective requirement for protecting against excessive PM emissions during conditions like
aggressive or high-speed driving behavior. Until the vehicles meet the stringent Tier 3 PM
standards for SFTP, PM emissions will generally be controlled to current Tier 2 performance
levels by maintaining the 4K SFTP requirements. Vehicles utilizing LEV III test fuel during the
phase-in may be considered Final Tier 3 if they meet the new PM standards and the 150,000 mile
useful life requirements.
Several commenters supported relaxed interim in-use SFTP NMOG+NOx standards for
light-duty vehicles. Also, one commenter recommended adding early credit provisions for the
SFTP NMOG+NOx standards. As discussed above, the levels of these standards are not
intended to force major new technological improvement, and will not be challenging for most
manufacturers for most vehicles. Based on our analysis of in-use emission performance from
IUVP testing, as discussed in the RIA, we do not believe that interim in-use standards are
necessary even with the increase in useful life requirements and the introduction of new
technologies in response to GHG and fuel economy requirements, and we are therefore finalizing
SFTP NMOG+NOx standards without associated interim in-use standards. In their comments,
CARB supported this approach and stated in their public comments their intention to propose
alignment with this approach once the Tier 3 program is finalized. Similarly, we do not believe
early credits will be necessary to facilitate compliance with the SFTP standards, and we are not
finalizing early credits for this purpose.
API and AFPM incorrectly state that there are no supporting data to substantiate
statements about SFTP NMOG+NOx feasibility. EPA analyzed in-use certification test results
for model years 2010 and 2011, which represent the most recent model years for which complete
IUVP data sets are available. This analysis can be found in RIA chapter 1.3. The analysis
indicates that most vehicles are already meeting the final SFTP composite standard of 50 mg/mi.
We concluded that the small number of vehicles currently not meeting the final standards will
need additional focus on the SFTP cycles and/or possible hardware improvements to better
manage temperature concerns. Additionally, we recognize that our analysis may not have
included some newer technologies that have just recently entered the market and that these
technologies may require diligent implementation to meet the Tier 3 standards.
4-29
-------
Chapter 4: Proposed Vehicle Emissions Program
4.1.2.5. Specific Comments on FTP PM Standards
What Commenters Said:
Alliance of Automobile Manufacturers (Alliance) and Association of Global Automakers
(Global)
Another point of concern is that the PM standards are flat standards that all vehicles must meet,
as compared to fleet average standards that are proposed for NMOG+NOx compliance. Thus the
risk that a particular type of technology would be precluded by the PM standards is increased
because such technology cannot be certified to a higher bin and offset by a different technology
certified to a lower bin.
It is also worth noting that the PM standards begin at a time when manufacturers enter the eighth
year of the most stringent GHG standards in history, when the number of ZEVs and PHEVs
jumps considerably in California and the Section 177 states, and while reducing exhaust
emissions by 75 percent on both the FTP and SFTP and phasing in zero evaporative emission
standards on all vehicles.
American Lung Association
Harmonize PM Standards with CARB now. We strongly urge EPA to harmonize the Federal
Test Procedure (FTP) PM emissions standard with CARB to 1 mg/mile by 2025. Diesel vehicles
and some gasoline-powered vehicles have the technology to meet a 1 mg/mile PM standard
today. According to the International Council on Clean Transportation (ICCT) in their comments
to CARB on the Low Emissions Vehicle program III, certifying vehicles to a low standard can
be challenging, but can be done. ICCT recommended focusing on the more precise solid particle
number measurement such as in the United Nations Particle Measurement Programme as an
alternative proxy, but not a replacement, for the gravimetric method. Technology exists today to
collect and report particle number emissions. Automakers in Europe currently do so, therefore,
given the lead time, this issue should be resolved by 2017.
Appalachian Mountain Club (AMC)
In setting the particulate matter emissions standard and the Supplemental Federal Test Procedure
(FTP) we urge EPA to use the lowest feasible mg/mi level supported by the EPA's own testing
information. In addition, the Agency should consider the current efforts by the California Air
Resources Board for the California Low Emission Vehicle 3 standards, which is phasing in a 1
mg/mi FTP standard starting in 2025.
California Air Resources Board (CARB)
CARB supports the Tier 3 PM standards proposed by U.S. EPA through the 2024 model year.
These standards, when fully implemented in 2022, represent a 70 percent reduction from current
PM standards and will ensure new vehicles and technologies will continue to perform at levels
achieved by the best of today's vehicle technologies. We also agree with U.S. EPA that these
standards are readily achievable with no additional technology or hardware.
4-30
-------
Tier 3 Summary and Analysis of Comments
However, the federal program does not include the LEV III 1 milligram per mile (mg/mi) PM
standard that phases-in between model years 2025-2028, stating "In order for U.S. EPA to
propose a standard at this level, there must be established methods to, reliably and consistently
measure PM below that standard, for compliance purposes.' While we understand U.S. EPA's
reluctance at this time due to concerns with current and proposed test methods, we encourage
U.S. EPA to consider adoption of such a standard as discussed below. CARB is in the process of
completing testing studies, which will show that existing methods can be modified in a
straightforward manner to achieve accurate and repeatable measurement of sub-1 mg/mi PM
emissions. At the same time, CARB is also pursuing additional research to explore alternative
methods for PM measurement in order to take advantage of promising, new commercially
available real-time instruments. We also urge U.S. EPA to re-examine the proposed PM
standards for the SFTP for the reasons discussed below.
CARB adopted the 1 mg/mi PM standard because PM emissions are a particular concern for
their multiple impacts on public health, air quality, and the global climate. In general, mobile
sources (mainly cars and trucks) are not major contributors to the statewide total PM mass
inventory. For instance, the PM emissions from light-duty vehicles add up to less than 5 percent
of the total California PM2.5 inventory. However, they do contribute significantly to urban
pollution and human exposure, such as the elevated concentrations of PM near heavily-travelled
roadways. Historically, PM emissions from diesel engines were of most concern because of their
high PM emission rates relative to gasoline engines. But as the modern diesel engine achieves
increasingly lower PM emissions, the interest in and relative contribution of PM emissions from
an ever increasing number of gasoline-fueled light-duty vehicles is growing. Thus, the need for
maximum mitigation of PM emissions at the source (i.e., tailpipe) using the most technically and
economically feasible approach is paramount.
Comment 1 - Harmonization of 1 mg/mi PM standard between Tier 3 and LEV III: As stated
above, based on studies CARB and others have conducted recently, we believe the measurement
method will not be an issue for implementing a 1 mg/mi PM standard for the 2025 model year or
sooner. We agree with U.S. EPA on the statement that meeting the 1 mg/mi PM standard
requires establishing methods to reliably and consistently measure PM below the standard.
CARB, U.S. EPA, and some industry laboratories have already measured and reported PM
emissions from light-duty vehicles well below 1 mg/mi using the existing three filter method
prescribed by 40 CFR Part 1065. In its draft Regulatory Impact Analysis, U.S. EPA reported
FTP composite PM emission results for 16 light-duty passenger cars and trucks (Table 1-8 of the
Regulatory Impact Analysis). Reported PM emissions from 11 of the 13 conventional port fuel
injection vehicles were below 1 mg/mi, ranging from 0.10 to 0.93 mg/mi. With the exception of
a few repeat results with comparatively high variability, the data in Figure 1-8 of the Regulatory
Impact Analysis suggests reasonably good precision and repeatability in the vast majority of the
tests. The reason for the high variability in a few results needs investigation, but may be
attributed to vehicle stability. Reproducibility between different tests cells and/or laboratories
also needs to be addressed so that consistent results for certification and compliance can be
generated within acceptable margins.
CARB has already developed and implemented a strict laboratory filter handling and weighing
procedure (MLD 145) that minimizes variability from the filter weighing contribution of PM
4-31
-------
Chapter 4: Proposed Vehicle Emissions Program
measurement. The background contribution from the Constant Volume Sampler dilution tunnel
at CARB facilities has been determined to be no more than 2.5 micrograms per filter. The
proposed test method in 40 CFR Part 1066 allows for a maximum background correction of 5
micrograms per filter, which is equivalent to 0.20 mg/mi PM with nominal Constant Volume
Sampler flow of 350 standard cubic feet per minute and no secondary dilution. Allowance for the
correction subtracts most or all of the system's contribution to a PM measurement. Therefore,
manufacturers can design closer to the standard and still have a reasonable probability of passing
a one test certification requirement.
CARB also agrees with U.S. EPA's proposed test procedure changes that will allow the use of
fewer filters when conducting PM measurements. Conceptually, this will reduce measurement
variability as well as the test burden associated with weighing multiple filters. CARB is
evaluating the various filter options proposed in 40 CFR Part 1066 and develop defined filter test
method procedures. We expect the additional options will show even further improvement in the
capability of the gravimetric technique for measurement of PM below 1 mg/mi.
Given the progress to date and ongoing work, CARB anticipates that improved test procedures,
in conjunction with the implementation of strict laboratory filter handing and weighing
procedures, will allow consistent and reliable gravimetric PM measurement well below 1 mg/mi.
CARB believes that the automotive industry can make the necessary changes to their emission
certification laboratories in a timeframe that allows for an earlier implementation of the adopted
1 mg/mi standard than the 2025 model year. To that point, CARB recommends that U.S. EPA
consider adoption of a 1 mg/mi PM standard to be consistent with the standards CARB adopted
in the LEV III regulations, and commit to evaluate in the near future, concurrently with CARB,
whether the standard could be implemented even earlier than the scheduled 2025 model year
start date.
Comment 4 — Optional Phase-In Formula for FTP and US06 PM Standards: U.S. EPA is
proposing to phase-in Tier 3 FTP and US06 PM emission standards beginning with the 2017
model year with 100 percent compliance required in the 2021 model year. U.S. EPA has also
proposed an optional alternative phase-in schedule that manufacturers may use to comply with
the Tier 3 PM emission standards for these model years. This alternative phase-in schedule is
based on a mathematical equation that provides credits for vehicles certified to the Tier 3 PM
standards based on the year in which they are certified. Accordingly, vehicles certified to Tier 3
PM standards in earlier years would be worth more than those certified in later years. This
alternative phase-in schedule is similar to one CARB adopted for vehicles meeting the LEV III
PM standards. However, while the LEV 111 regulations explicitly state that 100 percent
compliance is required at the end of the phase-in period, the proposed Tier 3 regulations do not.
We believe that in order to maintain the emission benefits of the Tier 3 program, it is necessary
to add language to the Tier 3 rule to clarify that a manufacturer that certifies its vehicles to this
alternative phase-in schedule must still meet the requirement that 100 percent of those vehicles
meet the applicable PM standards in the 2021 model year.
[See CARB's public comment document EPA-HQ-OAR-2011-0135-4919-A1, received by the
docket on October 23, 2013, for two graphics: CARB Light-Duty SFTP Program (PFI Vehicles)
and CARB Light-Duty SFTP Program (GDI Vehicles)]
4-32
-------
Tier 3 Summary and Analysis of Comments
Emissions Control Technology Association (ECTA)
Proposed Tier 3 Limits for PM is Too High: We disagree with EPA's Tier 3 PM limit value and
believe it is too high.
EPA concludes that: "The intent of the proposed 3 mg/mi standard is to bring all light-duty
vehicles to the PM level typical of that being demonstrated by most light-duty vehicles today'.
The EPA has proposed a Tier 3 PM limit value that is too high. The 3 mg/mile proposed FTP PM
standard is achievable in most of today's vehicles, and by EPA's own estimate, is not a
challenging value. Given the toxicity of PM, wherein there is no "safe" level in ambient air, that
some emerging engine concepts, especially GDI engines, increase and change the PM emission,
and that many technologies are readily available to achieve much lower levels, EPA needs to at
least match the California LEVIII 1 mg/mile limit value, harmonizing the regulation throughout
the United States, and more-closely matching the effective level that will be implemented in
Europe in 2017. The typical PM emission throughout the major markets in 2025 will not be 3
mg/mi, but substantially less given the particle number limits today in Europe and Beijing, and
being considered in Japan. Setting the bar at levels achievable today for a regulation that will be
phased in through 2025 on such a critical emission as PM is not sound public policy, especially if
other major markets are tightening much more.
Numerous approaches exist to bring the most difficult engine type, GDIs, into tighter PM
compliance; namely fuel injection, EGR, and air handling options (Hyundai, IQPC Gasoline
Emissions Conference, 2012; Mercedes, Vienna Motorsymposium, 2013; AVL CTI Emissions
Conference, Detroit, May 2011), and gasoline paniculate filters. All of these options will be on
vehicles in Europe in 2017.
At the very least, EPA should set a 1 mg/mile standard, and subject it to a technology review.
The public health benefits will be realized by California and Europe (and later, Japan and China)
when engine technology is available, and measurement technology will meet the need. Without
doing this, it is conceivable the majority of Americans will have particulate emissions on some
vehicles in 2025 that will be more than three times higher than those in California, Europe,
Japan, and China.
Environmental Defense Fund (EDF)
EOF applauds the Agency for proposing a 3 mg/mi FTP PM standard for light-duty vehicles to
"ensure that all new vehicles perform at the level already being achieved by well-designed Tier 2
emission control technologies." EPA estimates that PM2.5 emissions will be reduced by an
estimated 7,500 tons annually in 2030 as a result of these proposed tailpipe standards. EDF
agrees that it is important to lock in the lower PM emissions rates already being achieved by Tier
2 vehicles. Moreover, the standard will help prevent any backsliding that could occur as
manufacturers move toward low carbon technologies to meet increasing greenhouse gas and fuel
economy standards.
4-33
-------
Chapter 4: Proposed Vehicle Emissions Program
We also strongly urge the Agency to finalize a more protective PM standard of 1 mg/mi
beginning in 2025. The more stringent standard is technically feasible and would provide even
greater human health protections from harmful PM emissions. California has adopted a 1 mg/mi
standard to begin in 2025.
A 1 mg/mi FTP PM Standard is Technologically Feasible by 2025. In the preamble, EPA states
that the "FTP PM standards that we are proposing are the most stringent technically feasible
standards within the implementation timeframe of this proposal." However, the Agency does not
explain why a 1 mg/mi FTP PM standard, as finalized in LEV III, is not technically feasible. The
Agency only states that more research is needed on PM measurement methods. Nonetheless,
EDF believes that a 1 mg/mi standard is technically feasible beginning in 2025, as evidenced by
available technology and many Tier 2 vehicles already achieving such a low emission rate.
As evidenced by EPA and CARB testing, Tier 2 vehicles on the road today are already
achieving PM emissions levels at 1 mg/mi or lower. As described in the RIA, EPA conducted a
test program to measure PM emissions from a variety of Tier 2 light-duty vehicles to help
establish the feasibility of Tier 3 PM standards. The test program included 17 late model year
vehicles that represented a significant volume of annual light-duty sales and included vehicles
that ranged from small cars to trucks, as well as four vehicles with GDI engines. The results of
the FTP emissions tests found that all but one of the non-GDI vehicles had PM emissions under
1 mg/mi. And of those, all but one had PM emissions under 0.5 mg/mi. The results of EPA's
own tests confirm that existing Tier 2 non-GDI vehicles are already meeting PM emissions levels
well below the proposed 3 mg/mi standard and all but one are well below a 1 mg/mi PM
standard. Moreover, these are Tier 2 vehicles on the road today, and a 1 mg/mi PM standard
would not begin for another 12 years, giving manufacturers ample time to ensure all new
vehicles can achieve that low emission rate.
While the GDI engines in EPA's tests had higher levels of PM, there is ample evidence that GDI
engines and other low carbon engine technologies needed to meet the most stringent GHG and
fuel economy standards can also achieve a 1 mg/mi PM standard (135). GDI engines achieve
greater efficiency by mimicking a diesel engine, and therefore can have PM emissions close to
those from a conventional diesel engine (one without a DPF). Indeed, all of the Tier 2 GDI
vehicles tested by EPA had PM emissions over 2 mg/mi, with the highest result at just over 7
mg/mi. However, according to CARB, "[c]ar makers who choose to pursue gasoline-fueled, CO2
friendlier GDI internal combustion engines for their future vehicles will have two principal
technical solutions for further reduction of PM mass emissions" - optimized fuel-injection
systems and gasoline particulate filters (GPFs). Recent research by the Manufacturers of
Emission Controls Association (MECA) and Environment Canada found that GPFs can bring
GDI PM emissions down to levels in line with, or below, those of a typical port fuel injection
engine.
Moreover, CARB estimates that the "expected trend is for new GDI vehicles to move towards
spray-guided GDI engines" and therefore, "compliance with the proposed [LEV III] PM
standards is not expected to impose a cost increase to vehicle manufacturers' (138).
4-34
-------
Tier 3 Summary and Analysis of Comments
EPA and CARB test results, combined with the evidence that gasoline particulate filters are
capable of reducing GDI engine PM emissions in line with conventional gasoline engines,
indicate that a 1 mg/mi PM standard is technically feasible by 2025. Therefore, we urge the
Agency to finalize the more protective standard that would reduce the health burden on all
Americans.
EPA has Ample Time to Finalize a Testing Method for a 1 mg/mi PM Standard. The Agency
states in the preamble that it is not considering a 1 mg/mi standard because of the need for
"continuing PM measurement method development." While we agree with EPA that additional
research and development is needed to reliably measure a 1 mg/mi PM standard, there is ample
lead-time for such development, and the Agency should not delay the technically feasible, more
health-protective 1 mg/mi standard for this reason.
The current method for measuring vehicle and engine PM emissions is determining the mass of
the PM emissions. In the past, this approach was quite effective because of the large amounts of
PM emitted by most vehicles. However, as diesel and gasoline vehicles standards become more
protective, accurately measuring the PM mass becomes more difficult. Over the years, EPA has
continued to make improvements to its PM mass test methods. These improvements have been
reflected most recently in the heavy-duty GHG final rule in 2011 and this Tier 3 proposal for
measuring the proposed 3 mg/mi PM standard. At the same time, a consortium of government
and industry representatives are currently conducting further research on how to best measure
PM emissions of 1 mg/mi and lower, including further improvements to PM mass test methods
as well as new solid particle counting methods, like those used in Europe. California's LEV III
rulemaking documents have an extensive description of the advancements made to both
procedures as of 2012, and indicate that significant progress has already been made in
developing accurate and reliable measurement techniques for a 1 mg/mi standard. In addition,
more current data and information about advancements in test procedures were presented at
recent conferences, including the SAE Congress and Coordinating Research Council Real World
Emissions Workshop. It is clear that EPA, CARB, and other government and industry members
are well on their way to developing a reliable test method for more protective standards. And
with an additional decade before the 1 mg/mi standards would take effect, the Agency has more
than sufficient time to finalize the most effective and appropriate test method and procedure.
EDF strongly urges the Agency to finalize the technically feasible PM standard of 1 mg/mi
beginning in 2025.
Ferrari
Ferrari considers the proposed.. .PM standards feasible in 2021 MY. We therefore support this
aspect of EPA's proposed rule.
International Council on Clean Transportation (ICCT)
The ICCT recommends that EPA harmonize with both the CARB 1 mg/mile particulate mass
standard starting with 2025 and the European particulate number standards. Currently, both
requirements are hindered by the lack of measurement precision, but continuing research into
particulate measurement should resolve these issues in the future.
4-35
-------
Chapter 4: Proposed Vehicle Emissions Program
The FTP supports the proposed FTP particulate standards. These would largely harmonize
EPA's requirements with CARB's.
One key difference between the Tier 3 proposal and the LEV III rule is that under LEV III,
automakers must meet a tailpipe emission standard of 1 mg/mi beginning in 2025. Instead, the
Tier 3 proposal extends only to model year 2023, citing concerns expressed in the CA LEV III
rulemaking with regard to the state of PM measurement capability to enable testing and
compliance with a 1 mg/mi standard.. The ICCT strongly recommends that EPA harmonize with
the LEV III particulate standards beginning in 2025. This will allow plenty of time to develop
more accurate particulate measurement methods.
A notable omission from the proposed particulate standards is a particle number limit standard,
similar to already adopted requirements in Europe. Although the proposal cites the 2010 US EPA
Integrated Risk Assessment for Particulate Matter, which highlights evidence of a causal
association between PN exposure and adverse health impacts, the document notes a desire for
further research to find more robust associations between PN exposure and health impacts. The
ICCT strongly encourages EPA to investigate harmonization with the European particulate
number standards in the future. California is likely to pursue particle number measurement
methods to ensure compliance with 1 mg/mi, which carries the possibility of a transition to a
particle number limit in future rulemakings.
Manufacturers of Emission Controls Association (MECA)
MECA strongly supported and agreed with ARB's decision to include a 1 mg/mile particle
matter standard for light-duty vehicles over the FTP test cycle in their LEV III requirements. In
the Tier 3 proposal, EPA proposed only to harmonize with the LEV III 3 mg/mile FTP PM
standard and not propose a 1 mg/mile FTP PM standard. The 2012 decision by the European
Commission to establish a particle number emission standard for light-duty vehicles powered by
gasoline direct injection (GDI) engines as a part of their upcoming Euro 6 light-duty emission
standards provides a more stringent particle emission limit for these GDI vehicles in the same
time frame as the Tier 3/LEV III 3 mg/mile PM standard (proposed phase-in for the Tier 3, 3
mg/mile PM standard starts in 2017 and is fully phased-in with the 2021 model year;
implementation of the Euro 6 GDI particle number limit of 6 X 1011 particles/km [equivalent to
the Euro 5 light-duty diesel particle number limit], measured using the European PMP particle
measurement protocol, begins in September 2017; see:
ec.europa.eu/enterprise/sectors/automotive/documents/directives/motor-vehicles/index_en.htm).
This European light-duty GDI particle number limit will cause auto manufacturers to introduce
cleaner technologies, such as advanced fuel injection systems and/or gasoline particulate filters,
to comply with the European Euro 6 GDI particle number limit. Auto manufacturers are already
working to bring forward early introductions of these ultra-low PM, Euro 6-compliant gasoline
engines to the European market in the coming 12 to 18 months (European member states are
permitted to introduce tax incentives for early introductions of Euro 6 vehicles prior to the first
implementation dates of September 2014 for new models and September 2015 for all passenger
car models). Nearly all auto manufacturers that sell into the European market are working with
MECA members on potential applications of particulate filters on gasoline direct injection
vehicles.
4-36
-------
Tier 3 Summary and Analysis of Comments
Gasoline particulate filters (GPFs) are based on the same, wall-flow ceramic filters that have
been successfully applied on millions of light-duty and heavy-duty diesel vehicles in Europe and
the U.S. for more than 10 years. The performance and application of these gasoline particulate
filters has been highlighted in a number of recent technical publications in both the U.S. and
Europe (e.g., SAE paper nos. 2010-01-0365, 2011-01-0814, and 2013-01-0836; SAE paper no.
2013-01-0527 authored by Environment Canada and MECA). Like diesel particulate filters,
gasoline particulate filters are capable of reducing particle emissions by more than 85% over a
wide range of particle sizes, including high capture efficiencies for ultra-fine particulates. The
application of a GPF on a four-cylinder gasoline direct injection vehicle is expected to cost
approximately $100-120 (see ICCT's GPF cost estimate available here:
www.theicct.org/estimated-cost-gasoline-particulate-filters), making this emission control
technology a cost-effective solution for reducing particulate emissions from future gasoline
vehicles. When these filters are properly designed, the impact of a GPF installation on the
backpressure and fuel-efficiency of the vehicle is expected to be minimal.
EPA needs to make sure that these same ultra-low PM, Euro 6 GDI engine/emission
technologies are also utilized in the U.S. To that end, MECA believes that it is important for
EPA, at a minimum, to harmonize with ARB's LEV III, 1 mg/mile light-duty vehicle PM FTP
standard to maximize the public health benefits associated with reducing public exposure to
particulate emissions from future light-duty vehicles. Some consideration should also be given to
aligning with the European Euro 5/Euro 6 diesel/GDI particle number limit, especially if EPA
and ARB believe that there are measurement issues with a 1 mg/mile PM standard. Based on
information presented by ARB at the 2013 CRC On-Road Vehicle Emissions Workshop held in
San Diego in April 2013 and presented to MECA in May 2013, ARB believes that there is a
pathway to measuring PM emissions at levels below 1 mg/mile. ARB has published a revised
PM mass measurement protocol that is part of their pathway to measuring very low PM mass
levels from the exhaust of a vehicle (see ARB test method MLD 145 available at:
www.arb.ca.gov/testmeth/slb/exhaust.htm). Ford researchers have also developed a correlation
between particle number measurements and particle mass that can provide an alternative
pathway to measuring very low PM mass levels (see M. Matti Mariq and Ning Xu, Aerosol
Science 35 (2004), pp. 1251-1274). ARB adopted their 1 mg/mile PM standard to provide
additional public health protection to exposure to particulate emissions from vehicle emissions
and EPA needs to follow California's lead in harmonizing with this very tight PM standard. EPA
and ARB need to continue to work together and reach agreement on measurement protocols that
are acceptable for use with a 1 mg/mile FTP PM standard. [EPA-HQ-OAR-2011-0135-4675-A2,
p. 4] Finally, MECA would like to reiterate our strong support for EPA harmonizing with
ARB's LEV III 1 mg/mile FTP PM standard. In July 2013 MECA released a new report on
ultrafine particulate (UFP) emissions entitled, "Ultrafine Particulate Matter and the Benefits of
Reducing Particle Numbers in the United States." The report summarizes the current
understanding of the potential adverse health impacts of UFPs; outlines the various control
strategies and technologies that can be used to meet current and upcoming U.S. EPA and
California ARB emission standards (including LEV III and EPA's proposed Tier 3 standards);
and documents the success story of using diesel particulate filters (DPFs) to meet and exceed
U.S. and European emission standards. Notably, the report highlights a correlation between
particle number (PN) and PM that can be used in conjunction with PM-based health data to
estimate the health benefits of reducing particle number emissions, and indicates that a PN
4-37
-------
Chapter 4: Proposed Vehicle Emissions Program
measurement may offer a more robust unit for determining compliance at very low PM mass
levels. In addition, the report quantifies the health benefits of the additional emission reductions
that are realized when DPFs or gasoline particulate filters (GPFs) are used compared to only
engine-based strategies. With respect to light-duty vehicles, the report echoes many of the
comments made by MECA with respect to the expected dominant use of GDI engines in the U.S.
because of their improved fuel economy versus port injected gasoline engines, the higher particle
mass and number emissions of GDI engines relative to port injected engines, and the
recommendation that EPA follow California's lead in including a 1 mg/mile PM FTP limit in its
final Tier 3 standards. Gasoline particulate filters are a cost effective emission control technology
option for meeting a 1 mg/mile FTP PM standard, and GPFs are expected to be introduced in
Europe in the near future on some GDI models to meet the Euro 6 GDI PN limit of 6 X 1011
particles/km. As discussed in this report, compliance at the 1 mg/mile PM level provides
significant additional health benefits beyond the benefits included by EPA in their Tier 3
proposal. The full MECA report on ultrafine particulates is available on MECA's public website,
www.meca.org, under Resources » Reports. It is important for the United States to continue to
set the bar on light-duty vehicle emission standards in order to encourage the development and
use of best available control technologies for light-duty vehicles. EPA has a long history of
setting technology-forcing vehicle standards based on the public health benefits they provide and
this leadership needs to continue with respect to light-duty vehicle particle emission standards.
In addition, MECA asks EPA to harmonize with ARB's 1 mg/mile FTP PM standard and to set
tighter PM limits for the US06 test cycle.
Natural Resources Defense Council (NRDC)
EPA should establish a process to tighten particulate matter standards and incorporate particulate
matter number into future standards.
EPA's proposed particulate matter (PM) Federal Test Procedure (FTP) standard of 3 mg/mi is
less health-protective than a 1 mg/mi standard, which the California Air Resources Board
(CARB) recently adopted. EPA did not adopt a 1 mg/mi standard because the agency believes
that equipment that can reliably and consistently measure PM levels at and below the standard
needs further development. However, because CARB has set a 1 mg/mi standard such equipment
will exist by at least model year 2025.
EPA should establish a process with CARB to assess the development of needed measurement
technologies. If the joint agency assessment shows that sufficient measurement technologies are
feasible, EPA should tighten the PM FTP standard to be at least as stringent as California's with
regard to standard levels and phase-in schedules.
EPA should also evaluate the use of particle number as an additional basis for future PM
standards, especially to set standards that will reduce ultrafine particle pollution. As history has
shown with PM10 and then PM2.5 studies, attention to smaller and smaller particles has been
critical to protecting human health. Particle number measurements may serve to bolster
verification of controls of very low masses of fine particles (1-3 micrometers) but also become
the basis for control of ultrafine particles (<1 micrometer).
4-38
-------
Tier 3 Summary and Analysis of Comments
EPA acknowledges the potential significant health threat from ultrafmes, stating that the
agency's Integrated Science Assessment for Particulate Matter "concludes that there is
suggestive evidence of a causal relationship between short-term exposures and cardiovascular
effects, such as changes in heart rhythm and blood vessel function. It also concludes that there is
suggestive evidence of association between short-term exposure to ultrafme particles and
respiratory effects. Data are inadequate to draw conclusions regarding the health effects
associated with long-term exposure to ultrafme particles".
In its review of measurement and control technologies for tighter standards targeting PM2.5,
EPA should also be evaluating methods to count and control ultrafme particles in future
standards.
Northeast States for Coordinated Air Use Management (NESCAUM)
NESCAUM strongly supports the proposed PM standard of 3 mg/mi for all light-duty vehicles,
light-duty trucks, and medium-duty passenger vehicles for all model years through 2024.
However, we feel that the standard should be phased down to 1 mg/mi from 2025 to 2028,
consistent with the requirements of the California LEV III program. We note that encouraging
progress is being made with respect to reliability of advanced PM measurement techniques, and
we share CARS's view that lead time for the 1 mg/mi phase-in is sufficient for appropriate
measurement techniques to be perfected and validated. We urge EPA to fully harmonize its PM
standards with CARB's and to work with CARB to monitor and support continued progress in
the development of PM measurement techniques.
Sierra Club, Clean Air Watch, Respiratory Health Association
Fine particulate matter is recognized as extremely hazardous to health, causing lung disease and
contributing to premature death. The proposed standards seek to reduce emissions of particulate
matter by proposing a FTP PM certification standard of 3 mg/mi for all model years. The
proposed SFTP PM certification standard is much higher, either 10 mg/mi or 20 mg/mi,
depending on the vehicle.
It is important to note that as part of their Low Emission Vehicle III (LEV 3) standards, the
California Air Resources Board is proposing lower certification standards and will be phasing in
a 1 mg/mi FTP standard beginning in MY 2025.
We urge EPA to coordinate with CARB to determine how to measure ultra-fine particles and to
consider harmonizing PM standards with those set by CARB in LEV 3.
Union of Concerned Scientists (UCS)
Particulate Matter Emissions Standard: EPA should harmonize with the California Air Resources
Board's (CARB's) Low Emission Vehicle III (LEV III) standard for particulate matter
emissions.
Particulate matter (PM) can travel deep into the lungs and harm human health. Many scientific
studies have linked exposure to PM with respiratory ailments, chronic bronchitis, asthma attacks,
4-39
-------
Chapter 4: Proposed Vehicle Emissions Program
heart attacks and premature death, particularly within sensitive populations. Much progress has
been made in the last decade to limit PM emissions, but more must be done.
The Union of Concerned Scientists supported CARS's decision to issue a 1 mg/mile standard for
light duty vehicles in its Low Emission Vehicle III (LEV III) program and urges EPA to set the
most stringent PM exhaust emission standard possible as part of this rulemaking. This can most
easily be accomplished by working with CARB on its development of measurement protocols in
order to harmonize the Tier 3 program with LEV III.
Additionally, EPA should strongly consider the inclusion of a particulate matter number limit in
addition to the PM mass limit for exhaust emissions. The European Commission has recently
adopted particulate matter number limits in order to address technology adoption that has
resulted in low PM mass but high PM numbers due to an increase in the ultrafme (<100 nm) PM
emissions. Adopting a PM number limit (in coordination with CARB) will decrease harmful
emissions and also result in a standard that is easier to measure. A well-crafted PM number limit
will ensure the continued protection of public health by limiting exposure to particulate
emissions and driving advancements in vehicle and emission control technologies.
Our Response:
Most commenters were generally supportive of the proposed 3 mg/mi FTP PM standard.
CARB commented that they agree with EPA that these standards are readily achievable with no
additional technology or hardware. In addition, several commenters, including CARB and
several NGOs and auto industry suppliers, supported a more stringent standard of 1 mg/mi,
which the California LEV III program phases in beginning in MY 2025. Some commenters
agree with EPA that additional research and development is needed to reliably measure a 1
mg/mi PM standard, but believe that there is ample lead time for such development. Another
commented that, currently, both a 1 mg/mi US06 standard and a potential particle number (PN)
requirement are hindered by the lack of measurement precision, but they believe that continuing
research into particulate measurement should resolve these issues in the future. After careful
consideration of these comments and information available at this time, we continue to believe
that the PM standards that we are finalizing for the federal Tier 3 program are the most stringent
technically feasible standards within the implementation timeframe of this rule. (See Chapter
1.5.1 of the RIA describing EPA testing and PM emissions results.)
We will continue to work closely with CARB in this area. Specifically, our agencies will
continue our parallel evaluations of how improved gravimetric PM measurement methods can
reduce PM mass measurement variability at very low PM levels and how this relates to the
evolving technological capabilities of automakers to reach very low PM levels with sufficient
compliance margins. Also, as suggested by several commenters, EPA will continue to follow
research on the association between particle number (PN) and health effects and into the accurate
measurement of PN. Based on our future findings, we will determine whether any future action
is appropriate.
PM emissions over the FTP are generally attributed to cold-start operation, when PM
formation from combustion of the fuel is facilitated by the operating conditions, including a cold
4-40
-------
Tier 3 Summary and Analysis of Comments
combustion chamber and fuel enrichment. During cold-start operation, PM control is less
effective, especially the oxidation by the catalytic converter of semi-volatile organic compounds
from the lubricating oil. We believe that for vehicles that are not already at the Tier 3 levels, the
new standards can be achieved with improvements to the fuel controls during the cold start,
without the need for any new technology or hardware. We also expect that manufacturers will
pay close attention to maintaining low PM emissions during the implementation of newer
technologies like gasoline direct injection (GDI) and turbocharged engines. Improvements in
cold-start exhaust catalyst performance for NMOG+NOx control will also reduce emissions of
semi-volatile organic PM. For these reasons, cold start PM levels are relatively independent of
vehicle application and therefore we are finalizing a single FTP PM standard for all light-duty
vehicles, as proposed.
The Alliance and Global Automakers support fleet-average standards for PM, instead of
the proposed per-vehicle standards. We do not believe that fleet average standards are
appropriate in the case of FTP (and SFTP/US06 PM) because the purpose of the form and level
of the standards (i.e., per-vehicle versus fleet-average) is different than that of the NMOG+NOx
standards. For the NMOG+NOx standards, we designed the declining standards to require
increasing technological progress and gradually improving emissions. For the PM standards,
many or most manufacturers are already meeting the 3 mg/mi standard with well-performing
vehicles, and our intent is to bring all other vehicles to the same PM performance. An averaging
system in this context would not be appropriate, and possibly counterproductive, since fleet
averaging could allow some or all of the higher-emitting current vehicles to remain at those
unnecessarily high levels. For these reasons, we are finalizing the per-vehicle form of the FTP
PM standards.
CARB commented that we should require that a manufacturer that certifies its vehicles to
the proposed alternative phase-in schedule still meet the requirement that 100 percent of those
vehicles meet the applicable PM standards in the 2021 model year. The proposed Tier 3
alternative phase-in schedule includes MY 2021 as the final year when a manufacturer could
have the flexibility of not meeting the 100 percent target if they outperform the targets in
MY2017 through 2020. However, we believe that including MY2021 in the alternative phase-in
schedule provides a reasonable amount of time with which to achieve compliance. We do not
believe that the duration of the phase-in will negatively impact the PM emission reductions of
the overall program, because it will require manufacturers to exceed the obligations that would
have otherwise applied under the primary phase-in in an earlier model year. We are requiring
that manufacturers meet the 100 percent compliance requirement in MY2022.
4.1.2.6. Specific Comments on US06 PM Standards
What Commenters Said:
Alliance of Automobile Manufacturers (Alliance) and Association of Global Automakers
(Global)
4-41
-------
Chapter 4: Proposed Vehicle Emissions Program
The red text shows the discrepancy between the EPA Tier 3 and the ARB LEV III regulations.
We appreciate that EPA recognizes this disconnect between the regulations and allows
manufacturers to count vehicles in the 6,000-8,500 pounds GVWR weight class toward
compliance with the phase-in percentage (i.e., numerator only) in MY 2017 (we would note what
appears to be a typo in §86-181 l-17(b)(6)(i)). However, this would not benefit manufacturers
with a small volume of vehicles in the weight class. To harmonize with LEV III, we recommend
adding an option for manufacturers to comply with the MY 2017 phase in by producing 10
percent of their 0-8,500 pounds GVWR fleet to the revised PM Standard.
Recommendation: We recommend the following changes (including a correction to the typo
mentioned above) to §86-181 l-17(b)(6)(i):
'Tier 3 FTP standard for PM must also meet the Tier 3 US06 standard for PM. In model year
2017, the phase-in requirement does not applyjes only for to vehicles at or below above 6,000
pounds GVWR; however, if you certify these vehicles to the Tier 3 PM standards in the model
year 2017, you may count those projected U.S. sales toward your calculation for meeting the
phase-in percentage for that year (numerator only). Alternatively, you may comply in the model
year 2017 meeting the FTP and the US06 PM standards with 10 percent of your projected
nationwide sales of all vehicles (including those greater than 6,000 pounds GVWR).'
The NPRM proposes to adopt new US06 PM standards to control PM emissions under the high
speeds and loads encountered on the US06 cycle. The standards proposed harmonize with the
LEV III US06 PM standards of 10 mg/mile for vehicles <6,000 pounds GVWR, and 20 mg/mile
for vehicles >6,000 pounds GVWR. In §IV(A)(4)(b) of the Preamble to the Tier 3 NPRM, EPA
asks for comment on the proposed Tier 3 US06 PM standards, including whether EPA should
adopt a standard of 10 mg/mile for all vehicles (both less than and greater than 6,000 pounds
GVWR).
First, we would note that light-duty vehicles and light-duty trucks are a very small fraction of
PM emissions, accounting for less than two percent of total PM10 emissions and less than four
percent of PM2.5 emissions. The other 98 and 96 percent of PM emissions, respectively, come
from sources such as cooking, farming operations, construction and demolition, etc.
The US06 PM Standard is an entirely new test for which the industry and the agencies are still
gathering and analyzing the emissions data. Further complicating the standards are the likely
introduction of new technologies to meet the GHG standards. Given this, we are concerned with
the technical feasibility of achieving even the proposed SFTP US06 PM standard of 10 mg/mile
for passenger cars and trucks under 6,000 pounds GVWR and 20 mg/mile for light-duty trucks
over 6,000 pounds GVWR. In particular, technologies needed to meet the GHG requirements
may conflict with the US06 PM standards. Future low-powered/downsized technologies and
range extenders needed to meet the GHG requirements, for example, may not be able to comply
with the proposed PM standards.
Another point of concern is that the PM standards are flat standards that all vehicles must meet,
as compared to fleet average standards that are proposed for NMOG+NOx compliance. Thus the
risk that a particular type of technology would be precluded by the PM standards is increased
4-42
-------
Tier 3 Summary and Analysis of Comments
because such technology cannot be certified to a higher bin and offset by a different technology
certified to a lower bin.
Moreover, requiring full-size pickups and SUVs that make up the over 6,000 pounds GVWR
category to meet the same flat standard that smaller and lighter vehicles in the under 6,000
pounds GVWR category must meet is effectively making the full-size pickups and SUVs meet a
more stringent standard, given the higher engine throughput associated with these larger, heavier
vehicles.
It is also worth noting that the PM standards begin at a time when manufacturers enter the eighth
year of the most stringent GHG standards in history, when the number of ZEVs and PHEVs
jumps considerably in California and the Section 177 states, and while reducing exhaust
emissions by 75 percent on both the FTP and SFTP and phasing in zero evaporative emission
standards on all vehicles.
Recommendation: Taking all of this into consideration, industry does not oppose adopting the
proposed US06 PM standards that will harmonize the LEV III and Tier 3 standards. However,
we oppose adoption of 10 mg/mile PM standard on the US06 for all light-duty trucks over 6,000
pounds GVWR. We are concerned that such standards are not feasible given the technological
changes necessary to meet all of the other emission requirements manufacturers must meet in the
exact same timeframe. Moreover, such standards are not warranted.
American Lung Association
We believe the Supplemental FTP PM emissions standards are not sufficiently aggressive and
are, in reality, a non-standard. EPA has proposed a Supplemental FTP PM emissions standard of
10 mg/mile which would allow far more PM emissions than existing vehicles currently emit.
According to a memorandum describing EPA's own testing, no vehicles either above or below
6,000 pounds gross vehicle weight rating emitted more than 3.5 mg/mile and most were well
below that level. We urge EPA to set the tightest feasible Supplemental FTP PM emissions
standards, which would be no greater than 4 mg/mile.
California Air Resources Board (CARB)
Comment 3 - Set lower PM standards Over the Supplemental Federal Test Procedure — US06
Cycle: The proposed Tier 3 program includes US06 PM standards of 10 mg/mi for lighter
vehicles and 20 mg/mi for heavier vehicles. Based on recent PM emission data generated by both
CARB and U.S. EPA test programs, CARB believes that a more stringent US06 PM standard of
4 mg/mi applicable to both vehicle categories is feasible. The following is a summary of these
test programs as well as CARB's rationale for the recommendation.
Prior to California's LEV 111 rulemaking, eight California LEV II-certified light-duty vehicles,
which included five gasoline direct injection vehicles and three port fuel injection vehicles, were
tested for PM emissions over the US06 cycle in CARB's Haagen-Smit laboratory. Measured PM
emissions from seven out of eight vehicles were below 3 mg/mi, including all five gasoline direct
injection vehicles. A gasoline medium-duty passenger vehicle (gross vehicle weight rating
4-43
-------
Chapter 4: Proposed Vehicle Emissions Program
between 8,501-10,000 Ibs) with an odometer reading of approximately 50,000 miles was also
tested and achieved a US06 PM emission value of 0.8 mg/mi, comparable to those measured
from the lightduty test vehicles. One particular light-duty test vehicle had PM emissions in
excess of 3 mg/mi, but CARB suspects that because the vehicle was not designed or certified to a
stringent US06 PM standard, the calibration strategy may not have been optimized for PM
control over the US06 and further refinement to the calibration would likely address the issue
including minimizing any enrichment events that directionally increase emissions.
Recently, U.S. EPA shared PM emission data from its full useful life program with CARB. The
primary objective of the full useful life program was to establish the feasibility of Tier 3 PM
standards at vehicle full useful life and investigate the contribution of lubricating oil to PM
emissions in Tier 2 vehicles. More than two dozen passenger cars and light-duty trucks were
tested for PM emissions on the FTP and US06 cycles. Initial results showed high PM emissions
in excess of 40 mg/mi from some of the vehicles over the US06 cycle. U.S. EPA later
determined that silicone contamination from the use of silicone exhaust transfer tubes in the
emission sampling system was a major contributor to the measured PM mass emissions,
especially on test cycles that generate high exhaust temperatures such as the US06 cycle. As a
result, U.S. EPA re-tested a dozen vehicles and the new data show that all twelve vehicles,
including a suspected oil burner, emit at levels less than 4 mg/mi PM on the US06' cycle.
In response to U.S. EPA's recent US06 testing and the discovery of the silicone contamination
issue, CARB has initiated a test program to further investigate PM emissions on the US06 cycle
and determine the feasibility of a more stringent standard: The test program aims to gather US06
PM emission data on both current and future engine technologies, such as port fuel injection,
gasoline direct injection, turbocharging, and engine downsizing, and the test vehicles are being
selected accordingly. While the test program is on-going and vehicles are still being tested, thus
far, US06 PM data have been collected for eight port fuel injection vehicles, including one light-
duty truck and two gasoline direct injection vehicles. PM emissions over the US06 cycle from
seven out of the ten vehicles were measured below 1 mg/mi while nine out often were measured
below 4 mg/mi. The one vehicle that had PM emissions above 4 mg/mi also showed FTP
emissions that exceeded the LEV II FTP PM standard, so it is likely that the vehicle was not
operating within design specifications at the time of testing.
While the findings of the above test programs clearly show that current LEV II vehicles which
are not subject to PM standards under the LEV II SFTP program, are capable of complying with
a 4 mg/mi US06 PM standard, there have been some concerns regarding future technologies,
such as advanced gasoline direct injection and turbocharged, downsized engines. While these
future technologies are expected increase cylinder pressures and temperatures and possibly the
need for additional engine temperature control, CARB still believes the 4 mg/mi US06 PM
standard is feasible. This is because affected Tier 3 vehicles will be equipped with improved PM
control technologies needed to comply with the 3 mg/mi PM standard on the FTP (in lieu of the
10 mg/mi PM standard that LEV II vehicles are certified to) and because additional calibration
opportunities will exist as even today's LEV II vehicles, which are calibrated without PM control
considerations over the US06 cycle, generally have emission levels below 4 mg/mi with
compliance margin.
4-44
-------
Tier 3 Summary and Analysis of Comments
In summary, while CARB will continue to collect US06 PM emission data on future
technologies, CARB believes the data currently available sufficiently supports a more stringent
standard and recommends that U.S. EPA adopt a US06 PM standard of 4 mg/mi in the Tier 3
program. Additionally, should U.S. EPA adopt a more stringent standard in the final proposal,
CARB recommends that the corresponding in-use standards also be lowered in accordance with
any changes to the certification standards.
Comment 4 — Optional Phase-In Formula for FTP and US06 PM Standards: U.S. EPA is
proposing to phase-in Tier 3 FTP and US06 PM emission standards beginning with the 2017
model year with 100 percent compliance required in the 2021 model year. U.S. EPA has also
proposed an optional alternative phase-in schedule that manufacturers may use to comply with
the Tier 3 PM emission standards for these model years. This alternative phase-in schedule is
based on a mathematical equation that provides credits for vehicles certified to the Tier 3 PM
standards based on the year in which they are certified. Accordingly, vehicles certified to Tier 3
PM standards in earlier years would be worth more than those certified in later years. This
alternative phase-in schedule is similar to one CARB adopted for vehicles meeting the LEV III
PM standards. However, while the LEV 111 regulations explicitly state that 100 percent
compliance is required at the end of the phase-in period, the proposed Tier 3 regulations do not.
We believe that in order to maintain the emission benefits of the Tier 3 program, it is necessary
to add language to the Tier 3 rule to clarify that a manufacturer that certifies its vehicles to this
alternative phase-in schedule must still meet the requirement that 100 percent of those vehicles
meet the applicable PM standards in the 2021 model year.
[See CARB's public comment document EPA-HQ-OAR-2011-0135-4919-A1, received by the
docket on October 23, 2013, for two graphics: CARB Light-Duty SFTP Program (PFI Vehicles)
and CARB Light-Duty SFTP Program (GDI Vehicles)]
Environmental Defense Fund (EDF)
EPA Should Adopt Most Protective US06 PM Standard - Not Based on Vehicle Weight: EPA
seeks comment on the use of vehicle weight to establish separate US06 PM standards for cars
and trucks. 143 EDF asks the Agency to adopt one protective US06 PM standard for all vehicles
subject to the Tier 3 program, regardless of weight. [EPA-HQ-OAR-2011-0135-4355-A1, p. 27]
EPA is proposing a US06 PM standard of 10 mg/mi for vehicles at or below 6,000 Ibs GVWR
and a standard of 20 mg/mi for vehicles over 6,000 Ibs GVWR. EDF believes both of these
standards are too lenient and the Agency should adopt a single, more protective standard for all
vehicles subject to Tier 3. As EPA states in the preamble, "today's heavier vehicles are already
achieving PM emission levels well below our proposed 20 mg/mi standard and are
approximately equivalent to the performance of lighter vehicles." Indeed, EPA's data presented
in the RIA show that all of the vehicles tested, both above and below 6,000 Ibs, achieved a US06
PM emissions result of less than 4 mg/mi. And as illustrated in EPA's Figure below, similar
emissions rates were recorded for both light and heavy vehicles, indicating that heavier vehicles
are capable of meeting the same US06 PM requirements as the lighter vehicles. EPA's test
results also indicate that not only can all vehicles, regardless of weight, meet the same standard,
but that standard can and should be set well below the 10 mg/mi standard proposed by EPA for
4-45
-------
Chapter 4: Proposed Vehicle Emissions Program
vehicles under 6,000 Ibs GVWR. We recommend that EPA finalize a US06 PM standard for all
vehicles subject to the Tier 3 program of no greater than 4 mg/mi.
Ferrari
Ferrari considers the proposed PM standards feasible in 2021 MY. We therefore support this
aspect of EPA's proposed rule.
Ford Motor Company (Ford)
Consistent with the CA LEV III program, EPA has proposed two separate US06 paniculate
matter (PM) standards for light-duty vehicles: 10 mg/mi for vehicles less than 6,000 pounds
GVWR, and 20 mg/mi for vehicles over 6,000 pounds GVWR. EPA has asked for comment on
these standards and also whether a single, 10 mg/mi standard applied to all light-duty vehicle
(LDV) classes would be more appropriate. Ford agrees with the Alliance comments on this issue
and strongly opposes the application of a 10 mg/mi standard to vehicles over 6,000 pounds.
Unlike PM generation on the Federal Test Procedure (FTP) which is typically dominated by the
"cold-start" portion of the cycle-PM on the US06 is more closely related to the work the vehicle
is required to perform over the cycle, which naturally increases with vehicle mass. Requiring
heaver vehicles to meet the same PM limits as lighter vehicles would, in effect, increase the
stringency of the PM standards for these vehicles at a time when automakers must introduce new
technologies to meet challenging GHG standards. This increased stringency would be
disproportionally applied to full-line manufacturers. The higher 20 mg/mi standard will help
ensure that similar fuel economy / greenhouse gas advances can be applied to both lighter and
heavier vehicles. US06 PM emissions are also affected by several factors related to the high
exhaust temperatures characteristic of this drive cycle, including: 1) the need for catalyst
protection, 2) potential desulfurization of the catalyst, and 3) storage/release artifacts from the
exhaust and sampling systems. As vehicle size increases, so does the thermal energy in the
exhaust. Above 6,000 pounds GVWR, this leads to quicker and more extensive heating of the
catalyst and exhaust system during the US06 cycle that exacerbates these issues relative to
lighter vehicles.
Recommendation: Ford recommends that EPA adopt the light-duty US06 PM standards as
proposed in the Tier 3 NRPM, which specifies a 20 mg/mi standard for LDVs above 6,000 Ibs
GVWR.
International Council on Clean Transportation (ICCT)
As cold starts have a relatively small impact on particulate emissions, the same times two factor
found for SFTP NMHC+NOx emissions should also be applied to particulate emissions. This
means that the SFTP PM standard should be set at no more than 6 mg/mi.
Further, there is no reason why light-duty vehicles over 6,000 GVWR should be held to less
stringent particulate standards. This violates the premise established with the Tier 2 emission
standards that all light-duty vehicles should be held to the same emission standards. It is
4-46
-------
Tier 3 Summary and Analysis of Comments
especially important that GVWR not be used to discriminate between different standards, as
GVWR is easily gamed.
The problem is far worse for participate matter. As the proposed SFTP participate standard is
3.3 times larger for light vehicles and 6.7 times larger for heavy duties — heavier vehicles than
the FTP particulate standard.
Supplemental comments from ICCT:
The International Council on Clean Transportation (hereafter, "ICCT") is submitting
supplemental comments specifically on SFTP PM (particulate matter) standards. Our primary
comments were submitted to the docket with our letter of July 1, 2013 and the information in that
submission will not be repeated here.
In our initial submission, the ICCT discussed the reasons why the proposed SFTP standards are
too lenient and will not be effective. Setting the SFTP standards properly is especially important
for diesel engines, as diesel emission control hardware requirements are largely set by the high
load conditions on the SFTP. The ICCT recommended that the SFTP particulate standards be set
at no more than 6 mg/mi.
We are submitting additional comments on SFTP PM standards in response to the recent release
of test data on US06 PM2.5 emissions by CARS.1 CARB recently tested 11 light-duty vehicles
and trucks over the US06 and measured PM emissions. Results are presented in Figure 1 and the
vehicle descriptions are in Table 1.
Seven of the eleven vehicles tested had PM emissions much less than 2 mg/mile. The Camry and
the Explorer had PM emissions between 2 and 4 mg/mile. Only the Optima and the Juke
exceeded 4 mg/mile.
Both the Juke and the Optima had GDI (gasoline direct fuel injection) engines. The literature has
suggested that GDI engines with poor combustion design and wall-guided fuel injection could
have higher PM emissions. However, the literature has also suggested that GDI engines with
good design and spray-guided fuel injection could have low PM emissions. This is supported by
the test results on the Ford Fusion, which also has GDI and has PM emissions of just over 1
mg/mile.
In addition, the Juke failed particulate standards on the FTP. This means that this vehicle may
not have been operating properly or has a poorly designed GDI system. In any case, the fact that
this vehicle was out of compliance means that the data should not be used when setting the SFTP
PM standards.
While only two light duty trucks were tested, their emissions were also reasonably low,
suggesting that light duty truck standards should be set at the same level as light duty vehicle
standards. Further, there is no reason why light-duty vehicles over 6,000 GVWR should be held
to less stringent particulate standards. This violates the premise established with the Tier 2
emission standards that all light-duty vehicles should be held to the same emission standards. It
4-47
-------
Chapter 4: Proposed Vehicle Emissions Program
is especially important that GVWR not be used to discriminate between different standards, as
GVWR is easily gamed.
Finally, it is extremely important to note that these PM emissions were measured in vehicles with
existing emission controls. The proposed reductions in FTP PM emissions will require better
combustion chamber and fuel injection designs, resulting in lower SFTP PM emissions as well.
In summary, the recent CARB test data completely supports ICCT's recommendation in our
initial comments, that the SFTP PM standards should be set at no more than 6 mi/mile. In fact,
the test data suggests that even 6 mg/mile would be too lenient and too easy to meet.
Manufacturers of Emission Controls Association (MECA)
With respect to the proposed PM limits for the supplemental FTP test cycles, EPA acknowledged
that their Tier 3 proposal is flawed by a dataset that included contamination with respect to the
PM analyses that were done on Tier 2-compliant vehicles. ARB shared with MECA in late May
2013 some of the revised, corrected Tier 2/LEVIIUS06 test cycle results. These data support a
much tighter PM standard for the US06 test cycle than proposed by EPA for Tier 3. MECA
understands that ARB intends to propose that EPA set a Tier 3 US06 PM limit of approximately
4 mg/mile for all light-duty vehicles (as opposed to the proposed Tier 3 US06 PM limits that
depend on vehicle weight) based on the testing they are expected to complete before July 1,
2013. MECA is supportive of ARB's Tier 3 comments on this subject and asks that EPA (and
ARB) set the tightest, feasible US06 PM standard in their final Tier 3 regulation.
National Association of Clean Air Agencies (NACAA)
US06 PM Standards — For the Supplemental Federal Test Procedure EPA has proposed that the
standards for PM be met based on the US06 test, which represents aggressive highway driving,
since the greatest concern regarding PM formation and sensitivity of engine controls is due to
high-speed, high-load driving conditions. In particular, the agency has proposed a US06 PM
standard of 10 milligrams per mile (mg/mi) for vehicles at or below 6,000 pounds gross vehicle
weight rating (GVWR) and a standard of 20 mg/mi for heavier light-duty vehicles, to be phased
in over a five-year period beginning in 2017. However, given EPA data showing that
manufacturers appear to be controlling PM emissions from heavier light-duty vehicles over
severe duty cycles, EPA requests comments on whether it should adopt a common US06
standard of 10 mg/mi for all light-duty vehicles. NACAA has reviewed EPA US06 PM
emissions test data provided in a March 1, 2013, agency memorandum available in the Tier 3
rulemaking docket (4). According to the test data shown in Figure 8 (on final US06 PM emission
results) of that memorandum, it is clear that US06 PM emission results for vehicles under and
over 6,000 pounds GVRW are far below 10 mg/mi. In fact, no vehicle tested is over 4 mg/mi and
most are substantially lower. Given the significance for air quality and public health of reducing
PM emissions, NACAA recommends that EPA adopt US06 PM standards below 10 mg/mi for
all affected light-duty vehicles — under and over 6,000 pounds GVWR — as supported by the
agency's test data.
Union of Concerned Scientists (UCS)
4-48
-------
Tier 3 Summary and Analysis of Comments
Millions of Americans breathe cleaner air as a result of our nation's clean air laws but serious
challenges remain. More than 1 in 3 Americans still live in areas where air pollutant levels
exceed at least one of the health-based National Ambient Air Quality Standards. Passenger
vehicles remain the second largest emitters of nitrogen oxides (NOx) and volatile organic
compounds (VOCs) in the U.S. - the primary pollutants that form smog. These vehicles also
emit more than half of all carbon monoxide pollution and contribute to particulate matter
emissions. It is therefore essential that the Tier 3 standards for both PM and NMOG+ NOx be
sufficiently stringent as to provide quantifiable and real reductions in harmful emissions.
We urge EPA to consider levels for the SFTP... PM... that are sufficiently stringent to ensure
that they are suitably protective of human health and the environment.
Supplemental Federal Test Procedure: EPA should revisit the... particulate matter (PM) exhaust
emission requirements to ensure that the emission requirements of this test procedure are suitably
protective of human health and the environment.
The problem is similar for PM. The SFTP PM standard is more than three times higher than the
PM FTP standard in 2023 for light duty vehicles and more than six times higher for vehicles with
a gross vehicle weight rating of more than 6,000 pounds. As CARB explains in its comments
(10), there were flaws in the data collected during EPA's useful life testing due to silicon
contamination resulting in higher PM emissions. Retesting of some of these vehicles, as well as
additional testing undertaken CARB supports the setting of more a stringent PM standard for the
US06 test cycle than currently under consideration.
Our Response:
Comments from stakeholders representing states, including CARB, and several NGOs
urged EPA to finalize more stringent US06 PM standards than those proposed, in all cases
advocating for standards below 10 mg/mi for the entire light-duty fleet and in some cases
advocating for standards below 6 mg/mi. Conversely, auto industry commenters generally
supported the proposed standards (10 mg/mi and 20 mg/mi for lighter and heavier light duty
vehicles, respectively. We have concluded that the body of recent data clearly shows that the
long-term 6 mg/mi US06 PM standard that we are finalizing, is the appropriate level to prevent
any significant "backsliding" in US06 PM emissions as new vehicles and technologies enter the
fleet. At the same time, the 6 mg/mi standard provides a reasonable compliance margin - about
50% above the average levels of current vehicles, which are averaging about 3 to 4 mg/mi. A
long-term standard numerically lower than 6 mg/mi would run counter to our intent to bring the
emissions performance of all vehicles to that already being demonstrated by many vehicles in the
current light-duty fleet. We believe the long-term US06 PM standard we are finalizing is
appropriate based on all of the information available at this time and will not hinder introduction
of new technologies manufacturers may choose for compliance with the other Tier 3 standards or
other rules.
The short-term, less-stringent US06 standard of 10 mg/mi (applicable in MYs 2017 and
2018) responds to automaker concerns about uncertainties stemming from simultaneous
regulatory requirements and rapidly evolving exhaust and engine technologies in the coming
4-49
-------
Chapter 4: Proposed Vehicle Emissions Program
years. We recognize that vehicle control technologies for both criteria and GHG emissions are
evolving and will continue to do so, including an expected expansion of gasoline direct injection
(GDI) technologies (see Section IV.A.S.c of the preamble and the RIA). Also, the transition to
lower sulfur in-use gasoline required by this rule may create temporary additional challenges in
consistently achieving lower US06 PM emissions (see Section IV.A.6 of the preamble and in the
RIA). We believe that most manufacturers will implement similar if not identical emission
control strategies to comply (or, more often, to continue to comply) with both the 10 mg/mi and
the 6 mg/mi standards. In so doing, we expect them to use the temporary additional compliance
margin provided by the 10 mg/mi standard to reduce uncertainties about potential variability in
performance (in use and, in particular, later in vehicle life) during the early years of developing
and commercializing their control technologies.
The 10 mg/mi standard will expire after MY 2018, and the long-term standard of 6 mg/mi
will take effect. As the implementation of the program continues, we believe a limited degree of
relief for testing of in-use vehicles is appropriate. Manufacturers commented that because of the
industry's general lack of experience with stringent PM standards, especially as the newly-
designed vehicles age, less stringent standards for in-use testing would reduce near-term
concerns about performance variability early in the program. We agree, and we are finalizing a
separate standard of 10 mg/mi for in-use vehicle testing for the intermediate years of the
program, MYs 2019 through MY 2023. This standard is numerically lower than the proposed in-
use standards - again because of the availability of improved US06 test data as described above
- but the purpose of providing an in-use standard remains the same. The in-use standard, in
conjunction with the short-term 10 mg/mi standard, represents a longer duration of relief than we
had proposed, again based on comments from the industry about their compliance concerns with
new US06 standards. For MY 2024 and later, there will be no separate in-use standard and all
vehicles will need to meet the long-term standard at certification and in use.
EPA proposed that different US06 PM standards apply to lighter and heavier vehicles.
The US06 PM test data discussed above also makes clear that the US06 PM performance of
current vehicles is not closely related to vehicle weight, although the earlier data had indicated
that this might be the case. Several commenters urged EPA to finalize a single standard for
vehicles above and below 6,000 Ibs GVWR based on the newer data. At the same time, auto
manufacturers generally supported the proposed vehicle weight distinction, asserting a higher
degree of uncertainty about the emission performance of their larger vehicles, especially in the
early years of the program and in light of simultaneous technology challenges. The data clearly
show that larger vehicles today are generally achieving US06 PM levels very similar to smaller
vehicles, and well below the proposed standards. We are not finalizing separate US06 standards
for heavier and lighter vehicles because separate standards are unwarranted based on a review of
the data. However, we believe that the short-term 10 mg/mi standard, as well as the temporary
in-use vehicle testing standard, will significantly reduce manufacturer compliance uncertainties
in the early years of the program for all vehicles, as discussed above.
As with the FTP PM standards, manufacturers will comply with the US06 PM standards
with the same increasing minimum percentage of their vehicles. Also, in response to
manufacturer concerns about the transition to Tier 3, and as with the FTP PM phase-in, we are
providing the option for a manufacturer to choose to certify 10 percent of its total light-duty
4-50
-------
Tier 3 Summary and Analysis of Comments
vehicle sales in MY 2017 to the new US06 PM standards, including light-duty vehicles over
6,000 Ibs GVWR.
As discussed in Section IV.A.2.C of the preamble, for LDVs and LDTs more than 6,000
Ibs GVWR and MDPVs, EPA is also providing an alternative phase-in of the US06 PM
standards
The primary PM percent phase-in schedule for MY 2017 requires 20 percent of vehicles
6,000 Ibs. or less GVWR to comply with the new PM standards. In review of the comments
regarding the PM phase-in requirements for MY2017, we agree with the commenters that the
option we proposed may not by itself always apply equitably to manufacturers of different fleet
mixes. In response to the suggestions, in addition to the primary PM phase-in we are finalizing
the option that a manufacturer may comply in MY 2017 by meeting the FTP and the US06 PM
standards with 10 percent of its projected nationwide sales of all vehicles (including those greater
than 6,000 pounds GVWR). This approach is harmonized with the LEV III PM requirements in
MY 2017.
As in their comments on FTP PM standards discussed above, CARB commented that we
should require that a manufacturer that certifies its vehicles to the proposed alternative phase-in
schedule still meet the requirement that 100 percent of those vehicles meet the applicable PM
standards in the 2021 model year. See the response in Chapter 4.1.2.5 of this document.
Regarding the Alliance/Global comment about per-vehicle vs. fleet average standards for
PM, see our response in Chapter 4.1.2.5 of this Summary and Analysis of Comments document.
4.1. LD Exhaust Standards
4.1.3. Start Date and Phase-In Schedules and Lead Time
4.1.3.1. Need to Maintain 2017-18 Start Dates
What Commenters Said:
Boulder County Board of Commissioners and Boulder County Board of Health
Therefore, it is imperative that the federal government take action by December 31, 2013, to
adopt Tier 3 so it can be implemented with the 2017 model year of vehicles. If not, an entire year
of benefits will be lost, resulting in serious and adverse impact on human health and welfare.
Children's Environmental Health Network (CEHN)
EPA should eliminate the one-year delay of fleet average NMOG+NOx FTP standards phase-in
for vehicles with GVWR above 6,000 Ibs.
Motor & Equipment Manufacturers Association (MEMA)
4-51
-------
Chapter 4: Proposed Vehicle Emissions Program
The EPA's proposal is mindful of other regulatory frameworks and compliance regimes. Efforts
to harmonize Tier 3 vehicle emission requirements nationwide and to choreograph the timing of
Tier 3 implementation with the final rules for greenhouse gas (GHG) emissions and corporate
average fuel economy (CAFE) standards for light-duty vehicles are critical for interconnected
and highly complex vehicle manufacturing supply chains. Also, EPA's integrated systems
approach to vehicles and fuels, combined with the alignment to other parallel regulations, is
practical and cost-effective. Companies and government entities alike will benefit from the
resulting outcomes, such as streamlined costs for R&D and production and reduced burden for
multiple, overlapping testing and compliance protocols. Since the timing of this rule is critical, it
is important that EPA not delay its completion. If delayed, the benefits of synchronizing the
timing of the GHG emissions, CAFE standards, and other programs will be lost and may
negatively impact the states in achieving their respective air quality goals.
MEMA urges EPA to fully consider the public comments, particularly from the vehicle
manufacturers. In addition, EPA must continue to collaborate with states, like California, and
other stakeholders to avoid divergent policy pathways and competing regulatory regimes.
MEMA recommends that the Agency promulgate a final rule by end of 2013 in order to match
with other regulatory requirements standards affecting MY 2017 vehicles.
Manufacturers of Emission Controls Association (MECA)
[MECA] ... urges EPA to finalize these proposals by the end of this year.
Our Response:
As proposed and as discussed in Section IV.A.2.a of the Preamble, the declining fleet-
average NMOG+NOx FTP standards will begin in MY 2017 for light-duty vehicles and light-
duty trucks with a GVWR up to and including 6,000 Ibs and in MY 2018 for light-duty vehicles
and light-duty trucks with a GVWR greater than 6,000 Ibs and MDPVs. The standards apply to
the heavier vehicles a year later to facilitate the transition to a 50-state program for all
manufacturers. During this transition period, there will be two fleet-average NMOG+NOx
standards for each model year, one for LDVs and LDTls and one for all other LDTs (LDT2s,
LDT3s, and LDT4s) and for MDPVs that decline essentially linearly from MY 2017 through
MY 2025. At that point, the two fleet-average standards converge and stabilize for all later
model years at the same level, 30 mg/mi, which is identical to the LEV III final fleet average.
While the rule is be finalized in 2014 deadline we have maintained the requirement that
implementation of the Tier 3 standards start in MY 2017. This planned program start retains the
benefits synchronizing the phase-in of the Tier 3 rule with other related programs.
4.1.3.2. Statutory Concerns About Lead Time for Heavier LD Vehicles and Issues
With Alternative Phase-In Schedules
What Commenters Said:
4-52
-------
Tier 3 Summary and Analysis of Comments
Alliance of Automobile Manufacturers (Alliance) and Association of Global Automakers
(Global)
In several places throughout the regulations, EPA skirts the plain language of the CAA by
offering two alternatives - one that does not comply with the CAA lead time provisions but
offers a reasonable requirement, and another that provides the required lead time but is so
stringent that no reasonable company would ever choose it. The CAA does not authorize this
approach, nor is EPA following Congress' clear intent and direction. Moreover, complying with
the plain language of the CAA would achieve the same level of emissions reduction. While we
applaud and support EPA's desire to harmonize with California as expeditiously as possible, it is
not necessary to violate the CAA in the process.
The CAA clearly requires EPA to provide four years of lead time for changes to emission
standards for heavy-duty vehicles, which are defined in CAA § 202(b)(3)(C) as those vehicles
manufactured primarily for use on public roads that have a gross vehicle weight rating (GVWR)
of more than 6,000 pounds.(2) Specifically, CAA § 202(a)(3)(C) states, "Any standard
promulgated or revised under this paragraph and applicable to classes or categories of heavy-
duty vehicles or engines shall apply for a period of no less than 3 model years beginning no
earlier than the model year commencing 4 years after such revised standard is promulgated." (3)
EPA proposes both a primary path and an alternative path for complying with the FTP and SFTP
NMOG+NOx standards:
Primary Path - Fleet Average: Under this option, both FTP and SFTP NMOG+NOx standards
are phased in using a fleet average requirement. The manufacturer can certify vehicles to any of
a number of specified emission bins, but for each model year, the sales-weighted average
emissions must meet a fleet average standard that decreases from MY 2017 to MY 2025 for
vehicles less than 6,000 pounds GVWR, and from MY 2018 to MY 2025 for vehicles over 6,000
pounds GVWR.
Alternative Phase-In: Under this option, Tier 3 FTP and SFTP standards under the primary
option apply for vehicles under 6,000 pounds GVWR in MYs 2017 and 2018, but do not apply
for over 6,000 pound GVWR vehicles until MY 2019. Starting in MY 2019, 40 percent of all
vehicles (both less than and greater than 6,000 pounds GVWR) must meet the final FTP and
SFTP standards; in MY 2020, 70 percent of vehicles must meet the final standards; and in MY
2021- a full four model years ahead of the primary phase-in option - 100 percent of a
manufacturer's fleet must meet the final Tier 3 standards.
Assuming EPA issues a Tier 3 final rule in late 2013, as planned, MY 2014 already would have
begun. Therefore, the CAA requires that the regulations applicable to heavy-duty vehicles take
effect no earlier than the MY 2019 (four years after the MY 2014 would be MY 2018, and the
next MY would be 2019). Thus, EPA's proposed primary path for compliance with the FTP and
SFTP NMOG+NOx Phase-In, which would require manufacturers of heavy-duty vehicles to
begin complying with the more stringent Tier 3 standards in MY 2018, violates the lead time and
stability requirement in CAA § 202(a)(3)(C).
4-53
-------
Chapter 4: Proposed Vehicle Emissions Program
Although the alternative compliance path phase-in for NMOG+NOx does meet the lead time
requirement, it would require manufacturers to fully comply with the more stringent standards
four years earlier, imposing significantly more stringent requirements on the manufacturers'
fleet. This presents manufacturers with a Hobson's choice: either 1) give up the right to four
years of lead time under the CAA, or 2) assert the right to four years of lead time under the CAA,
and get punished with more stringent standards as a result. The alternative compliance path is so
burdensome that no reasonable manufacturer would choose this option; all manufacturers would
essentially be forced into excusing EPA's violation of the CAA and complying with the primary
path.
The lead time limits that the CAA imposes on EPA were never intended to be circumvented by
regulatory ultimatums of this nature. The CAA envisions that EPA will promulgate standards
taking into consideration all of the relevant factors and limits set forth in §202(a). Congress did
not intend to set up a process under which EPA could offer illusory options with the intent of
extracting lead time concessions from manufacturers.
When a regulatory scheme is so heavily weighted or creates significant economic incentives such
that only one of the options is reasonable, that option becomes a mandate and is no longer truly
an option. The Supreme Court has acknowledged the possibility that "acute, albeit indirect,
economic effects" of a state law could effectively restrict choices to the point that the law, while
not directly preempted, could result in a de facto mandate that was preempted.4 Similarly, the
U.S. District Court for the Southern District of New York held that the court had to examine the
effect of the New York City law to first determine whether incentives for the purchase of hybrid
taxis created a de facto mandate to purchase hybrids in order to decide whether the rules were
preempted by the federal fuel economy requirements in the Energy Policy and Conservation Act
(EPCA).5 Also in the vehicle context, the U.S. District Court for the Eastern District of
California concluded that while other options existed for compliance with California's 2002 ZEV
amendments, they were not viable alternatives to producing advanced gasoline hybrids, the cost
of which made likely irreparable injury supporting the auto manufacturer's request for a
preliminary injunction against the ZEV requirement.6 These cases make it clear that the courts
look with disfavor upon regulatory "alternatives" that are designed to drive the regulated
community in a particular direction. Here, the accelerated phase-in under the alternative
compliance path would be so stringent that it would effectively eliminate it as an option,
resulting in a de facto requirement to comply with the primary path, which is prohibited by the
CAA.
In this case, there is a very clear statutory directive at issue. CAA §202(a)(3)(C) unambiguously
requires four years of lead time for heavy-duty vehicles. This prohibits EPA from finalizing the
primary compliance path as proposed. Providing an alternative compliance path that meets the
statutory requirements does not, and cannot, remedy the fact that the primary compliance path
violates the statute.
There is a relatively simple solution to this issue that meets EPA's objectives while maintaining
compliance with the CAA. EPA can follow the lead time requirement of the CAA and begin the
phase-in requirement in MY 2019 for vehicles over 6,000 pounds GVWR at the currently
proposed fleet average emissions levels. In other words, the phase-in in MY 2019 would begin at
4-54
-------
Tier 3 Summary and Analysis of Comments
the level that would otherwise be in place under the currently- proposed primary path.7 Such a
rule would comply with the CAA, fleet emissions would be at an identical level starting in MY
2019, and there would be no meaningful environmental impact due to beginning the phase-in one
year later.
Recommendation: For vehicles more than 6,000 pounds GVWR, the Alliance and Global
Automakers recommend that Tier 3 fleet average requirements for both FTP and SFTP begin in
MY 2019 at the currently proposed MY 2019 value. This approach is in full compliance with the
CAA, completely harmonizes with LEV III in MY 2019, and achieves virtually the same level of
emission reduction as the "primary path" proposed by EPA.
242U.S.C§7521(b)(3)(C).
3 42 U.S.C. §7521(a)(3)(C) (emphasis added).
4 New York State Conference of Blue Cross & Blue Shield Plans v. Travelers Insurance Co., 514 U.S.
645, 668 (1995).
5 Metropolitan Taxicab Board of Trade v. Bloomberg, 633 F.Supp. 2d 83, 96 (S.D.N.Y. lune 22, 2009).
6 Central Valley Chrysler Plymouth v. California Air Resources Board, 2002 U.S. Dist. LEXIS 20403 at
* 15-16 (lune 11,2002).
7 If EPA does not release the final rule by the end of 2013, then an additional year of delay for vehicles
greater than 6,000 pounds GVWR is required, starting in MY 2020 rather than the 2019 MY.
Chrysler Group LLC
The NMOG+NOx fleet average requirements for vehicles over 6,000 pounds GVWR should
begin in the 2019 MY at the currently proposed 2019 MY fleet average emission level and
compliance should end with MY 2025.
Clean Air Act Requirements: Lead-Time for Vehicles More than 6,000 Pounds GVWR and
Useful Life Requirements - In several places in the proposed rule, EPA seeks to circumvent
plain limits on its authority in the CAA by offering automakers two compliance alternatives -
one that does not follow the CAA requirements but offers a reasonable compliance option, and
one that applies the required statutory protections but imposes compliance conditions that are so
stringent that no reasonable company would ever choose it. The CAA does not authorize this
"poison pill" approach, under which companies are essentially forced to waive their statutory
protections under the CAA to avoid an unreasonably burdensome compliance alternative. This is
particularly so where the burdens imposed with the option to retain the statutory protection bear
no rational relation to the statutory protection in question. In short, EPA's dual-path approach, as
described below, effectively and impermissibly eviscerates the statutory protections for industry
that Congress intended to provide. Chrysler believes that EPA does not have the authority to
create overt workarounds to clear statutory requirements and limitations.
It is instructive that EPA previously failed in an attempt to circumvent the requirements of the
CAA by proposing to adopt a similar favorable versus non-favorable dual-path approach. In
Virginia v. EPA, (24) the D.C. Circuit firmly rejected a regulatory scheme strikingly akin to the
proposals here: in the Virginia case, EPA promulgated a rule in 1995 that was intended to reduce
ozone pollution in the northeastern United States by requiring states in that region essentially to
adopt California's vehicle emission program. However, Section 202(b)(3)(C) of the CAA
prohibited EPA from adopting more stringent emission standards for new motor vehicles until
4-55
-------
Chapter 4: Proposed Vehicle Emissions Program
2004, providing that "the numerical emission standards . . . shall not be modified by the
Administrator . . . for any model year before the model year 2004" (25). In order to overcome
this explicit statutory limitation, EPA offered states in the region a choice: (1) adopt a vehicle
emission program that would achieve emission reductions from new motor vehicles in the
amount that would be achieved by California's Low Emission Vehicle program; or (2) develop
an acceptable "Substitute Program" that would reduce ozone precursors anywhere from 3.5 to
6.5 times more than would the California program (26). However, as even EPA acknowledged,
the Substitute Program was so burdensome that it was not a realistic alternative; in the words of
the D.C. Circuit, "EPA's alternative [was] no alternative at all, and, "only a very foolish state
would see EPA's offer to accept [the alternative compliance option] as a real alternative")."
The D.C. Circuit held that EPA's regulatory scheme, which effectively forced states to adopt
California's vehicle emission program in contravention of the CAA, was unlawful. In vacating
the rule, the D.C. Circuit flatly rejected EPA's position that because the states had an alternative
to adopting the California motor vehicle standards, EPA itself was not imposing more stringent
emission standards for new motor vehicles. In so holding, the Court stated: Here Congress's
policy and preference is loud and clear. It "is the intent of Congress that" EPA not modify the
"numerical emission standards" for "any model year before . . . 2004." 42 U.S.C. §
7521(b)(l)(C). EPA therefore may not require, mandate, order, or impose conditions demanding
that any state enact particular motor vehicle emission standards, even if those standards are
identical to California's. The time will come when EPA can make its case for tougher emission
limitations on motor vehicles. But that time is years from now and, under section 202, the case
must be made to Congress.
Likewise, here, EPA has proposed alternative compliance paths in the proposed rule that are "no
alternative[s] at all," effectively forcing manufacturers to waive their statutory protections. As
the D.C. Circuit held in Virginia v. EPA, such a regulatory scheme is prohibited by the CAA.
Two specific examples of this impermissible regulatory structure are addressed below: (1) the
proposed phase-in requirements for heavy-duty vehicles for compliance with the Tier 3 standards
for NMOG+NOx; and (2) the useful life durability requirements for vehicles less than 6,000
pounds gross vehicle weight rating (GVWR).
24 - 108 F.3d 1397 (D.C. Cir. 1997), modified on other grounds, 116 F.3d 499 (D.C. Cir. 1997).
25-42 U.S.C. § 7521(b)(3)(C).
26-108F.3dat 1403.
NMOG+NOx Phase-In Requirements for Heavy-Duty Vehicles: The CAA requires EPA to
provide manufacturers of heavy-duty vehicles, defined as vehicles in excess of 6,000 pounds
GVWR, (29) with four years of lead time (and a three-year period of "stability" with no changes
to the standards), to comply with new automobile emission standards. Specifically, Section
202(a)(3)(C) of the CAA, 42 U.S.C. § 7521(a)(3)(C), states: "Any standard promulgated or
revised under this paragraph and applicable to classes or categories of heavy-duty vehicles or
engines shall apply for a period of no less than 3 model years beginning no earlier than the model
year commencing 4 years after such revised standard is promulgated." (emphasis added). This
statutory requirement is intended to ensure that manufacturers of heavy-duty vehicles are
4-56
-------
Tier 3 Summary and Analysis of Comments
afforded sufficient lead time and stability to recoup their investments in new technology required
to comply with more stringent standards (30).
In the proposed rule, EPA attempts to thwart this statutory requirement by proposing two phase-
in alternatives for complying with the significantly more stringent Tier 3 NMOG+NOx tailpipe
emissions standards. Under the more favorable primary compliance option, manufacturers must
meet a fleet average standard that is phased-in between Model Years 2018 and 2025 for vehicles
over 6,000 pounds GVWR, but which contravenes the CAA's explicit four-year lead time
requirement. Under the less favorable alternative compliance option, manufacturers must phase
in 40 percent of their vehicles to the more stringent Tier 3 NMOG+NOx standards beginning in
Model Year 2019, which comports with the statutory four-year lead time requirement, but adds
the significant burden that all of the manufacturer's vehicles must meet the final standards by
Model Year 2021 — a full four model years ahead of the primary phase-in option (31).
Assuming EPA issues a Tier 3 final rule in late 2013, as planned, Model Year 2014 already
would have begun. Therefore, the CAA requires that the Tier 3 standards applicable to heavy-
duty vehicles take effect no earlier than Model Year 2019 (i.e. four model years after the revised
standards are promulgated per Section 202(a)(3)(C), that is, commencing with the 2015 Model
Year). EPA's proposed primary compliance option, however, would require manufacturers of
heavy-duty vehicles to begin complying with the stringent Tier 3 standards in 2018 (i.e., only
three model years after the revised standards are promulgated), and therefore would contravene
the lead time requirement in Section 202(a)(3)(C) of the CAA. Conversely, although EPA's
alternative compliance option does meet the statutory four-year lead time requirement, it
unreasonably imposes unduly and significantly more burdensome compliance requirements than
the primary compliance option, and does not present manufacturers with a realistic choice.
Moreover, these undue burdens are not rationally related to the four-year lead time requirement
and thus are clearly designed as a punitive stick to force manufacturers to "choose" the non-CAA
compliant primary option with only three years of lead time.
As described in greater detail in the comments provided by the Alliance of Automobile
Manufacturers and Association of Global Automakers, (32) the alternative compliance option,
which complies with the four-year lead time requirement but requires manufacturers to certify all
of their vehicles to the Tier 3 standards a full four model years sooner than the primary phase-in
option, is so burdensome that no reasonable manufacturer would ever choose it; all
manufacturers would essentially be forced into selecting the primary phase-in option and turning
a blind eye to EPA's blatant violation of the CAA's lead time requirements for heavy-duty
vehicles. When a regulatory scheme is so heavily weighted or creates significant economic
incentives such that only one of the options is reasonable, that option becomes a mandate and is
no longer truly an option. Providing a significantly more burdensome alternative compliance
path that meets the statutory requirements does not, and cannot, remedy the fact that the primary
compliance path violates the statute.
This dual-path approach that EPA proposes would violate the CAA for an additional reason, as
well. Under Section 202(a)(3)(A)(i) of the CAA, EPA is required to promulgate standards
"which reflect the greatest degree of emission reduction achievable through the application of
technology which the Administrator determines will be available for the model year to which
4-57
-------
Chapter 4: Proposed Vehicle Emissions Program
such standards apply, giving appropriate consideration to cost, energy, and safety factors
associated with the application of such technology" (33). Thus, the CAA implicitly requires EPA
to select the single standard that reflects the greatest degree of emission reduction achievable.
Here, it simply cannot be the case that both the primary compliance option (which begins in
Model Year 2018 and requires full compliance with the more stringent standards by 2025) and
the alternative compliance option (which begins in Model Year 2019 and requires full
compliance with the more stringent standards by 2021) reflect the greatest degree of emission
reduction achievable taking into account cost, energy, and safety. Rather, if the three-year phase-
in under the alternative compliance option is in fact achievable considering the specified factors,
then the primary compliance path, which affords manufacturers seven years to meet the more
stringent standards, cannot be reconciled with the CAA requirement that EPA promulgate
standards reflecting the greatest degree of emission reduction achievable. The more likely
explanation, however, is that the primary compliance option reflects the greatest degree of
emission reduction that EPA believes is truly appropriate, whereas the alternative compliance
option is intentionally so burdensome—and therefore far beyond reflecting the greatest degree of
emissions reduction achievable—that manufacturers will be forced to waive their statutory
protections and choose the primary path. This attempt to circumvent the statutory protections
embodied in the CAA cannot be reconciled with the plain provisions of the Act.
Further, Chrysler objects to the requirement that if a manufacturer chooses the alternative
compliance path for any of its vehicles, all of its vehicles, including light-duty vehicles and those
heavy-duty vehicles for which the manufacturer is willing to comply with three years lead time,
must nevertheless meet the final Tier 3 NMOG+NOx standards by 2021 (instead of 2025 under
the primary compliance option). This regulatory scheme, which sweeps in a manufacturer's
entire fleet if the manufacturer chooses to pursue the alternative compliance option for even just
a single vehicle model, cannot be characterized as an incentive to forego the four-year lead time
requirement; rather, it is a penalty imposed on manufacturers that refuse to forego their statutory
protections. There is no rational basis for penalizing a manufacturer's entire vehicle fleet simply
because the manufacturer chooses to exercise its statutory right to four years of lead time for one
heavy-duty vehicle model. Plainly, any emissions benefits of earlier compliance would accrue
for those models of vehicles for which the manufacturer might choose earlier compliance. But
precluding that choice and imposing more stringent standards even for those models that might
comply early, just because a different model or models cannot comply early, has no
environmental rationale except to penalize the choice of later compliance for the other models.
That design is arbitrary, unreasonable, and unlawful (34).
Recommendation: Chrysler recommends that, for vehicles over 6,000 pounds GVWR, the Tier 3
fleet average requirements for FTP and SFTP NMOG+NOx begin in MY 2019 at the currently
proposed Model Year 2019 fleet average emissions levels (under the primary compliance option)
and require full compliance only at Model Year 2025. This approach is in full compliance with
the CAA, completely harmonizes with LEV III in MY 2019, and will achieve the same level of
emission reduction as the proposed rule.
29 - See 42 U.S.C. § 7521(a)(3)(C). A "heavy-duty vehicle" is a "truck, bus, or other vehicle
manufactured primarily for use on the public streets, roads, and highways . . . which has a gross vehicle
weight... in excess of six thousand pounds." 42 U.S.C. § 7521(3)(C).
4-58
-------
Tier 3 Summary and Analysis of Comments
30 - See, e.g., Natural Res. Def. Council v. Thomas, 805 F.2d 410, 433 (D.C. Cir. 1986) (noting that the
four-year lead time requirement "was enacted for the benefit of the manufacturers, to allow time for them
to design and develop engines in compliance with newly promulgated standards") (citing H.Rep. No. 294,
95th Cong., 1st Sess. 274-75 (1977) (discussing "abundant lead time"), reprinted in 4 Leg. Hist. 2741-42,
J.A. at 19-20).
31 - EPA proposes to adopt a similar dual-path approach for "heavy-duty vehicles" as defined by EPA at
40 CFR §86.1803-01 (i.e., vehicles greaterthan 8,500 pounds GVWRbut less than or equal to 14,000
pounds GVWR). See 78 Fed. Reg. at 29,875-29,877. Under the proposed primary compliance option,
manufacturers must meet a declining fleet average standard for NMOG+NOx beginning in Model Year
2018, in contravention of the CAA's explicit four-year lead time requirement. Alternatively,
manufacturers can choose to comply beginning in Model Year 2019, which comports with the four-year
lead time requirement, but then are subject to a percent-of-vehicles phase-in approach rather than a
declining fleet average. Again, EPA's "poison pill" approach cannot be reconciled with the CAA.
Accordingly, Chrysler recommends that EPA start the declining fleet average in Model Year 2019,
consistent with the CAA's four-year lead time requirement.
32 - Alliance/Global Comments, at 4.
33 - 42 U.S.C. § 7521(a)(3)(A)(i),
Our Response:
The Alliance of Automobile Manufacturers (Alliance), and Association of Global
Automakers (Global) and Chrysler LLC commented on the alternative phase-in proposed for
vehicles over 6000 Ibs GVWR. These comments questioned whether the proposed structure of
and restrictions on the use of the alternative phase-ins were so onerous as to unduly restrict a
manufacturer from choosing the alternative phase-ins and their lead time and stability provisions
as set forth in the Clean Air Act. The commenters criticized the proposed requirement that a
manufacturer using the alternative phase-ins apply the alternative schedules to its entire light-
duty fleet, both below and above 6,000 Ibs GVWR. EPA had proposed this provision to
minimize the complexity of complying with the alternative phase-in if a manufacturer's heavier
and lighter light-duty vehicles had different compliance structures.
In considering these comments, EPA also considered that during the development of the
Tier 3 program and in their comments, the same auto industry commenters consistently urged
EPA to design the Tier 3 program to harmonize with the California LEV III standards as closely
and as early as possible. As discussed in detail in Section IV. A of the preamble, extensive data
that EPA has generated or received continue to support the conclusion that the primary fleet-
average standards provide a compliance path that is feasible across the industry and that closely
harmonizes with LEV III. EPA believes that we have reasonably resolved these somewhat
competing concerns - early harmonization vs. additional lead time - by finalizing the primary
declining fleet average standards as proposed while also finalizing revised alternative phase-in
compliance schedules (see preamble Section IV. A.2.c). In response to the comments, we have
revised the alternative phase-in schedules to reduce their associated burden for manufacturers,
while still maintaining environmental benefits that are equivalent to the primary program. We
also include provisions in the percent-of-sales phase-in alternatives that allow manufacturers to
exclude vehicle models that begin their 2019 model year production early in 2018, in order to
provide four years of lead time.
4-59
-------
Chapter 4: Proposed Vehicle Emissions Program
Specifically, we have removed from the alternative phase-in provisions the requirement
that a manufacturer apply the alternative schedules to its entire light-duty fleet including vehicles
below 6,000 Ibs GVWR. For the practical functioning of the program, the final rule requires that
any manufacturer choosing to use the alternative phase-in apply all four alternative phase-in
schedules to its entire light-duty fleet above 6,000 Ibs GVWR. We believe that the alternative
phase-ins allow manufacturers to comply with emission standards in a time frame that is clearly
feasible and fully compliant with the CAA requirements for lead time and regulatory stability.
To the extent that manufacturers choose to use them, the alternative would result in overall
emission reductions essentially identical to those of the primary program.
As explained in the preamble, the alternative phase-in schedules would begin to apply to
each vehicle for either MY 2019 or MY 2020, depending on exactly when the manufacturer
begins production of the vehicle. (See Section 86.181 l-17(b)(8)(i) for how we implement this
provision.) For models that begin MY 2019 production after the fourth anniversary of the
signing of the final rule, the alternative phase-in would provide four full years of lead time and
would first apply for MY 2019. The phase-in obligation would be calculated based only on
those vehicles beginning production after the fourth anniversary date. For models beginning
production before that date, the alternative phase-in would first apply for MY 2020, and the
phase-in percentage for MY 2020 would be based on the manufacturer's entire fleet of heavier
light-duty vehicles. Based on historical certification patterns, few models begin production
before mid-calendar-year, so we expect that the vast majority of MY 2019 vehicles will begin
production after the 4-year anniversary and thus the alternative phase-ins, if chosen, will
typically apply beginning in MY 2019.
At the time of certification for MY 2018, a manufacturer must declare whether it intends
to apply the alternative phase-in schedules to its heavier light-duty vehicles. A manufacturer
choosing the alternative phase-ins would be committed to this phase-in approach for the duration
of the phase-ins, and could not later choose the fleet-average approach for NMOG+NOx
standards. For all vehicles below 6,000 Ibs GVWR, the primary program will apply, beginning
in MY 2017. For a manufacture's vehicles subject to the alternative phase-ins, there would be no
new tailpipe emissions requirements beyond the Tier 2 program until the beginning of the
alternative phase-in schedules; that is, MY 2019 or 2020, as explained above.
As discussed above, a manufacturer choosing the alternative phase-in approach for its
heavier light-duty vehicles would be required to use all four phase-ins together. The next
paragraphs explain how each of the alternative phase-ins requires an increasing percent of the
manufacturer's sales to comply with the alternative standards. Thus, until the end of the phase-
ins, some percent of a manufacturer's affected vehicles will meet the new standard and the
remainder of that year's sales will not yet comply with Tier 3. For the practical functioning of
the program, a manufacturer choosing the alternative phase-ins would be required to comply
with exactly the same segment of their fleet in each model year for all four alternative phase-ins.
For example, a manufacturer that complies with the 70 percent MY 2020 requirement for the
FTP NMOG+NOx standard with a segment of its vehicle fleet must meet the 70 percent MY
2020 requirement for the FTP PM standard with the same set of vehicles.
4-60
-------
Tier 3 Summary and Analysis of Comments
For the FTP and SFTP NMOG+NOx alternative phase-in schedules, once the phase-in is
complete for a segment of a manufacturer's fleet, the standards continue for that set of vehicles
through MY 2024, after which the full Tier 3 program applies regardless of the phase-in strategy.
Thus, the fleet-average standards that decline through MY 2024 do not apply for these vehicles.
Although manufacturers would implement all four alternative phase-in schedules
together, as discussed above, each alternative phase-in has unique characteristics. The following
paragraphs explain the unique provisions of each.
Alternative Phase-In Schedule for the FTP NMOG+NOx Standard
Instead of the primary FTP NMOG+ NOx declining fleet average standards, a
manufacturer choosing the alternative phase-ins would comply with a stable fleet average FTP
NMOG+NOx standard of 30 mg/mi that would apply to an increasing percentage of a
manufacturer's combined sales of LDVs and LDTs above 6,000 Ibs GVWR and MDPVs. This
percent phase-in would match the percentages in the primary PM percent phase-in schedule, as
discussed above - specifically, 40 percent of MY 2019 heavier light-duty vehicles (excluding
those vehicles with production beginning before the 4-year anniversary), 70 percent of all of its
heavier light-duty vehicles in MY 2020, and 100 percent compliance in MY 2021 and later
model years.
Alternative Phase-In Schedule for the FTP PM Standard
Instead of the primary FTP PM percent phase-in schedule, a manufacturer choosing the
alternative phase-ins would postpone the beginning of its FTP PM phase-in for its LDVs and
LDTs above 6,000 Ibs GVWR and MDPVs until MY 2019 or 2020 (depending on the dates
production begins for its vehicle models, as discussed above). The manufacturer would then
comply with the 3 mg/mi per-vehicle FTP PM standard (and the 6 mg/mi in-use standard) on an
increasing percentage of these vehicles, following the 40-70-100 percentage phase-in of the
primary PM program — specifically, 40 percent of MY 2019 heavier light-duty vehicles
(excluding those vehicles with production beginning before the 4-year anniversary), 70 percent
of all of its heavier light-duty vehicles in MY 2020, and 100 percent compliance in MY 2021 and
later model years.
Alternative Phase-In Schedule for the SFTP NMOG+NOx Standard
As with the other alternative phase-ins, instead of the primary SFTP NMOG+ NOx
declining fleet average standards, a manufacturer choosing the alternative phase-ins would
comply with a stable fleet average SFTP NMOG+NOx standard of 50 mg/mi that would apply to
an increasing percentage of a manufacturer's combined sales of LDVs and LDTs above 6000 Ibs
GVWR and MDPVs. This percent phase-in again would match the percentages in the primary
PM percent phase-in schedule, as discussed above - specifically, 40 percent of MY 2019 heavier
light-duty vehicles (excluding those vehicles with production beginning before the 4-year
anniversary), 70 percent of all of its heavier light-duty vehicles in MY 2020, and 100 percent
compliance in MY 2021 and later model years.
Alternative Phase-In Schedule for the US06 PM Standard
4-61
-------
Chapter 4: Proposed Vehicle Emissions Program
Finally, instead of the primary US06 PM percent phase-in schedule, a manufacturer
choosing the alternative phase-ins would postpone the beginning of the US06 phase-in for its
LDVs and LDTs above 6,000 Ibs GVWR and MDPVs until MY 2019 or 2020 (depending on the
dates production begins for its vehicle models, as discussed above). The manufacturer would
then comply with the 10 mg/mi US06 PM standard for 40 percent of MY 2019 heavier light-duty
vehicles (excluding those vehicles with production beginning before the 4-year anniversary), 70
percent of all of its heavier light-duty vehicles in MY 2020, with 100 percent compliance in MY
2021, and then 100 percent compliance with the 6 mg/mi standard in MY 2022 and later model
years.
With regard to Chrysler's comments in paragraph "31" above concerning HD vehicles, reference
Chapter 4.2.1 of this Summary and Analysis of Comments document.
4.1.4. Feasibility
What Commenters Said:
Emissions Control Technology Association (ECTA):
We agree with EPA's conclusion that the emissions reduction requirements in the proposed rule
are technically achievable, and the pathways illustrated represent cost-effective solutions to meet
these reductions.
The technology choices are numerous. Many technologies are on some vehicles already, and can
readily be applied to others. Examples include hydrocarbon adsorbers (Ford SAE 2013-01-1297,
Nissan SAE2008-01-0449, SAE2009-01-1076), high cell density substrates (only a few 900-csi
applications), secondary air (Umicore SAE 2012-01-1245), and advanced thermal management
strategies.
International Council on Clean Transportation (ICCT)
The feasibility of the proposed Tier 3 standards has already been demonstrated by the numerous
vehicles that already meet the California LEV III standards and Tier 2 bin 2 standards. The two
keys to low emissions are precise air/fuel control and rapid catalyst light-off Since the Tier 2
standards were adopted there have been major improvements in both of these areas, making
compliance with the proposed Tier 3 requirements easier.
Manufacturers of Emission Controls Association (MECA)
MECA agrees with EPA staffs assessment that achieving the proposed Tier 3
exhaust.. .emission standards and associated emission reductions are both technically feasible
and cost-effective.
This fact is clearly demonstrated by the more than two million SULEV- and PZEV-compliant
light-duty vehicles that have been sold in the U.S. market since these near-zero emission,
4-62
-------
Tier 3 Summary and Analysis of Comments
gasoline vehicles were first introduced more than ten years ago. This technology base of
advanced three-way catalysts, exhaust hydrocarbon adsorber materials, high cell density
substrates, emission system thermal management strategies, secondary air injection systems,
advanced carbon canisters, advanced low fuel permeation materials, and air intake hydrocarbon
adsorber materials that have already been commercialized for a variety of PZEV gasoline vehicle
applications can be extended and further optimized to allow all light-duty .. .gasoline vehicles to
achieve the exhaust .. .emission reductions needed to comply with the Tier 3 vehicle emission
proposals put forward by EPA. On the exhaust side, Tier 3/LEV III emission technologies
(including advanced three-way catalysts, advanced high cell density substrates, and hydrocarbon
adsorber catalysts) are highlighted in a new MECA report entitled: "LEV III and Tier 3 Exhaust
Emission Control Technologies for Light-Duty Gasoline Vehicles" (available on MECA's
website, www.meca.org, under Resources » Reports). A recent SAE paper (SAE paper no.
2011-01-0301) demonstrates how advanced three-way catalysts utilizing high cell density
substrates can be combined to achieve Tier 3, Bin 20 or Bin 30, exhaust emission levels on a
four-cylinder, light-duty gasoline vehicle. A 2013 SAE paper (SAE paper no. 2013-01-1297)
provides insights into recent improvements in performance and durability of hydrocarbon
adsorber catalysts. These latest generation hydrocarbon adsorber catalysts show improved cold-
start hydrocarbon emission reductions with reduced precious metal content compared to earlier
generations of hydrocarbon adsorber catalysts. Hydrocarbon adsorber catalysts are available to
assist in difficult Tier 3 applications, such as larger weight trucks and SUVs.
In addition, advanced diesel emission control technologies, including diesel particulate filters
(DPFs), lean NOx adsorber catalysts, and selective catalytic reduction (SCR) catalysts, will be
combined with future, advanced diesel engines to allow light-duty diesel vehicles to achieve the
proposed Tier 3 emission limits, including EPA's proposed Tier 3, Bin 30 exhaust standards.
MECA agrees with EPA's projections that future hydrocarbon adsorber catalyst applications will
be targeted at larger light-duty vehicles. MECA believes that any Tier 3 applications of
hydrocarbon adsorber catalysts will rely on "passive" hydrocarbon adsorber approaches that
combine a hydrocarbon adsorber function with a three-way catalyst function on the same
substrate (typically in a layered coating architecture in an underfloor converter location), rather
than "active" hydrocarbon adsorber designs that utilize some type of an exhaust valve to direct
the exhaust through a hydrocarbon adsorber material during cold-start. Passive hydrocarbon
adsorber catalysts can provide effective cold-start hydrocarbon emission reductions at a
significant cost advantage versus active hydrocarbon adsorber system designs. MECA also
believes that manufacturers will have the opportunity to optimize emission system designs with
advanced powertrains as they move forward with their future light-duty vehicle greenhouse gas
compliance strategies. Strategies such as engine downsizing, vehicle weight reductions, and
improved engine combustion technologies will help to off-set emission system cost increases for
future Tier 3 compliance.
In summary, there are significant opportunities to reduce both criteria pollutant and greenhouse
gas emissions from the transportation sector through the design of fuel-efficient powertrains that
include advanced exhaust emission controls for meeting even the most stringent criteria pollutant
standards that are included in EPA's proposed Tier 3 program. MECA believes that advanced
4-63
-------
Chapter 4: Proposed Vehicle Emissions Program
emission control systems have a critically important role in future policies that aim to reduce
mobile source criteria pollutant and greenhouse gas emissions.
MECA believes that advanced emission control systems have a critically important role in future
policies that aim to reduce mobile source criteria pollutant and greenhouse gas emissions, and we
strongly supports EPA's Tier 3 emission proposal.
Pennsylvania Department of Environmental Protection (DEP)
DEP supports EPA's proposed Tier 3 emission standards for vehicles. The proposed standards
would be met by vehicle manufacturers beginning in model year 2017 and phasing in through
later model years. Vehicles sold in states such as Pennsylvania that have adopted light-duty
vehicle emission standards promulgated by California Air Resources Board (CARB) will be
meeting the same standards for model year 2015 and later vehicles. States that adopted CARB
light-duty vehicle standards comprise nearly 50 percent of the national market for these vehicles.
Technological developments in the automotive field that have already occurred or that are in the
developmental pipeline as a result of CARB and EPA efforts will allow automobile
manufacturers to effectively meet these more stringent standards for NOx, VOC, and PM2.5 by
the year 2021.
Sierra Club
The Draft RIA provides sufficient evidence that the proposed Tier 3 standards are
technologically feasible for a range of existing vehicles. Already, manufacturers are
implementing technologies that reduce NMOG + NOx emissions under more challenging
circumstances such as during cold starts, and it is anticipated that manufacturers can make
necessary technology improvements to comply with the Tier 3 standard.
EPA also demonstrates feasibility of meeting the proposed rule's PM exhaust emissions
reductions; some fleets already are achieving Tier 3 level reductions, and there are a number of
available controls and practices, such as reduced oil consumption, which will allow future fleets
to meet the proposed standards.
State of Utah
The bulk of pollutants from most of today's vehicles are emitted during cold starts—ignition and
the first few miles of driving before conventional catalytic converters warm up enough to be
effective—and during fueling. The Tier 3 standards would address these issues by requiring that
by 2017 new vehicles have larger catalytic converters to better remove NOx, hydrocarbons, and
carbon monoxide; heat pumps that warm the catalytic converter almost immediately upon
ignition to avoid an initial rush of emissions after a cold start.... These low-emission vehicle
technologies are in use today in California and other countries and have a proven record of
results.
Our Response:
4-64
-------
Tier 3 Summary and Analysis of Comments
All commenters that commented on the technological feasibility of the proposed Tier 3
standards pointed to the current or near-term availability of technological pathways to
compliance with all or most of these standards. For the FTP and SFTP NMOG+NOx standards
and the FTP PM standard, no commenters expressed significant concern about the feasibility of
the proposed levels (although some argued for even more stringent standards, as addressed in
Chapter 4.1.2 above). In the proposal, we concluded that all of the Tier 3 emissions standards
are technologically feasible in the time frame of the program. The technical conclusions we
reached at that time have been further reinforced by information we received in the public
comments or has otherwise become available and placed in the docket for this rulemaking. After
considering the comments received and with additional supporting information in Chapter 1 of
the RIA, we continue to conclude that the Tier 3 standards are feasible and reasonable,
considering lead-time provided and expected compliance costs.
See Section IV. A. 5 of the preamble and Chapter 1 of the RIA for more detailed
discussions of the feasibility of the Tier 3 light-duty vehicle standards.
4.1.4.3. Impact of Gasoline Sulfur on Feasibility of the Vehicle Tailpipe Standards
EPA Tier 2 In-Use Sulfur Test Program
What Commenters Said:
American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM)
The EPA in-use fleet test examined sulfur effects at two levels (5 and 28 ppm) and had a number
of questionable test practices including using an unrealistic base fuel (0% ethanol, high
aromatics, very low olefins at essentially 0 vol%, and high T50); requiring mild operation before
testing; test cycle; vehicle history and data analysis.
A more recent EPA study was conducted beginning in 2009 to understand gasoline sulfur effects
on the in-use Tier 2 fleet (7). For this EPA effort (hereinafter termed the "In-use sulfur study"),
in-use MY 2007-2009 vehicles from the state of Michigan were evaluated for sulfur reversibility,
instantaneous, and mileage accumulated effects at 28 ppm and 5 ppm sulfur. The fuel used for
these experiments contained 0 vol% ethanol and was doped with dibutyl disulfide to increase
sulfur content from 5 to 28 ppm. This test fuel is not representative of in-use fuels due to its low
ethanol content, high aromatics, very low olefins at essentially 0 vol%, and high T50. While this
study was not designed to evaluate fuel effects other than sulfur, the temperature of the exhaust,
emissions species present, and potential catalyst effects are not representative of real world
driving conditions. The sulfur history of the vehicles /catalysts is also unclear.
Vehicle operation prior to the testing process is also an area of concern. The civilian or NVFEL
drivers were instructed to "avoid hard accelerations and high speeds in an effort to preserve the
"as-received" state of the catalyst". This instruction contradicts the overall objectives of this
program and biased the "as-received" state of the catalyst.
4-65
-------
Chapter 4: Proposed Vehicle Emissions Program
The report of the EPA In-Use sulfur study contains a significant amount of statistical analysis,
however interestingly, raw data trends are largely absent from the body of the document. In
review of the raw data provided in the appendix, it is clear that there are issues concerning the
drive cycle chosen to evaluate sulfur effects. The pre-and post-catalyst clean out emissions on 28
ppm sulfur fuel show varying trends where, in some cases, the emissions levels post catalyst
clean out were higher than the pre-catalyst clean out. The drive cycles chosen for both the
catalyst clean-out and mileage accumulation are also only one set of potential driving conditions
and typically would not be seen in-use. Previous studies have shown that sulfur
accumulation/reversibility is sensitive to driving cycle.
EPA constructed models of emissions versus mileage for each vehicle, as shown in Figure 7-7 of
the In-Use Study report. Some models seem to track the data very well, while others do a poor
job. However, EPA did not provide any descriptive statistics for the model, but a number of the
vehicles (e.g., 0046 and 0178) seem to have particularly poor fit. EPA recognizes this (page 45)
and attributes it to the "variability in the emissions testing." However, this also calls into
question the parameters that EPA has assumed to be important with respect to fuel sulfur impacts
on emissions.
A primary issue with EPA's statistical analysis of the data from the In-Use sulfur study is that
the individual vehicles were treated as random variables. If this is true, a subset of the total
vehicle population would be expected to have the same responses as the entire population.
However, EPA did not present an analysis of whether the different fleets had an impact on the
analysis. This should have been done to determine whether EPA's assumption that "vehicle" is a
random variable is valid. This is of key importance because EPA used different sets of vehicles
to analyze different emission effects. For instance, the clean-out effect at 28 ppm sulfur was
measured on a fleet of 81 vehicles, and the clean out effect at 5 ppm sulfur was measured with a
fleet of 23 vehicles, which were a subset of the larger fleet.
If the vehicles were truly random variables as EPA assumed, then the responses of the individual
models would not be expected to be different, and each vehicle would be treated the same as any
other vehicle. However, EPA reported most of the results according to vehicle model, suggesting
that EPA itself doesn't believe in the random variable assumption. See, for example, Figures 7-2
and 7-5 of the In-Use Study—these plots, as well as plots in Appendix E, suggest that there are
major differences in how different vehicle models responded to the sulfur clean-out cycle with
28 ppm sulfur fuel. The statistical analysis that EPA conducted did not account for the possibility
that the variability of responses within models to fuel differences could be very different from
the variability between models. This is important statistical information and can help sharpen the
analysis. Treating cars as random variables is a good assumption when one sample of each model
is tested; however, when multiple samples of each model are tested, this assumption may not be
valid. In any case, this assumption should have been tested by EPA on the large 81-vehicle fleet
that tested 5 samples of most models.
EPA made a mid-test change to the procedure for mileage accumulation (from multiple dyno
FTP runs to on-road mileage accumulation); however there was no evaluation of the significance
of this change to the resulting emissions effects. In addition, the report references that between
and within vehicle, variances were significant. An influence analysis should have been
4-66
-------
Tier 3 Summary and Analysis of Comments
completed, such as was done in the EPAct study, on all emissions to show potential biases due to
specific vehicle types, including a sensitivity case removing those vehicles having significant
influence to the overall conclusions in the report. This was completed only for Bag 2 NOx. Also,
due to the vehicle variances, a number of statistical models were applied to the data which would
not converge. The final structure chosen was done such that "due to limited available options, we
acknowledge that there might be some limitations inherent in the assumption of constant distance
between two measurements" (assumes regularly spaced time intervals for all vehicles which is
not the case when emissions were measured at different mileage accumulation rates).
Finally, a large number of emissions concentration measurements taken from Bags 2 and 3 were
either lower or similar to the measured background concentrations. The percent differences
which are referenced in these cases represent a very small magnitude, which may have errors
associated with the analyzers' capability. The report does not indicate the measurement error at
levels below the analyzer calibration points for NOx in Bag 2 and does not mention the
calibration points or error for other low emitting species.
EPA presented the results from the data analysis of the In-Use Study in terms of percentages.
While this is a typical outcome of statistical analysis using the log transformation of emission
data, it can distort the apparent impacts of lowering gasoline sulfur content. Because mass
emission levels in Bag 1 of the FTP are substantially larger than emissions from Bag 2 and Bag 3,
even small percentage differences in Bag 1 may be similar in size to very large percentage
differences in Bag 2 and Bag 3 when mass emission rates are considered.
Marathon Petroleum Company LP (MPC)
The in-use fleet test examined sulfur effects at two levels (5 and 28 ppm) and had a number of
questionable test practices including using an unrealistic base fuel (0% ethanol, high aromatics,
very low olefms at essentially 0 vol%, and high T50); requiring mild operation before testing;
test cycle; vehicle history and data analysis.
Manufacturers of Emission Controls Association (MECA)
EPA has released a thorough and well-designed sulfur effects study on 81 in-use Tier 2 light-
duty gasoline vehicles that clearly showed significant reductions in criteria pollutants in
comparing emissions performance on gasoline with 28 ppm sulfur versus 5 ppm sulfur. Work
published in a 2011 SAE technical paper (SAE paper no. 2011-01-0300) shows similar,
significant emission benefits on a 2009 model year PZEV vehicle operated with 3 ppm sulfur
gasoline versus 33 ppm sulfur gasoline. In this gasoline sulfur effects study, on a 2009 PZEV
passenger car, the results clearly show that the underfloor converter used on the close-coupled +
underfloor PZEV catalytic converter system was susceptible to sulfur-related performance
degradation due to its cooler operating temperatures during the FTP test cycle using a 33 ppm
sulfur-containing gasoline. The loss in NOx performance of this underfloor PZEV converter in
successive FTP tests could be recovered to some extent, or avoided to a large degree, by either
purging stored sulfur off the underfloor converter with the use of a higher speed and load test
cycle (i.e., the US06 test cycle) sandwiched between FTP tests, or using a gasoline with
significantly lower sulfur levels (i.e., a 3 ppm sulfur-containing gasoline).
4-67
-------
Chapter 4: Proposed Vehicle Emissions Program
As Mr. Grundler stated, EPA has released a thorough and well-designed sulfur effect study on 81
in-use Tier 3 light duty gasoline vehicles that clearly show significant reductions in criteria
pollutants in comparing emission performance on gasoline with 28 ppm sulfur versus five ppm
sulfur.
Our Response:
The American Petroleum Institute (API), MFC and the Manufacturers of Emissions
Controls Association (MECA) all commented on EPA's analysis of the effects of gasoline sulfur
on in-use light-duty vehicle emissions. One of the comments received was with respect to the
fuel used in our in-use sulfur analysis and the interaction various fuel properties could have on
the overall emissions. There has been little or no published data in the literature pertaining to the
interaction of sulfur with other fuel properties such as aromatics or ethanol content. Given this
fact, we chose to conduct our study using Tier 2 (non-ethanol) certification fuel as a base. The
extent to which ethanol might raise or lower exhaust (and catalyst) temperatures faster after cold
start or reduce or increase the sulfur poisoning impacts on cold-start emissions to some extent is
not known and API does not provide substantive data on that issue. There is also a propensity
for ethanol to lean-out transients and open-loop operation that would also tend to increase NOx
and lower NMOG emissions. There does not appear to be a clear theoretical basis to assume a
bias in NMOG+NOx emissions results that would result solely from the use of a non-ethanol test
fuel when assessing sulfur impacts on exhaust emissions, or that using test fuel with ethanol
would have substantially changed the program's overall conclusions.
API also commented on the drive cycle used in our analysis of gasoline sulfur. The
objective of the instructions to the vehicle transport driver with respect to hard acceleration and
high speed driving was to limit the opportunity for the state of catalyst sulfur loading to change
appreciably (up or down) from how it was when the vehicle was acquired from the owner. This
does not indicate that we intended or continued to drive the vehicle in a manner that would result
in a change in the state of the catalyst. The transport driving in question covered relatively short
distances from the owner's location to the test facility, and may have involved rural, city and/or
highway driving. Aggressive and high speed driving was avoided because it is known to change
the state of the catalyst.
The drive cycles chosen to evaluate sulfur effects represent the types of driving that
manufacturers must perform in their certification testing and will need to protect for during in-
use emissions performance evaluations of their certified vehicles. No single cycle represents all
"typical" driving. We agree that both sulfur accumulation and reversibility are sensitive to
driving cycles; however this study was not designed to evaluate cycles but instead to establish
the potential for sulfur to degrade emission performance that compromises a vehicle's ability to
meet emission standards, particularly the new stringent Tier 3 levels.
With respect to the comment on model predictions and the vehicle parameters included in
the model, the intent of the study was to assess the aggregate behavior of a representative sample
of the Tier 2 fleet. While we agree that many catalyst and engine operation parameters interact
with sulfur to influence emissions and could be studied as part of an engineering analysis,
attempting to account for all of them across the test fleet would be very complex and
4-68
-------
Tier 3 Summary and Analysis of Comments
unnecessary for our purpose of assessing the impacts across a broad cross section of vehicles. In
terms of the accuracy of the mixed model, overall, the model predictions were in good agreement
with the data. Although there were some instances where the model overestimated the effect of
sulfur and some instances where the model underestimated the effect of sulfur, we believe that
the variability in predictions does not meaningfully affect the overall conclusions, since the
model predictions were not biased toward a single direction. Furthermore, the study was
designed to examine the effect of sulfur on in-use emissions from the fleet, rather than from
individual vehicles.
With respect to the treatment of vehicles as random variables, the mixed model accounts
for correlation in the data through the inclusion of random effects and modeling of the
covariance structure. The random effects represent parameters that are allowed to vary over
vehicles reflecting the natural heterogeneity in the vehicle fleet. In the draft report, the analyses
modeled each vehicle family as the random effect for the mixed model examining the effect of
clean-out, since multiple vehicles were tested for each make and model. For the mixed models
examining the effect of clean-out at 5 ppm and the effect of sulfur level, the analysis modeled the
individual vehicle as the random effects. However, since the draft report, additional vehicles
were tested and included in the analyses. Therefore, in the final report, the assumptions for the
mixed model have been modified to model vehicle family as the random effect for all three
datasets. By assuming each individual vehicle family as a random effect, we assumed the
presence of substantial between-vehicle family variability, based on the results from the
likelihood ratio test assessing the significance of the variation between vehicle families. In
addition, consistent with the change in the assumption of random effect, the data plots in the final
report are presented in terms of vehicle models. The method of treating vehicles as random
effects in order to properly model the covariance structure has been utilized in a similar study
examining the effects of fuel properties, including sulfur, on emissions.3 Furthermore, the
external peer-reviewers provided support by stating that "the statistical modeling approach seems
appropriate" considering the study design, and the structure and limitations of the emissions
dataset, since the approach accommodates missing data, irregularly spaced measurements, and
within-vehicle effects.4 In the final report, we also performed additional sensitivity analyses to
examine the extent to which the results of the analysis are affected by the vehicles in the dataset.
With respect to the comment regarding the variability of responses within and between
models, as discussed in other responses to comments, in order to account for both the within- and
between-vehicle model variability, in addition to modeling a covariance structure, each vehicle
family was treated as a random effect in the final analyses for all three datasets which included
the test results of multiple samples from each vehicle model.
Regarding API's comment that "EPA made a mid-test change to the procedure for
mileage accumulation," the additional on-road mileage accumulation was necessary to fully re-
3 Chapter 6 of the Regulatory Impact Analysis for the Control of Hazardous Air Pollutants from Mobile Sources
Final Rule, EPA 420-R-07-002, February 2007, last accessed on the Internet on 12/04/2013 at the following URL:
http://nepis.epa.gov/Exe/ZyPDF.cgi?Dockey=P1004LNN.PDF.
4 Four Peer Reviews in Support of the Tier 3 Rulemaking: Fuel Sulfur Effects Analysis draft report. U.S.
Environmental Protection Agency, Washington, D.C. Documents available at
http://cfpub.epa.gov/si/sijublic record report.cfm?dirEntryId=240145.
4-69
-------
Chapter 4: Proposed Vehicle Emissions Program
load the catalyst to a level that represents relatively mild in-use driving. The test results with
mileage accumulation constituted a part of the "sulfur level effect" dataset assessing the impact
of high and low sulfur fuel on emissions starting with an "unloaded" catalyst. We evaluated the
impact of mileage accumulation by including the 'mileage' term as well as the 'sulfur level' and
'mileage' interaction terms in our statistical modeling. The mixed model results showed that the
interaction term was not statistically significant, suggesting that the rate of sulfur loading does
not vary by accumulated mileages after the clean-out between high and low fuel sulfur levels.
The peer-reviewers supported our statistical modeling by stating that "overall, the methodology
for determining the in-use sulfur effect for models with and without the sulfur and mileage
interaction term appears to be sound."5
To evaluate the robustness of the statistical analyses assessing the overall in-use
emissions reduction between operation on high and low sulfur fuel, a series of sensitivity
analyses were performed to assess the impacts on study results of measurements from influential
vehicles for all pollutants, as documented in detail in the report.6 The sensitivity analyses
showed that the magnitude and the statistical significance of the results were not impacted and
thus demonstrated that the results are statistically robust.
When modeling the covariance structure, the first-order autoregressive structure was
selected. Although we acknowledged in the report that the spatial covariance structure did not
converge, we demonstrated that the selected first-order autoregressive structure is superior to the
compound symmetry matrix. Furthermore, as stated in the report, we expect the estimates of
fixed effects (i.e., the differences in mean emissions between high and low sulfur fuel) to be the
same for different covariance structures, differing only in the standard errors of these estimates.
The peer-reviewers also supported our methodology by stating that "the autoregressive
covariance structure is appropriate."7
Regarding emissions concentration measurements taken from Bags 2 and 3, manufacturer
specifications for analyzer performance are given in Appendix C of the test program report.
However, these are relatively general and do not provide the level of detail that is useful to
address the issue of error introduced by very low measurements. To address the increased level
of (relative) uncertainty in very low measurements, we performed the sensitivity analyses
discussed in Section 7.3 of the report. These exercises suggested that the overall results of the
program changed very little when very low-emitting vehicles or zero-emission observations were
removed from the dataset.
With respect to the comment on the presentation of the results from in-use sulfur program
in terms of percentages, as the commenter points out, it is a widely accepted method for analysis
of log-transformed data. In addition, since the statistical analyses were performed on FTP
5 The Presidential Memorandum is found at: http://www.whitehouse.gov/the-press-office/presidential-
memorandum-regarding-fuel-efficiency-standards.
6 The Effects of Ultra-Low Sulfur Gasoline on Emissions from Tier 2 Vehicles in the In-Use Fleet, EPA-420-R-14-
002.
7 The Presidential Memorandum is found at: http://www.whitehouse.gov/the-press-office/presidential-
memorandum-regarding-fuel-efficiency-standards.
4-70
-------
Tier 3 Summary and Analysis of Comments
composites, as well as each bag results, we do not believe that the effect of fuel sulfur on
emissions is distorted in any way.
American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM)
A second study used by EPA to support its 10 ppm sulfur proposal was the "MSAT (Mobile
Source Air Toxics) Study", conducted in 2005 with several automakers to examine the effects of
sulfur and other fuel properties on nine Tier 2 vehicles. It is interesting to note that this study has
not appeared in literature reviews of fuel effects on emissions including one of the most
comprehensive reports, Coordinating Research Council (CRC) Report E-84 "Review of Prior
Studies of Fuel Effects on Vehicle Emissions" published in August 2008. Several of the
participating MSAT Study automakers are also members of CRC.
The fuel set used in this study was limited and not well designed. It relied upon a base fuel to
which butane, benzene, and sulfur were sequentially added to produce the three main test fuels.
Sulfur levels were 6 ppm for the base fuel and 32 ppm for the sulfur-doped
base+butane+benzene fuel.
The 2005 MSAT study utilized Tier 2 vehicles and examined sulfur effects (6 versus 32 ppm)
with an unrealistic sulfur loading cycle for the 32 ppm fuel (3 hour cruise at 35 mph).
The Tier 3 Draft Regulatory Impact Analysis (DRIA) provides the following short discussion of
the MSAT Study: "In 2005 EPA and several automakers jointly conducted a program that
examined the effects of sulfur and other gasoline properties, benzene, and volatility on emissions
from a fleet of nine Tier 2 compliant vehicles, the "MSAT (Mobile Source Air Toxics) Study.
Reductions for FTP-weighted emissions for the sulfur changes in this program were 33 percent
for NOx, 11 percent for THC, 17 percent for CO, and 32 percent for methane. Given the prep
procedures related to catalyst clean-out and loading, these results may represent a "best case"
scenario that magnifies what would be expected under more typical driving conditions.
Nonetheless, these data suggested the effect of sulfur loading was reversible for Tier 2 vehicles,
and that there were likely to be significant emission reductions possible with further reductions
in gasoline sulfur level."
Having reviewed the publicly available documentation on the above program, we find the details
to be very limited. The vehicles are noted to be from model years 2004-2007 meeting the Tier 2
Bin 5 or Bin 8 emissions standards. At this point, without a list of vehicles tested, we have no
knowledge of the range of technologies tested, but given the model year information, this
suggests that manufacturers provided some pre-production vehicles for this 2005 test program. In
addition, these vehicles would likely not meet Tier 3 emission levels based on their Tier 2 Bin
designations.
EPA indicates that "Given the prep procedures related to catalyst clean-out and loading, these
results may represent a "best case" scenario that magnifies what would be expected under more
typical driving conditions", but the prep test procedure for sulfur loading is unrealistic as defined
in the MSAT RIA "Where a sulfur loading prep was indicated, a 3-hour 35 mph cruise was
4-71
-------
Chapter 4: Proposed Vehicle Emissions Program
conducted immediately before the final drain and fill. The purpose of this prep procedure was to
equilibrate the catalyst with higher sulfur fuel, simulating conservatively the conditions that
might occur in typical suburban driving" (5). No data were presented to indicate equilibration of
emissions before or after the sulfur clean-out and loading procedures.
Further expanding on the above point, an independent review of the MS AT study noted that the
use of different preconditioning cycles between the tests on the high and low sulfur fuels "... will
impact sulfur loading, and this makes the conclusions with respect to sulfur impact highly
questionable".
Without individual vehicle data we are constrained in our ability to provide informed comment
neither on the statistical analysis of this program nor on the impact of individual vehicles on the
overall fleet response. However, we note that EPA recognized the limited nature of this work in
its closing comments on the test program as described in Chapter 6 of the RIA for the MSAT2
rulemaking: "Clearly the data from this scoping study indicate that there may be benefits to
future fuel controls, though in many cases the size of the test program was not sufficient to
determine effects with statistical confidence. At this time, EPA is hoping to conduct a more
comprehensive fuel effects test program, as directed by the Energy Policy Act of 2005, "in
cooperation with stakeholders and other interested parties, to generate new data over the next
several years. We expect that work will produce updated emissions models, as well as sufficient
data to make decisions about future fuels programs."
Marathon Petroleum Company LP (MPC)
-The 2005 MSAT study utilized Tier 2 vehicles and examined sulfur effects (6 versus 32 ppm)
with an unrealistic sulfur loading cycle for the 32 ppm fuel (3 hour cruise at 35 mph)
Our Response:
The MSAT study was published by EPA as part of the MSAT2 rulemaking package. It
was not published separately within peer reviewed literature although it was subject to public
comment as part of MSAT2. Regarding the fuel set used in the study, this "one-dimensional"
fuel design was reasonable given the scope of the study. The fuel effect most relevant to Tier 3,
that of sulfur, was produced by adding a tiny amount of doping agent to the fuel that resulted in a
negligible change to the other fuel properties. Thus an A-B comparison between the doped and
undoped fuels provides a good assessment of emission impacts of sulfur.
With respect to other comments regarding the MSAT Study, EPA would like to clarify
that the MSAT study was merely a scoping program intended to confirm the magnitude of
certain fuel effects (benzene) and examine the potential range of certain others (sulfur) using Tier
2 vehicles. The program is mentioned in Sections III.A.2 and IV.A.6 of the preamble as
corroborating evidence of sulfur effects of significant magnitude, though the actual percentage of
the effects from the MSAT study were not used for inventory or benefits calculations (e.g., in the
MOVES model) within the Tier 3 Rule. The need for a larger vehicle fleet and a more
representative mileage accumulation approach was discussed in the background of the EPA In-
4-72
-------
Tier 3 Summary and Analysis of Comments
Use Sulfur Study, which has served as the primary source of EPA's data on in-use impacts of
gasoline sulfur under the Tier 3 program.
EPAct/V2/E-89 Study
What Commenters Said:
American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM)
The EPAct/V2/E-89 program did not look at sulfur effects (sulfur at 25 ppm). Instead, EPA
determined Tier 2 vehicle-fuel effects for 5 fuel properties: aromatics, ethanol content, RVP, T50
and T90. To determine sulfur effects, EPA then used a separate program (in-use fleet test) to
adjust for fuel sulfur sensitivity.
The EPAct/V2/E-89 gasoline fuel effects study resulted in a new fuel/emissions equation for
intended use in MOVES. This comprehensive assessment evaluated independent fuel effects on
fifteen MY2008 vehicles meeting Tier2 Bin5 standards. A statistically optimal study design was
developed to represent 5 fuel properties: aromatics, ethanol content, RVP, T50 and T90. These
properties were selected based on previous studies as having potential exhaust emissions
impacts. It is acknowledged that "sulfur also affects emissions but due to its impact on vehicle
catalyst, it is necessary to assess the effects of sulfur separately from those of other fuel
properties."
While the EPAct assessment is a detailed document on the experimental design, test procedure
development, and fuel effects summary, a key missing component is the reduction in sulfur as
proposed in this rulemaking. All test fuels, which were evaluated within the EPAct test program
contained sulfur levels of 25 +/-5 ppm, thus the robust fuel effects equation that was developed
based on statistical analysis of the study data has no sensitivity to sulfur.
Fuel effects, as developed by EPA from the EPAct data, differ in some cases from published
studies;
The EPAct Program vehicle test fleet is not representative of the in-use vehicle fleet;
The EPAct Program was not large enough to provide the data needed to resolve non-linear fuel
effects;
EPA's statistical analysis of the EPAct Program data does not address application of results to
in-use conditions;
Very low emission values observed from some vehicles create problems with the statistical
analysis;
Mathematical relationships relating fuel properties to emissions of some pollutants are too
complex and the quadratic form used by EPA is not optimal; and
The specific mathematical form chosen by EPA is not clearly superior to alternatives that
provide significantly different effects when tested on commercial fuels projected by EPA for
2030. [EPA-HQ-OAR-2011-0135-4276-A2.pdf, p. 19]
4-73
-------
Chapter 4: Proposed Vehicle Emissions Program
Marathon Petroleum Company LP (MPC)
The EPAct/V2/E-89 program did not look at sulfur effects (sulfur at 25 ppm). Instead, EPA
determined Tier 2 vehicle-fuel effects for 5 fuel properties: aromatics, ethanol content, RVP, T50
and T90. To determine sulfur effects, EPA then used a separate program (in-use fleet test) to
adjust for fuel sulfur sensitivity
Our Response:
While effects of fuel sulfur levels were not evaluated in the EPAct program, the large In-
use sulfur test program specifically addressed fuel sulfur effects on similar Tier 2 vehicle models
and included many cleaner and "Tier 3 like" vehicles. This dual approach was adopted because it
is not feasible to assess the effects of sulfur and other properties simultaneously, as the complex
behavior of sulfur can be expected to confound the effects of other properties during the course
of a study. Accordingly, the In-use sulfur test program was designed to examine the effects of
sulfur including the storage and release behavior of the catalyst. Accounting for this behavior
requires a different study design than that used to assess the non-sulfur properties. For example,
in studying sulfur effects, it is critical to account for driving history and to apply specific
preconditioning procedures that differ from those applicable to other fuel effects.
With respect to the comment that "fuel effects ... differ in some cases from published
studies," it is not surprising that EPAct results differ from those in other studies. Emissions are
sensitive to many details of how fuels are blended and how testing is conducted. In addition,
calibration strategies implemented differently by different vehicle manufacturers can cause
significant differences in fuel effects across vehicles. Invariably, studies are published based on
a wide variety of designs and procedures. The EPAct program used a large and representative
test fleet (15 high-sales vehicles) along with fuel blending and vehicle testing conducted by
experts to produce the highest quality data possible. More details on program design and data
analysis are available in the EPAct/V2/E-89 program reports available at
http://www.epa.gov/otaq/models/moves/epact.htm.
To address this comment in its original context, we make reference to the technical
analysis underlying it, appended to the API comments as Attachment 13.8
On this topic, the subsection begins with reference to a summary of results in a recent
literature review (CRC E-84).9 The overall approach in the E-84 report is to treat each fuel
property piecemeal, rather than in an integrated fashion. For example, in the chapter covering
"Aromatics and Benzene," test fuels and emissions differences are presented in terms of
differences in aromatics levels, and interpreted as though attributable to aromatics differences.
Similarly, in the chapter on "Vapor Pressure" effects, emissions differences in cited results are
discussed primarily solely in terms of RVP, with minimal reference to other fuel effects. In
addition, for all studies cited, results appear to be presented in terms of cycle aggregates, in
8 Sierra Research, Inc. Assessment of the Emission Benefits of the U.S. EPA 's Proposed Tier 3 Motor Vehicle
Emission and Fuel Standards. SR2013-06-01. Sacramento, CA. June 23, 2013.
9 Hochhauser, A.M. Review of Prior Studies of Fuel Effects on Vehicle Emissions. CRC E-84. Coordinating
Research Council, Alpharetta, Georgia. August, 2008.
4-74
-------
Tier 3 Summary and Analysis of Comments
which start and running emissions are averaged. This approach complicates simple comparisons
to the models derived from EPAct, which evaluated cold-start and hot-running emissions
separately, but did not attempt to combine or aggregate them.
To a large degree, the comparisons to previous studies show a basic misunderstanding of
what the EPAct model coefficients represent. As emphasized in the EPAct report, a model
coefficient for a given property, e.g., "aromatics," does not represent "the effect of aromatics on
emissions," but rather the effect of varying aromatics on emissions if the remaining four
properties were held constant. In reading Attachment 13, it appears that in most cases, the
comparisons listed are made by comparing single EPAct model coefficients to percent
differences obtained by averaging results across sets of fuel blends. However, in such "whole-
blend" results, differences observed reflect changes in all properties of the fuels, not only the
single property under consideration.
These considerations complicate attempts to make simplistic comparisons between
EPAct model coefficients and the averaged results of most fuels studies, which involve
measurements of small numbers of vehicles on sets of fuels in which selected properties are
modified, but without the intensive experimental design incorporated into programs such as
EPAct or the Auto/Oil study.
Nonetheless, despite these caveats, the studies summarized in E-84 appear broadly
consistent with EPAct in terms of aromatics effects on HC and CO. The chapter covering
aromatics effects for gasoline cites approximately 14 studies reporting HC or CO results, varying
greatly in scope and design, of which over half reported that reducing aromatics reduced HC and
CO (or both), with the remainder reporting mixed results or no effects. For the studies reporting
effects, these findings are generally consistent with the EPAct coefficients for aromatics, if we
assume that the cycle aggregates reported are dominated by cold-start emissions, for which the
EPAct aromatics coefficients are positive (i.e., reducing aromatics reduces emissions and vice
versa).
With respect to NOx, results are mixed. We disagree, however, with the conclusion that
the results obtained in EPAct are uniformly inconsistent with previous results. At the outset, it is
important to note that the three sets of analogous models produced prior to EPAct, namely, the
Complex model, the ARB predictive model, and the EPA predictive model, all have positive
main-effects aromatics coefficients for NOx.
In addition, in contrast to most studies, the design for the EPAct study is well balanced
between the two aromatics levels studied and the levels of the four remaining properties. When
the results are averaged by vehicle and by fuel properties, the results suggest a notable increase
in NOx for most vehicles, as shown in the figures below. For example, in Figure 1, which shows
cold-start (Bag 1) NOx results averaged by vehicle and aromatics level, 13 of 15 vehicles show
higher NOx at the higher aromatics level, with the remaining two vehicles showing no obvious
change. This effect is also clearly visible when the results are averaged by the two aromatics
levels and the levels of any of the four remaining fuel properties. Results are similar but not
identical for hot-running (Bag 2) results shown in Figure 2. For running emissions, behavior by
vehicle is more complex, with six of the vehicles showing a decline or no change by aromatics
4-75
-------
Chapter 4: Proposed Vehicle Emissions Program
level, but with the remaining vehicles showing increases. While this set of results suggests an
increase in NOx on average, it also shows variability in vehicle responses, which may be
attributable in some degree to measurement or random error, or also to differences in vehicle
design, calibration or emission-control strategies.
Figure 1. Cold-start NOx results for the EPAct Phase-3 Program, averaged by Aromatics
Level (vol.%) and by Vehicle.
vehicle
-AQ1WA
COHOLLACE
'ODYSSEY
'CALIBER
EXPLCflERXLT
'OlFTUOOKXE
Aromatics
"™CAMRY
F150
SIENNA
'CMC
• IMFWLALS
SILVERADO
'COBALT
"LIBERTY
4-76
-------
Tier 3 Summary and Analysis of Comments
Figure 2. Hot-Running NOx results for the EPAct Phase-3 Program, averaged by
Aromatics Level (vol.%) and by Vehicle.
0030:
O.OK-:
0.01*:
0.01*;
0.012:
MB;
OODB;
OJ006:
0004:
coos:
vdiide
tauter 6,2011 3
AromQtiGS
*A1TMA
EXFLORERXLT
CALIBER
PBO
'OtmOCKXE
FOCUS
SIENNA
•OUIC
•IIUFAIAin
SIL^FWDO
Attachment 13 also notes that according to the studies cited in E-84, reducing RVP generally
reduced HC emissions (i.e., NMHC), whereas the EPAct models suggest the opposite effect, i.e.,
the main-effects RVP coefficients are negative for both start and running emissions. We hold
that the coefficients for the HC and CO models are consistent with each other, and with
engineering judgment. However, in this context it is important to reemphasize that the
coefficient represents a change in emissions associated with a fuel property, if all others are held
constant.
The negative RVP coefficient for start emissions suggests that, other fuel properties held
constant, a less volatile fuel burns less efficiently under start conditions, resulting in emission of
more partially burned and unburned HC, hence elevated HC emissions. The negative coefficient
for running emissions may appear less intuitive at first glance, unless we recall that increasing
RVP while holding the remaining properties constant would require addition of heavier
components to prevent shifts in T50 and T90. The predicted increase in emissions would be
associated with a change in RVP, but probably attributable to the changes in other properties.
4-77
-------
Chapter 4: Proposed Vehicle Emissions Program
While keeping this caveat in mind, however, if concurrent changes in RVP and T50 are
viewed jointly, the fact that the EPAct coefficients for RVP and T50 are opposite in sign is also
internally consistent and easily understood. The interpretation is that heavier, less volatile fuels
(T50 increasing, RVP decreasing) tend to burn less cleanly than lighter, more volatile fuels (T50
decreasing, RVP increasing).
Attachment 13 notes the "mixed effects" for CO, i.e., that the linear coefficients for T50
are negative for cold-start emissions but positive for hot-running emissions. For start emissions,
the negative coefficient for T50 is consistent with the positive coefficient for HC, meaning that
because heavier fuels emit more unburned and partially burned hydrocarbons, they emit less CO.
However, during hot running operation, the situation is somewhat different. Under stabilized
conditions, heavier fuels burn somewhat less efficiently, but unlike start conditions, this reduced
efficiency manifests through increases in both HC and CO.
Commenting on previous fuel-effects models, Attachment 13 notes that the linear effect
HC coefficient for RVP is positive for the Complex and Predictive models, and negative for the
EPAct models. This may be explained by differences in technology between the Tier-0 and Tier-
1 vehicles underlying the Complex and Predictive models, and the Tier 2 vehicles underlying the
EPAct models. For example, improvements in air/fuel ratio control during fuel tank evaporative
canister purge, as well as the relative contributions of cold-start versus warmed-up operating
modes to test cycle composite results, could cause a shift in the primary impact of fuel volatility
on exhaust HC emissions.
Finally, with respect to particulate emissions, Attachment 13 notes that "It has been well
established that adding oxygenates to diesel fuel generally reduces PM emissions, which makes
the finding that oxygenates increase PM in gasoline vehicles interesting..." (page 36). This
conclusion reflects a misinterpretation of the EPAct coefficients, as mentioned above. The fact
that the linear coefficients for ethanol for the PM models are positive does not imply that the
model suggests that "oxygenates increase PM in gasoline vehicles." What the coefficients do
mean is that if ethanol is blended into gasoline with the four remaining properties held constant,
the increase in ethanol concentration by itself would be associated with an increase in PM.
Changes in the other fuel parameters that result from addition of ethanol also have to be taken
into consideration when assessing the potential impacts on PM. Furthermore, the model does not
necessarily show that the increase in ethanol causes the increase in PM. The impacts on PM may
be due to interactions of ethanol with other fuel properties or components.
With respect to the comment that the EPAct Program was not large enough to provide the
data needed to resolve non-linear fuel effects, the commenter's Attachment 13 notes that"...
aggregated datasets are not necessarily superior to a single well-designed dataset..." and that
"... it is very difficult to resolve a large number of non-linear fuel terms in a single program
design... . Given the limited number of fuels in the EPAct study, ... it is not clear that non-
linear effects can be properly resolved. ..."
4-78
-------
Tier 3 Summary and Analysis of Comments
We agree with the first comment and disagree strongly with the second and third. On the
contrary, we argue that for analyses of fuel effects, which tend to be tightly correlated, use of a
single coherent dataset reflecting a well-designed experiment is superior to meta-analyses
involving the combination of multiple datasets embodying differing designs (or no design). One
reason for preferring a single experimental dataset is that it allows the incorporation of vehicle as
a "blocking variable." A dataset including measurements for all fuels on each vehicle, in effect
replicating the experiment on each vehicle, allows variability due to vehicle differences to be
isolated from the fuel effects of interest. This feature is a major advantage, as the vast majority
of variability in any such dataset is contributed by vehicles, rather than by fuels.
The third comment is simply an unsubstantiated assertion. The EPAct design was
produced by a competent and experienced statistician with lengthy experience working in the
field of automotive and fuel emissions. The fuel set was accordingly designed using state-of-the-
art methods to allow characterization of two quadratic terms and four interaction terms. Granted,
as the use of classic factorials in estimation of fuel effects is not practical, the generation of an
optimal design required that these six effects be specified in advance, based on prior knowledge
of fuel effects. While it is not necessarily impossible to estimate additional effects not included
in the design, doing so requires caution. For this reason, we elected to retain models including
only subsets of the 11 terms included in the original design. Aside from other considerations, we
believe these two reasons are sufficient to rely on a single well-designed experiment to estimate
fuel effects. In addition, an independent peer review of the draft analysis report did not raise any
significant concerns with the methods used to produce the final models.10
With respect to the comment that EPA's statistical analysis of the EPAct Program data
does not address application of results to in-use conditions, the statistical analysis for EPAct
focused on producing the most robust findings possible from the test program itself. We
consider the assertion in Attachment 13 that "important questions about the applicability of the
results to the modeling of in-use vehicle emissions ... are not discussed ... include[ing] the
overall representativeness of the database to in-use vehicles and fuels..." to be unfounded.
During the design of the project, the vehicle sample was specifically and carefully designed to
cover the majority of vehicle sales for Tier-2 vehicles at the time (MY2008). In addition, the
fuel set was designed to span the fuel properties of 95% of summer fuels on the market, based on
current AAM fuel survey results.
With respect to the comment that "very low emission values observed from some
vehicles create problems with the statistical analysis", it would be more correct to say that
"censored" or "missing" measurements created issues in analysis, and that vehicles with "very
low" measurements were more likely to have values censored by limitations in the measurement
techniques. The issue of censored measurements and discussion of dilution and background
contamination as causal factors is thoroughly discussed in the project report. In addition, the
issue of censoring was addressed in analysis through the application of widely-accepted standard
10 Four Peer Reviews Supporting TierS: EPAct Analysis Draft Report. U.S. Environmental Protection Agency,
Washington, D.C. Related documents are available at
http://cfpub.epa.gov/si/sijublic record reportcfm?dirEntrvId=240069.
4-79
-------
Chapter 4: Proposed Vehicle Emissions Program
techniques, methods that were deemed appropriate by the peer reviewers.11 Results from this
program have advanced our knowledge base of best practices related to measurement.
API also commented that "mathematical relationships relating fuel properties to
emissions of some pollutants are too complex and the quadratic form used by EPA is not
optimal" and that "the specific mathematical form chosen by EPA is not clearly superior to
alternatives...". The text in Attachment 13 (page 44) notes that some EPAct models are more
complex than others. If we take "more complex" to mean "more terms" it is clear that the
observation is correct but the commenter gives no indication of how we can know when a model
is "too complex." As we make clear in the project report, the model-fitting process was
conducted so as to develop reduced models that contained no more terms than necessary to
achieve the best possible fit to the data, that is, were as simple as possible. An outcome of this
process was that models for hydrocarbons contained more terms than those for NOx and PM.
We have no technical basis on which to question this result. We can note, however, that on the
same page, Attachment 13 cites the recommendation of the DOE/NREL report that "the
complete fuels model" be used. As the complete (i.e., full) models contain all possible terms, this
citation appears to contradict the comment that the reduced models used are "too complex."
Under this heading, Attachment 13 also takes issue with the inclusion of quadratic terms
in the EPAct models, citing this form as "not optimal." On this point, we note that first, linear
models with the inclusion of quadratic terms have been used in development of predictive
models to date, including the Complex and Predictive models. Second, for EPAct, the
experimental design of the fuel set was based on a specified model including quadratic terms for
ethanol and T50. The design was reviewed by several fuel and emissions experts from the
automotive and petroleum industries, who did not raise similar questions about the proposed
model structure during the design phase. Third, it is improbable that given limitations in time and
budget, a sufficient number of experimental test points could have been included (and the
required fuels blended) to fit a more complex deterministic form, even had such a form been
proposed. Fourth, given the empirical, rather than theoretical approach to study design, it is not
clear that a more complex deterministic form would have given better results.
Finally, we can add that due to the application of hierarchy in model fitting, quadratic (or
interaction) terms are included only when the associated linear terms are also included. So, the
relevant unit to consider is not the quadratic term in isolation, but the linear and quadratic terms
taken together. The actual shape described by these two terms depends on their signs and the
sizes of the coefficients relative to each other. The second-order term is included only when it
improves fit and is necessary to describe curvature in the data where it is apparent.
Other Studies on Gasoline Sulfur Effects on Emissions
What Commenters Said:
11 Sierra Research, Inc. Assessment of the Emission Benefits of the U.S. EPA 's Proposed Tier 3 Motor Vehicle
Emission and Fuel Standards. SR2013-06-01. Sacramento, CA. June 23, 2013.
4-80
-------
Tier 3 Summary and Analysis of Comments
American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM)
EPA relies either on older vehicle studies designed to address different issues at the time or on
data generated from vehicles tested on fuels containing sulfur levels outside of the 10 to 30 ppm
range. For this proposal, the Agency makes numerous inaccurate assumptions and data
interpolations that are well outside the scope of those earlier studies. Furthermore, EPA makes
several assertions without supporting data.
EPA tested very few "Tier 3 - like" vehicles and did not distinguish between the sulfur effects
for these vehicles and those for older technology vehicles.
The "Umicore" study used only one PZEV vehicle and the gasoline sulfur (3 versus 33 ppm)
effects on emissions were confounded by changes in other fuel properties.
One study, the "Umicore" study, looks at a single PZEV vehicle (2009 Chevrolet Malibu)
operating on two fuels - a CARB Phase II Cert fuel at 33 ppm sulfur and a "zero-sulfur" EEE-
Lube certification fuel with 3 ppm sulfur. To infer any differences in NOx emissions between 10
ppm and 30 ppm sulfur fuels from this work is beyond our capability, unless we assume linearity
in NOx response to fuel sulfur levels. Confounding any NOx results from this limited sulfur
study are the underlying base fuel properties that although not noted by the authors, we believe
are different and therefore significantly impact the emissions results. Specifically, we believe
there to be differences in distillation properties, chemical composition and oxygenates. The
standard industry practice is to use sulfur dopants to avoid underlying base fuel changes
impacting the emissions. We would also note that results from testing two fuels on a single
vehicle do not provide much guidance on the potential impact of the proposed Tier 3 emissions
regulations. In particular, the "Umicore" paper did not provide a statistical analysis to show
whether the measurements made on the single vehicle tested were (a) statistically significant
and/or (b) broadly characteristic of the underlying technology represented. These concerns as
well as others regarding the Umicore study are further detailed in comments submitted by API to
EPA Administrator Lisa P. Jackson on November 11, 2011 and available in the Docket for this
proposed rule.
EPA proceeds to provide additional support for the "insight" into the fuel sulfur impacts on "Tier
3 like" vehicles by reviewing the data from the "Umicore" study (a single PZEV vehicle) which
we've already discussed above. Thus it would appear that EPA lacks data on Tier 3 emission
level vehicles at the sulfur levels of interest, namely between 10 ppm and 30 ppm, to fully
understand and comment on the appropriate and necessary fuel sulfur levels to allow compliance
with the Tier 3 emissions requirements. Proceeding to a formal decision without providing the
underlying supporting data prevents informed public comment, is a departure from accepted
scientific rigor and ultimately is a violation of the Administrative Procedures Act.
There are many published studies evaluating the impacts of extremely low sulfur levels on
vehicles, although, with the notable exception of the Umicore Study, these studies generally
tested vehicles with emission levels higher than the proposed Tier 3 exhaust standard. Of all of
4-81
-------
Chapter 4: Proposed Vehicle Emissions Program
the available studies, the RIA and the In-Use sulfur study report address only two—MSAT and
Umicore. EPA fails to explain why other available data were not included in its analysis, and the
Agency clearly should have performed the most comprehensive analysis possible. It also needs
to be stressed that even the two studies selected by EPA do not support the conclusions of EPA's
In-Use sulfur study or the proposed 10 ppm sulfur limit. As noted by the review conducted by
Sierra Research, had EPA conducted a more robust analysis of sulfur effects using all of the
existing data on late-model, low-emission vehicles, it would have likely observed significantly
lower responses for pollutants such as NOx and HC than those seen in the EPA In-Use Study.
MECA relies heavily on an SAE paper authored by Ball, Clark and Moser (Umicore) in
attempting to rebut our argument that there are sulfur-insensitive automotive technologies
available today that could be used to facilitate compliance with the proposed Tier 3 emissions
standards without the need to reduce average gasoline sulfur content to 10 ppm. This SAE paper
had reported measurements of the FTP NOx emissions response of a 2009 model year Chevy
Malibu PZEV to test gasolines with sulfur contents of 3 ppm and 33 ppm.
In reply to MECA, we reiterate concerns which API and AFPM had raised in a critique of the
Umicore SAE paper provided in our June 28, 2013 written comments on the Tier 3 NPRM and
also in an earlier November 11, 2011 API response to the AAM proposal to cap gasoline sulfur
content at 10 ppm. Specifically:
• The SAE paper fails to note the differences in the underlying properties of the base fuel used
and how this may confound the results. Specifically, when comparing a CARB Phase II
certification fuel (33 ppm sulfur) with a EEE-Lube certification gasoline fuel ("zero sulfur"),
differences in distillation properties, chemical composition and oxygenates can impact emissions
results. The standard industry practice is to use sulfur dopants to avoid underlying base fuel
changes impacting the emissions.
• Results from testing two fuels on a single vehicle do not provide much guidance on the
potential impact of fuel sulfur levels on the fleet of future Tier 3 vehicles. In particular, the
Umicore paper did not provide a statistical analysis to show whether the measurements made on
the single vehicle tested were (a) statistically significant and/or (b) broadly characteristic of the
underlying technology represented.
Marathon Petroleum Company LP (MPC)
-There were very few 'Tier 3 - like' vehicles tested and EPA did not separate out the effects for
these vehicles from older technology vehicles
-The 'Umicore' study used only one PZEV vehicle and the sulfur effects (3 versus 33 ppm) were
confounded by other base fuel property changes
Manufacturers of Emission Controls Association (MECA)
Numerous published studies have documented fuel sulfur-related deactivation of three-way
catalysts that are the primary exhaust emission control technology used on light-duty and
medium-duty gasoline vehicles. The negative impacts of gasoline fuel sulfur content on catalytic
4-82
-------
Tier 3 Summary and Analysis of Comments
emission controls are highlighted in a newly revised MECA report: "The Impact of Gasoline
Fuel Sulfur on Catalytic Emission Control Systems" (available on MECA's website,
www.meca.org, under Resources » Reports).
EPA has released a thorough and well-designed sulfur effects study on 81 in-use Tier 2 light-
duty gasoline vehicles that clearly showed significant reductions in criteria pollutants in
comparing emissions performance on gasoline with 28 ppm sulfur versus 5 ppm sulfur. Work
published in a 2011 SAE technical paper (SAE paper no. 2011-01-0300) shows similar,
significant emission benefits on a 2009 model year PZEV vehicle operated with 3 ppm sulfur
gasoline versus 33 ppm sulfur gasoline. In this gasoline sulfur effects study, on a 2009 PZEV
passenger car, the results clearly show that the underfloor converter used on the close-coupled +
underfloor PZEV catalytic converter system was susceptible to sulfur-related performance
degradation due to its cooler operating temperatures during the FTP test cycle using a 33 ppm
sulfur-containing gasoline. The loss in NOx performance of this underfloor PZEV converter in
successive FTP tests could be recovered to some extent, or avoided to a large degree, by either
purging stored sulfur off the underfloor converter with the use of a higher speed and load test
cycle (i.e., the US06 test cycle) sandwiched between FTP tests, or using a gasoline with
significantly lower sulfur levels (i.e., a 3 ppm sulfur-containing gasoline).
In a MECA study published in a 2007 SAE paper (SAE paper no. 2007-01-1261), an advanced
three-way catalyst system installed on a large 2006 V8-powered SUV showed clear evidence of
sulfur deactivation in successive FTP testing with aged catalysts using 17 ppm sulfur gasoline.
FTP emissions for this full-size SUV started at the proposed Tier 3, Bin 50 levels and increased
to slightly above proposed Bin 70 levels by the third FTP test, an emissions increase of more
than 80% over three FTP tests. Sulfur deactivation of three-way catalysts negatively impacts the
active precious metal catalysts, oxygen storage materials, and other activity promoters found in
these sophisticated catalysts. The coverage and negative impacts of sulfur poisons on a three-way
catalyst depends, in part, on the temperature history of the catalytic converter(s) found on the
vehicle. Fuel sulfur deactivation of three-way catalyst is most apparent at lower exhaust
temperatures (e.g., catalyst temperatures of less than about 500 degrees C). Sulfur deactivation of
catalysts can be reversed to some degree by exposing catalysts to higher exhaust temperatures.
Exhaust temperatures are expected to cool in the future as manufacturers reduce vehicle waste
heat to meet future vehicle fuel efficiency/greenhouse gas standards. These cooler catalytic
converter operating temperatures cause catalysts to accumulate higher amounts of sulfur poisons
with today's gasoline sulfur levels, resulting in higher emission levels of pollutants at the
tailpipe, including ozone-forming exhaust pollutants like hydrocarbons and NOx. Ultra-low
gasoline sulfur levels of 10 ppm on average are needed to ensure that manufacturers will be able
to meet the proposed Tier 3, Bin 30 fleet average emission standards over their 150,000-mile
useful life for the full range of light-duty vehicles that consumers wish to buy and manufacturers
want to produce.
Work authored by Umicore in a 2011 Society of Automotive Engineers Technical Paper shows
similar significant emission benefits on a 2000 model year PZEV vehicle, operated with three
ppm sulfur gasoline versus 33 ppm sulfur gasoline. In a MECA study published in a 2007 SAE
paper, an advanced three-way catalyst system installed on a large V-8 powered SUV showed
clear evidence of sulfur deactivation and successive FTP testing with aged catalysts using 17
4-83
-------
Chapter 4: Proposed Vehicle Emissions Program
ppm sulfur gasoline. FTP emissions for this full-size SUV started at the proposed Tier 3, Bin 50
levels, and increased to slightly above proposed Bin 70 levels by the third FTP test, and
emissions increased more than 80 percent over three FTP tests. Sulfur deactivation of three-way
catalysts negatively impacts the active precious metal catalysts, oxygen storage materials, and
other activity promoters found in these sophisticated catalysts. The coverage and negative
impacts of sulfur poisons on three-way catalysts depends in part on the temperature history of the
catalytic converters found on the vehicle. Exhaust temperatures are expected to cool in the
future as manufacturers reduce vehicle waste heat to meet future vehicle fuel efficiency and
greenhouse gas standards.
Our Response:
EPA evaluated every applicable previous study regarding gasoline sulfur impacts on
exhaust emissions known to us and we concluded that they were predominantly studies of pre-
Tier 2/LEVI vehicles with a very limited sample of LEV II or later vehicles. We determined that
the extensive EPA in-use sulfur study, which for the FRM has been supplemented with
additional vehicles representing "Tier-3-like" emissions levels, is the largest and most
representative data set of newer and cleaner vehicles albeit still dominated by current Tier 2 and
LEV II vehicles. In response to API comments that the in-use study did not represent future
vehicles not yet in production, EPA and a large manufacturer, Ford Motor Company, performed
sulfur testing on Tier 3 prototype vehicles representing the high volume and challenging LDT3/4
light duty trucks designed to meet the cleanest Tier 3 emission levels that have not previously
been included in any previous studies. Further, in response to API criticism of the original
Umicore study fuel differences, we repeated the Umicore study using the exact same vehicle
with a common base fuel at two sulfur levels, eliminating any doubt that the effects observed
were from anything other than gasoline sulfur. The increase in NOx emissions with increased
gasoline sulfur was approximately double that found in the original Ball/Umicore study.
Additional testing was also performed on other "Tier 3 like" vehicles with similar results. The
results can be found in the Preamble IV.A.6.d and RIA chapter 1 sulfur feasibility discussions.
Based on the new testing performed by EPA and data provided by Ford Motor Company on
"Tier-3-like" low emission vehicles at full useful life and meeting Bin 50 emissions or lower, the
percentage change in NMOG+NOx when going from 30 ppm S to 10 ppm S is much larger than
the overall results we reported from the In-Use Sulfur study in the NPRM. We agree entirely
with MECA's comments and the papers and report cited by MECA are also referenced and cited
within the RIA and within Preamble IV.A.6.
We disagree that the extensive in-use sulfur study supplemented with cleaner Tier 2
vehicles, the repeat of the Umicore study on a high volume passenger vehicle and the testing of
Tier 3 prototype vehicles does not properly support our analysis of the impact of sulfur levels
above 10 ppm. In fact, we believe that sulfur levels above 10 ppm on future vehicles designed to
meet the new GHG standards may result in a larger negative impact as these more efficient
vehicles will have less opportunity to prevent sulfur accumulation in the catalyst. Such an
assessment of light-duty vehicles subject to the 2017-2025 GHG standards is also supported by
comments from Alliance and Global Automakers, Honda and MECA.
4-84
-------
Tier 3 Summary and Analysis of Comments
Data from SGS provided by API also shows an emissions benefit for reducing gasoline
sulfur although the SGS testing alone is insufficient to determine feasibility of compliance with
the Bin 30 fleet-average exhaust emission standards. Testing by SGS was conducted entirely on
smaller mid-size and compact light-duty vehicles. More than one-half of light-duty vehicle sales
in the U.S. are for light-duty trucks. The procedures used by SGS for rapid thermal aging of the
catalysts also did not follow standardized EPA procedures or current industry practices and thus
the exhaust emissions tests are not representative of full-useful-life emissions on the tested
vehicles.12 The comments of the Alliance and Global Automakers regarding the difficulty of
bringing light-trucks into compliance with Tier 3 exhaust emission standards in the absence of
further gasoline sulfur control were in complete agreement with EPA's engineering assessment
within the draft RIA and the final Tier 3 RIA. Neither API, AFPM, nor MFC cited or provided
any data on gasoline sulfur impacts on the full range of vehicles, including light-duty trucks,
subject to the Tier 3 program and the Tier 3 Bin 30 fleet average exhaust emissions standard.
Sulfur Reversibility
What Commenters Said:
American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM) and Marathon Petroleum Company LP (MFC)
Sulfur's impact on catalysts is reversible, delaying rule implementation will not prevent vehicles
from meeting the Tier 3 standards throughout their useful life.
As explained above, January 1, 2017 implementation of the 10 ppm gasoline sulfur requirement
is not necessary to enable vehicles to meet the NMOG + NOx standards. But, even if it is true as
EPA claims that sulfur will negatively impact catalyst performance on such vehicles, that does
not justify the January 1, 2017 implementation date, because the effect of sulfur on catalysts is
reversible. A new study (described in detail in Section II.B of these comments) demonstrates that
exposure to gasoline fuels containing sulfur levels of 80 ppm sulfur has no lasting impact on the
performance of exhaust emission control systems on modern vehicles operated on 10 ppm sulfur
gasoline.
Our Response:
The observations from the In-use Sulfur study were that sulfur contamination is largely
reversible on the Tier 2 vehicles included in the study. However, in order to reverse the sulfur
contamination, a catalyst clean-out procedure involving very aggressive and high speed drive
cycles that push catalysts to thermal limits and sometimes into thermal protection modes were
required. Excessive and unnecessary thermal cycling of the catalyst can cause premature
deterioration. While a clean-out cycle largely returns the catalyst to a state of reduced emissions
levels, subsequent mileage accumulation on the higher sulfur fuel will return the system to a
contaminated state with elevated emission levels largely negating the emission reduction benefits
12 For further details regarding SGS's rapid thermal aging of exhaust catalysts, see the responses to "Current Tier 2
Vehicles Can Comply with Tier 3 Standards Using Tier 2 Fuels" later in this sub-section.
4-85
-------
Chapter 4: Proposed Vehicle Emissions Program
of the Tier 3 program. Further, some vehicles' emission levels after the cleanout were still
elevated on the high sulfur fuel compared with the emission levels following a clean-out with
low sulfur fuel. Thus, while reversibility may lessen the total negative effects of higher sulfur
fuel, it does not eliminate them. Vehicles will still run with elevated emissions until such time as
the vehicle encounters the very aggressive and high speed driving that would facilitate sulfur
removal from the catalyst. Depending on the driving habits of particular drivers, that type of
cycle may occur only rarely or never. Further, even after clean-out some vehicles experience
higher emission levels, and of course once the vehicle begins driving normally the emissions
would again begin to rise. Moreover, as emissions are increased for vehicles in use as well as
new vehicles, implementation of the 10 ppm requirement on January 1, 2017 will have
immediate results in terms of reduced emissions. EPA does not have data on sulfur reversibility
for vehicles in full compliance with the Tier 3 standards and post-2017 GHG standards. The
Tier 3 standards include additional provisions that impact the ability to use commanded
enrichment that previously facilitated sulfur removal during aggressive driving (see discussion of
commanded enrichment in Preamble IV.A.6). The GHG standards are generally expected to
reduce exhaust temperatures during in-use operation as vehicles become more efficient, resulting
in increased sulfur contamination of active catalyst surfaces (see discussion in Preamble IV.A.6
and response to comments regarding Gasoline Sulfur and GHG/Fuel Economy later in this sub-
section).
We discuss the implications of reversibility on lead time for sulfur reductions in Chapter
5.1.1.3 of this document.
Gasoline Sulfur and Sulfur-tolerant Cold Start Emissions Control
What Commenters Said:
American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM)
Sierra Research (10) recently conducted a comprehensive evaluation of the EPA's technical
justification for the proposed Tier 3 regulation as embodied in the emissions inventory estimates,
air quality modeling, emission control and vehicle technology assessments, and related studies
contained in the DRIA and in the public docket. A copy of the Sierra Research (hereinafter
referenced as Sierra) study is attached, in its entirety, to these comments in Attachment No. 13
In its review of EPA's DRIA, Sierra noted a range of technologies available to the automakers
that could be used to comply with the proposed rule, most of which would improve efficiency
even in the absence of any sulfur changes (Draft RIA, p. 1-28). These include the following:
- Increasing cell density;
- Using higher PGM loadings;
- Optimizing air fuel ratio control; and
- Limiting the amplitude of air fuel ratio excursions.
EPA, however, did not assess the actual need for additional reductions in gasoline sulfur content
in light of the emission control technology it expects automakers to deploy can be highlighted in
4-86
-------
Tier 3 Summary and Analysis of Comments
more detail using data from Chapter 1.4.1.4 of the DRIA. This section deals with EPA's
assessment of technology improvements required for large light-duty trucks, which EPA notes
"will be the most challenging light-duty vehicles to bring into compliance...."
Figure 1-6 of the DRIA shows a number of technologies that EPA expects to be deployed in
order to comply with the Tier 3 exhaust standard. EPA did not explicitly discuss the sulfur
sensitivity of these technologies. However, based on the brief descriptions provided, it is
reasonable to assume that the following technologies shown in Figure 1-6 will have little or no
sensitivity to sulfur:
- Hydrocarbon adsorbers;
- Reduction in the thermal mass of catalyst substrates and exhaust system piping; and
- Secondary air injection
EPA notes that "90% of NMOG emissions occur during the first 50 seconds after cold start," and
that "about 60% of the NOx emissions occur in these early seconds". The three technologies
listed above are designed to reduce or eliminate emissions that occur during the early period of
operation, when the engine is cold and the catalyst has not yet reached operating temperature. If
these technologies are employed to provide the bulk of the reduction necessary, then clearly the
need for sulfur reduction is lessened. Unfortunately, it appears EPA did not consider or even
discuss this possibility. These listed technologies are exclusive of changes in engine design that
EPA expects could and would be made to achieve compliance, and which are also not sensitive
to gasoline sulfur levels.
Figure 2 [of EPA-HQ-OAR-2011-0135-4276] shows that if manufacturers could completely
eliminate cold-start emissions, the current level of catalyst technology would allow compliance
with the Tier 3 standards. Furthermore, given that cold-start emissions and cold-start emission
control technologies are not likely to be very sulfur sensitive, little or no reduction in gasoline
sulfur content should be required to achieve compliance. While it is unlikely to be able to
eliminate this source of emissions fully, the table clearly shows that warmed-up emissions are
already at a level that is compatible with the standard, which again calls into question the need
for additional reduction in gasoline sulfur content and EPA's arbitrary selection of the proposed
10 ppm limit. Regardless, it is clear that EPA should have identified the possible emission
control steps that are sulfur sensitive and those that are not sulfur sensitive, and evaluated
whether 10 ppm sulfur was necessary and cost effective.
MECA completely misses the point which API and AFPM were attempting to make with respect
to the use of emission control technologies with lower sulfur sensitivity to reduce cold-start
emissions in order to comply with the proposed Tier 3 standards. MECA asserts that API and
AFPM "...assume that the use of three cold start emission control technologies [hydrocarbon
adsorbers, reduced thermal mass substrates and exhaust piping, and secondary air injection]
.. .could be used to completely eliminate cold-start emissions with today's gasoline sulfur
levels." This assertion takes the API and AFPM comments completely out of context. If one
were to read our comments in their entirety, one would note that we never stated that these three
technologies had zero sulfur sensitivity, nor did we assume that they would be used to
completely eliminate cold start emissions with today's gasoline sulfur levels. Rather, we noted
that cold-start technologies "are not likely to be very sulfur sensitive" and that".. .if
4-87
-------
Chapter 4: Proposed Vehicle Emissions Program
manufacturers could completely eliminate cold-start emissions, the current level of catalyst
technology would allow compliance with the Tier 3 standards" primarily because "warmed-up
emissions are already at a level that is compatible with the standard." A full rendition of our
comments on this point follows (from p. 20 of our detailed comment submittal):
"The three technologies listed above are designed to reduce or eliminate emissions that occur
during the early period of operation, when the engine is cold and the catalyst has not yet reached
operating temperature. If these technologies are employed to provide the bulk of the reduction
necessary, then clearly the need for sulfur reduction is lessened. Unfortunately, it appears EPA
did not consider or even discuss this possibility. These listed technologies are exclusive of
changes in engine design that EPA expects could and would be made to achieve compliance, and
which are also not sensitive to gasoline sulfur levels.
Figure 2 shows that if manufacturers could completely eliminate cold-start emissions, the current
level of catalyst technology would allow compliance with the Tier 3 standards. Furthermore,
given that cold-start emissions and cold-start emission control technologies are not likely to be
very sulfur sensitive, little or no reduction in gasoline sulfur content should be required to
achieve compliance. While it is unlikely to be able to eliminate this source of emissions fully, the
table clearly shows that warmed-up emissions are already at a level that is compatible with the
standard, which again calls into question the need for additional reduction in gasoline sulfur
content and EPA's arbitrary selection of the proposed 10 ppm limit. Regardless, it is clear that
EPA should have identified the possible emission control steps that are sulfur sensitive and those
that are not sulfur sensitive, and evaluated whether 10 ppm sulfur was necessary and cost
effective."
Manufacturers of Emission Controls Association (MECA)
With respect to the need for a 10 ppm average sulfur gasoline standard, MECA has had the
opportunity to review written Tier 3 comments submitted by the American Petroleum Institute
(API) and the American Fuel & Petrochemical Manufacturers (AFPM). In these comments API
and AFPM argue that EPA did not adequately justify the need for a 10 ppm average sulfur
standard for gasoline in EPA's Tier 3 proposal (API & AFPM Tier 3 comments dated June 28,
2013). In their discussion of the technical need for lower gasoline sulfur levels, API and AFPM
refer to three emission control technologies that target cold-start emissions on gasoline light-duty
vehicles and that EPA noted in their Tier 3 Draft Regulatory Impact Analysis (DRIA) as
technologies they expect to be deployed in order to comply with proposed Tier 3 exhaust
standards. These three cold-start emission control technologies are hydrocarbon adsorbers,
reduced thermal mass catalyst substrates and exhaust piping, and secondary air injection. API &
AFPM make the statement in their comments that based on the description of these technologies
provided by EPA in their DRIA, it is reasonable to assume that these technologies have little or
no sensitivity to fuel sulfur levels. Using this assumption of no fuel sulfur sensitivity, API and
AFPM then go onto to assume that these sulfur insensitive cold-start technologies could be used
to completely eliminate cold-start emissions with today's gasoline fuel sulfur levels. They then
go on to show that with zero cold-start emissions of NMOG and NOx, typical warmed-up
emissions are already at a level on light duty vehicles that would allow proposed Tier 3
emissions of 30 mg/mile NMOG+NOx to be achieved with a 50% compliance margin without
4-88
-------
Tier 3 Summary and Analysis of Comments
the need for a lower gasoline sulfur level (Figure 2 in the detailed comments section of the API
& AFPM, June 28, 2013 Tier 3 comments).
This analysis of a Tier 3 compliance pathway that utilizes sulfur insensitive cold-start
technologies is severely flawed in its initial assumption that cold-start technologies like
hydrocarbon adsorbers, low thermal mass catalyst substrates/exhaust components, and secondary
air injection are not sensitive to fuel sulfur levels. Each of these cold-start technologies still relies
on a precious metal-based catalyst to oxidize hydrocarbons or reduce NOx, and these precious
metal-based catalysts have well known sensitivities to fuel sulfur levels. Hydrocarbon adsorbers
utilize zeolite-based materials to adsorb exhaust hydrocarbon constituents under relatively cold
exhaust conditions and then release these stored hydrocarbon species at elevated exhaust
temperatures. Once an adsorber releases these hydrocarbons back into the exhaust gas, a precious
metal-based catalyst is required to oxidize these hydrocarbon species. The adsorber primarily
functions as a temporary hydrocarbon "sponge" that provides time for the catalyst to heat-up and
"activate" the catalytic oxidation reaction. The catalyst must "light-off or activate to oxidize
hydrocarbons during the cold-start portion of the emissions test cycle. There is an extensive body
of literature that clearly shows that precious metal-based catalyst hydrocarbon light-off
characteristics are negatively impacted by fuel sulfur levels. As sulfur accumulates on the active
catalyst surface, the hydrocarbon light-off temperature increases. Hydrocarbon adsorber
effectiveness in reducing cold-start hydrocarbon emissions is tied to the catalyst hydrocarbon
light-off properties which, in turn, are impacted by fuel sulfur levels.
In a similar manner, cold-start technologies like low thermal mass substrates and secondary air
injection help to accelerate the heat-up of the active catalyst but the catalyst is still the agent that
facilitates the chemical reactions of hydrocarbon oxidation and reduction of NOx. Just as in the
case of hydrocarbon adsorbers, the catalyst still needs to be activated or lit-off for the oxidation
and reduction reactions to occur. Low thermal mass substrates or exhaust piping 3 and secondary
air injection only impact the catalyst heat-up process. The catalyst still needs to accomplish the
oxidation and reduction reactions and the catalyst activity/light-off temperature is impacted by
fuel sulfur levels. An example of this sulfur dependence is found in SAE paper number 2013-01-
0300 (authored by Ball and Moser) that reports on the sulfur sensitivity of FTP NOx emissions
using a 2009 model year Chevy Malibu PZEV vehicle. This Malibu PZEV vehicle utilizes
secondary air injection and high cell density, low thermal mass substrates. Figure 11 of this
paper summarizes the NOx FTP emission performance results for each portion of the test cycle:
Bag 1 (cold-start), Bag 2 (warmed-up performance), and Bag 3 (hot start performance). For tests
run with a 33 ppm sulfur gasoline, NOx emissions for each phase of the test cycle, including the
cold-start portion of the test increased with each subsequent FTP test run with the vehicle (a total
of three FTP tests run successively). Cold-start NOx emissions increased from 6.6 g/mi to 8.7
g/mi to 9.2 g/mi over three FTP tests using 33 ppm sulfur gasoline - a 39% increase in NOx
cold-start emissions for the third FTP test compared to the first FTP test. The use of secondary
air and low thermal mass substrates did not make this vehicle insensitive to cold-start sulfur
poisoning. In this case sulfur is accumulating on the available active catalyst surfaces and
negatively impacting the catalysts' cold-start NOx performance (and the catalysts' NOx
performance in the other two phases of the test cycle). Triplicate FTP tests run on this same
PZEV vehicle using a 3 ppm sulfur gasoline did not show any negative Nox emission trends in
the cold-start phase or any other phase of the test cycle. The negative impacts of sulfur on NOx
4-89
-------
Chapter 4: Proposed Vehicle Emissions Program
performance observed with the 33 ppm sulfur gasoline where also largely erased by running
higher speed US06 test cycles between each FTP test. In this case, the higher speed operation of
the vehicle between FTP testing creates higher catalyst temperatures that can purge sulfur that
accumulates on the active catalysts during the cooler FTP test cycle.
MECA is unaware of any cold-start emission control technology that is not impacted by fuel
sulfur levels since ultimately the cold-start emission performance is tied to the precious metal-
containing three-way catalyst performance. API's and AFPM's premise that cold-start emissions
can be zeroed out by a sulfur insensitive technology has no basis in fact. As indicated in our
earlier comments, MECA agrees with EPA's assessment that a critically important element to
ensuring that future gasoline vehicles will be able to comply with EPA's proposed Tier 3
emission limits is EPA's proposed reduction of gasoline fuel sulfur levels to a 10 ppm national
average starting in 2017. Numerous published studies have documented fuel sulfur-related
deactivation of three-way catalysts that are the primary exhaust emission control technology used
on light-duty and medium-duty gasoline vehicles. The negative impacts of gasoline fuel sulfur
content on catalytic emission controls are highlighted in a newly revised MECA report: "The
Impact of Gasoline Fuel Sulfur on Catalytic Emission Control Systems" (available on MECA's
public website, www.meca.org, under Resources » Reports). This MECA gasoline fuel sulfur
report includes the Toyota 2000 SAE paper reference that showed strong sulfur sensitivity on the
emissions performance of a prototype SULEV vehicle that employed a close-coupled three-way
catalyst and an underfloor converter that utilized a combination three-way catalyst plus
hydrocarbon adsorber design. In their published test results both hydrocarbon and NOx FTP
emissions increased significantly when the gasoline fuel sulfur level was increased from 8 ppm
to 33 ppm (additional large increases in hydrocarbon and NOx FTP emissions were observed
when the fuel sulfur level was increased to 150 ppm). The reference for this Toyota paper is SAE
paper number 2000-01-2019. API and AFPM note in their comments that EPA neglected to
include a reference for this work in their Tier 3 proposal.
Our Response:
EPA agrees with the comments submitted by MECA regarding cold-start emissions.
EPA disagrees with the comments submitted by API and AFPM. The technologies listed by API
and AFPM are not as suggested immune to the sulfur penalty. Thus, use of higher sulfur fuel in
combination with these technologies will reduce their ability to control cold-start emissions. API
and AFPM also ignore the engineering limitations within their analysis. For example, substrate
cell density is limited by backpressure constraints. The amplitude of air/fuel ratio excursions is
limited by control systems, feedback sensors and the design of the fuel injection system. The
technologies described in the RIA to address cold start emissions also will not eliminate cold
start emissions and largely do not reduce engine out emission levels. They are generally
designed to allow the usage of the catalyst sooner following the cold start by increasing the
catalyst temperature to more optimal levels earlier in the operation of the engine. As stated in
MECA's supplemental comments, all of these technologies ultimately rely on catalytic activity
and thus all have performance that is negatively impacted by gasoline sulfur. It is important to
note that the effect of sulfur is specifically on the catalyst's ability to perform necessary
functions at almost all temperatures including cold starts therefore the effect of sulfur will still be
present during cold start even for these specific technologies. It is also well documented within
4-90
-------
Tier 3 Summary and Analysis of Comments
the literature cited in the preamble and RIA that gasoline sulfur directly increases PGM catalyst
light-off temperatures (temperature at which the catalysts can effectively oxidize hydrocarbons
and CO and reduce NOx). Additionally, the hydrocarbon adsorber ultimately also uses the
catalyst when the hydrocarbons are later released into the catalyst therefore sulfur will ultimately
also impact the effectiveness of this technology. Furthermore, hydrocarbon adsorbers do not
contribute to reduction of cold-start NOx emissions and one of the technologies in the API and
AFPM list, secondary air injection, can increase cold-start NOx emissions since lean excursions
during air injection can result in an OSC oxidation state that is detrimental to NOx reduction
reactions over the exhaust catalysts.
Finally, in order to meet the stringent Tier 3 NMOG+NOx standards, manufacturers will
need to employ every available technology in some applications. To eliminate or reduce the
effectiveness of a single technology, such as the exhaust catalyst, would result in the inability of
certain vehicles to meet the standards.
Current Tier 2 Vehicles Can Comply with Tier 3 Standards Using Tier 2 Fuels
What Commenters Said:
American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM)
In Table 1-1 of the Tier 3 DRIA, EPA demonstrates that lower gasoline sulfur levels are not
necessary for compliance with the full useful life NMOG+NOx standard as the Tier 2
certification fuel sulfur levels can vary from 15 to 80 ppm and yet there are a number of 2009
model year vehicles that were already certified below the final 2025 standard of 30 mg/mi.
Tables 1-1 and 1-2 demonstrate that there are a range of domestic and foreign vehicles sizes and
types that already meet the 2022 NMOG+NOx standards by being below 50 mg/mi.
EPA indicates that a number of 2009 MY Tier 2 vehicles easily meet the 2025 NMOG+NOx
target of 30 mg/mi even when running on current cert fuel (15-80 ppm sulfur).
As we've already shown, EPA's own data indicate current vehicles can comply with the Tier 3
standards even when using Tier 2 fuels.
Marathon Petroleum Company LP (MPC)
EPA indicates that a number of 2009 MY Tier 2 vehicles easily meet the 2025 NMOG+NOx
target of 30 mg/mi even when running on current cert fuel (15-80 ppm sulfur)
Our Response:
The assessment in Table 1-1 of the Tier 3 DRIA simply reported our finding that a
limited set of vehicle models are able to certify to the Tier 3 standards while operating on Tier 2
4-91
-------
Chapter 4: Proposed Vehicle Emissions Program
or LEV II certification fuels conceivably with sulfur levels above 10 ppm . While the allowed
range of sulfur in current certification fuel can vary, we believe that typical certification fuel is
blended at sulfur levels well below the current 30 ppm average. However, it cannot be
concluded that all vehicles can achieve these low levels or meet the fleet average of 30 mg/mi
with any sulfur level above 10 ppm. In fact, while certification emission levels seeming to
comply with Tier 3 standards may be observed with some manufacturer's current Tier 2 test
vehicles, these vehicle models would be subject to in-use testing where they would be required to
meet full useful life standards. If they experience fuel sulfur levels above what was considered
when they were originally designed, they would likely no longer meet applicable Tier 3
standards. In fact, due to the risk of Tier 3 vehicles encountering higher sulfur level fuels
particularly during the transition of in-use fuel to 10 ppm average, we have implemented a
provision to allow manufacturers to perform a sulfur cleanout procedure for in-use testing similar
to the temporary Tier 2 cleanout provision.
API, AFPM and MFC commented that "EPA indicates that a number of 2009 MY Tier 2
vehicles easily meet the 2025 NMOG+NOx target of 30 mg/mi even when running on current
cert fuel (15-80 ppm sulfur)." Detailed comments from API and AFPM mention that 7 of 81
vehicles in EPA Tier Sulfur Test Program met the Tier 3 Bin 30 mg/mi standard on higher sulfur
gasoline. Tier 3 Bin 30 emissions of 30 mg/mi NMOG+NOx represent full useful life emissions
at 150,000 miles. Upon reviewing data from the test program, EPA identified 6 individual
vehicles (not 7) with NMOG+NOx test results below 30 mg/mi. The test results were from only
2 vehicle families - 2007 Honda Odyssey and 2008 Ford Focus. Emissions results of the
vehicles tested from these two vehicle families are summarized in the table below:
Make
Honda
Honda
Honda
Honda
Honda
Ford
Ford
Ford
Ford
Ford
Model
Odyssey
Odyssey
Odyssey
Odyssey
Odyssey
Focus
Focus
Focus
Focus
Focus
Model
Year
2007
2007
2007
2007
2007
2008
2008
2008
2008
2008
Vehicle ID
M503ASD-
0122S
M503ASD-
0187S
M503ASD-
0194L
M503ASD-
0122S
M503ASD-
025 5 S
N513ASD-
0174S
N513ASD-
0035I7M
N513ASD-
0089S
N513ASD-
017817M
N513ASD-
022 IS
Accumulated
Mileage
35,553
37,693
35,742
34,149
36,434
24,864
35,067
21,607
27,737
24,917
Average
NMOG +
NOx
(mg/mi)
32
33
37
32
28
18
21
18
15
20
NMOG+
NOx
95% CI
(mg/mi)*
±3
±13
±7
±5
±3
±8
±17
±11
±4
±6
Tier 2
Certification
Bin
4
4
4
4
4
3
3
3
3
3
Tier 2
NMOG
Standard @
120,000
miles
70
70
70
70
70
55
55
55
55
55
Tier 2 NOx
Standard @
120,000
miles
40
40
40
40
40
30
30
30
30
30
Equivalent
FTP
NMOG±NOx
@ 120,000
miles
110
110
110
110
110
85
85
85
85
85
*95 % confidence interval based upon a 2-sided student's t-test.
Note that none of the tested vehicles within these two vehicle families approach the
120,000 mile or 150,000 mile full useful life of the of the Tier 3 Bin 30 exhaust emissions
standards or the 120,000 mile full useful life of the Tier 2 emissions standards to which these
vehicles were certified. The accumulated mileages represent approximately 18-32% of a
4-92
-------
Tier 3 Summary and Analysis of Comments
120,000 mile full useful life or approximately 15-25% of a 150,000 mile full useful life and thus
none of the vehicles tested "easily meet the 2025 NMOG+NOx target of 30 mg/mi" unless the
full useful life mileage requirements of either the Tier 3 or the previous Tier 2 standards are
completely ignored. The comments appear to infer that emissions degradation that occurs over
the last two-thirds to three-fourths of a vehicle's life should be ignored or are negligible. The
comments also imply that the approximately 50% Tier 2 compliance margins previously used by
auto manufacturers to ensure emissions compliance that take into consideration production
variations among vehicles and compliance with IUVP can be ignored within an analysis of
whether a particular vehicle is reasonably expected to comply with the Tier 2 Bin 30 exhaust
emissions standards.
While exhaust emissions testing at accumulated mileages between 20,000 miles and
38,000 miles are often useful for comparing emissions differences due to fuel changes or
changes in emission control hardware, emissions at such relatively low mileage are not
representative of the full useful life exhaust emissions to which these vehicles are certified.
Of the five 2007 Honda Odyssey vehicles tested, only one vehicle had average
NMOG+NOx emissions lower than 30 mg/mi (28 ± 3 mg/mi). The remaining vehicles had
average NMOG+NOx emissions above 30 mg/mi. Even if this individual vehicle could
demonstrate NMOG+NOx exhaust emissions at this level at 150,000 miles instead of 36,434
miles, there would still be insufficient compliance margin for a manufacturer to demonstrate
compliance with the standard. All of the other four, nearly identical 2007 Honda Odysseys
tested had average NMOG+NOx emissions over 30 mg/mi even with limited mileage
accumulation (approximately 34,000 to 38,000 miles). This family of vehicles was originally
certified to Tier 2 Bin 4 emissions of 70 mg/mi NMOG and 40 mg/mi NOx (combined
equivalent NMOG+NOx of 110 mg/mi) at a full useful life of 120,000 miles. The average
NMOG+NOx emissions of 28 to 37 mg/mi for these vehicles compared with the level of the
standards to which they are certified does not represent over-compliance with the Tier 2 Bin 4
emissions standards to which this vehicle family was certified. The test results represent typical
emissions for Tier 2 Bin 4 vehicles of this type when considering the vehicles' relatively low-
mileage and the Tier 2 full useful life exhaust emissions standards to which they were originally
certified. Emissions at relatively low mileage are not representative of full useful life exhaust
emissions because they do not account for the deterioration of exhaust catalysts and other
emission control systems as the vehicle continues to accumulate mileage up to the regulatory full
useful life. Even if the emissions results were from tests closer the end of useful life, emissions
at approximately 30 mg/mi do not represent exhaust emissions consistent with a vehicle
complying with Bin 30 since there would be insufficient compliance margin for the manufacturer
to ensure in-use compliance with Tier 3 exhaust emissions standards even assuming a smaller
20-40% compliance margin.
Of the five 2008 Ford Focus vehicles tested, all five vehicles had average emissions that
were below 30 mg/mi, although the result was not statistically different from 30 mg/mi at a 95%
confidence level for one of the vehicles (N513ASD-0035L/M). Vehicle mileage ranged from
just over 21,000 miles to just over 35,000 miles. This family of vehicles was originally certified
to Tier 2 Bin 3 emissions of 55 mg/mi NMOG and 30 mg/mi NOx (combined equivalent
NMOG+NOx of 85 mg/mi) at a full useful life of 120,000 miles. The test results represent
4-93
-------
Chapter 4: Proposed Vehicle Emissions Program
reasonably typical emissions for Tier 2 Bin 3 vehicles of this type when considering the vehicles'
relatively low-mileage and the Tier 2 full useful life exhaust emissions standards to which they
are certified. As with the Honda Odyssey emissions results, the average NMOG+NOx emissions
reported for the 2008 Ford Focus are not representative of exhaust emissions that would need to
be achieved in order for a manufacturer to comply with Tier 3 Bin 30 full useful life exhaust
emissions standards.
API and AFPM also cited emissions data from six additional Tier 2 vehicles tested by
SGS to further assert that current vehicles could comply with the Tier 3 standards even when
using Tier 2 fuels.13 EPA does not believe the data to be representative of the full-useful-life
emissions of future Tier 3 vehicles for the following reasons:
1. The vehicles selected for testing by SGS were not representative of the broad range of
vehicles that will need to comply with Tier 3. All of the vehicles chosen were mid-size
or compact automobiles with 4-cylinder engines and no attempt was made to choose
high-volume vehicles or to sales-weight the selection of vehicles. No light trucks were
tested even though light-trucks represent more than 50% of vehicle sales.
2. The rapid thermal aging cycle used to simulate mileage accumulation and the parameters
selected for the aging cycle appear to have been arbitrarily chosen and did not achieve
exhaust catalyst aging conditions that would allow emissions testing representative of
full-useful-life emissions for the tested vehicles.
All of the vehicles tested were relatively low-mileage, with accumulated mileage of
between 5,000 miles and 11,000. SGS relied upon rapid thermal aging of exhaust system
components using the RAT-A cycle for 225 hours with a catalyst inlet temperature of 825 +/- 20
°C for all vehicles to achieve conditions that they indicated would represent full-useful-life
emissions. Catalyst aging by SGS was not conducted using the EPA Standard Bench Cycle
(SBC) catalyst durability procedure (Title 40 CFR § 86.1823-08 "Durability demonstration
procedures for exhaust emissions") or using an alternative, demonstrated equivalent catalyst
aging procedure. The single catalyst inlet aging temperature selected for all of the vehicles'
exhaust systems does not appear to be representative of an aging temperature that would be
representative for these vehicles during operation over the EPA Standard Road Cycle as required
by 40 CFR § 86.1823-08. The vehicles tested by SGS ranged from a PFI 1.8L 140 bhp
naturally-aspirated vehicle to a midsize, turbocharged, GDI 2.4L 200 bhp vehicle. It would be
extremely unlikely that a single, identical catalyst-inlet aging temperature and identical
cumulative aging time would be appropriate for rapid thermal aging across all six vehicle models
or that such aging would also achieve the same equivalent cumulative mileage for all six
vehicles. Data shown in figure 18 of the SGS report clearly showed remarkably different closed-
coupled catalyst bed temperatures between the vehicles tested by SGS, further indicating that a
single bench-aging temperature was very likely not appropriate for all of the vehicles. It is
impossible to determine how the 225-hour cumulative aging time was arrived at or how it was
13 Vertin, K., Reek, A. "Reversibility of Gasoline Sulfur Effects on Exhaust Emissions From Late Model Vehicles."
API Contract No. 2012-106409, June 20, 2013.
4-94
-------
Tier 3 Summary and Analysis of Comments
determined to be equivalent to 120,000 to 150,000 miles for all of the vehicles tested since the
cumulative hours do not appear to have been determined using the Bench Aging Time equation
from 40 CFR § 86.1823-08 and the rationale behind the choice of aging parameters was not
reported by SGS. SGS stated in their report to API that "Manufacturers have developed
proprietary catalyst aging tests based on RAT-A to correlate the test results with real-world
experience. For this study, it was not feasible to use proprietary aging cycles for each of the
vehicle systems." This statement by SGS did not demonstrate a full understanding of currently
acceptable rapid thermal catalyst aging procedures. Manufacturers can only use proprietary
catalyst aging cycles for which they have demonstrated equivalence to the standardized EPA
SBC procedures (see 40 CFR 86.1823-08). It would have been entirely feasible for SGS to
simply use the EPA SBC procedures with vehicle-specific aging times and temperatures to
achieve an equivalent of 150,000 miles of driving and thus it is not clear why the RAT-A
procedure was chosen for catalyst aging or if the cycle, cycle times and catalyst inlet
temperatures were at all appropriate for aging exhaust system hardware to near full useful life
equivalent mileage for each of the vehicles tested.
Moreover, as the comments show, the vast number of vehicles in the EPA test program
did not meet the Tier 3 standards. As EPA discusses elsewhere, the required reductions are of a
magnitude that for the standards to be met across the regulated fleet, EPA expects manufacturers
to employ advances in technology in all of the relevant areas of emissions control - reducing
engine-out emissions, reducing the time to reach catalyst light-off temperatures, improving
exhaust catalyst durability at 120,000 or 150,000 miles and improving efficiency of fully
warmed up exhaust catalysts. All of these areas of emissions control need to be improved, and
gasoline sulfur reduction to a 10 ppm average is a critical part of achieving Tier 3 levels through
these emissions control technology improvements. The use of 10 ppm average sulfur fuel is an
essential part of achieving Tier 3 levels while applying an array of advancements in emissions
control technology to the regulated fleet. The testing of Tier 2 and Tier 3 type technology
vehicles, as well as other information, shows that sulfur has a very large impact on the
effectiveness of the control technologies expected to be used in Tier 3 vehicles. Without the
reduction in sulfur to a 10 ppm average, the major technology improvements projected under
Tier 3 would only result in a limited portion of the emissions reductions needed to achieve Tier 3
levels.
Gasoline Sulfur and PM Emissions
What Commenters Said:
American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM)
For this proposal, the Agency makes numerous inaccurate assumptions and data interpolations
that are well outside the scope of those earlier studies. Furthermore, EPA makes several
assertions without supporting data. For example, EPA offers no test data that compares PM
emissions from 10 ppm and 30 ppm sulfur fuels and instead just relies on light-duty vehicle
4-95
-------
Chapter 4: Proposed Vehicle Emissions Program
emissions measurements on a 7 ppm sulfur test fuel to support an assertion that the proposed PM
standards are feasible. Detailed analysis and critique of the scientific shortcoming in EPA's
analysis are provided in our specific comments.
EPA offers no test data that compares PM emissions from 10 ppm and 30 ppm sulfur fuels and
instead just relies on light-duty vehicle emissions measurements on a 7 ppm sulfur test fuel to
support an assertion that the proposed PM standards are feasible.
We see no justification for EPA's comment that "FTP PM emissions increased with CO2
emissions for the PFI vehicles".
When reviewing PM feasibility, EPA notes "Sulfur and nitrogen compounds are emitted
primarily as gaseous species (SO2, NO and NO2). Sulfate compounds can be a significant
contributor to PM emissions from stratified lean-burn gasoline engines and diesel engines,
particularly under conditions where PGM-containing exhaust catalysts used for control of
gaseous and PM emissions oxidize a large fraction of the SO2 emissions to sulfate (primarily
sulfuric acid). Sulfate compounds do not significantly contribute to PM emissions from spark-
ignition engines operated at near stoichiometric air-fuel ratios due to insufficient availability of
oxygen in the exhaust for oxidation of SO2 over PGM catalysts." Given that we see little
penetration of stratified lean-burn gasoline engines and diesel fuel sulfur is not under discussion
for this Tier 3 rulemaking, it is not surprising that EPA offers no test data in the PM feasibility
section comparing PM emissions from 10 ppm and 30 ppm sulfur fuels and instead just rely on
data from a 7 ppm sulfur test fuel to support emissions compliance.
Marathon Petroleum Company LP (MPC)
EPA offers no test data in the PM feasibility section comparing PM emissions from 10 ppm and
30 ppm sulfur fuels and instead just rely on data from a 7 ppm sulfur test fuel to support
emissions compliance.
EPA indicate that sulfate compounds can be a significant contributor to PM emissions from
stratified lean-burn gasoline engines and diesel engines but does not support this with any data.
We see no justification for EPA's comment that "FTP PM emissions increased with CO2
emissions for the PFI vehicles".
Our Response:
The purpose of the PM test program on Tier 2 light-duty vehicles was to demonstrate the
feasibility of Tier 3 PM emission standards using Tier 2 vehicles at full useful life. The purpose
was not to investigate the effect of fuel sulfur level on Tier 2 light-duty PM emissions. Sulfate
emissions were measured from the Tier 2 light-duty study. The largest sulfate emissions
measured in the study were 0.13 mg/mile on the FTP cycle, and 0.32 mg/mile on the US06 cycle.
The average sulfate emissions across vehicles were only .021 mg/mile on the FTP cycle, and
0.077 mg/mile on the US06 cycle.
4-96
-------
Tier 3 Summary and Analysis of Comments
Measurements from the Kansas City Light-Duty Vehicle Emissions Study tested in-use
light-duty gasoline vehicles with much higher fuel sulfur content, and found that the sulfate
contribution to the PM emissions was still relatively minor for light-duty gasoline vehicles. The
sulfate emissions constituted less than 1% of PM start emissions, and less than 10% of running
emissions14. As such, even if the fuel sulfur content was increased to 30 ppm in the Tier 2 light-
duty PM emission study, the increase in PM emissions is expected to be relatively minor, and
would not impact the demonstration of feasibility of achieving the proposed Tier 3 PM
standards.
The comment from API regarding FTP PM emissions in relation to CO2 emissions from
PFI vehicles originated from text describing a preliminary graph of the results from the test
program. The text in question does not accurately describe the data. However, the text was only
observational in nature, and was inconsequential relative to the purpose of the associated
Figures, which was to show the technological feasibility of the proposed PM standards. In
response, we have removed the text in question regarding a CO2/PM relationship from the
figures in the final RIA.
Regarding Marathon's comment of there being no supporting data to indicate that sulfate
compounds can be a significant contributor to PM emissions from stratified lean-burn gasoline
engines and diesel engines, it should be very clear that all lean NOx catalytic reduction systems
with sufficient activity to meet Tier 3 with either lean-burn gasoline or diesel vehicles (e.g.,
compact urea SCR systems, NOx adsorption catalysts and variations/combinations of these
systems) all rely heavily on Pt/Pd oxidation catalysts as key system components. As a petroleum
refiner, Marathon should be well aware of the oxidation reactions that form sulfate from SO2
over Pt/Pd catalysts under net oxidizing conditions. These reactions and subsequent hydration
and formation of sulfate particulate are concisely summarized within the text by Heck et al. and
also by numerous other sources.15
Evaluation of Emissions at Gasoline Sulfur Levels of 10 ppm and 30 ppm
What Commenters Said:
Marathon Petroleum Company LP (MPC)
EPA utilized very few studies none of which looked at sulfur effects over the range of 10 to 30
ppm.
14 Sonntag, D. B., R. W. Baldauf, C. A. Yanca, C. R. Fulper C. R. Particulate Matter Speciation
Profiles for Light-Duty Gasoline Vehicles in the United States. J. Air & Waste Manag. Assoc.
Published online December 13, 2013. DOI: 10.1080/10962247.2013.870096.
15 Heck, R.M., Farrauto, R.J., Gulati, S.T. Equation # 8.2 and surrounding text. "Catalytic Air Pollution Control -
Commercial Technology, 2nd Edition." John Wiley and Sons, Inc., 2002.
4-97
-------
Chapter 4: Proposed Vehicle Emissions Program
American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM)
EPA relies on very few studies to support its proposal to lower the average annual sulfur
standard for gasoline below the current level of 30 ppm. None of these studies evaluate vehicle
emission control system response to changes in gasoline sulfur content either between or within
the range defined by the proposed level of 10 ppm S and the current standard (30 ppm S). As
such, EPA can only conjecture on what is necessary to meet the proposed Tier 3 emissions
standards.
In developing the 10 ppm gasoline sulfur average, EPA relied on very few studies to support
their position, none of which look at the linearity of sulfur effects between 10 and 30 ppm to
understand vehicle emission system response over the range of interest.
EPA proceeds to provide additional support for the "insight" into the fuel sulfur impacts on "Tier
3 like" vehicles by reviewing the data from the "Umicore" study (a single PZEV vehicle) which
we've already discussed above. Thus it would appear that EPA lacks data on Tier 3 emission
level vehicles at the sulfur levels of interest, namely between 10 ppm and 30 ppm, to fully
understand and comment on the appropriate and necessary fuel sulfur levels to allow compliance
with the Tier 3 emissions requirements. Proceeding to a formal decision without providing the
underlying supporting data prevents informed public comment, is a departure from accepted
scientific rigor and ultimately is a violation of the Administrative Procedures Act.
Even after all the above data have been discussed, EPA indicated in Chapter 1.2.3.4 of the DRIA
that "A gasoline sulfur standard of 10 ppm also represents the highest level of gasoline fuel
sulfur that will allow compliance with a national fleet average of 30 mg/mi NMOG+NOx."
Nowhere has EPA even tested levels above 10 ppm sulfur other than the Tier 2 baseline
comparison at the current average sulfur level of 30 ppm. This is a serious deficiency in the
overall technical justification for the 10 ppm sulfur level.
Pennsylvania Department of Environmental Protection (DEP )
Neither the proposed rule nor the regulatory impact analysis adequately explains how emission
reductions were estimated. The proposed rule mentions that studies were performed that
examined emission reductions occurring in vehicles when 3 ppm, 5 ppm, and 6 ppm sulfur
gasoline are used, but not 10 ppm sulfur gasoline. How did the EPA determine what reductions
would occur at 10 ppm sulfur? Regarding catalytic converter operation, the proposed rule states
that reactions ' ... can be blocked by sulfur blinding and may be responsible for observation of
reduced NOX activity over [palladium] Pd/ceria catalysts even with exposure to fairly low levels
of sulfur (equivalent to 15 ppm in gasoline)' (78 FR 29864). EPA also indicates that Pd catalysts
will be important to meeting Tier 3 standards. Emissions in vehicles at 6 ppm sulfur gasoline are
low according to the studies mentioned, but at 15 ppm sulfur reactions that reduce NOX are
blocked, which indicates that the catalysts' efficiency may be responding non-linearly to sulfur
levels in gasoline. Did EPA consider that catalyst operation may not be nearly as efficient at 10
ppm sulfur in gasoline as it is at 3 ppm, 5 ppm, or 6 ppm? DEP cannot discern from the
4-98
-------
Tier 3 Summary and Analysis of Comments
explanations given in the proposed rule how the control efficiency of the catalyst using 10 ppm
sulfur gasoline was derived. This is very important because the monetized health benefits are
directly calculated using the estimated emissions reductions.
Our Response:
There have been a number of published studies of the effects of gasoline sulfur levels on
NMOG and NOx emissions, particularly those for Tier 2/LEV II and cleaner emission control
technologies. Although limited, most of the cited studies included testing of fuel sulfur levels
that cross the range of 10 ppm to 30 ppm fuel sulfur and thus provide important insight into the
emissions impacts of reducing gasoline sulfur levels from a national average of 30 ppm sulfur
under Tier 2 to an average of 10 ppm sulfur under Tier 3 (Takei et al, 2000; Ball et al., 2011;
EPA 420-R-07-002, 2007; EPA 420-R-13-002, 2013). At the time of the NPRM, these were the
most recently available studies on effects of gasoline sulfur on exhaust emissions. After the
NPRM, EPA reviewed additional, supplemental studies and data submissions and cites this
information within Preamble IV.A.6 of the Tier 3 final rule. This includes a contract report from
SGS Environmental testing submitted by API as part of its Tier 3 comments (Vertin and Reek,
2013), an updated, peer-reviewed EPA report on emissions impacts of gasoline sulfur on high-
sales-volume Tier 2 vehicles (EPA 420-R-14-002, 2014), data from EPA's repeat testing of the
previously tested (Ball et al., 2011) PZEV Chevrolet Malibu, data from EPA's developmental
Tier 3 Bin 30 Chevrolet Silverado testing and supplemental data submitted by Ford Motor
Company on a Tier 3 Bin 50 Ford Explorer. This data is summarized within the RIA and
Preamble IV.A.6. All of the data was analyzed to characterize emissions changes when reducing
average gasoline sulfur from 30 ppm S to 10 ppm S. Most of the additional data generated by
EPA or submitted to EPA by industry focused on emissions from low-emitting Tier 2 (bins 3 and
lower) or developmental prototype Tier 3 vehicles.
EPA has drawn from all of the analyses identified above to determine that the
relationship between changes in gasoline sulfur content and NOx, HC, NMHC and NMOG
emissions is typically linear. The linearity of sulfur impacts on NOx, NMHC and NMOG
emissions is supported by past studies with multiple fuel sulfur levels all of which compare
gasoline with differing sulfur levels that are below approximately 100 ppm (e.g., CRC E-60 and
2001 AAM/AIAM programs as well as comments submitted to this rulemaking by MECA cited
within Preamble IV.A.6). As stated within Preamble IV.A.6, the relative linearity of the effect of
gasoline sulfur level on NMOG and NOx emissions allows exhaust emissions results generated
within EPA and other studies of gasoline sulfur at levels immediately above or below either 10
ppm or 30 ppm to be normalized to either 10 ppm sulfur or to 30 ppm sulfur. This allowed EPA
to evaluate vehicle emission control system response to changes in gasoline sulfur content
adjacent to or within the range defined by the Tier 3 Standard of 10 ppm S and the current
standard (30 ppm S) for the data cited within Preamble IV.A.6 and the RIA.
EPA Citation of Toyota Technical Paper on Fuel Property Requirements for Advanced
Technology Engines
What Commenters Said:
4-99
-------
Chapter 4: Proposed Vehicle Emissions Program
American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM)
To provide additional support for the proposed change in fuel sulfur levels, EPA comments on
the impact of sulfur on "Tier 3 like" vehicles. In the first instance, EPA indicates that "Emissions
of vehicles certified to the SULEV standard of the California LEV II program, or the equivalent
Tier 2 Bin 2 standards, can provide some insight into the impact of fuel sulfur on vehicles at the
very low proposed Tier 3 emissions levels. Vehicle testing by Toyota of LEV I, LEV II, ULEV
and prototype SULEV vehicles showed larger percentage increases in NOx and HC emissions
for SULEV vehicles as gasoline sulfur increased from 8 ppm to 30 ppm, as compared to other
LEV vehicles they tested." EPA does not include a reference for this work nor does it appear at
first glance to be in the docket (searching the 600+ docket titles for Toyota). Given the lack of
reference and supporting information, we cannot evaluate, provide informed comment, nor
accept the Toyota program outcomes mentioned to support fuel sulfur effects on proposed Tier 3
emission level vehicles.
EPA references test data from Toyota to support their position but do not provide any details or
the reference material.
Marathon Petroleum Company LP (MPC)
EPA references test data from Toyota to support their position but do not provide any details or
the reference material.
Manufacturers of Emission Controls Association (MECA)
The negative impacts of gasoline fuel sulfur content on catalytic emission controls are
highlighted in a newly revised MECA report: "The Impact of Gasoline Fuel Sulfur on Catalytic
Emission Control Systems" (available on MECA's public website, www.meca.org, under
Resources » Reports). This MECA gasoline fuel sulfur report includes the Toyota 2000 SAE
paper reference that showed strong sulfur sensitivity on the emissions performance of a
prototype SULEV vehicle that employed a close-coupled three-way catalyst and an underfloor
converter that utilized a combination three-way catalyst plus hydrocarbon adsorber design. In
their published test results both hydrocarbon and NOx FTP emissions increased significantly
when the gasoline fuel sulfur level was increased from 8 ppm to 33 ppm (additional large
increases in hydrocarbon and Nox FTP emissions were observed when the fuel sulfur level was
increased to 150 ppm). The reference for this Toyota paper is SAE paper number 2000-01-2019.
API and AFPM note in their comments that EPA neglected to include a reference for this work
in their Tier 3 proposal.
Our Response:
4-100
-------
Tier 3 Summary and Analysis of Comments
The specific citations were included within the Preamble to the Proposed Rule,
specifically, preamble citation footnote numbers 229 and 23016:
229 Takei, Y., Kungasa, Y., Okada, M., Tanaka, T. Fujimoto, Y. (2000). Fuel Property
Requirement for Advanced Technology Engines. SAE Technical Paper 2000-01-2019.
230 Takei, Y., Kungasa, Y., Okada, M., Tanaka, T. Fujimoto, Y. (2001). "Fuel
Properties for advanced engines." Automotive Engineering International 109 12, 117-
120.
A search containing search terms for the primary author, "Takei", would also have yielded the
specific citations in question.
Gasoline Sulfur and Lean-burn Gasoline Spark-ignition Engines
What Commenters Said:
American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM)
EPA asserts that sulfate compounds can be a significant contributor to PM emissions from
stratified lean-burn gasoline engines and diesel engines with no supporting data on technology
deployment(s) . .
Lean burn gasoline direct injection (GDI) engine technology is specifically identified in the auto
industry white paper referenced by EPA in the NPRM as demonstrating the need for 10 ppm
sulfur in the United States. However, as we have noted in previous comments on the white paper
(included as Attachment No. 3), the penetration of this technology into the market place in areas
such as Japan and Europe (where gasoline sulfur is capped by regulation at 10 ppm) has been
limited and is not expected to grow. Instead, it is expected that the automakers will rely on other,
more cost-effective technologies which will not require the highly sulfur sensitive and costly
exhaust aftertreatment devices needed with lean-burn engines. In the US, it is expected that the
maximum potential share for lean-burn engines will reach -3% between 2015 and 2020 and
decline thereafter as observed in Japan and Europe, according to research by The Martec Group
(Executive Summary is provided as Attachment No . 8).
This issue was addressed at length in Section LA. EPA boldly asserts that reducing gasoline
sulfur levels to 10 ppm will enable newer technologies that could improve fuel economy. 78
Federal Register 29820. EPA claims this benefit in a single sentence in the proposed rule
concerning lean-burn engines without any rationale or justification whatsoever. Similarly, the
draft RIA contains exactly one sentence regarding lean-burn engines with no supporting data or
evidence. DRIA at 1-31. This cannot be taken seriously as a justification for the proposed
reduction of sulfur to 10 ppm. If EPA does seriously intend for this to be a justification for the
rule, EPA should re-issue the proposal providing its rationale and data for such an assertion so
16 U.S. Federal Register, Vol. 78, No. 98, Tuesday, May 21, 2013, Proposed Rules, footnotes at the bottom of page
29862.
4-101
-------
Chapter 4: Proposed Vehicle Emissions Program
that it can be properly evaluated and commented upon by the public, as required by section
307(d) of the Clean Air Act.
In any event, there is no basis to claim that lean-burn technology is likely to expand in the U.S.
and that lowering gasoline sulfur will enable such expansion. In other regions of the world where
such technology has been introduced, the automobile manufacturers have indeed been scaling
back its use, not expanding its use.
The American Petroleum Institute (API) recently became aware that the Manufacturers of
Emission Controls Association (MECA) submitted "supplemental" comments to the EPA Tier 3
Rulemaking Docket that were critical of comments which we had filed jointly with the American
Fuel and Petrochemical Manufacturers (AFPM) on June 28, 2013 concerning the Tier 3 Notice
of Proposed Rulemaking (NPRM). Specifically, MECA stated that the intent of its submittal
was:
"...to provide additional information and comments on four important topics: 1) the importance
of 10 ppm average sulfur gasoline to meeting proposed Tier 3 emission limits; 2) the synergy
between ultra-low sulfur gasoline and the introduction of cost effective, lean-burn, gasoline
direct injection technology with improved fuel consumption; 3) costs associated with Tier 3
compliance on gasoline light-duty vehicles; and 4) harmonization with ARB's 1 mg/mile LEV
III PM standard."
In this submittal, MECA attempts to rebut the API/AFPM written comments concerning: (a)
EPA's inadequate justification of a technical need for a 10 ppm gasoline S standard, and (b) the
future market penetration prospects for lean-burn GDI technology in the US light-duty vehicle
fleet.
The supplemental comments from MECA are based upon very limited data that have not been
independently validated and that were generated using test fuels with potentially confounded
properties. In addition, MECA's outlook for lean-burn GDI technology in the US ignores the
real-world experience of this technology in the European and Japanese markets where 10 ppm
gasoline sulfur regulations have been in place. Our detailed response regarding each of these
topics is further elucidated below.
MECA asserts that lean NOx adsorber catalysts are the preferred strategy for reducing NOx on
lean GDI light-duty engines and claims that 10 ppm sulfur gasoline is necessary for the use of
this technology in the U.S. However, MECA does not explain why lean-burn GDI technology
failed to enter the European and Japanese markets in significant volumes after these two regions
adopted 10 ppm gasoline sulfur standards, (i.e., market penetration of lean-burn GDI peaked at
-2% and has since been declining.) MECA also alludes to two European OEMs who want to
introduce this technology into the US market. Given the relatively low US gasoline light-duty
vehicle market penetration by European OEMs, combined with the relatively high incremental
cost of this technology, MECA's comments do not contradict an analysis performed by the
Martec Group which estimates that the market opportunity for lean-burn GDI is limited to at
most -3% by 2020.
4-102
-------
Tier 3 Summary and Analysis of Comments
We further note that the Agency agreed with Martec's evaluation of lean-burn GDI technology in
its regulatory impact assessment conducted for the final rulemaking, "2017 and Later Model
Year Light-Duty Vehicle Greenhouse Gas Emissions and Corporate Average Fuel Economy
Standards." Specifically, the EPA did not use lean-burn GDI in the vehicle technology packages
that were evaluated for that rulemaking as it was found less cost effective than other available
technologies. In short, the Agency agreed with API's comments on this issue.
Marathon Petroleum Company LP (MPC)
In any event, there is no basis to claim that lean-burn technology is likely to expand in the U.S.
and that lowering gasoline sulfur will enable such expansion. In other regions of the world where
such technology has been introduced, the automobile manufacturers have indeed been scaling
back its use, not expanding its use.
Manufacturers of Emission Controls Association (MECA)
API and AFPM in their Tier 3 comments are also negative concerning the future application of
lean burn, gasoline direct injection engine technology in the U.S. to comply with future EPA
greenhouse gas/fuel economy standards. Their comments indicate that the market share for lean-
burn engines in the U.S. will reach only about 3% between 2015 and 2020 and decline thereafter
(based on research done by The Martec Group). The auto industry has generally expressed
significant interest in lean GDI engines in comments made at both Tier 3 public hearings held in
Philadelphia and Chicago in April 2013. This auto industry interest is specific to the potential for
lean GDI engines to deliver up to 20% improvements in fuel consumption relative to
stoichiometric GDI engines (which in turn have lower fuel consumption than port fuel injection
gasoline engines). A recent analysis of EPA's estimates of costs to meet their future light-duty
2017-2025 greenhouse gas emission standards by Dr. Timothy Johnson of Corning, Inc. (see
SAE Int. J. Engines 6(2):2013, doi: 10.4271/2013-01-0538) indicates that CO2 reductions will
cost about $1007% CO2 reduced in the 2020-2025 timeframe. According to recent Ricardo
estimates, lean GDI engine technology can deliver about 20% lower fuel consumption at a cost
of about $30/% CO2 reduced. This relatively attractive cost for reducing CO2 emissions relative
to EPA's estimates for costs to reduce CO2 emissions in the 2020-2025 timeframe should drive
auto industry interest and adoption of lean GDI engines to meet future U.S. greenhouse gas
standards.
As MECA and the auto industry have both pointed out in public comments, a 10 ppm sulfur
average gasoline standard is an important enabler for allowing lean GDI engines to meet
proposed Tier 3 emission limits and deliver cost effective CO2 reductions. In our previous Tier 3
comments, MECA noted that lean NOx adsorber catalysts are the preferred strategy for reducing
NOx on lean GDI light-duty engines and NOx adsorber catalysts have known strong sensitivities
to gasoline fuel sulfur levels (see for example Toyota's SAE paper number 2000-01-2019
referred to previously). At least two European auto manufacturers offer lean GDI vehicles that
utilize lean NOx adsorber catalysts for NOx control in Europe and the current gasoline sulfur
levels present in the U.S. market prevent these manufacturers from offering these lean GDI
vehicles in the U.S. market. Emission control manufacturers are working with their automotive
4-103
-------
Chapter 4: Proposed Vehicle Emissions Program
customers to make lean NOx adsorber catalysts more effective and less costly (as shown in the
recent SAE paper number 2013-01-1299, referenced in MECA's earlier Tier 3 comments), but
lower sulfur gasoline is necessary to make lean GDI a viable future option in the U.S. market.
Without 10 ppm sulfur gasoline, manufacturers will be forced to use more costly approaches for
reducing CO2 emissions from future light-duty vehicles.
In addition to the need for a 10 ppm national average sulfur limit on gasoline for compliance
with EPA's proposed Tier 3 emission standards, the availability of ultra-low sulfur gasoline will
also open up opportunities for vehicle manufacturers to develop and commercialize lean gasoline
engines that can provide improved fuel economy benefits relative to stoichiometric gasoline
engines. A number of manufacturers are offering lean gasoline engine options in Europe and are
interested in using this lean combustion approach to meet more stringent, future U.S. fuel
economy/greenhouse gas emission standards. Lean gasoline engines will require the use of a lean
NOx emission control technology to comply with proposed Tier 3 emission standards. Lean NOx
adsorber catalysts are being used in Europe (where the gasoline sulfur cap is 10 ppm) on lean
gasoline engines to reduce NOx emissions from these lean engines. Lean NOx adsorber catalyst
performance is significantly impacted by gasoline fuel sulfur levels - the NOx adsorber function
of these catalysts also strongly adsorb sulfur constituents present in the exhaust. Ultra-low sulfur
gasoline is an important enabler for maximizing the performance of lean NOx adsorber catalysts,
minimizing the duration and frequency of NOx adsorber desulfation events, and maximizing the
potential fuel economy benefits of lean engine operation. A recent SAE publication (SAE paper
no. 2013-01-1299) describes recent efforts to optimize the performance and desulfation
characteristics of lean NOx adsorber catalysts that are targeted for a light-duty lean gasoline
engine application. A national 10 ppm gasoline sulfur average requirement will provide
manufacturers with the opportunity to use lean gasoline engine technology as an option for
meeting future U.S. fuel efficiency/greenhouse gas standards.
Alliance of Automobile Manufacturers (Alliance) and Association of Global Automakers
(Global)
OEMs need a low sulfur content of fuel in order to introduce advanced lean burn gasoline direct
injection (GDI LB) engines that will enable better fuel system efficiency. LB engine technology
requires the use of highly efficient Lean NOx Absorbers or Traps (LNT) because of the
technology's inherent higher "engine out" NOx rates. The LNT is more sensitive to gasoline
sulfur poisoning than conventional automotive three way catalysts because of its specific catalyst
chemistry which binds more strongly with the sulfur. As little as a few parts per million of fuel
sulfur will start to bind to the catalyst active sites and require high temperature (-500-600 C),
fuel enriched regeneration events to restore the LNT efficiency. If fuel sulfur is not near zero,
then more energy is needed to regenerate the catalyst than may be conserved by the lean burn
technology itself, jeopardizing its cost effectiveness.
Some groups have asserted that EPA has not identified any automotive technologies that would
benefit from lower sulfur gasoline and that will be utilized to comply with CAFE/GHG standards
and Tier 3 standards. This assertion is false. The lower sulfur fuel will allow for improved
technology crucial for meeting the new criteria pollutant standards. For example, the use of GDI
technologies is referenced by EPA in the MY 2017 rule technical support document, which
4-104
-------
Tier 3 Summary and Analysis of Comments
notes: "EPA and NHTSA's current assessment is that the availability of ultra-low sulfur (ULS
less than 15 ppm sulfur) gasoline is a key technical requirement for lean burn GDI engines to
meet EPA's Tier 2 NOx emission standards". 10 ppm sulfur will result in increased used of
optimized GDI and similar technologies in gasoline engines, as it has done for diesel engines.
While we recognize that despite 10 ppm sulfur cap standards, adoption levels of complete
GDI/LB technology in Europe and Japan have been low to date, some components of the
technologies have been widely used. The full suite of technologies has not been adopted simply
because emission limits have been less stringent than Tier 3, and therefore not all of the pollution
control equipment is required. Because Tier 3 will require more stringent reductions in criteria
emissions, past GDI/LB adoption in Europe and Japan is not predictive of the developing U.S.
response.
Mercedes-Benz USA, LLC on behalf of Daimler AG
The 10 ppm retail average and 20 ppm refinery gate/25 ppm retail caps are necessary to optimize
fuel efficiency and maintain the function of advanced engine and exhaust aftertreatment
technologies. In 2006, Mercedes-Benz first capitalized on Ultra low sulfur gasoline and the
concomitant fuel consumption that resulted when it introduced in the EU, the CLS 350CGI
which is equipped with stratified lean burn combustion. Depending on the driving mode, engine
speed and load, gains in fuel efficiency can range between 5-15% as compared to conventional
stoichiometric combustion. Fuel efficiency gains and CO2 reductions of this magnitude, required
by the GHG mandate and enabled by lean burn combustion, illustrate the importance of this
technology and the need for universal availability of low sulfur gasoline in the US to the
Mercedes-Benz GHG Compliance Plan.
While it is possible to dislodge the embedded sulfur through various mechanisms which raise the
exhaust gas temperature sufficiently to 'burn off such deposits, in practice, it is far more
difficult to burn off sulfur deposits given the nature of emerging, higher-efficiency combustion
technologies which extract more energy from the fuel resulting in lower operating temperatures
of exhaust devices. These 'sulfur burn off events are required to maintain pollutant conversion
efficiency of exhaust aftertreatment devices, but they come with a substantial fuel economy
penalty. When market fuel sulfur concentrations exceed 20 ppm, fuel efficiency gains can be
completely negated in the effort to maintain regulated emission control function.
Lean burn combustion among those technologies necessary to accomplish a nearly 50 percent
reduction in greenhouse gas emissions requires low sulfur gasoline for peak efficiency. Fuel
efficiency gains in CO2 reductions of this magnitude required by the greenhouse gas mandate are
enabled by lean burn combustion, illustrate the importance of this technology and the universal
availability of low sulfur gasoline in the U.S. to the Daimler Greenhouse Gas Compliance Plan.
Our Response:
There are two primary rationales for reducing gasoline sulfur to an average of 10 ppm As
discussed in the preamble and elsewhere in this document. The first is to provide an immediate
reduction in emissions from the existing Tier 2 fleet. The second is to enable compliance with
4-105
-------
Chapter 4: Proposed Vehicle Emissions Program
the Tier 3 Bin 30 fleet-average emissions standard, with the assumption that the majority of
light-duty vehicles would use stoichiometric-combustion, spark-ignition gasoline engines.
Although both the oil and the automotive industry commented on lean-burn technology and low
sulfur gasoline, we did not rely on the broader commercialization of lean-burn engine technology
as a potential pathway to compliance with the Tier 3 standards, and we do not address in this rule
the impact of fuel sulfur on the performance of vehicles with this technology.
Gasoline Sulfur and GHG/Fuel Economy
What Commenters Said:
Alliance of Automobile Manufacturers (Alliance) and Association of Global Automakers
(Global)
Sulfur Reductions Facilitate More Fuel Efficient Vehicle Technology. As engine efficiency
increases, more work is extracted from the fuel during the combustion process and less energy is
rejected as "waste" heat to the cooling system and in the exhaust system. Certain technologies
such as lean burn combustion and turbocharged induction, promote cooler exhaust temperatures
as compared to naturally aspirated induction and stoichiometric combustion. Such cooler exhaust
gas temperatures both facilitate sulfur deposition and inhibit its removal during periodic "burn
off' cycles. These events are required to maintain conversion efficiency of exhaust after
treatment devices and come at a substantial fuel economy penalty. In the extreme, depending on
the sulfur level present in the fuel, fuel efficiency gains can be completely negated in the effort to
maintain emission control function.
Some groups have asserted that EPA has not identified any automotive technologies that would
benefit from lower sulfur gasoline and that will be utilized to comply with CAFE/GHG standards
and Tier 3 standards (54). This assertion is false. The lower sulfur fuel will allow for improved
technology crucial for meeting the new criteria pollutant standards. For example, the use of GDI
technologies is referenced by EPA in the MY 2017 rule technical support document, which
notes: "EPA and NHTSA's current assessment is that the availability of ultra-low sulfur (ULS
less than 15 ppm sulfur) gasoline is a key technical requirement for lean burn GDI engines to
meet EPA's Tier 2 NOx emission standards". 10 ppm sulfur will result in increased used of
optimized GDI and similar technologies in gasoline engines, as it has done for diesel engines.
American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM)
EPA has failed to demonstrate that reducing gasoline sulfur levels to 10 ppm is necessary to
enable newer technologies that EPA claims could improve fuel economy
Marathon Petroleum Company LP (MPC)
EPA has entirely failed to demonstrate that reducing gasoline sulfur levels to 10 ppm is
necessary to enable newer technologies that EPA claims could improve fuel economy
4-106
-------
Tier 3 Summary and Analysis of Comments
This issue was addressed at length in Section II. A. EPA boldly asserts that reducing gasoline
sulfur levels to 10 ppm will enable newer technologies that could improve fuel economy. 78
Federal Register 29820. EPA claims this benefit in a single sentence in the proposed rule
concerning lean-burn engines without any rationale or justification whatsoever. Similarly, the
draft RIA contains exactly one sentence regarding lean-burn engines with no supporting data or
evidence. DRIA at 1-31. This cannot be taken seriously as a justification for the proposed
reduction of sulfur to 10 ppm. If EPA does seriously intend for this to be a justification for the
rule, EPA should re-issue the proposal providing its rationale and data for such an assertion so
that it can be properly evaluated and commented upon by the public, as required by section
307(d) of the Clean Air Act.
American Honda Motor Co., Inc.
It should also be noted that the Tier 3 standards have an important relationship with other
recently finalized vehicle regulations, the model year 2017-2025 fuel economy and vehicle
greenhouse gas standards. As a result of the latter set of standards, numerous efforts are now
underway to improve the thermal efficiency of advanced internal combustion engines. These
efforts are successfully reducing waste heat, delivering more energy to the wheels. Yet because
catalysts use wasted engine heat to reach operating temperatures, cooler engines mean lower
catalyst operating temperatures. These catalysts are more prone to being poisoned by sulfur in
the fuel. Countermeasures to ensure aftertreatment operability in a sulfur-rich environment
would require burning unnecessary excess fuel, eroding mpg and putting additional greenhouse
gas emissions into the atmosphere - exactly the opposite of the intended fuel economy and
greenhouse gas regulations.
Manufacturers of Emission Controls Association (MECA)
Sulfur deactivation of three-way catalysts negatively impacts the active precious metal catalysts,
oxygen storage materials, and other activity promoters found in these sophisticated catalysts.
The coverage and negative impacts of sulfur poisons on three-way catalysts depends in part on
the temperature history of the catalytic converters found on the vehicle. Exhaust temperatures
are expected to cool in the future as manufacturers reduce vehicle waste heat to meet future
vehicle fuel efficiency and greenhouse gas standards. These cooler converter operating
temperatures cause catalysts to accumulate higher amounts of sulfur poisons with today's
gasoline sulfur levels, resulting in higher emission levels of pollutants like hydrocarbons and
NOx.
Mercedes-Benz USA, LLC on behalf of Daimler AG
As engine efficiency increases, more work is extracted from the fuel during the combustion
process and less energy is rejected as 'waste' heat in the cooling system and in the exhaust
stream. Certain technologies especially lean burn combustion and turbocharged induction,
promote cooler exhaust gas temperatures as compared to naturally aspirated induction and
stoichiometric combustion. Such cooler exhaust gas temperatures facilitate sulfur deposition.
During the combustion process, sulfur, in various compound forms, is present in the exhaust
stream and is readily deposited on the surface of exhaust aftertreatment devices including
4-107
-------
Chapter 4: Proposed Vehicle Emissions Program
oxidation and reduction catalysts as well as NOx adsorption components. It is on these surfaces
that precious metals are placed by design to provide the conversion sites for the chemical
reaction which converts pollutants to harmless gases. This surface contamination with sulfur
reduces the conversion efficiency of exhaust aftertreatment components. In fact, even low levels
of sulfur impede the function of these devices especially if previously subjected to higher sulfur
fuels. However, the cooler exhaust gas temperatures brought about through increased efficiency
associated with lean burn and other advanced technologies also inhibit sulfur removal during
periodic 'burn off cycles.
Our Response:
EPA generally agrees with the comments by Alliance, Global Automakers, MECA, and
Mercedes-Benz regarding the reduction in exhaust temperatures of stoichiometric-combustion
engines as part of GHG emissions compliance and disagrees with the statements by API and
MFC, based on the analysis in preamble Section IV.A.6 and RIA Chapter 1. As described in
preamble Section IV.A.6, EPA agrees that as engine efficiency is improved via reductions in
pumping losses, friction and other means, the general trend will be towards lower exhaust
temperatures that can both increase sulfur adsorption onto active catalytic surfaces and make
removal of sulfur from those surfaces more difficult.
Comments on GHG, CAFE, and CC>2 emissions are beyond of the scope of the Tier 3
rulemaking.
4.1.5. LD Exhaust Standards: Other Issues
4.1.5.1. Early Credits/Early Tier 3 Compliance
What Commenters Said:
Alliance of Automobile Manufacturers (Alliance) and Association of Global Automakers
(Global)
EPA has proposed a mechanism that would allow for the generation of early credits from two
model years prior to the beginning of the Tier 3 standards. There are, however, certain aspects of
the proposal that are unclear. Additionally, manufacturers' efforts to earn early credits can be
frustrated because of the differences in the bin structure between the Tier 2 program and the LEV
Ill/Tier 3 programs.
Rather than dividing the fleet between vehicles over and under 6,000 pounds GVWR, EPA could
simply combine the < 8,500 pounds GVWR emission credits federally and compare those to the
CA+177 State pooled emission credits (appropriately scaled to a nationwide sales). This would
simplify the calculations.
While the language of the preamble and proposed regulation make it clear that the cap is to be
computed based upon the ratio of nationwide (50-state) sales to California-only sales, the
language related for the computation of the potential early credits is less clear. For the latter, it is
4-108
-------
Tier 3 Summary and Analysis of Comments
not clear whether the language was intended to indicate whether this computation should be
based upon nationwide sales or whether it should be based upon only the sales of the "federal"
fleet (i.e., excluding the CA+177 State fleet (i.e., 37-state sales)). EPA staff have subsequently
verbally clarified that the intent is to base this early credit computation only upon the federal
fleet excluding CA+177 States (or 37-state sales). This expectation should be clarified in the
final rule, since it is needed to implement the early credits in just a few years and clarification in
the final rule would avoid EPA being required to later issue a separate guidance document or
technical amendment.
Additionally, the proposal was to compute the "cap" by multiplying California credits by 50-
state sales divided by California sales. However, given that the California LEV III regulations
allow the option (which most, if not all, manufacturers are expected to use) to base compliance
upon the pooled CA+177 State fleets, EPA should revise the cap computation to specify CA+177
State credits multiplied by 50-state sales divided by CA+177 State sales.
For the possible but unlikely case that a manufacturer may not choose to use the CA+177 State
pooling option, the EPA proposed cap based upon the ratio of 50-state to California only sales
could work. However, for consistency it would be best to also have such manufacturers do the
computation on a CA+177 State basis. In this case, the manufacturer could sum all of the credits
from the separate fleets to create a CA+177 State total and then apply the ratio based upon total
sales in the combined states.
Allow LEV Ill/Tier 3 Bins in MYs 2015-2017. Due to the differences in the Tier 2 and LEV
Ill/Tier 3 bin structure, some manufacturers could find it difficult to earn early credits unless
they created unique vehicle designs and/or calibrations solely for the purposes of earning early
credits. The result could be that they would need to sell a fleet mix that would end up having
lower fleet average emissions than their respective California/177 State fleets. However, we
think it should be reasonable to earn early Federal credits for selling vehicles that are cleaner
than required under the Tier 2 program and which would already be designed and sold for
compliance with the California program applicable to MYs 2015 and 2016.
For example, a number of manufacturers would expect to comply with the required LEV III fleet
averages for MYs 2015 and 2016 by selling a fleet mix containing a significant fraction of
ULEVs (i.e., Bin 125) averaged with SULEV-PZEVs (i.e., Bin30). However, there is no Bin 125
or equivalent bin available under the Tier 2 program. This means California ULEVs would either
have to certify to Bin 4 or 5 under the Tier 2 program. If certified to Bin 5 (equivalent to 160
mg/mi on an NMOG+NOx basis), they would not earn any early credit. If certified to Bin 4
(equivalent to 110 mg/mi), the vehicles would have to be designed for lower emissions than
required by the California program. It would be inefficient to have to design unique vehicles for
this purpose, plus it would be challenging (and perhaps even infeasible) to do so given the short
available lead time and the need to comply with the separate 0.04 mg/mi NOx standard
associated with Bin 4.
We understand, however, that EPA does not intend to reopen the Tier 2 program, but we still
believe it is possible for EPA to provide early credits for interim Tier 3 vehicles certified in MYs
2015-2017. The conditions for certification should be the same as available in the California
4-109
-------
Chapter 4: Proposed Vehicle Emissions Program
program (i.e., EO and 120K useful life, although a manufacturer should not be precluded from
using E10 or El 5 and/or a useful life of 150K if they had a reason to do so) If such conditions
are acceptable in MY 2017 there should be no reason for not allowing certification to these
conditions prior to MY 2017.
EPA should make all of the LEV Ill/Tier 3 bins below Bin 160 available for the purposes of
earning early Federal credits as interim Tier 3 vehicles. This would allow manufacturers who
choose to use bins such as Bins70, 50, 30 or 20 on an NMOG+NOx basis to earn early credits in
the Federal program if they sold these same vehicles nationally. Again, the California
certification conditions would apply (i.e., E10 and 150K useful life for Bin 70 and below unless
it was a carryover SULEV meeting special conditions in California that would allow continued
use of EO and/or 120K useful life).
Allowing certification to these NMOG+NOx LEV Ill/Tier 3 bins raises a question as to how
such vehicles should be incorporated into the Tier 2 fleet averaging program which is based only
upon NOx levels. The simplest solution would be to exclude these vehicles from the Tier 2 fleet
average NOx computation. Since the above proposal would not include use of the LEV Ill/Tier 3
bin!60, all of the bins involved would have more stringent equivalent NOx emissions than the
required Tier 2 fleet average level.
Tier 2 Vehicles Tested on California E10: Finally, §86.113-04(a)(2)(ii) allows manufacturers to
"certify 50-state vehicles based upon testing used to meet California's LEV III standards" in MY
2015 and later. However, paragraph §86.113-04(a)(2)(ii)(E) states that "such vehicles are
considered to be Tier 2 vehicles for EPA certification; however, manufacturers must exclude
them from the fleet-average NOx calculation in subpart S of this part." On the surface, this
proposal sounds similar to the option discussed above where vehicles that would actually be
certified to LEV Ill/Tier 3 NMOG+NOx bins and then would be excluded from the Tier 2 fleet
average. However, this provision applies to a totally different situation.
This provision applies to the case where a manufacturer would be certifying Federal vehicles to a
Tier 2 bin using E10 test data that was used for certification of the same vehicle in the California
LEV III program. This would allow the manufacturer the ability to avoid a second test on
Federal EO just to demonstrate compliance with a less stringent Tier 2 bin.
But EPA's proposal to exclude such a vehicle from the Tier 2 NOx average would be
counterproductive. This would prevent a manufacturer from being able to earn Tier 2 credits for
vehicle being certified to Tier 2 bins 2, 3, and 4 simply because they chose the option to use the
California test data to demonstrate compliance. Hence, if the manufacturer needed those credits
to offset other vehicles certified to available Tier 2 bins above bin5, they would have to retest
this vehicle on Federal EO. This defeats the purpose of allowing the use of California E10 test
fuel in the first place.
Also note that the above situation should only apply to vehicles that were certified in California
to bins70 and below since in the affected MYs it is only these bins where California requires the
use of E10. Vehicles in bin!25 and 160 would still be allowed to certify on California phase 2
EO, which EPA has historically accepted for Tier 2 certification purposes. Hence there would be
4-110
-------
Tier 3 Summary and Analysis of Comments
no relaxation of stringency in Tier 2 compliance if vehicles tested on California E10 where
allowed to be counted in the Tier 2 fleet average. These would be vehicles designed for
compliance with California bin70 or below but which would be certified in higher available Tier
2 bins (i.e., most likely bin 3 or 4).
The provision in §86.113-04(a)(2)(ii)(E) should be deleted as it defeats the whole purpose of
having the (2)(ii) in the first place.
Recommendation: EPA should consider simply comparing the sum of the 0-6,000 pound GVWR
and 6-8,500 pound GVWR credits to the cap at the end of MY 2017. Also, EPA should revise
the language on the computation of the early credit cap to be based upon the CA+177 State credit
total and total sales rather than be based only upon only California credits and sales. Given the
complexity of this issue, EPA should consider providing an example in the preamble to the final
rule.
To facilitate the ability to earn early credits, EPA should allow early certification to all of the
LEV III/Tier3 bins below Bin 160 under the same conditions that would be available in the
California program for MYs 2015 and 2016. Such vehicles could be excluded from the Tier 2
NOx fleet average to avoid issues with the fact that these vehicles would not have a separate
NOx standard.
Additionally, the provision in §86.113-04(a)(2)(ii)(E) should be deleted.
American Honda Motor Co., Inc.
Another important aspect related to the harmonization of the Tier 3 and LEV III programs is the
treatment of Tier 3 early credits. Honda strongly urges EPA to allow not just federally-certified
Tier 2 vehicles, but interim Tier 3 vehicles, to generate early credits (calculated against the Tier 2
standard) in model years 2015 and 2016. Doing so would accelerate nationwide adoption of
cleaner emissions control technology prior to the formal introduction of Tier 3, improve
automakers' flexibility in meeting the standards in the early years of the Tier 3 program, and
ease the transition toward more stringent standards as the program fully phases in.
As EPA is aware, many automakers - including Honda - produce models certified to ULEV125
in California (and Section 177 states), and certified to Tier 2 bin 5 (or Bin 160) federally. Given
the short lead time, we believe it is not practicable to certify these models to the Tier 2 Bin 4
(Bin 110) level in model years 2015 and 2016 while still maintaining a sufficient margin to
ensure compliance across a mass-produced line. We do, however, believe there would be
sufficient compliance margin to meet interim Tier 3 (Bin 125) levels, and that those models
could thus see an accelerated nationwide deployment strategy at the cleaner Bin 125 level. For
the purpose of generating early Tier 3 credits during the 2015 and 2016 model years, we see no
sound reason for treating interim Tier 3 models differently than models certified to formal Tier 2
bins.
It is our understanding from conversations between EPA and OEM trade associations that the
agency does not intend to reopen the Tier 2 standards; Honda understands the technical rationale
4-111
-------
Chapter 4: Proposed Vehicle Emissions Program
for that position, such as the lack of a separate NOx standard at the Bin 125 level. Barring
changes to Tier 2 levels, we believe that an equitable, alternative resolution to this early credit
issue is to allow early credits for interim Tier 3 vehicles, which could comply with Tier 3 using
E15 or with LEV III using E10 and the accompanying 150,000-mile full useful life, and we urge
the agency to do so. (We would further request that the agency, for clarification purposes,
articulate whether interim Tier 3 vehicles earning early credits could use EO and 120,000-mile
full useful life, as those values are also consistent with the Tier 3 phase-in provisions.)
The agency notes in the proposed rule that the early credit program is designed to accomplish
three goals:
(1) Encourage manufacturers to produce a cleaner federal fleet earlier than otherwise required;
(2) provide needed flexibility to the manufacturers to facilitate the "step down" from the current
Tier 2 Bin 5 fleet average required in MY 2016 to the LEV Ill-based declining fleet average in
MY 2017; and (3) create a Tier 3 program that is equivalent in stringency to the LEV III program
such that manufacturers will be able to produce a 50-state fleet at the earliest opportunity.
Honda believes that allowing interim Tier 3 vehicles to generate early credits will improve the
agency's ability to obtain its stated early credit-related objectives, accelerating a 50-state
deployment strategy, and bringing cleaner vehicles to market sooner than would otherwise occur.
BMW of North America, LLC
CARS's LEV III program allows the option to use combined NMOG+NOx for LEV II vehicles
included in a manufacturer's fleet average. The proposed Tier 3 language addresses the inclusion
of the so called interim Tier 3 vehicles; however, allowing a combined NMOG+NOx standard is
not entirely clear in the proposed language. BMW kindly requests clarification of the language
such that MY 2015 and 2016 Tier 2 vehicles may certify to the combined NMOG+NOx
standard.
General Motors LLC (GM)
For the exhaust FTP NMOG+NOx fleet average, we support EPA's innovative approach of
allowing early credits and then capping them entering the 2018 model year in proportion to
credits a manufacturer has in its California credit bank (or, as requested in the Alliance/Global
comments, California +177 State pooled credit bank). We believe this approach will address
EPA's three goals stated in the preamble: 1) encourage a cleaner Federal fleet earlier than
required; 2) give manufacturers needed flexibility to go from Tier 2 to LEV Ill/Tier 3; and 3)
make Tier 3 equivalent in stringency to LEV III to facilitate a 50-state fleet.
Under 86.113-04(a)(2) of the NPRM, EPA is suggesting a method for manufacturers to obtain
nationwide certifications prior to the introduction of Tier 3. Essentially EPA will accept LEV III
certifications as early as model year 2015, as long as the specified requirements are met. GM
appreciates EPA's allowance for manufacturers to execute nationwide applications; it's a
manpower savings and a reduction in complexity and is consistent with the goal of transit!oning
to a harmonized national program. The only concern GM has with the proposal is under
4-112
-------
Tier 3 Summary and Analysis of Comments
paragraph (ii)(B), which states: (B) The manufacturer must also use this E10 fuel for fuel
economy measurements, with any appropriate corrections related to ethanol content in the fuel.
[EPA-HQ-OAR-2011-0135-4288-A1, pp.9-10] (ii)(B).
Our Response:
Several auto industry commenters suggested additional provisions to facilitate earlier
harmonization between Tier 3 and LEV III and streamlining of development and certification of
vehicle models. Specifically, these commenters requested the ability to have vehicles certified to
the Tier 3 standards in MYs 2015 and 2016. They commented that this would allow them to
develop, certify and sell a vehicle model for all 50 states, reducing the complexity of potentially
different federal and California requirements in MYs 2015 and 2016. Additionally, auto industry
commenters noted that the Tier 3 program provides more flexibility in the certification bin
structure compared with the existing Tier 2 program, providing them additional opportunities to
generate early credits.
We are finalizing a provision to allow manufacturers to certify to Tier 3 standards starting
in MY 2015 as "Early Tier 3" vehicles. Manufacturers will have the option to certify their
vehicle models to meet the Tier 3 emission requirements, including requirements for test fuel and
useful life, in MY 2015 and 2016, and MY 2017 for vehicles over 6000 Ibs. GVWR, for all
LDVs, LDTs, and MDPVs, which otherwise would be required to begin in MY 2017 under the
primary program. As an example, a manufacturer choosing to certify a vehicle as Early Tier 3
can bring the same vehicle models certified to LEV III standards in MY 2015 or 2016 into the
Early Tier 3 program by meeting all the same requirements under the primary Tier 3 schedule.
There would not be a Tier 3 fleet average requirement for FTP or SFTP NMOG+NOx in MY
2015 or 2016 (and 2017 for vehicles over 6,000 Ibs GVWR) if all the same vehicle models
certified to LEV III are also certified as the Early Tier 3 vehicles meeting the same LEV III
emission standards and also the Tier 3 additional requirements (high altitude, and cold CO and
hydrocarbons). These Early Tier 3 vehicles would replace any Tier 2 offering of the vehicle
model consistent with the LEV III offering replacing the LEV II models. If a manufacturer
chooses to certify only a portion of their LEV III vehicle models as Early Tier 3 vehicles in a
given MY, they will be required to meet the LEV III fleet average requirements in that MY for
those models certified as Early Tier 3 vehicles.
The same carry-over provisions that begin in MY 2017 will also apply in MY 2015 and 2016.
This includes the ability to carryover Tier 2 test results using Tier 2 fuel into the Tier 3 Bins
above Bin 70.
The early credit program we are finalizing includes several distinct provisions. The first
provision allows manufacturers to generate early Tier 3 credits against the current Tier 2 Bin 5
requirement in MYs 2015 and 2016 for vehicles under 6,000 Ibs GVWR and MYs 2016 and
2017 for vehicles greater than 6,000 Ibs GVWR. We proposed and are finalizing a provision
limiting the application of the early Tier 3 credits to the following conditions:
• Early Tier 3 credits generated as described above could be used without limitation in
MY 2017 on the portion of the fleet entering the Tier 3 program in that MY.
4-113
-------
Chapter 4: Proposed Vehicle Emissions Program
• Early credits generated from all vehicles in the light-duty program in MY2015
through MY 2017 used for compliance in MY 2018 and beyond will be capped at an
amount equal to the lesser of the manufacturer's federal credits as calculated above or
the manufacturer's LEV III credits scaled up by the ratio of 50-state sales to
California and LEV III required states sales. This limitation accounts for the fact that
some LEV III credits may have begun to expire and will no longer be eligible as a
basis for Tier 3 early credits.
By capping the available federal Tier 3 early credits, we believe that the two programs,
LEV III and Tier 3 will be at parity in terms of relative stringency starting in MY 2018. In
addition, because the number of Tier 3 early credits that can be used is based on the number of
LEV III credits that the manufacturer has generated, there may be additional motivation for
manufacturers to over-perform in California during the initial model years, accelerating emission
reduction benefits.
The provision in §86.113-04(a)(2)(ii)(E) was deleted as suggested because the provisions
we are finalizing provide for both Tier 2 and early Tier 3 in MY 2015, 2016 and 2017
certification paths eliminating the issue of the LEV III combined NMOG+NOx standards not
matching the Tier 2 independent NMOG and NOx requirements.
4.1.5.2. Credit Life and ABT
What Commenters Said:
Alliance of Automobile Manufacturers (Alliance) and Association of Global Automakers
(Global)
Credit Life (FR29868): We support the 5-year credit life and harmonization of this provision
with California. We support the ABT program, which maintains program stringency while also
providing flexibility. We also support harmonization of the credit life between EPA and
California.
Hyundai Motor Group
We also appreciate flexibilities included in the proposal, such as the ability to average across the
fleet for most pollutants. Additionally, we support the averaging banking and trading provisions,
which allow companies to obtain or sell emissions credits with other manufacturers, and the
carry forward and back of credits, which are helpful flexibilities.
New York State Department of Environmental Conservation
Tier 2 vehicle emissions credits should not carry over to Tier 3. EPA proposes to prevent the use
of existing Tier 2 credit balances to meet Tier 3 requirements. EPA also proposes to cap the use
of early credits to comply with Tier 3. New York fully supports these provisions, which will
expedite harmonization with LEV III and ensure that compliant vehicles are actually produced in
the early years of the program.
4-114
-------
Tier 3 Summary and Analysis of Comments
The averaging, banking, and trading program is an important means of providing vehicle
manufacturers with compliance flexibility. To that end we support credit carry forward and carry
back provisions identical to those used in the LEV III program.
Pennsylvania Department of Environmental Protection (DEP)
DEP supports flexibility for automobile manufacturers.. .through averaging, banking, trading
programs, or other mechanisms.
Our Response:
We did not receive adverse comments on our proposed design of the ABT program. We
are finalizing these provisions as proposed, with one exception. Based on conversations with
representatives of the auto industry, we determined that, with certain restrictions, Tier 3 credit
life can be temporarily extended in a way that would address uncertainties about the middle
years of the program, with no adverse impacts on the overall emission reductions of the program.
Specifically, we are finalizing a credit life of 8 years for credits generated in MYs 2017-2022 for
the FTP and SFTP NMOG+NOX fleet average standards. For the heavier light-duty vehicles,
the 8-year credit life begins for credits generated in MY 2018. (Note that, as proposed, credits
generated under the Early Tier 3 Credit provision (preamble Section IV.A.T.a) are limited to 5-
year life, and are not affected by the longer credit life provision.)
For credits generated in MYs 2023-2025, the credit life declines by one year of credit life
annually, with credit life stabilizing at 5 years for credits generated in MYs 2025 and later. That
is, credits generated in MY 2023 have a 7-year life, in MY 2024 a 6-year life, and in MY 2025
and later a 5-year life. However, while credits can be generated, banked, and used internally for
the extended time periods, credits cannot be traded to other manufacturers after 5 years.
4.1.5.3. Useful Life for Cold CO and Cold NMHC Standards
What Commenters Said:
Alliance of Automobile Manufacturers (Alliance) and Association of Global Automakers
(Global)
Cold CO and NMHC Useful Life: The preamble of the Tier 3 regulation states, "We are not
proposing new emission requirements for any vehicle or fuel over the cold temperatures test
cycles (i.e., the 20 °F cold CO and NMHC tests)'. We understand this language to mean that in
addition to the exhaust emission standards for the cold CO and cold NMHC tests, there would
not be any change in the useful life for cold CO and cold NMHC. The current useful life is 5
years or 50,000 miles for cold CO and for cold NMHC is 10 years or 120,000 miles for LDV and
LLDT and 11 years or 120,000 miles for HLDT and MDPV. The proposed regulations, however,
do not carry over the language that specifies the current useful life for these standards. Based on
the preamble language, it is not EPA's intent to extend the useful life of the cold CO emission
4-115
-------
Chapter 4: Proposed Vehicle Emissions Program
test to 10 years or 120,000 miles (or 15 years/150,000 miles) under Tier 3, and therefore, the
cold CO useful life language from §86.1805-12(d) should be added to the new section,
§86.1805-17, to clarify the useful life for cold CO. Likewise, based on the preamble language,
we believe that it is EPA's intent to retain the cold NMHC standards at the current useful life
requirements and would not apply the voluntary 15 years/150,000 miles standards for cold
NMHC even if the vehicle is certified to 15 years/150,000 miles for the FTP and SFTP
standards.
Thus, EPA should also add clarifying language to §86.1805-17 that cold NMHC useful life is not
impacted.
Recommendation: EPA should add the following language to §86.1805-17:
"(f) Where cold CO standards are applicable, the useful life requirement for compliance with the
cold CO standard only, is 5 years or 50,000 miles, whichever occurs first.
(g) Where cold NMHC standards are applicable, the useful life requirement for compliance with
the cold NMHC standard, only, is 10 years or 120,000 miles, whichever occurs first for LDV and
LLDT, and 11 years or 120,000 miles, whichever occurs first for HLDT and MDPV."
Our Response:
The useful life requirements for cold CO and cold NMHC standards are unchanged for
Tier 3 and will remain the same as under the Tier 2 program. EPA has modified the language in
§86.1805-17 to clarify the useful life requirements for cold CO and cold NMHC standards in a
way that is equivalent to the commenter's suggested language.
4.1.5.4. Enrichment Limitation
What Commenters Said:
California Air Resources Board (CARB)
Comment 1 - SFTP Enrichment Limitation Requirement
CARB supports U.S. EPA's Tier 3 proposal to reduce the tolerance of additional enrichment
from six to four percent of the air to fuel ratio of Lean Best Torque. CARB believes the reduced
tolerance is appropriate and feasible due to the improved fuel control utilized by today's
vehicles. CARB intends to propose alignment with this proposal once the Tier 3 program is
finalized.
Ford Motor Company (Ford)
Enrichment Limitation for Spark-Ignition Engines
EPA has proposed a new Lean Best Torque (LET) definition, tolerance, and fixed spark mapping
procedure designed to limit excessive in-use enrichment. Ford has several major concerns with
this proposal: (1) there is no universally accepted "textbook" definition of LET, (2) there is no
4-116
-------
Tier 3 Summary and Analysis of Comments
standardized engine mapping procedure to measure LET, and (3) LET can vary significantly
depending on engine technology and the reference test fuel. For these reasons, we recommend
that EPA carry over existing Tier 2 regulatory language (or adopt the LEV III LET
requirements) and establish a joint government / industry committee to determine a universally
accepted SAE LET mapping procedure. This would help to ensure consistent, repeatable results,
and enable an unambiguous interpretation of LET between EPA and manufacturers even as
engine and vehicle technologies evolve over time.
Recommendation: Ford recommends that EPA retain the Tier 2 LET requirements, or
alternatively, adopt the LEV III requirements, until a universally accepted LET definition and
mapping process can be established by SAE.
Our Response:
The requirement to not exceed LET +6% when SFTP standards were first adopted was
based on variation in fuel and emission control hardware at that time. More stringent emission
standards that have occurred since that time have resulted in reduced part variability that allows
us to reduce the 6% tolerance to 4%. The LET definition included in the FRM is to provide
clarity regarding expected enrichment amounts and eliminate the risk that enrichment may
interact with the control of ignition timing. Without this clarification, enrichment limitations can
be interpreted differently and result in excessive emissions. The FRM allows for alternative
definitions of LET with approval if the finalized definition does not properly achieve the
enrichment limitation goals of a unique technology.
4.1.5.5. AECD Requirements
What Commenters Said:
General Motors LLC (GM)
Expansion of AECD Requirements: EPA is proposing adding three additional requirements
under paragraph (11) of 86.144-01: "Information requirements: Application for certification and
submittal of information upon request."
(i) For any AECD uniquely used at high altitudes, EPA may request engineering emission data to
quantify any emission impact and validity of the AECD.
(ii) For any AECD uniquely used on multi-fuel vehicles when operated on fuels other than
gasoline, EPA may request engineering emission data to quantify any emission impact and
validity of the AECD.
(iii) For Tier 3 vehicles with spark-ignition engines, describe how AECDs are designed to
comply with the requirements of §86.181 l-17(d). Identify which components need protection
through enrichment strategies; describe the temperature limitations for those components; and
describe how the enrichment strategy corresponds to those temperature limitations. We may also
require manufacturers to submit this information for certification related to Tier 2 vehicles.
4-117
-------
Chapter 4: Proposed Vehicle Emissions Program
While the first two proposed requirements are within the scope of EPA's current AECD
regulations ("EPA may request additional data"), the third requirement (1 l)(iii) is an expansion
of the current AECD required submissions. GM is opposed to this additional work for several
reasons. This adds additional burden on manufacturers to produce this detail for every
submission to EPA. This also falls under the same logic as the first two proposed additions -
EPA already has the authority to ask for this information from manufacturers as needed. Our
biggest concern is the precedent this sets going forward for on-road spark-ignited applications.
Today, the heavy duty diesel engine applications require a highly burdensome and detailed
AECD submission. The work necessary to complete our heavy duty diesel engine AECD's is
easily a factor of 10 times the man-hours to complete compared to a similar on-road spark-
ignited application. Even with all the detail submitted for the heavy duty diesel engine
applications, it's often that EPA will ask for further clarification/detail. Our concern is the given
proposal in the Tier 3 NPRM under 86.144-01 (1 l)(iii) is the beginning of a continual expansion
of AECD submissions for light-duty that is unnecessary considering EPA already has the
authority to ask for the given information. To place the whole industry in a position of spending
the manpower to provide that information up front for every application is unnecessary and
burdensome.
Our Response:
EPA currently has the authority to request additional data or information regarding any
AECD to determine the validity of its use and to determine the impact on emissions. This
information is something manufacturers should already have available as part of the certification
process they perform in order to justify any AECD. The requirement that we are finalizing to
report information that should be readily available to the manufacturer will add minimal
additional work and allow the agency to better determine if industry-wide guidelines are being
followed.
4.1.5.6. Carryover of Small Volume Test Group Data
What Commenters Said:
BMW of North America, LLC
To reduce the financial burden for small volume models and test groups, we propose allowing
small volume test groups in the fleet to carryover certifying to Tier 2 standard and test fuel
through MY 2021.
Our Response:
To reduce the burden of testing in the first years of the Tier 3 program, EPA has provided
a mechanism that largely allows carryover of Tier 2 test data into the Tier 3 program through
MY 2019 for exhaust and MY2021 for evaporative emissions. Consistent with LEV III, the new
E10 test fuel will be required for all vehicle exhaust emission certification for MY 2020 and
4-118
-------
Tier 3 Summary and Analysis of Comments
later. Further delay of the Tier 3 program beyond the allowed provisions would restrict the
opportunity for a harmonized program with LEV III, an important goal of the Tier 3 program.
4.1.5.7. DOR and Extended Warranty Credits
What Commenters Said:
Alliance of Automobile Manufacturers (Alliance) and Association of Global Automakers
(Global)
DOR Credit (FR29873): We support continuation of the DOR technology credit.
Extended Warranty Credit (FR29873): We support the optional credit opportunity of 5 mg/mi for
extending the vehicle's warranty to 150,000 miles but do not believe that EPA needs to adopt
California's 4 percent recall provision in conjunction with this optional credit.
Our Response:
There were no adverse comments about either the DOR or Extended Warranty
opportunities. We are finalizing these provisions as proposed, including no changes to the EPA
recall provisions.
4.1.5.8. Compliance Demonstration
What Commenters Said:
Alliance of Automobile Manufacturers (Alliance) and Association of Global Automakers
(Global)
The Alliance and Global Automakers share the goal of one national program that would allow a
vehicle to be tested once, certified once and sold everywhere. Automaker and agency resources
are limited, and it is critical that compliance efforts not be duplicated. To this end, we appreciate
that the proposed Tier 3 regulations base fleet average emission requirements on nationwide
vehicle sales rather than on non-Section 177 State sales. Given the current level of fleet average
emissions, this is a reasonable flexibility that significantly reduces the resources needed to track
and manage separate fleets. Similarly, LEV III allows manufacturers to comply with fleet
average emission requirements based on sales in California and the Section 177 states. We intend
to request that ARB modify the LEV III regulations to allow compliance on a nationwide basis
consistent with Tier 3.
4-119
-------
Chapter 4: Proposed Vehicle Emissions Program
Compliance Demonstration Based on National Sales (FR29873): We support harmonization
between EPA and California, including the goal of a single fleet, tested once, certified once and
sold in all 50-states. Additional comments are provided in the written comment.
National Automobile Dealers Association (NADA)
Harmonized Emissions Mandates, Lead Time, Durability and Other Technical Standards: Last
year, the California Air Resources Board (CARB) finalized its Low Emission Vehicle (LEV) III
regulations. In order to minimize compliance costs and maximize compliance flexibilities, EPA's
final Tier 3 regulations should encourage if not mandate that CARB harmonize with the federal
scheme. Among other things, this means that fleet average emissions compliance should be
based on a manufacturer's nationwide sales.
Our Response:
We received no adverse comments on our proposal to base fleet-average sales on a
manufacturer's nationwide sales, and we are finalizing this provision.
4.1.5.9. Aftermarket Catalysts
What Commenters Said:
Ozone Transport Commission (OTC)
California's vehicle program includes updated aftermarket part rules that require catalytic
converters to meet a 50,000 mile warranty and rely on a mass based standard, among other traits
as outlined in OTC's "Recommended Revisions to the Federal Aftermarket Catalytic Converter
(FACC) Program." EPA's Tier 3 regulatory proposal does not include such an update to the
FACC program. Without updating its policy, EPA would be allowing use of replacement
converters that will not guarantee emission reductions as long as the original converters. To
ensure emission reductions occur when catalysts fall outside of the warranty period, the final
Tier 3 regulation should include a more stringent aftermarket catalytic converter policy as
recommended previously by OTC.
Our Response:
We did not propose and are not finalizing aftermarket catalyst requirements as suggested
by the commenter. If EPA decides to pursue such a program, it will be through a separate public
rulemaking process.
4.1.5.10. Treatment of FFVs
What Commenters Said:
4-120
-------
Tier 3 Summary and Analysis of Comments
Alliance of Automobile Manufacturers (Alliance) and Association of Global Automakers
(Global)
Flex-Fuel Vehicle (FFV) SFTP Certification Requirements - Testing on Gasoline and E85: The
Tier 3 NPRM contains a new requirement that FFVs be tested using the SFTP on both gasoline
and E85 (i.e., 85% ethanol). FFVs operating on gasoline almost always have higher SFTP
emissions than when operating on E85. Consequently, to reduce what is a very burdensome test,
current EPA Tier 2 and ARB LEV III regulations only require manufacturers to test FFVs using
gasoline. Today's new vehicles are already extraordinarily clean. Although we recognize the
need for and support further emission reductions, automakers simply do not have the resources to
pour into unnecessary tests such as those suggested in this portion of the NPRM.
If EPA is concerned that FFV SFTP emissions would be higher if tested using E85 rather than
gasoline, it could require that manufacturers test on the fuel that will result in worst-case
emissions (based on the manufacturer's good engineering judgment). This would satisfy EPA's
desire to ensure the vehicle would meet standards even when using the worst-case fuel.
Finally, if EPA is unwilling to eliminate unnecessary tests and accept either of the above
recommendations, it should at least allow manufacturers to substitute FTP emissions for SC03
emissions. The SC03 test is particularly burdensome, and FTP emissions will be higher than
SC03 emissions; allowing the substitution of FTP emissions instead of conducting SC03 testing
would result in a worst-case condition. In fact, for these reasons, both EPA and ARB allow
vehicles >8,500 pounds GVWR to substitute FTP emissions for SC03.
Recommendation: EPA should eliminate the new requirement for manufacturers to conduct
SFTP certification testing on E85. Instead, it should harmonize the Tier 3 and LEV III
regulations specifying that SFTP testing of FFVs should be conducted using gasoline only.
Alternatively, EPA could require manufacturers to certify over the SFTP on gasoline and then
the manufacturer would attest that the vehicle will comply with the SFTP standards if operated
onE85.
Regardless, for E85 testing, EPA should allow manufacturers to substitute FTP emissions for
SC03 emissions. For in-use compliance testing, the manufacturer should have the option of
running the actual SC03 rather than using FTP data for any test group.
California Air Resources Board (CARB)
SFTP Test Requirements for Fuel-Flexible Vehicles: CARB supports U.S. EPA's Tier 3 proposal
to require fuel-flexible vehicles to certify to SFTP standards using both gasoline fuel and the
highest ethanol content fuel that a. given vehicle is designed to operate on. CARB believes this
proposed requirement would ensure that emissions are controlled with all possible fuel blends
and will propose inclusion of this provision in the LEV III program once the Tier 3 program is
finalized.
Renewable Fuels Association (RFA)
4-121
-------
Chapter 4: Proposed Vehicle Emissions Program
In an effort to make Federal standards more consistent with California's LEV III program, EPA
is proposing new light-duty vehicle exhaust standards for NMOG, NOx and PM, as well as new
evaporative emissions standards. Vehicle manufacturers have expressed serious concerns about
the inability to certify emissions of FFVs under California's LEVIII standards when those
vehicles are operating on flex fuel. At issue is the fact that control of NMOG emissions during
cold start conditions is more difficult on flex fuel due to the fuel's volatility characteristics. Thus,
when operating on flex fuel, NMOG emissions from FFVs tend to exceed NMOG standards
before the catalyst is warmed up. The inability to certify FFVs under the California LEVIII
program has resulted in greatly restricted sales of FFVs in the state.
As EPA itself has acknowledged, the increased availability of FFVs is paramount to the
successful implementation of the RFS. Thus, we encourage EPA to carefully consider how
certification of emissions from FFVs should be handled.
Our Response:
Because of the physical and chemical differences in how emissions are generated and
controlled between vehicles operating on different blends of gasoline and ethanol, manufacturers
of vehicles designed for high-percentage blends of ethanol (usually called Flexible Fuel
Vehicles, or FFVs) may face unique compliance challenges under the Tier 3 program.
Historically, under the Tier 2 program, FFVs have only been required to meet all Tier 2 emission
standards, FTP and SFTP, while operating on gasoline (EO); when operating on the alternative
fuel (generally this means a blend that is nominally 85 percent ethanol, or E85), they have only
been required to meet the FTP emission standards.
However, E85 use may rise considerably in the future as ethanol use increases in
response to the Renewable Fuels Standards (RFS). Thus, as the Tier 3 program is implemented,
it is increasingly important that FFVs maintain their emission performance when operating on
E85 across different operating conditions.
We believe that at standard test conditions, requiring manufacturers to meet the Tier 3
standards on any blend of gasoline and ethanol will not be significantly more challenging
technologically than compliance on lower ethanol blends, including the E10 Tier 3 test fuel we
are adopting. We are thus finalizing, as proposed, the requirement that in addition to complying
with the Tier 3 requirements when operating on Tier 3 test fuel, FFVs also comply with both the
FTP and the SFTP emission standards when operating on E85. This includes the requirement to
meet emission standards for both Tier 3 test fuel and E85 for the FTP, highway test, and SFTP
emission standards at standard test temperatures (i.e., 68 °F to 86 °F). Since FFVs can operate
on any blend of gasoline and ethanol (up to a nominal 85 percent ethanol), the emission
requirements apply to operation at all levels of the alternative fuel that can be achieved with
commercially available fuels. However, for exhaust emission compliance demonstration
purposes, we will test on Tier 3 test fuel and on fuel with the highest available ethanol content.
Also, because gasoline and E85 have very different emission profiles, we are not allowing
attestation of compliance on E85.
4-122
-------
Tier 3 Summary and Analysis of Comments
To reduce the potential increase in test burden, EPA is (as recommended by the Alliance
and Global) finalizing the provision to allow manufacturers to substitute FTP emissions for
SCOSemissions when submitting test data for certification to the SFTP composite standard for
FFVs operating on E85. Manufacturers are still required to meet the SFTP standards if the SC03
test cycle is used for the composite calculation.
4.1.5.11. Applicability of FTP NMOG+NOx Standards to Highway Fuel Economy
Testing
What Commenters Said:
Alliance of Automobile Manufacturers (Alliance) and Association of Global Automakers
(Global)
We support the proposed highway NMOG+NOx standards harmonized with California LEV III
standards.
Our Response:
We did not receive adverse comment on this proposed provision and we are finalizing it as
proposed.
4.1.5.12. Preconditioning for Catalyst Desulfurization
What Commenters Said:
California Air Resources Board (CARB)
In response to reduced fuel sulfur levels in California and federal commercial gasoline, use of a
sulfur purge preconditioning to reduce catalyst sulfur deposits prior to IUVP emission testing
was disallowed by both CARB and U.S. EPA beginning in 2007. These purge procedures, which
consist of high speed driving cycles not normally experienced as typical driving patterns, are
designed to reduce sulfur and carbon deposits in the catalytic converter. Decreasing carbon build
deposition in the exhaust system may have a positive impact on emissions but because carbon
deposits are a result of real world driving conditions and fuel use, use of a sulfur purge procedure
may produce a false representation of true in-use emissions. Nonetheless, the NPRM proposes to
allow for conducting a sulfur purge cycle over the US06 test cycle for those manufacturers
requiring a sulfur preconditioning test.
Comment 2 - In-Use Verification Program Sulfur Purge Procedure for Sulfur Preconditioning.
As noted above, sulfur content of federal commercial fuel has been significantly reduced and,
based on past IUVP test data, manufacturers have not been allowed to perform sulfur purge
cycles since 2007 and have not experienced any test vehicles failing to meet the exhaust
emission standards that were attributed to high sulfur deposits. CARB believes running
4-123
-------
Chapter 4: Proposed Vehicle Emissions Program
additional sulfur cleanout procedures is unnecessary and should not be included as part of the
IUVP or any in-use confirmatory test program.
Ford Motor Company (Ford)
In-Use Preconditioning: While Ford agrees with the need for controls and specifications for
IUVP preconditioning on light-duty vehicles and MDPVs, EPA's proposed modifications to this
process places additional restrictions on a manufacturer's ability to mitigate sulfur contamination
on in-use vehicles due to exposure to high-sulfur market fuel. These new restrictions come at a
time when new, more stringent SFTP and PM requirements must be met and high sulfur fuel
(»10 ppm) will still be widely available. Also, this proposal does not provide the option to
conduct a sulfur purge cycle for any emission constituent other than NMOG + NOx on the FTP
or FIFET cycle. Ford believes that sulfur contamination also poses significant risks for CO and
PM, on the FTP cycle, and NMOG +NOx, CO, and PM on the SFTP.
Additionally, the Tier 3 IUVP preconditioning proposal requires an in-tank fuel sulfur
assessment before allowing additional preconditioning cycles. Since most contract and OEM labs
do not have in-house fuel sulfur measurement capabilities, fuel samples must be sent to off-site
laboratories for testing, which could result in impractically long delays for IUVP retests on
customer vehicles.
Finally, because sulfur contamination on gasoline catalysts is a cumulative effect, a spot check of
the fuel in the customer's tank just prior to an IUVP test captures at best the last 0-400 miles of
operation, and is insufficient to ensure that a vehicle has not been impacted by exposure to high-
sulfur fuel.
Recommendation: In light of these issues, Ford proposes that EPA carry over existing Tier 2
IUVP protocol, eliminate the fuel sulfur check requirement, and include new, forward-looking
IUVP retest provisions for NMOG+NOx, CO, PM, and N2O with new Elx test fuel on all IUVP
test cycles . Additionally, for IUVP FTP and SFTP PM measurements, we request that EPA
allow OEM's to use good engineering judgment to optionally average multiple tests to determine
PM compliance thru 2025MY.
General Motors LLC (GM)
Preconditioning for In-Use Testing: 86.1845-04 (a)(3) outlines provisions on how to address
potential residual effects of sulfur contamination on in-use vehicle testing. While we're grateful
EPA recognizes the impact of sulfur poisoning on in-use vehicles, these provisions either do not
adequately address the sulfur poisoning or are impractical to implement.
EPA goes into great detail about the impact of sulfur poisoning on in-use vehicles in section
(IV)(A)(6)(a) of the NPRM, and in sections (IV)(A)(6)(b) through (c), EPA discusses the effects,
impacts and controls regarding emission standards on vehicles impacted by sulfur. The
information provided by EPA is very compelling and obvious - sulfur does impact in-use
vehicles negatively and can prevent manufacturers from meeting standards. EPA also
demonstrates that, while not completely effective, preconditioning vehicles to properly burn off
4-124
-------
Tier 3 Summary and Analysis of Comments
some (not all) of the sulfur poisoning effects can significantly improve catalyst performance.
Without a preconditioning step, EPA states that "Any degradation in catalyst performance due to
gasoline sulfur would reduce or eliminate the margin necessary to ensure in-use compliance with
the proposed Tier 3 emissions standards."
With this in mind, GM does not understand the resistance of EPA to allow preconditioning of in-
use vehicles prior to formal in-use testing. Granted EPA is proposing that manufacturers could
take a fuel sample of the as-received candidate vehicle and use that as potential evidence that
retesting is applicable, but this raises several questions. What if the sample is lost? What if the
sample is mislabeled? Do we really want to add the burden of in-use testing by handling a very
volatile and hazardous liquid? By no means is this fuel sample conclusive of the vehicle's sulfur
poisoning level. The customer could have easily poisoned the catalyst on previous fuel, and now
have a tank of low sulfur fuel.
Moreover, there are numerous practical issues associated with EPA's proposed approach. For
example, taking a fuel sample of every vehicle that is brought in for in-use testing would add
time and cost to an in-use test program.
The final concern is that sulfur levels in the U.S. will be phased down over multiple years, while
this requirement takes effect with the final rule. This means field fuel will continue to have high
sulfur levels in the market. Even when low sulfur levels have been fully phased-in, how can we
be sure that a given refinery hasn't had a "spill" and released fuel with high levels of sulfur?
When vehicles are to meet the stringent Tier 3 standards, any degradation such as sulfur
poisoning can have significant effects.
We believe EPA wants "real world" testing results, but manufacturers should not be held to
comply based on factors they cannot control (i.e., high sulfur fuels). As for real world testing
results - that's a different scope than manufacturer in-use testing for compliance. In-use
compliance testing should be conducted in a technically sound, repeatable and accurate manner
similar to the original certification process, and manufacturers should not be held accountable for
factors beyond their control.
GM proposes allowing manufacturers to conduct either on-road or lab preconditioning of in-use
vehicles. Not only will this help mitigate the effects of sulfur poisoning, it will also allow
manufacturers to check for any other unforeseen issues with an as-received customer vehicle
(e.g., mechanical soundness, operating properly, etc.).
Volkswagen Group of America, Inc.
Tier 3 proposes to limit the OEM ability to use fuel enrichment in calibration. This restriction
effectively increases the stringency of Tier 3 because it is now harder to clean the sulfur off the
catalyst. If EPA is unwilling to mandate a 10 ppm cap in market fuel sulfur, OEMs must
continue to be allowed to condition the sulfur off the catalyst prior to any In-Use exhaust testing.
The market fuel will be high enough in sulfur to poison the catalyst. Two US06 cycles on the
proper cert fuel will be required to overcome the market fuel and its effects. Vehicles should not
be punished because they are operating in an environment of suboptimal fuels. EPA's own data
4-125
-------
Chapter 4: Proposed Vehicle Emissions Program
shows the need for this preconditioning. Mirroring the sulfur purge cycles that EPA used before
conducting its sulfur studies, purge cycles are needed for In-Use compliance on high sulfur
market fuel. With credits and the high proposed allowable pump sulfur levels, two USO6 cycles
are necessary to remove the catalyst poisoning sulfur. This should be the procedure implemented
without any monitoring of the fuel in the tank of the vehicle or other burdensome requirements.
Mercedes-Benz USA, LLC on behalf of Daimler AG
Mercedes-Benz requests that the EPA also include in its final Tier 3 rulemaking, a pre-test
catalyst conditioning procedure to be performed prior to in-use vehicle testing, which design is
mutually agreed upon by Mercedes-Benz and the EPA, in order to minimize the magnitude and
frequency of 'sulfur burn off cycles required in normal operation. This will provide real-world
benefit to the consumer in optimized fuel economy and reduced GHG emissions.
Alliance of Automobile Manufacturers (Alliance) and Association of Global Automakers
(Global)
The Tier 3 rulemaking is the correct place to set the near-term standards for sulfur and also
create a roadmap of how the agency will further reduce sulfur in the coming years. EPA should
specify that catalyst burnoff cycles are allowed for Tier 3 In-Use Testing.
Our Response:
Based on our findings from the In-use Sulfur Test program, we firmly believe that in
order for Tier 3 vehicles to perform at their designed emission levels, they will need to operate
in-use on fuels at sulfur levels that average 10 ppm or less. However, during the transition from
30 ppm average sulfur levels and the possibility of higher sulfur levels that could be encountered
due to the Tier 3 fuel 95 ppm sulfur cap extending past the transition, vehicles emission
performance could be adversely impacted which would be demonstrated during IUVP emission
testing. To address this issue, we are including a provision that allows manufacturers to retest a
vehicle if it has failed the initial "as received" tests during IUVP emission testing. The provision
allows for a limited and reasonable sulfur clean-out based on Tier 2 experiences, and an
additional requirement to provide potential proof of high sulfur exposure, typically from a fuel
sample of "as received" vehicle fuel.
We do not believe allowing sulfur related preconditioning, either EPA prescribed or
manufacturer determined procedures prior to "as received" testing is appropriate as it may distort
the "real world" emission performance evaluation related to sulfur or other vehicle factors. This
initial "as received" emission performance information is critical to evaluating the emission
performance of the vehicle on design intent fuel and additionally to determine if in-use sulfur
levels are supporting expected emission reductions. We acknowledge that the requirement to
provide evidence that the vehicle has operated on elevated fuel sulfur levels potentially adds an
additional step in the IUVP process, however we believe it is important to attribute any emission
increases or failures to the proper root cause so that we may determine if future actions are
required, including tighter control of in-use fuels.
4-126
-------
Tier 3 Summary and Analysis of Comments
4.1.5.13. Cold-Start Emissions; Start-Stop as Potential Tier 3 Technology
What Commenters Said:
Private Citizen
Going forward, there were a couple of things I had concerns with. A lot of the car companies
don't seem to be interested in dealing with the pollution before a catalytic converter warms up,
the so-called open loop phase. And I'm wondering if that could be included in some of your
thinking. Perhaps something could be included where an auto maker might get credit for a start
and stop system whereby if someone is at a drive-thru, instead of polluting, the car would just
stop automatically, and thereby wouldn't pollute, and perhaps could get credit if they didn't
completely meet some of these other tough standards.
Our Response:
The Tier 3 standards, particularly the NMOG+NOx standards, require manufacturers to
pay careful attention to cold-start emissions and design their control systems to address
emissions prior to catalyst light-off Regarding start-stop technology, the impact of these
systems on CC>2 emissions can be significant, but NMOG+NOx and PM emissions that are the
focus of the Tier 3 program are less impacted by reducing idle operation and will be accounted
for in the Federal Test Procedure.
4.1.5.14. Cold Durability Testing
What Commenters Said:
Alliance of Automobile Manufacturers (Alliance) and Association of Global Automakers
(Global)
Cold Testing Requirements for Emissions Durability Vehicle: For emission data vehicle (EDV)
selection, Tier 2 currently requires that manufacturers select the vehicle expected to emit the
highest CO emissions at 20°F on candidate in-use vehicles from the test vehicles selected. The
proposed Tier 3 regulations add another test based on the worst-case cold NMHC in §86.1828-
01(g). EPA recognizes that the expected worst-case cold CO vehicle could be the same as the
expected worst-case cold NMHC vehicle, and only testing on one vehicle would be required.
However, we see little benefit, but significant cost, to potentially testing two separate vehicles
based on different criteria pollutants. The 75°F FTP and SFTP tests are based on a more general
"worst-case tailpipe emissions' (11). We believe this is appropriate for 20°F testing as well. The
recommendation below parallels the SFTP and FTP language in §86.1828-01(a).
Recommendation: We recommend deleting proposed new paragraph §86.1828-01(g), and
instead modifying §86.1828-0l(c) as follows:
4-127
-------
Chapter 4: Proposed Vehicle Emissions Program
(c) Cold CO 20 degrees F testing. For cold temperature CO 20 degrees F exhaust emission
compliance for each durability group, the vehicle expected to omit tho highest CO omissions at
20 degrees F be worst-case for exhaust emission compliance on candidate in-use vehicles shall
be selected from the test vehicles selected in accordance with paragraph (a) of this section.
Our Response:
We do not believe this modification is appropriate for evaluating worst case candidates
and are finalizing these provisions as proposed. We do not believe that this change will
necessarily require the added burden of separate testing for each of the emission requirements
however it is clear to us that the relationship between CO and NMHC is not as clear as other
worst case emission determinations and involves a variety of factors. These factors may include
design difference that asymmetrically impact one emission and not the other (e.g. battery size
and cranking system impacts NMHC but not CO). Additionally, the relative compliance margin
and approach to these two requirements are very different with Cold CO standard as a fixed cap
with typically large compliance margins and cold NMHC as an FEL possibly requiring
manufacturers to test different vehicles.
4.1.5.15. Diesel Bench Aging
What Commenters Said:
Alliance of Automobile Manufacturers (Alliance) and Association of Global Automakers
(Global)
Due to the lack of diesel bench aging, certification of diesel vehicles under Tier 2 is resource and
time intensive as a result of maintaining and testing early prototype vehicles for 120,000 miles.
A regulatory provision for diesel bench aging approval is requested to support the incremental
durability work associated with 150,000 mile standards under Tier 3. The Alliance and Global
Automakers recognize that before individual manufacturers can obtain approval for diesel bench
aging, the EPA will need the industry to develop and submit a publically available bench aging
procedure that the agency approves. A comprehensive industry group has been formed to
develop this procedure and work with the EPA to create a mutually agreeable bench aging
procedure for diesel components. Since the industry has now accepted the responsibility for
developing the required procedure, the Alliance and Global Automakers respectfully request that
the EPA insert language into the Tier 3 regulation that specifically allows diesel bench aging
pending approval of the industry developed bench aging procedure. Allowing diesel bench aging
in the Tier 3 regulation, avoids the need for an additional rulemaking, streamlining the process
for diesel bench aging approval, and saving time and resources in the future.
Mercedes-Benz USA, LLC on behalf of Daimler AG
A bench aging durability procedure should be allowed for diesel-fueled vehicles as it is allowed
for other light-duty vehicles, light-duty trucks, and medium-duty passenger vehicles. Under the
current Tier 2 provisions, certification of diesel vehicles is both time- and resource-intensive due
4-128
-------
Tier 3 Summary and Analysis of Comments
to the lack of diesel bench aging. In order to achieve the necessary mileage without bench aging,
early prototype vehicles must be maintained and tested for 120,000 miles, resulting in additional
costs and testing time. These costs would only be higher for a useful life of 150,000 miles under
the Tier 3 provisions. As noted in the comments submitted by the Alliance of Automobile
Manufacturers, which we support, a comprehensive industry group has already been formed to
develop information regarding the feasibility of the standard certification test procedure.
In the Tier 3 Proposed Rule, EPA references the existing exhaust emission bench aging
durability programs allowed under 40 C.F.R § 86.1823-08 and the evaporative bench aging
durability programs allowed under the provisions of 40 C.F.R § 86.1824-08. 78 Fed. Reg. at
29,908. Allowing diesel bench aging would simply be an expansion of existing programs,
tailored as appropriate to diesel-fueled vehicles and similarly requiring the use of good
engineering judgment in the design of the bench aging program.
Since the introduction of EU5 standards in the European Union ('EU') in 2009, the EU reviewed
and fully adopted the EPA bench aging procedures for gasoline light-duty vehicles. This process
allowed maintaining of separate EU standards but coordination of the EU and U.S. procedures.
The EU, with a larger diesel share, also adopted a diesel bench aging procedure which has been
used by manufacturers and confirmed by the EU since 2009. Moreover, from the development of
the new world harmonized light-duty test procedure ('WLTP'), where EPA is also a member,
real world driving data of U.S. citizens are certainly demonstrated as being comparable to EU
driving patterns. Accordingly, the EU diesel bench procedure has proven it is able to demonstrate
real world aging of aftertreatment systems under U.S. real world driving conditions, as is the
case for the gasoline procedure. EPA's inclusion of similar provisions would be consistent with
past EU practice and would help implement this Administration's global harmonization initiative
as evidenced by the Transatlantic Trade and Investment Partnership. Development testing of
U.S. diesel engines to U.S. standards using EU bench aging protocol shows that diesel bench
aging is an effective way of ensuring compliance efficiently.
For all of these reasons, diesel bench aging should be allowed under the Tier 3 regulatory
provisions. If EPA has any concerns regarding issuance of a bench aging approval in the final
Tier 3 rule, subject to manufacturer guidance in accordance with existing bench aging regulatory
requirements, EPA should immediately initiate a supplemental notice or direct final rulemaking
to allow diesel bench aging, the completion of which should be timed to coincide with the
finalization of the Tier 3 provisions.
Volkswagen Group of America, Inc.
Diesel bench aging is necessary for streamlining resource-intensive activities. OEMs need a
procedure that is well developed with a collective of manufacturers. Tier 3 needs to allow
options like these for the future. VWGoA specifically requests that Diesel bench aging be
included in the Final Tier 3 rule. We understand that EPA needs an Industry supported and
publically vetted procedure. A group of more than 8 OEMs have agreed to work together to
create a logical, useful bench aging procedure. With the increase to 150k miles full useful life,
demonstrating emissions durability using on-road procedures is too expensive and cumbersome.
Currently Diesel vehicle families are not equivalent to gasoline families and require a full,
separate demonstration. In the future, compliance to GHG and Fuel Economy rules may mean a
4-129
-------
Chapter 4: Proposed Vehicle Emissions Program
larger portion of our fleet may include Diesel options, and creating a streamlined approach to
bench aging components will be necessary.
Our Response:
Although EPA appreciates the industry interest in a diesel bench aging protocol, in the
absence of an industry-consensus approach, we do not believe it would be appropriate to take
action as a part of the Tier 3 rule. EPA will monitor industry progress in this area and consider
action at an appropriate future time.
4.1.5.16. Diesel Emissions Fluid Refill Intervals
What Commenters Said:
Volkswagen Group of America, Inc.
Diesel Emissions Fluid (DEF) is the primary method of controlling diesel emissions and marks
the first time the driver is integral to the function of the vehicle emission controls. In VW Group
vehicles the driver receives multiple escalating warnings including preventing vehicle start for
not keeping the fluid reservoir at proper levels. A recent Heavy Duty study in California (!)
shows that vehicles (drivers) are filling up the tanks, and when they don't, the inducements are
working. Mandating a large mileage interval between fills results in a large fluid capacity and
adds unnecessary weight to the vehicle. Current regulations tie the oil change interval to the DEF
refill interval; decoupling DEF fill from oil change interval allows more compliance options and
easier packaging of smaller systems. The stringency of Tier 3 will require more DEF
consumption. This means that the storage tanks will either have to increase in capacity or we
must needlessly shorten oil change interval, and thus require the customer to change their oil
more often. Allowing a shorter refill interval will help optimize vehicle design for the multitude
of functional objectives. VW views this as a critical issue, as it risks not being able to give the
driver the best ownership experience. We do understand that this is not directly included in the
Tier 3 NPRM, but feels that it can be included in the Final Rule. [EPA-HQ-OAR-2011-0135-
4299-A1, main comment pp. 1-2]
Our Response:
As noted by the commenter, this comment is not within the scope of the Tier 3 rule.
However, EPA plans to finalize a separate rule to address this issue. On June 8, 2012, EPA
proposed a rule to codify SCR maintenance (77 FR 34149). The EPA is close to finalizing that
rule and addressing all comments received, including comments about the connection between
Tier 3 compliance and DEF tank sizes, as well as comments about de-linking DEF refill intervals
from oil change intervals.
4.1.5.17. Manufacturer In-Use Verification Testing (IUVP) Requirements regarding PM
4-130
-------
Tier 3 Summary and Analysis of Comments
What Commenters Said:
Alliance of Automobile Manufacturers (Alliance) and Association of Global Automakers
(Global)
In-Use Verification Program (IUVP) PM Testing at Low Mileage: We do not believe low
mileage IUVP PM testing is an efficient or effective use of resources. If there are going to be
compliance problems, it is either going to be with fundamental vehicle calibrations where
compliance can be demonstrated as sufficient during certification testing or it is going to be due
to a high mileage deterioration issue, which would not show up during low mileage IUVP testing
anyway. However, EPA has said it wants to see low mileage IUVP data for PM to be assured
inadequate designs (calibrations) did not make it past certification especially given the newness
and stringency of the proposed PM standards.
We appreciate that EPA has attempted to reduce the burden of IUVP PM testing by proposing
that automakers only be required to test roughly half of all IUVP test vehicles for PM, and we
support this approach. However, given the substantial burdens of PM testing, we have two
significant concerns with the proposed low-mileage PM IUVP testing requirements. One is the
excessive burden of having to perpetuate low mileage PM testing after manufacturers have
demonstrated that certain fundamental vehicle designs will perform well below the standards
based upon certification and initial low mileage in-use testing. The other - and more immediate -
- concern is the near term cost and lead time issues with converting IUVP testing facilities so
quickly after doing a major conversion of development testing facilities and virtually all
certification testing facilities.
While we do not agree that low mileage PM testing will be cost-effective, if EPA is going to
insist on low mileage in-use testing for an initial screening purpose, it does not seem that regular
low mileage IUVP testing needs to be perpetuated. The current IUVP regulations allow the
agency the discretion to reduce IUVP testing as a manufacturer has demonstrated good in-use
performance. However, for non-PM IUVP testing, manufacturers have demonstrated excellent
in-use compliance now for over 10 years of IUVP testing, and the agency has never exercised
authority to reduce testing burdens despite industry requests. Hence, without more specific test
burden reduction criteria being built into the regulations, we would have to assume the same
thing would occur with PM testing, whereby the agency would never move to reduce burdens,
even after a number of years of demonstrated good in-use performance even at low mileage.
The Agency should be able to conclude that a general design approach (i.e., a given generic fuel
injection/engine design approach) assures compliance at low mileage based upon certification
testing followed by fairly limited low mileage IUVP confirmation. One specific design approach
where low mileage IUVP testing will certainly result in limited benefits is gasoline PFI vehicles.
In fact, almost all reported data indicates that the gasoline PFI vehicles have remained well
below the 3 mg/mi standards, and historically EPA has waived all testing for these vehicles.
Adding testing oversight that will occur in certification alone should be sufficient for these
vehicles. The very few concerns that have been noted with PFI vehicles have been associated
with higher mileage deterioration, such as potentially with oil consumption, which would not be
4-131
-------
Chapter 4: Proposed Vehicle Emissions Program
seen during low mileage IUVP tests. Hence, low mileage testing should be completely waived
for PFI vehicles.
Additionally, other fundamental design approaches should be able to forgo additional low
mileage PM testing after they have gone through a single round of low mileage IUVP testing
without there being any indication of a problem (i.e., if initial IUVP testing validates what was
observed in certification testing). At the very least, a test group that is certified based upon
carryover data and which has been validated once in low mileage IUVP testing should not be
subject to additional low mileage testing as long as carryover continues. Language should be
included in the regulations that would allow the manufacturer to request a waiver of the IUVP
low mileage PM testing for test groups that it can demonstrate use essentially the same fuel
injection, engine design, and PM control characteristics as another test group which has
undergone successful low mileage IUVP testing once. Actually, we recommend that similar
language be included in IUVP that would similarly allow reduction of low mileage IUVP testing
for all other pollutants as well. But it is most important for PM testing given the added burdens
and added risks of voided tests associated with the much more challenging PM test procedures.
Furthermore, the IUVP low mileage requirement follows so quickly after the beginning of the
Tier 3 program that manufacturers will have to begin conversion of IUVP testing facilities before
they have even completed the substantial phase in of new testing facilities needed for
development and certification. Given the standards phase in with only 10 to 20% of a
manufacturer's fleet needed to comply in MY 2017, we acknowledge that there will not be many
tests groups that will be subject to low mileage IUVP testing at the end of that first one or two
compliance years. However, even with only a few tests being needed, the test facility
modifications to accommodate PM testing will have to be in constructed. Many manufacturers
perform IUVP testing at physically different locations than where development or certification
testing is performed. Hence, the "few" initial low mileage test groups cannot be easily tested
using the new facilities that would have just been put into place for certification testing. To
minimize the burdens associated with converting so many facilities at nearly the same time, we
ask that EPA grant extra lead time for establishing IUVP testing capabilities by waiving the low
mileage PM testing requirement for the first two model years as the PM standards phase in.
High mileage IUVP testing would be retained as proposed, so minimization of low mileage
testing would not let manufacturers escape ultimate recall liability. Hence, manufacturers cannot
afford to be careless or try to game their way through certification and minimal low mileage
testing. Additionally, if low mileage testing were waived for the first two years of the program,
the test groups that would be waived would not totally escape testing prior to high mileage
testing, since many such test groups would likely be carry-over groups for the next couple years
(i.e., many of these would end up being tested under subsequent model year low mileage testing
requirements).
Recommendation: We recommend that EPA include language in the low mileage IUVP
regulations that would minimize the need to perpetuate low mileage PM testing for vehicles that
have demonstrated good performance via certification and initial low mileage testing.
Specifically we recommend:
• Completely waiving of the requirement for low mileage testing of gasoline PFI vehicles.
4-132
-------
Tier 3 Summary and Analysis of Comments
• Waiving testing of carryover test groups that demonstrated compliance in their first model year
of low mileage testing. We specifically recommend this test reduction for low mileage PMIUVP
tests but suggest that it would be equally important and appropriate for low-mileage IUVP
testing of other criteria emission pollutants.
• Including a provision where a manufacturer may request the waiver of low mileage testing for
test groups having similar PM reduction design characteristics as other test groups that have
undergone successful low mileage testing.
Additionally, EPA should delay the applicability of low mileage IUVP PM testing requirement
until after the first two model years of the PM standards phase-in.
Our Response:
As noted by the commenter, EPA proposed and is finalizing a provision to significantly
reduce the burden of IUVP testing PM measurement by requiring roughly half of all IUVP test
vehicles measure PM emissions. We believe that it is important to closely monitor the PM
emissions from the future Tier 3 vehicles given the importance of PM emissions reductions to
the Tier 3 program and the associated health benefits. The requirements that we are finalizing
for low mileage testing are reasonable and necessary to provide an early opportunity to ensure
that PM levels observed at certification are consistent with in-use performance. As described in
the comment, the current IUVP regulations allow the agency the discretion to reduce IUVP
testing as a manufacturer has demonstrated good in-use performance.
While we observed a general trend of certain mature technologies (i.e. gasoline PFI)
performing better than others in our feasibility study, we believe
that the changes in vehicle designs in response to items such as GHG/FE standards and customer
preferences will not guarantee that future vehicle designs will maintain the same PM
emission performance regardless of technology maturity. We feel that enough changes can
occur year to year such as in the emission controls and
calibrations to justify the reasonable and limited low mileage PM testing finalized for Tier 3.
Additionally, manufacturers have not provided sufficient evidence in certification or in-use that
the levels of the new PM standards, particularly for SFTP US06, are consistently being
demonstrated by any technology, including the mature technologies discussed in their comments.
In other words, statements regarding the relative maturity of technology and its associated PM
performance are based solely on the limited data made public by EPA and CARB and
commenters did not provide a single piece of low mileage test data on any technology that would
conclude additional reductions in PM measurement are justified, particularly for the SFTP US06
test cycle
We reject the notion that manufacturers require more time to prepare emission
measurement testing facilities for the already reduced amount of PM testing that will be required.
In fact, the phase-in of the Tier 3 PM standards is structured in recognition of the issue of testing
facilities availability and associated burden in the early years as discussed in the preamble and
less concern about feasibility since manufacturers have the ability to select the vehicles and
technologies in which they have the most confidence in the early years.
4-133
-------
Chapter 4: Proposed Vehicle Emissions Program
4.2 Tailpipe Emissions Standards for Heavy Duty Vehicles
4.2.1. HDV Program Lead Time
What Commenters Said:
Truck and Engine Manufacturers Association (EMA):
Leadtime, Stability and Technical Feasibility:
As established by the Clean Air Act, any mobile source emission standards adopted by EPA for
on-highway engines and vehicles above 6,000 Ibs. GVWR must be technologically feasible and
may be implemented, among other things, only if the requisite leadtime and period of stability
are provided to manufacturers.
CAA Section 202(a) requires, among other things, that emission standards for heavy-duty
engines must be technologically feasible:
[Standards must reflect the greatest degree of emission reduction achievable through the
application of technology ... determine[d to] be available for the model year to which such
standards apply, giving appropriate consideration to cost, energy, and safety factors associated
with the application of such technology.
CAA Section 202(a)(3); 42 U.S.C. §7521. See Motor & Equip. Mfrs. Ass'n v. EPA, 627 F.2d
1095, 1111 (D.C. Cir. 1979) (consistency with the CAA requires standards to be
"technologically feasible"). EPA has failed to justify the technological feasibility of many of the
proposed requirements, specifically those relating to the lack of interim in-use and high-altitude
standards for medium- and heavy-duty vehicles and engines.
Engine manufacturers also need sufficient time to develop technology that is feasible and
practical. Section 202(a) of the CAA specifically requires EPA to provide sufficient leadtime
and an adequate period of stability for any standards affecting new on-highway engines and
vehicles above 6,000 Ibs. GVWR:
Any standard promulgated or revised under this paragraph and applicable to classes or categories
of heavy-duty vehicles or engines shall apply for a period of no less than 3 model years
beginning no earlier than the model year commencing 4 years after such revised standard is
promulgated.
In other words, any new emission standards may go into effect only four or more full model
years after the year in which they were promulgated. And, all new or previously-adopted
standards must stay in effect for at least three full model years before EPA may establish another
applicable standard.
For vehicles and engines over 6,000 Ibs. GVWR, which are subject to those leadtime and
stability requirements, EPA has proposed both "primary" and "alternative" paths to compliance:
4-134
-------
Tier 3 Summary and Analysis of Comments
one path that phases in fleet average standards (which become more stringent in successive
model years) over a period of several years and an alternative path that requires a percentage of a
manufacturer's product to be phased in to far more stringent emission standards along a much
more aggressive timetable. The standards and phase-ins for the categories of vehicles and
engines above and below 8,500 Ibs. GVWR are different, but each category has similar
"primary" and "alternative" paths.
In both categories (above and below 8,500 Ibs. GVWR), the "primary" path fails to provide four
full model years' leadtime and three years of stability. Meanwhile, while the "alternative" path
meets minimum leadtime and stability requirements, the "alternative" path standards are not
technologically feasible. EPA has provided manufacturers with a "choice" that is effectively no
choice at all: manufacturers must choose one of two alternatives, neither of which meets the
statutory requirements of the C AA.
The comments of the Alliance outline in greater detail the issues and impacts associated with
EPA's failure to provide four full model years' leadtime for those categories of vehicles under
8,500 Ibs. GVWR. For vehicles and engines in the 8,501-14,000 Ibs. GVWR range ("Heavy-
Duty Vehicles"), EPA has proposed declining fleet average standards that begin in 2018 through
2022, with an "alternative" phase-in percent of sales requirement that begins in 2019 for "any
manufacturer who prefers a stable standard and four full years of lead time, as specified in the
[CAA]." (78 Fed. Reg. at 29876; emphasis added.) EPA's failure to provide sufficient leadtime
and stability in those proposed declining fleet average standards is not a matter of "preference"
but a matter of law - as established in the CAA. And, while standards that begin in 2019
technically would provide the legally-required four model years' leadtime (assuming the
Proposed Tier 3 Standards are finalized before the end of 2013), EPA has failed to demonstrate
that the proposed standards of the "alternative path" are technically feasible for Heavy-Duty
Vehicles.
For example, in support of its decision to exclude interim in-use standards for vehicles over
8,500 Ibs. GVWR, EPA only used data generated from two medium-duty vehicles which were
not aged to 150,000 miles full useful life and were not representative of the vehicles that will be
needed to meet new, stringent GHG standards for heavy-duty vehicles. EPA also referred to
light-duty vehicles with powertrains similar to their heavy-duty counterparts as support for the
lack of interim in-use standards. Yet, in apparently relying on such light-duty data, EPA did not
account for the additional technical challenges associated with meeting the proposed standards at
adjusted loaded vehicle weight (ALVW). As EPA has noted in the preamble to the Proposed
Rule, emissions from Heavy-Duty Vehicles are tested with loaded vehicles to ensure the
emissions are controlled when such vehicles are performing their core function: hauling heavy
loads. (78 Fed. Reg. at 29874.) Heavily-loaded vehicles have different emission characteristics
from vehicles under light load. Thus, light-duty vehicles not tested at ALVW cannot and should
not be relied on for demonstrating the technological feasibility of standards applied to heavier
vehicles at ALVW.
EPA must meet its obligations to propose and finalize regulations that it can demonstrate are
both technologically feasible and provide the necessary leadtime and period of stability required
4-135
-------
Chapter 4: Proposed Vehicle Emissions Program
by the CAA. EPA has not done so with either the primary or alternative standards for vehicles
greater than 6,000 Ibs. GVWR.
Chrysler Group LLC:
NMOG+NOx Phase-In Requirements for Heavy-Duty Vehicles:
The CAA requires EPA to provide manufacturers of heavy-duty vehicles, defined as vehicles in
excess of 6,000 pounds GVWR, (29) with four years of lead time (and a three-year period of
"stability" with no changes to the standards), to comply with new automobile emission
standards. Specifically, Section 202(a)(3)(C) of the CAA, 42 U.S.C. § 7521(a)(3)(C), states:
"Any standard promulgated or revised under this paragraph and applicable to classes or
categories of heavy-duty vehicles or engines shall apply for a period of no less than 3 model
years beginning no earlier than the model year commencing 4 years after such revised standard is
promulgated." (emphasis added). This statutory requirement is intended to ensure that
manufacturers of heavy-duty vehicles are afforded sufficient lead time and stability to recoup
their investments in new technology required to comply with more stringent standards.
In the proposed rule, EPA attempts to thwart this statutory requirement by proposing two phase-
in alternatives for complying with the significantly more stringent Tier 3 NMOG+NOx tailpipe
emissions standards. Under the more favorable primary compliance option, manufacturers must
meet a fleet average standard that is phased-in between Model Years 2018 and 2025 for vehicles
over 6,000 pounds GVWR, but which contravenes the CAA's explicit four-year lead time
requirement. Under the less favorable alternative compliance option, manufacturers must phase
in 40 percent of their vehicles to the more stringent Tier 3 NMOG+NOx standards beginning in
Model Year 2019, which comports with the statutory four-year lead time requirement, but adds
the significant burden that all of the manufacturer's vehicles must meet the final standards by
Model Year 2021 — a full four model years ahead of the primary phase-in option.
Footnote: EPA proposes to adopt a similar dual-path approach for "heavy-duty vehicles"
as defined by EPA at 40 CFR §86.1803-01 (i.e., vehicles greater than 8,500 pounds
GVWR but less than or equal to 14,000 pounds GVWR). See 78 Fed. Reg. at 29,875-
29,877. Under the proposed primary compliance option, manufacturers must meet a
declining fleet average standard for NMOG+NOx beginning in Model Year 2018, in
contravention of the CAA's explicit four-year lead time requirement. Alternatively,
manufacturers can choose to comply beginning in Model Year 2019, which comports
with the four-year lead time requirement, but then are subject to a percent-of-vehicles
phase-in approach rather than a declining fleet average. Again, EPA's "poison pill"
approach cannot be reconciled with the CAA. Accordingly, Chrysler recommends that
EPA start the declining fleet average in Model Year 2019, consistent with the CAA's
four-year lead time requirement.
Assuming EPA issues a Tier 3 final rule in late 2013, as planned, Model Year 2014 already
would have begun. Therefore, the CAA requires that the Tier 3 standards applicable to heavy-
duty vehicles take effect no earlier than Model Year 2019 (i.e. four model years after the revised
standards are promulgated per Section 202(a)(3)(C), that is, commencing with the 2015 Model
Year). EPA's proposed primary compliance option, however, would require manufacturers of
4-136
-------
Tier 3 Summary and Analysis of Comments
heavy-duty vehicles to begin complying with the stringent Tier 3 standards in 2018 (i.e., only
three model years after the revised standards are promulgated), and therefore would contravene
the lead time requirement in Section 202(a)(3)(C) of the CAA. Conversely, although EPA's
alternative compliance option does meet the statutory four-year lead time requirement, it
unreasonably imposes unduly and significantly more burdensome compliance requirements than
the primary compliance option, and does not present manufacturers with a realistic choice.
Moreover, these undue burdens are not rationally related to the four-year lead time requirement
and thus are clearly designed as a punitive stick to force manufacturers to "choose" the non-CAA
compliant primary option with only three years of lead time.
As described in greater detail in the comments provided by the Alliance of Automobile
Manufacturers and Association of Global Automakers, the alternative compliance option, which
complies with the four-year lead time requirement but requires manufacturers to certify all of
their vehicles to the Tier 3 standards a full four model years sooner than the primary phase-in
option, is so burdensome that no reasonable manufacturer would ever choose it; all
manufacturers would essentially be forced into selecting the primary phase-in option and turning
a blind eye to EPA's blatant violation of the CAA's lead time requirements for heavy-duty
vehicles. When a regulatory scheme is so heavily weighted or creates significant economic
incentives such that only one of the options is reasonable, that option becomes a mandate and is
no longer truly an option. Providing a significantly more burdensome alternative compliance
path that meets the statutory requirements does not, and cannot, remedy the fact that the primary
compliance path violates the statute.
This dual-path approach that EPA proposes would violate the CAA for an additional reason, as
well. Under Section 202(a)(3)(A)(i) of the CAA, EPA is required to promulgate standards
"which reflect the greatest degree of emission reduction achievable through the application of
technology which the Administrator determines will be available for the model year to which
such standards apply, giving appropriate consideration to cost, energy, and safety factors
associated with the application of such technology". Thus, the CAA implicitly requires EPA to
select the single standard that reflects the greatest degree of emission reduction achievable.
Here, it simply cannot be the case that both the primary compliance option (which begins in
Model Year 2018 and requires full compliance with the more stringent standards by 2025) and
the alternative compliance option (which begins in Model Year 2019 and requires full
compliance with the more stringent standards by 2021) reflect the greatest degree of emission
reduction achievable taking into account cost, energy, and safety. Rather, if the three-year
phase-in under the alternative compliance option is in fact achievable considering the specified
factors, then the primary compliance path, which affords manufacturers seven years to meet the
more stringent standards, cannot be reconciled with the CAA requirement that EPA promulgate
standards reflecting the greatest degree of emission reduction achievable. The more likely
explanation, however, is that the primary compliance option reflects the greatest degree of
emission reduction that EPA believes is truly appropriate, whereas the alternative compliance
option is intentionally so burdensome—and therefore far beyond reflecting the greatest degree of
emissions reduction achievable—that manufacturers will be forced to waive their statutory
protections and choose the primary path. This attempt to circumvent the statutory protections
embodied in the CAA cannot be reconciled with the plain provisions of the Act.
4-137
-------
Chapter 4: Proposed Vehicle Emissions Program
Further, Chrysler objects to the requirement that if a manufacturer chooses the alternative
compliance path for any of its vehicles, all of its vehicles, including light-duty vehicles and those
heavy-duty vehicles for which the manufacturer is willing to comply with three years lead time,
must nevertheless meet the final Tier 3 NMOG+NOx standards by 2021 (instead of 2025 under
the primary compliance option). This regulatory scheme, which sweeps in a manufacturer's
entire fleet if the manufacturer chooses to pursue the alternative compliance option for even just
a single vehicle model, cannot be characterized as an incentive to forego the four-year lead time
requirement; rather, it is a penalty imposed on manufacturers that refuse to forego their statutory
protections. There is no rational basis for penalizing a manufacturer's entire vehicle fleet simply
because the manufacturer chooses to exercise its statutory right to four years of lead time for one
heavy-duty vehicle model. Plainly, any emissions benefits of earlier compliance would accrue
for those models of vehicles for which the manufacturer might choose earlier compliance. But
precluding that choice and imposing more stringent standards even for those models that might
comply early, just because a different model or models cannot comply early, has no
environmental rationale except to penalize the choice of later compliance for the other models.
That design is arbitrary, unreasonable, and unlawful.
Recommendation: Chrysler recommends that, for vehicles over 6,000 pounds GVWR, the Tier 3
fleet average requirements for FTP and SFTP NMOG+NOx begin in MY 2019 at the currently
proposed Model Year 2019 fleet average emissions levels (under the primary compliance option)
and require full compliance only at Model Year 2025. This approach is in full compliance with
the CAA, completely harmonizes with LEV III in MY 2019, and will achieve the same level of
emission reduction as the proposed rule.
Our Response:
Chrysler commented on Clean Air Act requirements for lead time and stability for
vehicles over 6000 Ibs GVWR. Excerpts from these comments pertaining to vehicles over 8500
Ibs GVWR are addressed here. Those dealing with vehicles 6000-8500 Ibs GVWR are
addressed in Chapter 4.1.1.
Comments we received on the proposed HDV standards did not specifically address our
analysis of their technical feasibility. The Manufacturers of Emission Controls Association
(MECA) outlined diesel and gasoline-engine technologies that they expect will be used to
achieve the Tier 3 standards cost-effectively, generally consistent with our RIA. These
comments are discussed in Chapter 4.1.4. Vehicle and engine industry commenters argued that
the case we made for feasibility relied too heavily on extending light-duty truck test data,
supplemented by testing of only two HDVs, neither of which were fully aged or representative of
future vehicles designed to meet our new GHG standards. However, these comments did not
question the feasibility, durability, implementability, or effectiveness of the technologies we
identified, or their ability to achieve the proposed standards. Instead, the focus of these
comments was on statutory provisions for lead time and stability, and on how relaxed standards
for in-use testing and testing at high altitudes would help to implement the standards within the
allotted lead time. These issues, including changes we are making in response to the comments,
are addressed in Sections IV.B.2.C, IV.B.6.a, and IV.B.6.f, of the preamble to this final rule.
Related additional discussion can be found in Chapters 4.2.4 and 4.2.6.1 below.
4-138
-------
Tier 3 Summary and Analysis of Comments
Contrary to the assertions of these commenters, HDV manufacturers clearly have a
choice between the primary LEV Ill-harmonized phase-in and the alternative percent-of-sales
phase-in. As discussed in the preambles to the proposed and final rules, we are providing for the
LEV Ill-harmonized phase-in to allow manufacturers to more easily market the same HDV
models in all 50 states. We are also providing for the alternative percent-of-sales phase-in so
that manufacturers may choose to benefit from four years of lead time and three years of stability
as specified in CAA section 202(a)(3)(C). That a commenter might view one alternative as
"more favorable" and the other "less favorable" does not mean that a manufacturer does not have
a choice of phase-in alternatives under the Tier 3 program.
As acknowledged by the commenters, the alternative phase-in clearly meets statutory
lead time and stability requirements, but commenters argue that the alternative phase-in of the
Tier 3 HDV standards is not technologically feasible. Commenters who disagreed with our
assessment of the feasibility of the alternative phase-in for HDV manufacturers did not explain
the reasons for their disagreement, beyond referring to similar comments they had on the
analogous light-duty (above 6000 Ibs GVWR) alternative phase-in. However, the proposed
light-duty alternative differs from the one we proposed for HDVs, and the elements in it that
were found objectionable by the manufacturers are not present in the HDV alternative. The
HDV alternative does not impose "unduly and significantly more burdensome compliance
requirements than the primary compliance option", and we see no reason to conclude that it does
not present manufacturers with a realistic choice.
Commenters objected that the proposed percent-of-sales alternative has not been shown
by EPA to be feasible, or in fact is infeasible because it mandates the early phase-in of low-
emitting vehicles certified to the final standards. In fact, the percent-of-sales phase-in is no more
stringent than the LEV Ill-harmonized phase-in, so a manufacturer meeting the percent-of-sales
alternative would not have any more difficulty meeting the stringency levels than a manufacturer
using the other approach. EPA chose the phase-in percentages in the percent-of-sales alternative
to result in a fleet average NMOG+NOx level that is equal to the LEV Ill-harmonized standards.
For example, a manufacturer choosing the percent-of-sales approach must demonstrate
(including through use of credits) that 60% of its MY2019 Class 3 fleet meets the final Tier 3
NMOG+NOx standard of 247 mg/mi; the remaining 40% may be certified to pre-Tier 3 NMHC
and NOx standards adding up to 630 mg/mi. On average this Class 3 fleet attains (40%)x(630) +
(60%)x(247) = 400 mg/mi, which is the fleet average standard in this year under the LEV Ill-
harmonized alternative.
In response to the comments about EPA's assessment of feasibility for the percent-of-
sales alternative standards, the feasibility analysis provided in the Final RIA, which expressly
addresses the LEV Ill-harmonized phase-in, serves to demonstrate the feasibility of both
alternatives. Contrary to commenters' contentions, neither alternative mandates the early phase-
in of low-emitting vehicles certified to the final standards, due to the existence of identical
averaging, banking, and trading (ABT) provisions in both alternatives. In fact, with ABT, every
manufacturer can produce the same mix of vehicles in any model year to comply with either
HDV phase-in alternative, with the exception that MY 2018 is a voluntary phase-in year under
4-139
-------
Chapter 4: Proposed Vehicle Emissions Program
the alternative phase-in and a required year under the LEV Ill-harmonized phase-in. By no
means are manufacturers forced to make only vehicles certified to the final standards.
Some comments seem to assert that the percent-of-sales construct for the alternative was
chosen by EPA to make this alternative so stringent (by forcing some vehicles to meet final
standards four years early) that no reasonable company would use it. This is incorrect, both in
regard to its actual effect (discussed above), and in regard to our intent. Rather, the percent-of-
sales construct for the alternative was proposed and is being adopted to provide manufacturers
with a phase-in alternative that explicitly meets the applicable Clean Air Act stability
requirement, consistent with the commenter's legal statements.
We are making one change to the percent-of-sales alternative, necessitated by the fact
that this final rule is being signed in 2014, not 2013 as envisioned in the proposal. HDV models
for which the 2019 model year begins before the fourth anniversary of the signature date of this
final rule may be excluded from the Tier 3 fleet average compliance calculations and all other
Tier 3 requirements. These excluded vehicles would instead need to comply with the applicable
pre-Tier 3 standards and requirements for the entire production of these models throughout the
2019 MY. This limited allowance ensures that the alternative meets EPA's obligation for four
years of lead time under the Clean Air Act. It is similar to a phase-in alternative we provided in
the light-duty vehicle Tier 2 rule (see 65 FR 6747, February 10, 2000).
4.2.2. HDV FTP Exhaust Standards
What Commenters Said:
Truck and Engine Manufacturers Association (EMA):
Chassis-Dynamometer-Certified Heavy-Duty Vehicle NOx Standards:
The current NOx standards for heavy-duty, chassis-certified vehicles are 0.2 g/mile and 0.4
g/mile for Class 2b and Class 3 vehicles, respectively. Under the Proposed Tier 3 Standards,
EPA is proposing two interim bins, one for Class 2b and one for Class 3, both of which are
intended to be "carry-over bins" for existing Tier 2 vehicles. This would allow manufacturers to
continue certifying vehicles that are designed to meet the current standards without any
modifications. However, as proposed, the NOx standards for these "interim bins" are expressed
as three significant digits (0.200 g/mile and 0.400 g/mile for Class 2b and Class 3, respectively).
Such a change represents a significant increase in the stringency of the standards due to the
applicable rounding rules. Having three significant digits (instead of one) means that existing
vehicles would be subject to significantly more stringent NOx standards, which is not consistent
with the intent of creating "interim bins."
EMA believes that this proposed change in significant digits was unintentional on EPA's part.
Either way, EPA should not change the number of significant digits for the NOx standard in the
"interim bins" for chassis dynamometer certified heavy-duty vehicles. Rather, the Agency
should - indeed, must - retain the NOx standard at 0.2 g/mile and 0.4 g/mile for Class 2b and
4-140
-------
Tier 3 Summary and Analysis of Comments
Class 3 interim bins, respectively. Otherwise, EPA's proposed standards are inconsistent with its
expressed intent - which EMA supports - to allow "carryover" bins.
Chassis-Dynamometer-Certified Heavy-Duty Vehicle PM Standards:
The current PM standard for heavy-duty, chassis-certified vehicles is 0.02 g/mile for both Class
2b and Class 3 vehicles. Under the Proposed Tier 3 Standards, EPA is proposing two interim
bins, one for Class 2b and one for Class 3, both of which are intended to be "carry-over bins" for
existing Tier 2 vehicles. This would allow manufacturers to continue certifying vehicles that are
designed to meet the current standards without any modifications. However, as proposed (Table
5 of 86.1816-18), the PM standards for these "interim bins" do not match the values in 86.1816-
08. Instead of 0.02 g/mile, the PM standards for the two bins in each class are listed as 0.012
g/mile. Such a change represents a significant increase in the stringency of the standard due to
both the lower value and to the inclusion of three significant digits. The lower numeric value
and the three significant digits (instead of two) means that existing vehicles would be subject to
significantly more stringent PM standards, which is not consistent with the intent of creating
"interim bins."
EMA believes that this proposed change was unintentional on EPA's part. Either way, EPA
should not change the numeric level or the number of significant digits for the PM standard in
the "interim bins" for chassis dynamometer certified heavy-duty vehicles. Rather, the Agency
should - indeed, must - retain the PM standards at 0.02 g/mile for both Class 2b and Class 3
interim bins. Otherwise, EPA's proposed standards are inconsistent with its expressed intent -
which EMA supports - to allow "carryover" bins.
[EMA recommends that EPA undertake the following:]
- Correct the number of significant digits in the NOx standards for chassis-certified heavy-duty
vehicles to maintain the standards at 0.2 g/mi and 0.4 g/mi for Class 2b and Class 3 vehicles,
respectively
- Correct the numeric value and the number of significant digits in the PM standard for chassis-
certified heavy-duty vehicles to maintain the standard at 0.02 g/mi for Class 2b and Class 3
vehicles
Our Response:
We agree with EMA's comments regarding the numerical levels of the NOx and PM
standards for the interim bins. We have corrected the PM standard accordingly in the
regulations. In response to another comment (made by EMA and others—see Chapter 4.2.2.1),
we have combined the NMOG and NOx standards for these interim bins, with an additional
requirement capping the NOx level in certification testing. This capped level is specified in the
regulation with a single significant digit, as suggested by EMA. This, along with the flexibility
for NOx emissions provided under a combined NMOG+NOx standard, effectively addresses the
EMA comments regarding NOx.
What Commenters Said:
4-141
-------
Chapter 4: Proposed Vehicle Emissions Program
California Air Resources Board (CARB):
Medium Duty Vehicle Fleet Average Requirement: Unlike the LEV III program that requires
manufactures to certify a defined percentage of their medium-duty vehicle's to increasingly
stringent emission standards, the proposed Tier 3 program imposes a fleet average emission
requirement for this class of vehicles. Staff has determined that, while different in structure, the
LEV III and proposed Tier 3 emission requirements for medium-duty vehicles are identical.
Accordingly, to provide manufacturers with additional flexibility, CARB will propose an
identical fleet average option for compliance in the LEV III program.
Our Response:
We support CARB's intent to propose an identical declining fleet average standard
option.
What Commenters Said:
Alliance of Automobile Manufacturers (Alliance) and Association of Global Automakers
(Global Automakers):
The following list areas from the NPRM preamble for which EPA has requested comment on
exhaust emission related items. This includes comments and recommendations on items that are
not addressed in our main comment letter.
HDVs (FR29874): [EPA] request[s] comment on the usefulness of creating additional bins
between Bin 0 and the next lowest bin in each vehicle class, as a means of encouraging clean
technologies and adding flexibility. We see no need for additional bins beyond the bins needed
to harmonize with the California program.
HDVs (FR29876): [EPA] request[s] comment on extending the voluntary compliance
opportunity to the 2015 model year. Extending voluntary compliance to 2015 as proposed by
EPA provides no meaningful assistance to manufacturers as they transition to Tier 3. Instead,
EPA should pattern its HDV early credit provisions after those proposed for light-duty.
HDVs (FR29878): EPA requests comment on this proposed optional [PM] phase-in mechanism.
We support the optional phase-in mechanism.
Our Response:
We acknowledge the comments from the Alliance and Global Automakers regarding
additional bins, 2015 voluntary compliance, and the proposed optional PM phase-in mechanism,
and are not making changes to the associated final Tier 3 program provisions from what we
proposed. We are finalizing the proposed optional PM phase-in mechanism for which the
commenters expressed support. See Chapter 4.2.5 of this Summary and Analysis of Comments
document for further discussion of comments on early credits.
4-142
-------
Tier 3 Summary and Analysis of Comments
4.2.2.1. Combined NMOG + NOx Standards for High Bins
What Commenters Said:
Truck and Engine Manufacturers Association (EMA):
Bin Standards: EPA has proposed separate NMOG and NOx standards for the two highest bins
in both the 8,501-10,000 Ibs. GVWR range and the 10,001-14,000 Ibs. GVWR range of vehicles
and engines. EPA states that its proposal is "equivalent" to California's LEV II program. EPA
also proposes that those "interim Tier 3" bins be available only through model year 2021. (78
Fed. Reg. at 29875.)
EPA's proposal is not "equivalent" to the LEV II program nor does it harmonize with California
for the two highest bins in both the 8,501-10,000 Ibs. GVWR range and the 10,001- 14,000 Ibs.
GVWR range of vehicles and engines. Under California's requirements for the transition from
LEV II to LEV III standards, manufacturers may choose whether to certify such vehicles and
engines to separate NMOG and NOx or combined NMOG+NOx standards at the same levels as
EPA has proposed for the two highest bins in Classes 2b and 3 (395 (LEV) and 340 (ULEV) for
Class 2b; 630 (LEV) and 570 (ULEV) for Class 3). Moreover, the California LEV III
regulations allow manufacturers to certify to those combined NMOG+NOx standards for as long
as manufacturers are able to do so while still meeting the fleet-wide phase-in requirements.
Manufacturers certifying to those combined standards also will meet combined OBDII
thresholds as part of certification.
Meanwhile, EPA also is proposing to adopt ARB's OBDII regulations. Even though ARB's
OBDII threshold for NMOG+NOx will be combined, manufacturers certifying products in those
bins will not be able to certify to a combined standard or OBDII threshold, eliminating the
possibility of certification to a 50-state vehicle.
EPA should provide manufacturers with the same option to certify to combined NMOG+NOx
standards for all bins in the 8,501-14,000 Ibs. GVWR range and should allow such option
beyond the 2021 model year. Manufacturers who have done the development and testing
necessary to certify such vehicles to the ARB combined standards will have done so also with
the expectation that EPA would align its standards to assure harmonization nationwide. EPA's
Proposed Rule would force certification to separate NMOG and NOx standards for those bins
and undermine manufacturers' ability to certify a 50-state vehicle. Such a result would be
contrary to EPA's stated goal in the Proposed Rule.
Consistent with its intent to align the EPA program with the ARB standards generally, and to
support harmonization where technically feasible and reasonable, EPA should allow
manufacturers the option to certify to combined NMOG+NOx standards for the "interim" bins
for as long as manufacturers are able to do so.
4-143
-------
Chapter 4: Proposed Vehicle Emissions Program
Alliance of Automobile Manufacturers (Alliance) and Association of Global Automakers
(Global Automakers):
HDVs (FR29875): [EPA proposes] that vehicles in the interim bins meet separate NMOG and
NOX standards, as indicated in Table IV-13, rather than combined NMOG+NOX standards...
[EPA requests] comment on this issue and the proposed approach to addressing it.
We support combined NMOG+NOx for the interim HDV bins, consistent with the California
program and all other Tier 3 bins.
Our Response:
Industry commenters objected to both the proposed sunsetting of the interim bins and the
proposed separate NOx and NMOG standards for these bins, arguing that they overly restrict
manufacturer flexibility and work against harmonization with LEV III. However, commenters
did not address EPA's concern expressed in the NPRM regarding increased NOx emissions at
the interim bin levels.
After considering the comments, we believe a modified approach to the interim bins can
at least partly address the industry concerns regarding harmonization while still precluding
backsliding on NOx levels. We are finalizing the interim bins with combined NMOG+NOx
standards as requested by the commenters, but are adopting a restriction on deterioration-
adjusted NOx levels in certification testing, to the levels allowed under the current standards in
40 CFR 86.1816-08. These are 0.2 and 0.4 g/mi for Class 2b and Class 3, respectively. This
restriction will not apply to vehicles in use, and does not impose a parallel NMOG restriction.
Given our continuing concerns about NOx increases that would be allowed by the combined
standards at the interim bin levels, we believe that this approach and the associated certification
burden are reasonable, noting that manufacturers already must obtain NOx test results in
certifying to an NMOG+NOx standard, and the differing NOx and NMOG deterioration
mechanisms will likely dictate that they be considered separately in obtaining deteriorated
NMOG+NOx levels for certification.
We believe that making the interim bins available indefinitely would run counter to their
limited purpose as an aid to making the transition to Tier 3 emissions levels. Making these bins
permanent would, we believe, necessitate that they take on other key elements of the Tier 3
program such as longer useful life, SFTP compliance, and the use of Tier 3 certification fuel.
These requirements in turn would negate the usefulness of these bins in helping to carry over
some pre-Tier 3 vehicle designs during the transition years in which the declining fleet average
standard levels are high enough to accommodate their continued sale. By MY 2022, the fleet-
wide standard will be stringent enough to effectively eliminate the ability of manufacturers to use
interim bins while meeting the declining fleet average standard levels. We are therefore
adopting the sunsetting of the interim bins as proposed, making them available only through MY
2021.
4.2.3. HDV SFTP Exhaust Standards
4-144
-------
Tier 3 Summary and Analysis of Comments
What Commenters Said:
California Air Resources Board (CARB):
Linking of Medium-Duty Vehicle FTP and SFTP Emission Standards: CARB supports the
proposal to require medium-duty vehicles certifying to a Tier 3 ultralow- emission vehicle or
super-ultra-low-emission vehicle FTP-equivalent emission category to certify to the equivalent
SFTP emission category as well. CARB believes this requirement is appropriate because the
SFTP emission standards were designed to be met using the same hardware required to meet the
FTP emission standards. Accordingly, CARB intends to propose alignment with this proposal
once the Tier 3 program is finalized.
International Council on Clean Transportation (ICCT):
ICCT fully supports extending the supplemental FTP requirements to complete vehicles between
8,500 and 14,000, which were previously exempt.
Our Response:
We support CARB's intent to propose alignment with our requirement linking SFTP and
FTP compliance.
4.2.4. HDV In-Use Standards
What Commenters Said:
Alliance of Automobile Manufacturers (Alliance) and Association of Global Automakers
(Global Automakers):
Interim In-Use Standards: For the > 8,500 pounds GVWR vehicles and for the SFTP standards
for LDVs, the proposed Tier 3 regulations do not contain the interim in-use standards that were
adopted in LEV III. Interim in-use standards do not change the certification value, and therefore
should not have any environmental impacts. However, manufacturers must certify much bigger
vehicles to much lower standards across the board at the same time they're introducing new
GHG reduction technologies. For example, HDVs 8,500 - 10,000 pounds GVWR currently
meet standards in the range of 340 mg/mile NMOG+NOx, but vehicles will need to meet
standards that are half that as Tier 3 and LEV III phase in. The same is true for vehicles 10,000
- 14,000 pounds GVWR where the standards are currently about 560 mg/mile but will have to
meet standards almost half that under Tier 3. While these standards appear numerically higher
than the LDV standards, they are comparable in terms of stringency and emission control
hardware required given that they apply to larger vehicles that are tested under much heavier
loads including at adjusted loaded weight (ALW) instead of loaded vehicle weight (LVW).
4-145
-------
Chapter 4: Proposed Vehicle Emissions Program
The manufacturers do not have extensive in-use experience at these levels for these larger
vehicles and could face substantial jeopardy as the standards are adopted across a large segment
of their fleet. Interim in-use standards, which have historically been granted, simply reduce this
jeopardy and allow manufacturers to gain in-use data and experience with new, more stringent
requirements.
Similarly, interim in-use standards are appropriate for SFTP for both the MDV and LDV classes.
For the first time ever, the SFTP standards will apply at a 150,000 mile useful life. And the
required levels will be far lower than current standard levels. These two factors, coupled with
the rapid introduction of new technologies driven by the GHG and fuel economy requirements,
justify the need for interim in-use SFTP standards to help manufacturers manage their in-use
compliance risk.
Recommendation: We recommend EPA continue the historic practice of allowing interim in-use
standards by harmonizing with the LEV III interim in-use standards for the HDV FTP and SFTP
requirements and the LDV SFTP requirements.
Heavy Duty Vehicle FTP, SFTP, Altitude and PM Interim In-use Standards:
We propose that EPA align with the CA LEV III program by adopting the same interim in-use
provisions for the FTP, SFTP and PM that are provided in LEV III. These interim standards are
necessary to mitigate the risks (both known and unknown) as manufacturers phase in to the new
lower standards, a task which is complicated by: (1) rollout of new GHG-enabling engine
technologies; (2) first-time application of SFTP standards to HDVs; (3) new test cycle (LA-92);
(4) new PM test procedures (Part 1066); and (5) differences between federal and CA market fuel
quality (i.e., sulfur).
In support of its decision to exclude the LEV III interim in-use provisions from the Tier 3
program, EPA used data generated from only two vehicles. These vehicles were not aged to
150,000 mile full useful life, and were not representative of the vehicles that will be needed to
meet new stringent HDV GHG standards. EPA also referenced vehicle data taken from the
existing light-duty fleet to support the view that interim standards were not needed. The selected
vehicles had powertrains similar to their heavy-duty counterparts; however, these LDVs were not
run at adjusted loaded vehicle weight (ALVW) and cannot be properly used to make this
determination.
Recommendation: We recommend: (1) Harmonizing with LEV III by adopting the interim in-use
standards for vehicles >6,000 pounds GVWR for both FTP and SFTP.
California Air Resources Board (CARB):
SFTP Medium-Duty Vehicle Interim In-Use Emission Standards: CARB supports U.S. EPA's
proposal to not include interim in-use emission standards for NMOG+NOx and PM in the
medium-duty vehicle SFTP program. CARB agrees that the technologies required for SFTP
compliance are well-established and sufficient lead time is provided such that in-use interim
emission standards, which are typically reserved for new technologies are not needed.
4-146
-------
Tier 3 Summary and Analysis of Comments
Accordingly, CARB intends to propose alignment with this proposal once the Tier 3 program is
finalized.
Ford Motor Company (Ford):
HDV Interim In-Use Standards: In the Tier 3 NPRM, EPA did not adopt the LEV III interim in-
use provisions for vehicles in the 8,500-14,000 pound weight classes. Ford believes that EPA's
reluctance to harmonize with this aspect of the LEV III program is due, at least in part, to a lack
of data supporting the need for such provisions. Interim standards are typically intended to
provide a temporary mechanism for manufacturers to handle the known challenges and
unforeseen difficulties associated with the implementation of new, more stringent requirements.
Since manufacturers cannot be expected to supply data to substantiate all such future concerns
(the relevant vehicles and/or technologies often do not yet exist or have not yet been realized in
"hardware"), we believe that there is sufficient uncertainty in the proposed HDV program alone
to justify the adoption of the interim standards finalized in the LEV III program. These
uncertainties include an entirely new testing mode for heavy-duty vehicles (i.e. SFTP), new
application of LA-92 test cycle, and the roll-out of new test procedures in Part 1066. These
concerns are further intensified by the need to implement the new, innovative technologies and
vehicle configurations that will be required to meet the challenging GHG standards in this same
timeframe. It should also be noted that the case for interim standards is particularly strong in
regards to PM, not only because of the unique PM challenges for heavier vehicles detailed in the
preceding section, but also because EPA has proposed a new PM test procedure in Part 1066.
Recommendation: Ford proposes that EPA adopt the FTP and SFTP interim in-use standards
(including PM) that were finalized in California's LEV III MDV program.
General Motors LLC (GM):
Another important area for harmonization relates to in-use compliance. In-use compliance is the
biggest technical hurdle for manufacturers, and also the greatest risk to manufacturers due to all
of the factors that can impact in-use emission testing results. To help manage this in-use
compliance risk during the transitional years, the LEV III regulations provide interim in-use
standards. The Tier 3 regulations adopt some but not all of the LEV III interim in-use standards,
and as detailed in the Alliance/Global comments, GM requests that EPA adopt all the types of
interim in-use standards adopted in LEV III to ensure harmonization.
Truck and Engine Manufacturers Association (EMA):
Interim In-Use Standards: For vehicles above 8,500 Ibs. GVWR, the Proposed Tier 3 Standards
do not include the interim in-use standards that were adopted in the ARE LEV III program. EPA
should continue the historic practice of allowing interim in-use standards by harmonizing with
the LEV III interim in-use standards for the Heavy-Duty Vehicle FTP and SFTP requirements.
The interim standards are necessary to address challenges (both known and unknown) that
manufacturers will face as they phase products in to the new, more-stringent Tier 3 standards.
As manufacturers design, research, test, and produce Tier 3-compliant vehicles and engines that
4-147
-------
Chapter 4: Proposed Vehicle Emissions Program
are commercially viable for introduction into the market, manufacturers also must address the
following challenges: (1) rollout of new GHG-enabling engine technologies; (2) the first-time
application of SFTP standards to Heavy-Duty Vehicles (and at 150,000-mile useful life); (3) a
new test cycle (LA-92); (4) new PM test procedures; and (5) differences between federal and
California market fuel quality (i.e., sulfur). Without interim in-use standards, manufacturers face
greater risk that they will not be able to produce fully-compliant vehicles and engines or meet
customer demands for robust and reliable product.
The Proposed Rule would require manufacturers to certify much larger vehicles to much lower
standards across the board at the same time they are introducing new GHG reduction
technologies. For example, Heavy-Duty Vehicles 8,501-10,000 Ibs. GVWR currently meet
standards in the range of 340 mg/mile NMOG+NOx, but such vehicles will need to meet
standards that are half that as Tier 3 and LEV III phase in. The same is true for vehicles 10,001-
14,000 Ibs. GVWR, where the standards currently are about 560 mg/mile but will be reduced to
half that under Tier 3.
EPA has failed to demonstrate that its proposal to exclude interim in-use standards for Heavy-
Duty Vehicles above 8,500 Ibs. GVWR is technologically feasible. In support of its proposal to
exclude the LEV III interim in-use provisions from the Tier 3 program for vehicles over 8,500
Ibs. GVWR, EPA used data generated from only two medium-duty vehicles which were not aged
to 150,000 miles full useful life and were not representative of the vehicles that will be needed to
meet new, stringent GHG standards for heavy-duty vehicles. In addition, EPA referred to light-
duty vehicles with powertrains similar to their heavy-duty counterparts as support for the lack of
interim in-use standards. Yet, as discussed above, in apparently relying on such light-duty data,
EPA did not account for the additional technical challenges associated with meeting the
proposed standards for vehicles and engines over 8,500 Ibs. GVWR at ALVW.
Heavy-Duty Vehicles are designed to do work. Manufacturers must design engines used in such
"work-capable" vehicles with characteristics that allow the vehicle to do the work of hauling or
towing when such work is required. The design aspects of work vehicles impose laws of physics
and thermodynamics which result in work-capable vehicles having different emission
characteristics than light-duty vehicles. As a result, heavy-duty vehicles must be compliant
under much heavier loads than their light-duty counterparts. Light-duty vehicles cannot and
should not be relied on for demonstrating the technological feasibility of standards applied to
heavier vehicles at ALVW. As a result, EPA has failed to show that excluding interim in-use
standards is technologically feasible for manufacturers.
Manufacturers do not have extensive in-use compliance experience at the proposed standard
levels for larger, heavier-duty vehicles and could face significant challenges. Any time there is a
substantial change in regulatory requirements, as standards are adopted across a large segment of
their fleets, manufacturers not only must meet compliance obstacles but also address customer
concerns as new products and new technologies are introduced into the market. Interim in-use
standards, which have historically been allowed, reduce the potential compliance risks and
impediments to customer acceptance and permit manufacturers to use in-use data and experience
with the new, more stringent requirements to make any adjustments necessary to assure
compliance with the fully-implemented final standards.
4-148
-------
Tier 3 Summary and Analysis of Comments
In order to make its program practical and implementable, EPA should align with the California
LEV III program by adopting the same interim in-use provisions for the FTP, SFTP and PM
standards as are provided in LEV III.
- Include interim in-use FTP, SFTP, and SFTP PM standards for vehicles above 8,500 Ibs.
GVWR
Our Response:
After considering the comments we have concluded that relaxed interim in-use standards
are appropriate for HDVs, both for FTP and SFTP testing. The comments we received from
C ARB, offered in the spirit of fostering harmonization, do not provide new technical information
in support of going forward with our proposal not to set interim in-use standards. We are
adopting HDV in-use standards levels that are identical to those adopted for LEV III. We
consider these levels reasonable, in line with relaxed in-use standards adopted in past programs,
and helpful toward harmonization.
We are not applying interim in-use NMOG+NOx standards to the interim (two highest)
bins for the FTP standards, because these bins are intended for carry-over of existing designs,
and there should be little uncertainty over their in-use emissions performance. Interim bin
vehicles certified to the Tier 3 PM standards shall, however, be subject to the relaxed in-use PM
standards in the same way as for HDVs in other bins. Bin 0 standards are driven by specific
zero-emissions technologies for which in-use margins would not be appropriate, and so we are
not setting in-use standards for Bin 0. We are also adopting the general approach taken in LEV
III of making these interim standards available during the phase-in period (model years 2016-
2022) for the first two model years that a test group is newly certified to a Tier 3 NMOG+NOx
or PM standard.
4.2.5. HDV Emissions Averaging, Banking, and Trading Program
What Commenters Said:
Alliance of Automobile Manufacturers (Alliance) and Association of Global Automakers
(Global Automakers):
Early Opt-in vs. Early Credits:
Historically, agencies have provided a mechanism to earn early credits in order to facilitate the
transition into more stringent fleet average requirements and to safeguard against unforeseen
circumstances that could cause a deficit situation. Although the proposed HDV program allows
manufactures to opt-in to the program early, this is not equivalent to providing early credits. The
early opt-in option is also more stringent than the LEV III program, which will allow for
carryover of LEV II credits into the LEV III phase-in. Given the short lead time, extending this
opt-in period to the MY 2015 would not provide significant additional flexibility to
manufacturers.
4-149
-------
Chapter 4: Proposed Vehicle Emissions Program
Recommendation: We propose that EPA align the HDV credit provisions with the light-duty
program by allowing early credits to be generated in MYs 2016 and 2017 vs. a Bin 395/630
average. National HDV credits could then be proportionally capped at the CA level in MY 2018.
VEC vs. Fleet average:
EPA is proposing a fleet-average based compliance for HDVs. This is different than ARB's
approach in LEV III, which uses Vehicle Equivalent Credits (VEC) to determine compliance.
Adoption of these incongruent approaches for compliance demonstration and reporting would
prevent the harmonization of LEV III and Tier 3 for HDVs, even though the stringency of the
EPA and California standards are equivalent. Since the stringency is equivalent, manufacturers
have no preference for the method. ARB staff have indicated that ARB would consider allowing
optional compliance with the EPA fleet average. Although this would have the effect of
harmonizing the requirements, ARB is not currently in rulemaking and cannot consider
providing this option until after EPA finalizes Tier 3.
Recommendation: Since EPA is in rulemaking, we recommend that EPA provide an option for
manufacturers to comply using California's VEC system. After EPA finalizes Tier 3 and ARB
adopts the optional compliance with the fleet average approach in Tier 3 (assuming ARB does
so), manufacturers would not object to EPA eliminating the California VEC option.
Our Response:
Industry commenters argued that EPA should align the HDV credit provisions with the
light-duty program by allowing early Tier 3 credits to be generated in MYs 2016 and 2017,
calculated relative to the highest Class 2b and Class 3 bin NMOG+NOx levels (395 and 630
mg/mi, respectively), and capped at a level proportional to the California level in MY 2018.
However, these highest bin levels correspond to those of the existing HDV standards for NMHC
and NOx, and are significantly higher than the MY 2016 and 2017 LEV III levels. Thus vehicles
designed to just meet the LEV III standards in these years could generate a large preliminary
number of credits under the industry's Tier 3 early credits proposal, credits they would not earn
in LEV III, thereby potentially thwarting the harmonization of the two programs, undermining
the stringency and benefits of Tier 3, and bringing into question whether the standards should be
revised to meet the requirements of Clean Air Act section 202(a)(3)(A).
We considered whether truncating that credit bank for each manufacturer in 2018 such
that it is proportional to their LEV III balance, combined with additional restrictions on trading
and banking, would address these concerns and restore a harmonized credit status in that year.
However, such an approach would constitute an unnecessarily complex and uncertain pathway to
the same result as that achieved under EPA's early opt-in provisions. We are also not we
providing for the conversion of pre-Tier 3 HDV credits for use in Tier 3, as we believe that by
providing an early Tier 3 opt-in program for HDVs, capable of generating credits for two model
years before the mandatory standards take effect (even longer under the alternative percent-of-
sales phase-in approach), we are giving ample opportunity for the manufacturers to accumulate
early credits. HDV manufacturers are currently certifying their vehicles to existing standards
without generating or using NOx or NMHC credits, and the levels we set for Tier 3 standards are
4-150
-------
Tier 3 Summary and Analysis of Comments
not based on any assumption of credit transfers into Tier 3. Allowing such transfers would also
introduce new complexities in dealing with the conversion of NOx and NMHC credits into
NMOG+NOx credits.
Manufacturers commented that the proposed fleet average compliance approach is
incongruous with California's LEV III method based on vehicle equivalent credits (VECs).
Although stating that they do not have a preference for the method (since the stringency is
equivalent), they recommended that EPA foster harmonization by providing a compliance option
based on VECs. We believe that such an option would add unnecessary complexity to the Tier 3
program, and is made even more unnecessary by the intent expressed in CARS's written
comments to propose a fleet average option for LEV III that is identical to EPA's approach.
4.2.6. Other HDV Provisions
What Commenters Said:
Alliance of Automobile Manufacturers (Alliance) and Association of Global Automakers
(Global Automakers):
Chassis vs. Dyno Cert:
We request that EPA align with LEV III and allow the option to choose chassis or dyno
certification for complete diesel engines 10,000-14,000 Ibs. GVWR (Class 3). Without this
option, manufacturer's may be required to dual certify vehicle models that include variants both
under and over 14,000 pounds.
International Council on Clean Transportation (ICCT):
The ICCT supports maintaining fuel-neutral criteria emissions standards for heavy-duty vehicles.
We also fully support extending chassis-based emission requirements to all complete vehicles up
to 14,000 gross vehicle weight.
It is ICCT's position that technology-neutral performance standards are critical in any
transportation policy, especially one that involves multiple and alternative fuels that can all be
utilized as part of meeting emissions and energy objectives. Thus, we are supportive of
maintaining fuel neutral criteria emissions standards for heavy-duty vehicles.
The ICCT fully supports extending chassis-based emission requirements to all complete vehicles
up to 14,000 gross vehicle weight. The trend since the first standards were adopted in the 1970s
has been to increase the GVW of pickup and other light trucks above the threshold for light-duty
emission standards. This has especially been a problem for diesel engines in pickup trucks,
which are only sold above 8,500 GVW in order to avoid the light-duty emission standards.
Extending the threshold to 14,000 GVW will ensure emission standards are applied appropriately
to all complete vehicles.
Truck and Engine Manufacturers Association (EMA):
4-151
-------
Chapter 4: Proposed Vehicle Emissions Program
Chassis- and Engine-Dynamometer Certification of Complete Vehicles:
EPA has proposed to "codify" the practice of most, if not all, manufacturers that currently
chassis-certify their Heavy-Duty Vehicles by requiring all diesel-fueled complete vehicles to be
chassis-certified under the Proposed Rule. Furthermore, although incomplete Heavy-Duty
Vehicles are not required to be chassis-certified, EPA has requested comment on requiring
chassis certification of Class 2 incomplete vehicles as is required under the California Air
Resources Board's ("ARE") LEV III program for vehicles up to 10,000 Ibs. GVWR.
EMA and its members oppose mandatory chassis-certification for any class of engines or
vehicles over 8,500 Ibs. GVWR. Even if manufacturers have more frequently chosen to certify
such vehicles on the chassis test, EPA should not remove the flexibility manufacturers currently
have with respect to certification of diesel-fueled complete vehicles. Rather, EPA should
finalize a rule that allows engine manufacturers the continued option to choose either chassis-
dynamometer or engine-dynamometer certification for both complete and incomplete
engines/vehicles over 8,500 Ibs. GVWR. Maintaining the option allows manufacturers
continued flexibility in certification decisions, thereby minimizing unnecessary costs and
certification burdens.
EMA recognizes that its recommendation for optional chassis-certification for complete vehicles
is one area where we are not recommending harmonization with ARB, which mandates chassis-
dynamometer certification for complete vehicles with GVWRs between 8,501 and 10,000 Ibs.
GVWR. Such lack of harmonization in this area is appropriate, however, as it provides
manufacturers greater compliance flexibility without adverse emissions impact.
If EPA nevertheless proceeds with mandatory chassis-certification, EPA, at a minimum, should
align with ARB's LEV III Rule and allow manufacturers the option of engine or chassis
dynamometer certification for complete vehicles with GVWRs between 10,001 and 14,000 Ibs.
or the engines used in them.
- Allow engine manufacturers the continued option to choose either chassis dynamometer or
engine-dynamometer certification for both complete and incomplete vehicles between 8,501 Ibs.
and 14,000 Ibs. GVWR.
- Adopt the recommended high-altitude standards for heavy-duty vehicles.
- Engage in further review of optional chassis- or engine-dynamometer certification for gasoline-
or diesel-fueled vehicles over 14,000 Ibs. GVWR.
Our Response:
Industry commenters opposed mandatory chassis certification for complete diesel
vehicles, especially for vehicles in Class 3. ICCT supported it. It is difficult to assess the degree
to which a desire to avoid light-duty regulations has driven a shift in the diesel market toward
heavy-duty pickups as asserted by ICCT. Nevertheless, although sensitive to the issues raised by
the manufacturers, we remain concerned that the fleet average standard program we are
finalizing would not work well if a major fleet component, such as complete Class 3 diesel
trucks, can be left in or taken out of the fleet calculation based on what each manufacturer
4-152
-------
Tier 3 Summary and Analysis of Comments
considers to be most advantageous. We believe the resulting competitive issues and
uncertainties would be problematic, given the wide variance in gasoline/diesel HDV sales among
the manufacturers, our provision for averaging across each manufacturer's entire Class 2b/3
fleet, and the overwhelming preponderance of diesels in the Class 3 market. It would also create
uncertainties in the Tier 3 environmental benefits, given the pronounced difference between
these Tier 3 standards and the heavy-duty diesel engine standards we set 13 years ago, which we
expect to remain in effect for the foreseeable future.
As a result, we are finalizing the proposed requirement for chassis certification of
complete (but not incomplete) diesel Class 2b/Class 3 HDVs, except that we are providing that
manufacturers of complete diesel Class 3 HDVs may, instead of certifying these vehicles, install
diesel engines that are engine-certified in any model year that the engine family has less than half
of its sales being installed in such complete Class 3 vehicles. This provision is intended to help
address manufacturers' concern about dual certification, while at the same time ensuring a
coherent fleetwide standards regimen in this vehicle class. It also better harmonizes with
California's LEV III program which does not mandate chassis certification for diesel Class 3
vehicles. By only allowing for engine-certified vehicles when they use engines primarily
produced for other purposes, we believe this approach adequately guards against potential
abuse. In the case of complete diesel Class 3 HDVs produced by a company other than the
engine certifier, the responsibility for ensuring the sales limit is not exceeded remains with the
vehicle manufacturer, who will need to coordinate with the engine supplier.
4.2.6.1. High Altitude Standards
What Commenters Said:
Truck and Engine Manufacturers Association (EMA):
High-Altitude Standards:
EPA requires high-altitude testing for Heavy-Duty Vehicles (8,501-14,000 Ibs. GVWR). While
EPA provides high-altitude testing relief in the Tier 3 Rule (relative to low-altitude standards) for
light-duty vehicles, EPA does not provide similar relief for Heavy-Duty Vehicles. (78 Fed. Reg.
29870, Table IV-9.) Separate high-altitude standards such as those provided in the light-duty
Tier 3 program also are needed for HDVs, which must test at ALVW even at high altitude.
EMA recommends that EPA adopt separate high-altitude standards for Heavy-Duty Vehicles in
the same proportion as it provides for LDVs. The table below (marked "Table IV- 13B" to
indicate its proposed placement in the Preamble) [The table can be found on p. 8 of Docket
number EPA-HQ-OAR-2011-0135-4314-A1] shows the high-altitude FTP Standards
recommended by EMA. Consistent with the proposed relief for light-duty vehicles, no
adjustment is made for the interim bins.
EMA has recommended high-altitude standards for Class 2b Bin 250 and for Class 3 Bin 400
that are higher than the high-altitude standard for Class 2b Bin 340 and for Class 3 bin 570,
respectively. The explanation for this apparent anomaly is that the useful life for the non-interim
4-153
-------
Chapter 4: Proposed Vehicle Emissions Program
bins is higher (150,000 miles) than it is for the interim bins (120,000 miles), thus requiring
higher standards for the immediately lower bins.
Ford Motor Company (Ford):
HDV High-Altitude Provisions:
In the proposed rule, EPA acknowledged that fundamental physical challenges exist at high
altitude that typically results in higher emissions during cold starts. EPA accordingly proposed
altitude relief, but only for light-duty vehicles. Heavy duty vehicles have similar, if not greater,
challenges at high altitude. These vehicles are designed to operate under load and trailer tow
conditions, and as a result the emissions systems are located further downstream from the
exhaust manifold relative to a light duty vehicle, further delaying catalyst light-off To provide
manufacturers with sufficient compliance margin, we propose that EPA provide altitude relief
for the most stringent HDV bins.
Ford proposes that EPA also provide the following high-altitude standards for HDVs:
[Specific proposed high-altitude standards for HDVs follow.]
Alliance of Automobile Manufacturers (Alliance) and Association of Global Automakers
(Global Automakers):
Separate altitude standards such as those provided in the light-duty Tier 3 program are also
needed for HDVs, which must test at ALVW even at high altitude.
We recommend: Adopting separate high-altitude standards for vehicles >6,000 pounds GVWR
in the same proportion as LDV.
Our Response:
Manufacturers argued in their comments that the reasons EPA cited in proposing relief at
high altitudes for light-duty vehicles apply for HDVs as well, and requested that relaxed
NMOG+NOx standards be adopted in the more stringent bins for testing of HDVs at high
altitudes. Ford argued that the challenges could be even greater for HDVs because they are
designed to operate at high altitudes with heavy payloads and towed trailers, and this may
necessitate the locating of emissions systems farther from exhaust manifolds, thereby increasing
catalyst lightoff delays.
Although we agree to a certain extent about the performance of gasoline-fueled HDVs at
high altitudes and their similarity to LDVs, the comments did not alter our view that the
compliance margins provided in the HDV FTP bin standards compared to what the control
technologies can achieve, and the freedom manufacturers have to shift to the more stringent bins
gradually as the program phases in, are adequate to account for these effects at altitude. The
manufacturers provided no data to counter this view.
We note that our adoption of relaxed interim in-use standards for vehicles in these bins
will be directonally helpful to address any remaining concerns by manufacturers regarding
4-154
-------
Tier 3 Summary and Analysis of Comments
emissions at altitude (preamble Section IV.B.6.a). This is because testing at high altitudes is
often not required for certification (typically manufacturers use an engineering analysis instead),
and thus the relaxed in-use standards will help to facilitate Tier 3 implementation for any HDV
designs in which in-use problems at high altitudes surface in the initial model years.
4.2.6.2. Vehicles Over 14,000 Lbs.
What Commenters Said:
Truck and Engine Manufacturers Association (EMA):
Optional Certification for Vehicles Over 14,000 Pounds GVWR:
Currently, heavy-duty gasoline vehicles (HDGVs) above 14,000 Ibs. GVWR (or the engines
used in them) have the option of certifying on either an engine dynamometer or chassis
dynamometer to demonstrate compliance with GHG requirements. As EPA notes in the
Preamble, manufacturers choosing the chassis dynamometer option to certify a family of
HDGVs for GHGs must use the engine dynamometer test to certify to g/hp-hr standards for all
other emissions. (78 Fed. Reg. at 29883.) EPA requests comment on extending this HDGV
option for the demonstration of compliance with criteria pollutant requirements.
In addition, heavy-duty diesel vehicles (HDDVs) above 14,000 Ibs GVWR (or the engines used
in them) currently must certify on an engine dynamometer to demonstrate compliance for both
criteria pollutant and GHG requirements. EPA requests comment on whether to provide the
option to certify on either an engine dynamometer or chassis dynamometer to demonstrate
compliance with criteria pollutant and GHG requirements to HDDVs (or the engines used in
them).
While EMA generally supports compliance options that provide flexibility to manufacturers in
meeting regulatory requirements, there are still significant unknowns regarding EPA's intent in
possibly offering the options. EMA recommends that EPA engage heavy-duty engine and
vehicle manufacturers in a thorough review of the issues and their potential impacts. That has
not yet been done. In addition, we will continue to review EPA's request for comment and will
strive to provide EPA supplemental comments.
Cummins Inc.:
Optional Chassis Certification for Vehicles above 14,000 Pounds GVWR:
The existing heavy-duty greenhouse gas (GHG) regulations allow manufacturers to certify
heavy-duty gasoline vehicles (both complete and cab-complete) on either the engine
dynamometer or chassis dynamometer to demonstrate compliance with GHG requirements.
However such vehicles must be certified on an engine dynamometer for all other emissions
standards (g/hp-hr). EPA requests comment on allowing a manufacturer to certify a heavy-duty
gasoline vehicle (HDGV) for all emissions standards using chassis dynamometer procedures.
4-155
-------
Chapter 4: Proposed Vehicle Emissions Program
In addition, engines for heavy-duty diesel vehicles (HDDVs) above 14,000 Ibs GVWR currently
must certify on an engine dynamometer to demonstrate compliance for both criteria pollutant and
GHG requirements. EPA requests comment on whether to provide the option to certify on either
the engine dynamometer or chassis dynamometer to demonstrate compliance with criteria
pollutant and GHG requirements for engines for HDDVs.
Cummins does not support an option of chassis certification for GHG or criteria emission
standards for vehicles above 14,000 Ibs GVWR. Cummins urges EPA not to finalize these
options.
In the NPRM, EPA has not provided any regulatory language or discussed this option in detail in
the Preamble. It is not clear how such an option would work and what the limits are relative to
vehicle size. For example, EPA has not specified whether such an option would be limited to
Class 4 vehicles only or if it would extend to even bigger vehicles. In absence of such details, it
is impossible for Cummins or any entity for that matter to consider the implications of these
provisions.
For vehicles above 14,000 Ibs GVWR, EPA already has a robust regulatory framework for both
GHG and criteria emissions based on engine dynamometer test procedures. The engine
dynamometer test cycles (e.g. FTP) are appropriate for such vehicles, and EPA would need to
study the applicability of chassis dynamometer test cycles if the Agency wants to move forward
with consideration of these options.
Under the current GHG rule, HDGVs above 14,000 Ibs can certify to chassis standards based on
a 'complete sister vehicle' concept. Vehicles above 14,000 Ibs (e.g. Class 4-6 vehicles) are
typically engineered and marketed to meet vocational requirements. Many of the vehicles in this
class are not sold as complete, ready to be placed into service, vehicles. The common sales path
to the end-user is through a specialty body or utility function builder. While the vehicle OEM
can communicate parameters such as aerodynamic design standards to which the finish builders
must comply, the finish body may exceed the frontal projection and/or may affect air flow quite
differently than on a complete pickup truck or even on the bare "cab-complete" vehicle. Hence
it may not be appropriate to use coastdown and test weight information of a complete sister
vehicle (in Class 2b/3) for chassis certification of vehicles above 14,000 Ibs GVWR.
EPA also requested comments on whether manufacturers of such vehicles that are certified to a
Final Tier 3 bin should be allowed to exclude them from the fleet average NMOG+NOX
calculation. Cummins does not support this provision. Excluding these vehicles from fleet
average calculations could potentially lead to unintended consequences. For example, a
manufacturer could increase GVWR of their heavier Class 3 vehicles to just above the threshold
(e.g., 14,001 Ibs GVWR) and then exclude them from fleet average calculations.
Recommendation: Cummins urges EPA not to finalize these proposed options for chassis
certification for vehicles above 14,000 Ibs GVWR or exclusion of these vehicles from fleet
averaging. If EPA wants to consider these options, Cummins would work with the Agency and
other heavy-duty engine and vehicle manufacturers to engage in a complete review of the issues
and potential impacts of such a rule.
4-156
-------
Tier 3 Summary and Analysis of Comments
Our Response:
Cummins and EMA responded to our request for comment on optionally extending
chassis-based certification to vehicles over 14,000 Ibs. Both commenters felt that a more
thorough review of the issues and potential impacts involved needs to occur before EPA takes
action on this matter. We agree and are not making this change in this rule.
4.3. Evaporative and Refueling Emission Standard Comments
See 78 FR 29884-29899, May 21, 2013 for full detail on the proposal
4.3.1. Tier 3 Evaporative Emission Standards
What Commenters Said:
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
Evaporative Emissions 2017 Phase in Requirement:
The Tier 3, LEV III, and PZEV zero evaporative emissions standards are identical. The phase-in
is also identical in all but MY 2017. For MY 2017, California does not require a specific
percentage of a manufacturer's fleet meet the LEV III evaporative emissions standards. Instead,
MY 2017 vehicles in California that generate PZEV credits are required to certify to the zero
evaporative emission standard. The proposed Tier 3 regulations allow two options for
compliance in MY 2017. The first option requires a manufacturer to meet the Tier 3 evaporative
emissions standard on 40 percent of its fleet less than 6,000 pounds GVWR. The second option
requires the manufacturer to make the PZEV evap systems available nationwide (i.e., for all
California PZEVs, the evaporative emissions control system would need to be sold nationwide).
The Tier 3, LEV III, and PZEV evaporative emissions standards require manufacturers to make
significant hardware changes to the vehicle. These hardware changes trickle down throughout
the manufacturers' development, testing, certification, and supply chain and require substantial
lead time to avoid very costly disruptions. While the Tier 3 proposal would have provided
sufficient lead time had Tier 3 been finalized in late 2011 (i.e., the 2012 MY), meeting those
same requirements when manufacturers have lost two years of lead time would be very costly
and disruptive. In addition to the early leak emission standard option described above, the option
to make PZEVs available nationwide isn't necessarily as clear as it could be. We understand that
this option would require that if a manufacturer made a single vehicle model with three option
packages (4-cylinder, 6-cylinder, and hybrid) but only the hybrid was a PZEV, then the
manufacturer would need to make the evaporative system on the hybrid available nationwide.
The manufacturer would not have to meet Tier 3 evap (or PZEV evap) on the 4-cylinder and 6-
cylinder vehicles nor restrict sales of those vehicles. However, the regulatory requirements in
§86-1813-17 could be read to require the manufacturer to produce Tier 3 evap in all three
vehicles, which we understand is not EPA's intent. Furthermore, we believe EPA's ultimate
intent with this provision is that manufacturers choosing this option produce about the same
proportion of zero evaporative emission vehicles in the non-Section 177 states as they produce in
4-157
-------
Chapter 4: Proposed Vehicle Emissions Program
the Section 177 states without being subject to sales mix differences between the two regions
which are beyond the manufacturer's control. We recommend language below to clarify this
option.
We recommend revising §86-1813-17(g)(3) as follows:
(3) You may disregard the percentage phase-in specified in paragraph (a)(5) of this section for
2017 if you choose 50-state certification for all your vehicles meeting the LEV III PZEV
evaporative standards in 2017. Under this option, you may not produce a higher-emitting version
of those vehicle models for sale outside of California or the section 177 states. For example, if a
manufacturer produces Model X in four configurations (4-cylinder (50-State non-PZEV), 6-
cvlinder (50-State non-PZEVX hybrid CCA PZEV). and hybrid (Federal non-PZEVX then you
could comply with the 2017 model year requirement by putting the CA-PZEV evap system on
the hybrid (Federal, non-PZEV) vehicle: you would not have to make changes to the 4-cvlinder
or 6-cylinder vehicles. Such vehicles may be certified using carryover data under the California
program; however, they may generate or use emission credits only if they are certified to meet
the emission standards of paragraph (a)(2) of this section. Vehicles that comply under this
paragraph (g)(3) may not generate allowances under paragraph (g)(l) of this section, regardless
of the calculated percentage of compliant vehicles in model year 2017. Furthermore, you will be
deemed in compliance if the fraction of 2017 MY vehicles with LEV III PZEV evaporative
systems sold in the non-Section 177 states is within three percent of the fraction of 2017 MY
vehicles you sold in California.'
Our Response:
This option was proposed to permit the manufacturers to simply extend their current
SULEV or PZEV Zero Evap offerings nationwide as a basis for providing the non section 177
states the same emission reduction benefits as would occur in California and the section 177
states. One of several benefits of using the PZEV zero evap only option was that there would be
no end of model year accounting required at the Federal level for the 2017 model year. EPA
recognizes that the fleet mix in each state is not identical and that in phase-in programs such as in
Tier 3 (or LEV III) there could be very minor year-to-year differences in the number of Tier 3
compliant vehicles in any given state. Thus, we see no need for setting the three percent
requirement suggested by the commenter.
Central to this option is the principle that any PZEV zero evap model (e.g., fuel
system/powertrain configuration) offered for sale in California and/or any of the section 177
states must be offered for sale nationwide. This does not preclude a manufacturer from offering
different fuel system/power train configurations of the same base model for sale nationwide but
only if they are offered for sale in California and the section 177 states. Stated differently, to
qualify for this option in the 2017 model year, a manufacturer must offer all fuel system/power
train /configurations/evaporative control system configurations of a base model it sells in
California and the section 177 states in the non-section 177 states also.
High-Altitude testing-FTTP and 3-Day (FR29892)
4-158
-------
Tier 3 Summary and Analysis of Comments
What Commenters Said:
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
EPA indicates the following regarding high altitude 3-day evap.: "We are proposing to keep that
requirement but to allow for an adjustment of 5°F in the temperatures related to the running loss
test within the 3-day test cycle. Thus, the applicable fuel and ambient temperatures at §86.134-96
(f) and (g) would 90±5 °F instead of 95±5 °F for high altitude testing."
We recommend EPA clarify that all temperature values within the Fuel Tank Temperature
Profile (established for sea level testing) will be adjusted down by 5°F for high altitude testing.
Additionally, the hot soak and diurnal temperatures should be adjusted down by 5 °F for high
altitude testing.
To help reduce the testing burden on industry, we recommend eliminating the high-altitude 3-day
requirement. The 3-Day test includes higher temperature inputs that are not typically
encountered at high altitude.
In fact, to further reduce testing burden on industry, we would support removing the 3-Day
requirement at sea-level as well.
Our Response:
We agree with the commenter regarding the technical rationale for a 5F° downward
adjustment in the fuel tank temperature profile for testing at altitude. This has merit because of
the nature of the evaporative process for 7.8 RVP fuel at the higher tank temperature and lower
atmospheric pressure a as compared to sea level. The 5°F downward adjustment in the sea level
fuel tank temperature profiles for high altitude testing is still representative for high altitude and
control will still occur using the control systems/technical approaches designed for low altitude
operations. This argument does not apply to the temperature cycle used for the diurnal test
because temperatures of 96°F or more are encountered in high altitude areas in the ozone
season17 and manufacturers provided no information supporting changes to these test elements at
altitude or sea level. Therefore, we are not eliminating the three day test or adjusting test
temperatures at low or high altitude.
Organization: American Honda Motor Co., Inc.
High Altitude Evaporative Emission Standards:
Under the Tier 3 rule, EPA allows vehicles certified to LEV III to be carried over into Tier 3.
However, LEV III does not specify high altitude standards. Tier 3 sets less stringent
standards/requirements for high altitude conditions, due to many, well-understood reasons (e.g.
lower atmospheric pressure on vapor generation rates, canister loading and purge dynamics, etc).
In §86.113-04(a)(2)(ii)(C), EPA states that LEV III vehicles must also meet "standards at high-
17 See for example, http://coolweather.net/statetemperature/colorado_temperature.htm.
4-159
-------
Chapter 4: Proposed Vehicle Emissions Program
altitude conditions." Honda believes this requirement is too stringent, and inconsistent with
EPA's intention to relax high altitude obligations. Vehicles that certify to LEV III standards
should be allowed to meet the relaxed high altitude EVAP standards of Tier 3.
Our Response:
We concur with the commenters view regarding hot soak plus diurnal evaporative
emission standards for high altitude. The intent of this provision is for any LEV III or PZEV
zero evap carryover vehicles to meet the Tier 3 high altitude standards if it is to be sold outside
of California. For PZEV zero evap carryover the manufacturers can use Tier 2 or Tier 3 fuel to
meet the Tier 3 high altitude standards through the 2019 model year. For vehicles meeting the
LEVIII option 1 or option 2 standards in the 2015 model year and using carryover for future
model years they can use Tier 2 or Tier 3 fuel to meet the Tier 3 high altitude standards. For new
Tier 3, LEV III option 1 or option 2 certifications in the 2016 and later model years they must
use Tier 3 test fuel for the Tier 3 high altitude standards.
What Commenters Said:
Organization: Aston Martin Lagonda Ltd
For evap capability AML recognizes the reasons for EPA wanting to keep all three tests and even
adding a leak test; however the 3 day test still allows point source running loss testing to be
accepted rather than full enclosed chamber measurement. In this respect I would propose that
SVMs certify only using the ORVR test and the 2 day test. The ORVR test effectively drives
canister capacity requirements. The 2 day test is the best method for validating purge capability
(backed up with the bleed test) and for driving a diurnal capable system design. The 3 day test
does not add anything to the validation of an evap system design and in fact with point source
running loss measurement the 3 day test with its extended diurnal does not prove system
capability any further than the 2 day test but does incur significant testing burden and cost for
SVMs. Additionally Aston Martin acknowledges and approves of the removal of the need for the
leak test at time of certification.
Our Response:
The leak test and standard apply at time of certification and in-use, but can be met by
attestation at time of certification. EPA does not concur with the commenter with regard to the
three day evaporative emissions test. The commenter's technical assertions are often but not
consistently true. For example, for an ORVR system with a mechanical seal, canister volume and
working capacity may be driven by the three day test. The same is the case for sealed fuel tanks
used on some hybrid vehicles and for any non-integrated evaporative and refueling system
canister approaches. Also, it should be noted that another commenter encouraged EPA to
consider diurnal emissions control for more than three days, not less.
What Commenters Said:
4-160
-------
Tier 3 Summary and Analysis of Comments
Organization: California Air Resources Board (CARB)
Evaporative canister bleed test standard in-use requirement:
CARB supports U.S. EPA's proposal to implement an in-use requirement for the bleed test
standard. This in-use requirement would ensure that canister performance is maintained as
vehicles age. Accordingly, CARB intends to propose alignment with this proposal once the Tier
3 program is finalized.
Our Response:
We acknowledge CARB's support and intention to propose revisions to align with EPA's
canister bleed test standard and in-use requirements.
4.3.2. Program Structure and Implementation Flexibilities
What Commenters Said:
Organization: Volkswagen Group of America, Inc.
VWGoA would also like to request that the Family Emission Limits (FEL) for evap Tier 3 be set
at a value much lower than the currently proposed 25 mg. This large increment does not provide
the needed flexibility to ensure success on a fleet wide level. VW requests an FEL of 5 mg.
Our Response:
EPA understands that a lower or unfixed increment for the FEL for the hot soak plus
diurnal standard would potentially lead to greater credits or smaller debits. However, we believe
the increment we proposed is appropriate. Setting a smaller increment creates the potential for
false credits which could occur by shaving the compliance margin or by taking credit for the
decay in non-fuel hydrocarbon emissions which occurs over time on real world vehicles.
A 25 mg increment means there will be a compliance margin of zero to 24 mg relative to
the FEL depending on the measured emission level. This is up to 8 percent of the level of the
standard of 300 mg for LDVs/LDTls, which we believe is not an overly high percentage. In
2013 VWGoA certified with compliance margins of 20-80 percent relative to the MS AT
standards, and the differences between the measured levels and the certified levels were very
small ranging from zero to 20 percent. The 25 mg increment generally results in a compliance
margin larger than that for a 5 mg increment, but this is reasonable since, for example, in 2013
all of VWGoA's evaporative families certified with a compliance margin of 20 percent or more
of the standard.18 Finally, it should be noted that participation in ABT is voluntary.
1818 Passavant, G. (February, 2014). "Volkswagen Group of America Evaporative Emissions Certification Data for
the 2013 Model Yeaf. Memorandum to the docket.
4-161
-------
Chapter 4: Proposed Vehicle Emissions Program
What Commenters Said:
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
Cert (FR29887):
Beginning in the 2017 MY, all new evaporative/refueling emission family certifications would
have to meet the proposed EPA Tier 3 certification requirements for both test procedure and
certification test fuel for evaporative and refueling emission standards. This statement should
apply for all new evaporative/refueling certification testing for families certified to the Tier 3
requirements, not all new evaporative/refueling testing. This requirement should not apply to
evaporative/refueling families certified to the Tier 2 requirements.
Our Response:
EPA concurs with the comment. As specified in the preamble, Tier 3 evaporative and
refueling test procedures and certification test fuel apply to vehicles brought into the Tier 3
evaporative emission program. As discussed in the preamble, these requirements vary depending
on the program option and the applicable standards. They do not apply to vehicles still meeting
Tier 2 evaporative emission requirements.
What Commenters Said:
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
EPA asked for comment on a third option that would require that a manufacturer meet the Tier 3
evaporative emissions standard on 20 percent of its fleet less than 6,000 pounds GVWR and also
meet the leak check standard on those vehicles. We support this option and appreciate that EPA
understands our concerns. We agree that manufacturers could meet the leak check standard on 20
percent of their MY 2017 fleet; however, since these are independent requirements, there is no
reason that manufacturers need to overlap the vehicles meeting Tier 3 evaporative emission
standards with vehicles meeting the leak check standard.
Recommendation:
1. We recommend adding the following as new paragraph §86-1813-17(g)(2).
"(2) For 2017 model year only, you may reduce the percentage phase-in specified in paragraph
(a)(5) of this section to 20 percent of projected vehicle sales of vehicles at or below 6,000 pounds
GVWR, if you also certify 20 percent of your vehicles at or below 6,000 pounds GVWR to the
Tier 3 leak emission standard in paragraph (a)(4) of this section. Also, if you certify vehicles
above 6,000 pounds GVWR to the Tier 3 evaporative emissions standard in model year 2017,
you may count projected U.S. sales of those vehicles toward your calculation for meeting the 20
percent Tier 3 evaporative emissions phase-in requirement in 2017 (numerator only)."
What Commenters Said:
4-162
-------
Tier 3 Summary and Analysis of Comments
Organization: General Motors LLC (GM)
For phasing in the evaporative emission standards, we support a 20% requirement in 2017
coupled with an alternative phase-in approach starting in 2017 to facilitate the ramp-up to
national volumes while maintaining equivalent stringency with the California LEV III program.
This is detailed further below.
Evaporative Emission Phase-in
EPA asked for comment on the Tier 3 evaporative emission phase-in schedule options. One of
the phase-in options listed would require OEMs to meet the Tier 3 evaporative emissions
standards in the 2017 model year for 20% of their fleet for passenger cars and LDTs < 6,000 Ibs.
GVWR. These vehicles would also have to comply with the leak check standard. Phase-in
schedules greater than the 20% option are problematic due to insufficient lead-time and zero
evaporative system component supplier constraints. Therefore, GM supports a 20% requirement
in the 2017 model year.
The evaporative emission phase-in rate is especially aggressive, and the combination of a 20%
requirement in 2017 along with the allowance of an alternative phase-in approach beginning in
2017 will provide needed flexibility for expanding zero evaporative system technology
throughout the U.S.
Our Response:
The 20/20 option for the 2017 model year is included in the final rule. The number of
vehicles needed to meet the 20 percent value is based on multiplying 0.2 times the passenger car
and light truck sales < 6,000 Ibs GVWR for the non-section 177 states and excluding California.
Compliance with this value can be from any vehicle category covered by the Tier 3 program.
Under the 20/20 program the manufacturer must meet or exceed each of the two values: 20
percent Tier 3 evaporative hot soak plus diurnal and 20 percent leak standard. These standards
may be net on the same or different vehicles. Manufacturers utilizing this option will have to
demonstrate that they meet the 20 percent requirement based on actual sales after the end of the
2017 model year. There is no flexibility to under comply with one percentage but offset by over
complying with the other. These vehicles must also meet the 0.020" evaporative system leak
monitoring requirement which also takes effect in the 2017 model year.
4.3.3. Alternative Phase-in Percentage Approach
What Commenters Said:
Organization: General Motors LLC (GM)
In addition, GM supports the allowance of an alternative phase-in approach, starting in 2017
model year, which allows flexibility from model year to model year in the phase-in percentage as
4-163
-------
Chapter 4: Proposed Vehicle Emissions Program
long as the manufacturer achieves a minimum number of phase-in points equivalent to the
baseline phase-in percentage.
The evaporative emission phase-in rate is especially aggressive, and the combination of a 20%
requirement in 2017 along with the allowance of an alternative phase-in approach beginning in
2017 will provide needed flexibility for expanding zero evaporative system technology
throughout the U.S.
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
Alternative Phase-in Percentage Approach: EPA proposes that manufacturers may use an
alternative phase-in approach that is equivalent to the normal percentage phase-in requirement
measured over the course of the phase-in period but allows more flexibility between model
years. Specifically, in lieu of meeting the 60/60/80/80/100% phase-in requirement over MYs
2018-2022, EPA proposes allowing a manufacturer to meet a phase-in that adds up to a value of
1040 or greater using the formula 5 x 2018MY% + 4 x 2019MY% + 3 x 2020MY% + 2 x
2021MY% + 1 x 2022MY%. In addition to providing manufacturers added flexibility between
model years, this approach also provides an incentive to over-achieve relative to the normal
percentage phase-in requirement in the early years of the phase-in. We support the availability of
this alternative phase-in approach.
EPA also asks for comment on whether this alternative phase-in approach should include the
2017 model year, based on a 2017 percentage requirement of either 40% or 20%. We support
including MY 2017 in the alternative phase-in approach based on a MY 2017 requirement of
20%. As EPA notes in the preamble, the above equation would be modified as follows to
accommodate this change: 6 x 2017MY% + 5 x 2018MY% + 4 x 2019MY% + 3 x 2020MY% +
2 x 2021MY% + 1 x 2022MY% must be greater than or equal to 1160. We believe this
alternative phase-in approach would give auto manufacturers the flexibility they need to ramp-up
Tier 3 evaporative emission systems to a nationwide level across 100% of their models by MY
2022.
Recommendation: We support inclusion of an alternate phase-in, as well as an alternate phase-in
that includes MY 2017.
Our Response:
EPA is incorporating an alternative phase-in scheme which includes the 2017 model year.
As stated by the commenter, we proposed the alternative phase-in percentage approach and are
finalizing it including the option to include the 2017 model year for the 20 percent option.
However, for the 20 percent option in 2017 under the alternative phase-in scheme, the vehicles
must meet the hot soak plus diurnal, leak standard, and the OBD 0.020" evaporative system leak
monitoring requirements. Over compliance with the 20 percent hot soak plus diurnal requirement
could earn "points" toward the required total in the alternative phase-in scheme.
4-164
-------
Tier 3 Summary and Analysis of Comments
EPA is also including the 40 percent of passenger cars and light trucks < 6000 Ibs
GVWR option in the alternative phase-in percentage approach if a manufacturer elects that
option for the 2017 model year. In either the 20 or 40 percent options, compliant vehicles can be
from any class covered by the Tier 3 evaporative emission standards, but the number of vehicles
is based on the percentage (i.e., 20% or 40%) of passenger cars and light trucks < 6000 Ibs
GVWR excluding sales in California and the section 177 states.
What Commenters Said:
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
End of year validation for non-ABT (FR29887):
EPA indicates the following requirement regarding the Tier 3 Evap. Phase-in: "At the end of the
model year they would be expected to show that the percentages were met and if not they would
either use additional allowances or bring more vehicle families into the calculation."
EPA indicates that the end of year validation with actual production volumes is necessary
because they allow for fleet averaging.
We recommend EPA clarify that end of year validation with actual production volumes is not
required if the manufacturer does not participate in the evaporative emissions ABT program (i.e.,
the manufacturer does not utilize FELs or credits/allowances).
Our Response:
We do not concur with the commenter. The end of year validation with actual production
volumes is required to demonstrate that the required sales percentages were met. This is true
whether or not the manufacturer participated in ABT during the phase-in.
What Commenters Said:
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
Final Tier 3 Cert (FR29871): "...100 percent of vehicles would need to meet the all Tier 3
requirements and would be considered 'Final Tier 3' vehicles."
We assume this is in reference to "Final Tier 3" exhaust and not final Tier 3 evap, which will not
be at 100% until 2022.
Our Response:
4-165
-------
Chapter 4: Proposed Vehicle Emissions Program
The commenter's interpretation of the NPRM preamble text is correct. "Final Tier 3" has
no meaning in the regulatory context for evaporative emissions, but the preamble text here was
referring to the exhaust emission standard program requirements. Evaporative emissions
requirements are fully phased-in in the 2022 model year.
What Commenters Said:
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
Credit Trading (FR29889):
Credit trading should be allowed between OEMs. Credit trading between OEMs will not reduce
the overall stringency of the Tier 3 program.
Our Response:
We agree with the commenter and have included credit trading between manufacturers in
the final program.
What Commenters Said:
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
LEV 3 Option 2 (FR29885, FR29886, et.al): Vehicles certified to evaporative emission LEV 3
option 1 provisions are accepted as compliant to the Tier 3 requirements and are included in the
percentage calculations. Vehicles certified to the LEV 3 option 2 provisions should be included
as well.
Our Response:
We agree with the commenter that vehicles certified to the LEV 3 option 2 provisions are
considered compliant and have clarified the description in the preamble (see Section IV.C.l.d) to
the final rule of the Tier 3 evaporative emissions requirements for vehicles in various categories.
4.3.4. Technological Feasibility
What Commenters Said:
Organization: Michigan Department of Environmental Quality (MDEQ)
The MDEQ, Air Quality Division supports the more stringent evaporative, leak and refueling
emissions standards, as well as the new test procedures described in the proposed rule. Control
of these significant sources of volatile organic compounds, particulate matter, and toxic air
contaminants have been demonstrated to be cost effective and attainable. California appears to
4-166
-------
Tier 3 Summary and Analysis of Comments
have already implemented a program with many similar requirements, showing that the limits
can be met.
Organization: Manufacturers of Emission Controls Association (MECA)
MECA agrees with EPA staffs assessment that achieving the proposed Tier 3 exhaust and
evaporative emission standards and associated emission reductions are both technically feasible
and cost-effective.
Advanced evaporative emission technologies (including advanced carbon canisters and air intake
hydrocarbon adsorbers) that are available to meet Tier 3/LEV III evaporative emission standards
for light-duty or medium-duty gasoline or flex-fuel vehicles are discussed in the MECA report:
"Evaporative Emission Control Technologies for Light-Duty Gasoline Vehicles" (available on
MECA's website, www.meca.org, under Resources » Reports). MECA worked closely with
ARB in developing the LEV III canister bleed emission testing protocol (also included in EPA's
Tier 3 proposal) that provides a cost-effective means of defining the bleed emission performance
characteristics of carbon canisters used on light-duty or medium-duty gasoline vehicles.
Including this canister bleed emission test procedure in the Tier 3/LEV III requirements ensures
that vehicle manufacturers have to meet a minimum canister performance level in complying
with the "zero" evaporative standards that California first put in place for PZEV-certified
vehicles. As on the exhaust side, the millions of PZEV-certified vehicles operating on U.S.
highways today form a solid evaporative technology base that can be extended to all future Tier
3/LEV III light-duty and medium-duty vehicles.
Our Response:
EPA agrees with the commenters. The canister bleed standard helps to ensure that vehicle
evaporative emissions are near zero. As discussed in Chapter 1 of the RIA there are many
technology approaches which could be used to reduce evaporative emissions and there are over
50 models certified to PZEV zero evap requirements for the 2013 model year.
What Commenters Said:
Organization: Ferrari
Ferrari considers the proposed evaporative emission and PM standards feasible in 2021 MY. We
therefore support this aspect of EPA's proposed rule.
Organization: Northeast States for Coordinated Air Use Management (NESCAUM)
I'd like to quickly mention two other provisions that would contribute to improved air quality
and reduce public exposure to toxic contaminants in gasoline: reducing evaporative emissions to
near zero levels from all affected vehicles as is currently the case with hundreds of thousands of
California certified vehicles on the road in the northeast.
4-167
-------
Chapter 4: Proposed Vehicle Emissions Program
Organization: Sierra Club
Further, EPA demonstrates that there are a number of technologies in use today that can be
integrated into motor vehicles to fully comply with the proposed Tier 3 evaporative emissions
standards.
Our Response:
We acknowledge the support of the commenters for this aspect of the proposal.
4.3.5. Evaporative Emission Requirements for Heavy-duty Vehicles (HDV) Requirements
What Commenters Said:
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
Chassis-Certified HDV Evaporative Emission Test Procedures: For chassis certified HDV
evaporative emission test procedures, industry recommends that EPA reference the light-duty
evaporative emissions test procedures and remove the HDV test procedures from the heavy-duty
subpart. This would simplify and streamline the regulations and prevent possible inconsistencies
between the light- and heavy-duty subparts.
Our Response:
EPA concurs with this suggestion and we are revising the regulatory text to reference the
light-duty evaporative emissions test procedures and removing the HDV test procedures from the
heavy-duty subpart. It will also help to facilitate the introduction of ORVR for HDGVs over
10,000 IbsGVWR.
What Commenters Said:
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
HDGV Cert (FR29895): OEMs should be allowed to certify HDGV >14K GVWR to Tier 3
(voluntarily) and get credits (allowances). OEMs should also be allowed to optionally group
>14K HDGV with 10 - 14K GVWR vehicles for certification.
Our Response:
In the NPRM EPA asked for comment on allowing voluntary certification to Tier 3
standards for HDGV>14,000 Ibs GVWR if we did not adopt specific requirements. Tier 3
includes revised evaporative standards for these vehicles, as discussed in the preamble and
4-168
-------
Tier 3 Summary and Analysis of Comments
further in Chapter 4.4 of this summary and analysis of comments, so the comment on that point
is moot. We agree that HDGVs >14,000 Ibs GVWR can be included with those between 10,000
and _14,000 Ibs GVWR for certification purposes, such as in optional chassis tests, but they must
meet all requirements related to vehicles less than 14,000 Ibs GVWR vehicles such as the leak
standard and OBD requirements.
4.3.6. HDGV Certification Flexibility
What Commenters Said:
Organization: Truck and Engine Manufacturers Association (EMA)
Use of Engineering Analysis for Vehicles Above 14,000 Pounds GVWR:
EPA is proposing to permit manufacturers to demonstrate compliance to evaporative emission
standards using engineering analysis in lieu of direct testing and supplying test results for
vehicles above 14,000 Ibs. GVWR. (78 Fed. Reg. at 29895; proposed 40 CFR 1037.103(c), 78
Fed. Reg. at 30098.) This is an important change from the current allowance, which starts for
vehicles with GVWR above 26,000 Ibs. In its proposal, EPA points out that this is the same cut
point allowed by ARB and would allow one certification method. EPA requests comments on
this proposal.
If EPA determines not to exclude HDGVs over 14,000 Ibs. GVWR from the evaporative
emission requirements (see above discussion), then EMA supports this proposal. As EPA points
out, the proposed change would provide consistency with ARB, and would allow one test
method, which is of paramount importance to EMA and its members. In addition, few facilities
exist to test evaporative emissions for those large vehicles. As a result, such an allowance would
provide needed compliance demonstration flexibility. EMA recommends that EPA add the words
"above 14,000 Ibs. GVWR" after the first use of the word "vehicle" in the opening sentence of
paragraph 1037.103(c) to more clearly communicate that engineering analysis is available for
vehicles starting with GVWRs above 14,000 Ibs.
- To the extent evaporative standards are in place for vehicles over 14,000 Ibs GVWR, allow the
use of engineering analysis in place of test data for evaporative emissions compliance
Our Response:
EPA acknowledges the conditional support of the commenter and is incorporating the
proposed change to 40 CFR 1037.103(c) in the final rule.
What Commenters Said:
Organization: California Air Resources Board (CARB)
4-169
-------
Chapter 4: Proposed Vehicle Emissions Program
Comment 7 - Evaporative certification method for heavy-duty vehicles over, 14,000 Ibs. Gross
Vehicle Weight Rating: The NPRM requests comment on a proposal to modify existing
certification requirements for heavy-duty vehicles over 14,000 Ibs. gross vehicle weight rating so
that evaporative certifications would rely solely on design parameters and engineering analysis
instead of emissions testing. LEV III test procedures require engineering analysis and data to
certify vehicles in this weight category for evaporative emissions, but do not specifically
preclude emission testing as U.S. EPA is suggesting. Because some auto manufacturers may
prefer to do actual testing, CARB recommends that Tier 3 include actual testing as a means of
meeting the evaporative emission requirement. Nonetheless, CARB anticipates providing
reciprocity to federal evaporative emission certification for these vehicles once Tier 3 is
finalized.
Organization: Truck and Engine Manufacturers Association (EMA)
Further Reliance on Design Parameters and Engineering Analysis:
EPA also requests comments on taking an additional step to rely on design and engineering
analysis. (78 Fed. Reg. at 29895.) Under this approach, manufacturers would be allowed to
demonstrate that the design of their purge strategy, canister capacity, and overall control system
would control emissions to the same degree as similar (or comparable) Class 2b or Class 3
vehicles that meet emission standards when tested over the established measurement procedures.
This additional compliance demonstration flexibility - specifically, being able to reference
evaporative emission control strategies used on similar or comparable Class 2b or Class 3
vehicles to demonstrate compliance with a performance-based standard - clearly is needed for
large vehicles. Accordingly, EMA supports EPA taking such additional steps. It should be noted,
however, that EMA's support of this proposal should not be taken in any way as diminution of
EMA's support for performance-based emission standards, of which EMA has been a long-time
advocate.
- Provide additional compliance flexibility through the use of design analysis for evaporative
system compliance to a performance standard
Our Response:
EPA agrees with CARB. Testing using designated test procedures and test fuels should
not be precluded as a means to show compliance with the emission standards. . However, For
HDGVs over 14, 000 Ibs GVWR, certification based on comparison of design parameters and
engineering analysis relative to certified configurations is permissible. Of course, the results of
any application for certification are subject to EPA approval.
With regard to EMA's comments, EPA solicited comment on the idea of a design- based
certification approach for HDGVs over 14,000 Ibs GVWR. EPA acknowledges the conditional
support of this commenter, but has decided to not to finalize such a provision. We believe the
use of engineering analysis and data from other HDGVs as prescribed in 40 CFR 86.1037(c)
provides an equivalent flexibility for certification without precluding the use of testing for
determining compliance.
4-170
-------
Tier 3 Summary and Analysis of Comments
What Commenters Said:
Organization: Truck and Engine Manufacturers Association (EMA)
More Descriptive Provisions Related to Using Engineering Analysis: EPA proposes to add more
descriptive provisions relating to the use of engineering analysis to demonstrate compliance with
evaporative and refueling emissions standards. (78 Fed. Reg. at 29895 and 30098; proposed 40
CFR 1037.103(c)). The more descriptive language, and the addition of an example related to
evaporative emissions from certain fuel system components that is set forth in the regulatory
language, provide manufacturers helpful guidance as to EPA's expectations. EMA recommends
that the proposed language be finalized.
- Adopt the proposed descriptive provisions related to the use of engineering analysis to
demonstrate compliance with evaporative and refueling emissions standards.
Our Response:
EPA acknowledges the support of the commenter and is finalizing the proposed
language.
What Commenters Said:
Organization: California Air Resources Board (CARB)
Comment 10 - Evaporative option to use design based certification on gaseous-fueled vehicles:
Subsequent to adopting evaporative emission standards for gaseous-fueled vehicles, U.S. EPA
developed designed-based certification procedures for other sources (i.e., for marine vessels
pursuant to 40 Code of Federal Regulation (CFR) 1060.240). These designed-based certification
procedures would minimize or eliminate certification emission testing. The NPRM requests
comment on changing the certification requirements for gaseous-fueled vehicles to a design-
based approach, which U.S. EPA contends would allow for a simpler assessment when certifying
these vehicles. CARB does not support this approach and believes that testing vehicles is still the
most effective way to ensure system integrity and evaporative emission control. Whole vehicle
testing also ensures that the evaporative system as installed in the vehicle, rather than its
individual components, is properly evaluated such that the evaporative emission targets are met.
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
CNG/LPG Design Based Cert (FR29907): Industry supports a design based certification
approach for gaseous fueled vehicles. The evaporative emissions from metallic high pressure
gaseous fuel systems are very low.
4-171
-------
Chapter 4: Proposed Vehicle Emissions Program
Our Response:
EPA agrees that whole vehicle testing is preferable and would not preclude
manufacturers from doing so. The test procedures and standards in the CFR still apply to these
vehicles and could be used by EPA even with this flexibility for design-based certification
approaches. With regard to gaseous fueled vehicles we do not think that requiring emission
testing using current test procedures is necessary and meeting design standards is sufficient to
conclude that the vehicles are meeting emission standards. The fuel storage system designs for
these vehicles are greatly different than for gasoline-fueled vehicles and they do not use the same
emission control strategies as used on a gasoline-fueled vehicle. The test procedures are not as
well suited to evaluate their performance since some aspects of the test procedure are intended to
evaluate the characteristics and performance of control system approaches which are not used on
gaseous fueled vehicles.
4.3.7. Evaporative Emission Requirements for FFVs
What Commenters Said:
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
FFV Test Fuel (FR29884): Footnote 277 indicates that splash blended fuel should be used for
FFV evap testing for Tier 3 certification. To simplify the regulations and reduce the number of
required test fuels Tier 2 and Tier 3 certified FFVs should be allowed to optionally conduct evap
testing with the new Tier 3 test fuel. Data carryover on splash blended Tier 2 test fuel should be
allowed for FFVs certified to Tier 2 regulations.
Organization: General Motors LLC (GM)
Additionally, the use of splash blended 10 psi test fuel for flex-fuel vehicle (FFV) evaporative
emission testing, as proposed by EPA, represents an evaporative emission stringency increase for
FFVs complying with the Tier 3 requirements. This would serve as a roadblock for
manufacturers to designing and offering FFVs and would also undermine the harmonization with
California's LEV III program.
GM recommends that EPA adopt a 9 psi Tier 3 test fuel, and that this same fuel be used for both
conventional and FFV evaporative emission testing. This would facilitate the timely adoption of
the Tier 3 regulations and preserve the harmonization with CARB and the LEV III regulations.
Our Response:
We have considered the comments on the appropriate test fuel for FFV evaporative
emission testing, taking into account the vapor pressure of Tier 3 E10 test fuel. EPA has decided
to use the same evaporative emissions test fuel for conventional vehicles and FFVs. This would
provide harmonization with LEV III and Tier 3 evaporative emission requirements for these
4-172
-------
Tier 3 Summary and Analysis of Comments
vehicles. More detail on our approach to certification fuel for evaporative and refueling
emissions testing for FFVs can be found in Section IV.C.5 of the preamble. For our response to
comments on test fuel for non-FFV evaporative emissions testing, see Chapters 4.5.1 and 4.5.2
of this Summary and Analysis of comments.
4.3.8. Test Procedures
Vehicle Preconditioning before the Hot Soak plus Diurnal SHED Test
What Commenters Said:
Organization: Volkswagen Group of America, Inc.
Among the exclusive brands in the Volkswagen group are a group of vehicles that are treated
exceptionally by developers and owners. These exotic vehicles are still subject to the some
emission standards regardless of the production volume expected. Therefore these vehicles are
rare and require special handling to account for vehicle background emissions that may not
decay us in other models. VW requests exceptions for specialty vehicles in the evap vehicle
preparation sections of Tier 3. Tires may have to be conditioned even after 12 months, as well as
the vehicles themselves. A chassis may be several years old with current fuel system components
removed and installed. This process may contaminate areas of the vehicle that are only 'cleaned'
by baking off residual hydrocarbon. This should be allowed for all of Tier 3, and requires
modification to section 1066.814(e).
Our Response:
EPA agrees that there are some unique vehicle models that may require special handling
in vehicle preparation prior to evaporative emission SHED testing for certification. This may be
the case for specialty vehicles such as when a chassis used in one year's certification is
"reconditioned" for re-use in a subsequent year's certification. In these circumstances, a
manufacturer may seek EPA pre-approval for revised vehicle preconditioning measures.
What Commenters Said:
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
Baking tires and to remove contamination (FR29898)
EPA indicates that no preconditioning shall be permitted for any vehicle aged twelve months
from the date of manufacture.
We recommend EPA revise this limitation to address the following contingencies:
4-173
-------
Chapter 4: Proposed Vehicle Emissions Program
1) If a manufacturer determines that a vehicle with >12 months has been contaminated (e.g., fuel
spill, refrigerant leak, washer fluid leak, etc) then baking shall be allowed to clean up the
contamination;
Our Response:
We do not agree with the commenter's request. Spot cleaning of confirmed spills or leaks
may be requested and approved on a case-by-case basis before evaporative emission testing of
the vehicle. However, baking of the entire vehicle would create the potential for the removal of
non-fuel hydrocarbons from the test vehicle which should otherwise be accounted for in the
vehicle emission certification level.
2) If a manufacturer needs to install tires with less than 12 months from date of manufacture (as
indicated by the tire manufacture date on tire sidewall) on a >12 month vehicle, baking of the
tires alone shall be permitted prior to installation on the vehicle.
Our Response:
EPA generally concurs with this comment, but this allowance must be requested and
approved on a case-by-case basis before evaporative emission testing of the vehicle.
What Commenters Said:
Organization: California Air Resources Board (CARB)
Comment 6 - Evaporative emission testing spare tire removal allowance: The proposed Tier 3
regulatory language (§1066.814) contains an allowance to remove the spare tire during
evaporative emission testing. This provision conflicts with the text in the NPRM that only allows
exchanging the vehicle's spare tire with one that has been aged. CARB supports the proposed
allowance to exchange the spare tire with an aged one, but does not agree with allowing spare
tire removal. Such an allowance would reduce the stringency of the whole vehicle emission
standards, which already account for background emissions, including those from the spare tire.
Our Response:
EPA concurs with this comment. The final rule provides that the spare tire may be
exchanged for a used tire or baked but not removed. This is because non-fuel hydrocarbon
emissions related to the spare tire were considered when the Tier 3 hot soak plus diurnal
evaporative emission standards were developed.
SHED FID Ethanol Adjustment Factor
What Commenters Said:
4-174
-------
Tier 3 Summary and Analysis of Comments
Organization: Volkswagen Group of America, Inc.
1. The direct measurement of criteria pollutants has been widely regarded as the technical
pathway of choice in complying with regulations. Therefore, regulating the use of a FID factor
instead of allowing direct measurement is alarmingly contrary to this philosophy. Allowing the
use of photo acoustic or impinger methods, used to measure precisely the amount of alcohols in
the SHED, provides a robust, direct measurement of fuel-based evaporative emissions. Providing
a FID factor, which may be arbitrary, also will effectively increase the stringency of the
standards. Manufacturers will have to account for this factor in their design criteria to minimize
the confirmatory and In-Use testing risk of the Agency's desire to use a FID factor. CARB
provides an optional factor, 1.08, which is clearly not harmonized with EPA. Analysis of real
vehicle data shows this factor to be smaller than this number. VW requests that a test program be
developed and executed prior to any factor being introduced in the regulation. This program
would examine this factor using several OEM and Agency labs, as well as vehicles of differing
technologies. We look forward to further discussions in this area.
Organization: General Motors LLC (GM)
2. Evaporative Emission Testing Ethanol Adjustment Factor
In the Tier 3 NPRM, EPA proposes the use of a 10% mass adjustment factor to account for
alcohol based emissions during evaporative emission SHED testing. In the current proposal, EPA
specifies the use of a flame ionization detection analyzer coupled with the use of the adjustment
factor as the only viable option to account for alcohol emissions. GM believes that the 10%
adjustment factor is far greater than the true magnitude of alcohol based emissions. OEMs
should be afforded the opportunity to utilize analyzers (i.e., photo acoustic) that measure the
alcohol contribution during certification and in-use evaporative emission SHED testing.
Furthermore, GM believes that certification confirmatory and in-use evaporative emission SHED
testing should be conducted using the same test methodology that was used for the initial
certification testing. Regarding an adjustment factor, while GM supports the availability of a
factor as an option, it should be based on a more accurate assessment of the impact of alcohol
emissions on total evaporative emissions. Therefore, GM recommends that EPA specify the
actual measurement of alcohol emissions as the baseline for evaporative emission testing, and
allow use of an adjustment factor as an option. However, EPA should review the available
industry data to determine a more representative adjustment factor value.
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
3. Ethanol Adjustment Factor - Photoacoustic or Impinger
Total hydrocarbon evaporative emissions are measured in a SHED using a flame ionization
detector (FID). To measure ethanol evaporative emissions, a photo acoustic analyzer (e.g.,
INNOVA™) or a midget impinger sampler with gas chromatograph analysis (hereafter,
impinger) may be implemented. The FID is less responsive to ethanol emissions than to gasoline
emissions, requiring that a "response factor" be determined for ethanol. To correctly measure
4-175
-------
Chapter 4: Proposed Vehicle Emissions Program
total evaporative emissions with a fuel that contains ethanol, the photo acoustic or impinger
methods are required to measure the ethanol evaporative emissions independently, and then
subtract the ethanol contribution from the FID measurement (with the ethanol response factor
applied). Finally, the FID measurement, after correction for ethanol response, is added to the
photo acoustic or impinger measurement for ethanol to determine the total evaporative
emissions.
In the preamble to the Tier 3 NPRM, EPA proposes to use an ethanol adjustment factor of 1.10
for evaporative emission testing conducted using the proposed El 5 certification fuel.
Manufacturers would use a FID and multiply the measured results by 1.10 to obtain the
certification or in-use value. EPA admits that the 1.10 ethanol adjustment factor is based on
worst-case El5 data. The LEV III regulations allow manufacturers to measure actual emissions
using impingers or photo acoustic analyzers, or an adjustment factor of 1.08 (based on the LEV
III E10 test fuel).
We would also note that our data suggest that the ethanol adjustment factor should be
significantly lower than the proposed 1.10 for El5 or 1.08 for El0. Moreover, the factor is likely
to change based on the test being conducted. Industry data suggest that for the two- and three-
day diurnal tests the adjustment factor should be about 1.03; and no adjustment is necessary for
the other tests (spit back, BETP, ORVR, and running loss). (See attached Ford FFV Evap and
ORVR Ethanol Adjustment Data)
The proposed Tier 3 regulations state that the adjustment factor would apply to all evaporative
emission tests - hot soak plus diurnal, refueling, canister bleed, and spit back.
For clarity, the ethanol adjustment factor should not be applied to the spit back emissions test
since this is mass-based and does not use a FID. Additionally, the SUED ethanol adjustment
should not be applied to the canister bleed emissions test procedure (BETP) for the reasons
outlined in the LEV III rulemaking (see Section 12.9, "Hydrocarbon Mass Determination. There
is no requirement to separately measure for alcohol emissions in this bleed emission test.")
While the use of a FID would streamline the test procedures, the Tier 3 evaporative emissions
standards are too stringent to design evaporative systems that provide headroom for both the
standard and another 10 percent for the ethanol adjustment factor. Consequently, manufacturers
need the option to test the actual emissions (using either a photo acoustic or impinger) rather than
using a FID with an adjustment factor.
Recommendation: We recommend that EPA harmonize with ARB requirements allowing the
measurement of actual emissions using photo acoustic or impingers for all emission
measurements. If EPA wanted to streamline its own procedures, it could test using FIDs and an
ethanol adjustment factor: vehicles that pass in this way would pass overall, but compliance
failures would need to be verified by measuring actual emissions.
In addition, EPA and industry should collaborate to determine and adopt a more representative
ethanol adjustment factor.
4-176
-------
Tier 3 Summary and Analysis of Comments
IUVP & EtOH Adj (FR29899)
4. For IUVP testing use an ethanol adjustment factor. Industry is opposed to the use of the
adjustment factor to account for alcohol during IUVP evaporative emission testing. EPA should
provide the ability for manufacturers to determine OMHCE results for evap, running loss and
ORVR using impingers or Innova in place of the 1.10 factor (as we do today for ethanol FFV
programs which test evap, running loss and ORVR with ethanol based fuels). This allowance is
necessary because of concerns that the 1.10 factor provided by EPA is too high.
Our Response:
The commenters generally acknowledge that a correction for the FID ethanol response is
needed. EPA agrees with the commenters that they should have the option to measure and
correct the FID measurement or use a fixed correction factor. However, to prevent arbitrary and
inconsistent results by using the lower of the correction values on a test-by test basis, the final
rule stipulates that all testing of any vehicle using the certification evaporative emissions data
must be based on the approach used for certification testing. This applies to EPA confirmatory
testing as well as all IUVP and IUCP testing. Also, if the data is approved for carry across for
certification of other evaporative families or carryover for a subsequent model year's
certification the same rules regarding which method will be used to adjust the FID data will
apply. The commenters also presented the view that EPA's proposed 1.10 factor for El5 was too
large based on their data. Although not conducting a test program as requested by VWGoA,
EPA reviewed the data provided by the Alliance and Global in their comments and supplemental
information provided by GM. Presuming the use of a test fuel with E10, EPA also reviewed the
test data provided by CARB to support their adoption of the 1.08 value. Taking the information
in the datasets together, we believe CARB's 1.08 value to be conservative but reasonable for the
intended purpose. Four of the fifteen values calculated by CARB were 1.07 or larger.19 The data
provided by GM showed seven of twenty points were 1.07 or larger. All of the data provided by
Ford was less than 1.07.20 The correction factor depends on the vehicle and FID response factor
and to some degree the FID instrument used (analog or digital) which may explain some of the
differences.
Furthermore, EPA considered the manufacturers' comments that the correction should
only apply to the hot soak plus diurnal measurement. EPA agrees that the correction should
apply to the hot soak and diurnal measurements and is not needed for the canister bleed,
refueling, or spit back measurements. There was no data for the SHED rig test provided by the
commenters and the running loss data showed some need for correction. Thus, to align with
CARB, we are applying the requirement for measurement correction to the running loss and
SHED rig test results, in addition to those from the hot soak and diurnal test.
SHED Rig Test
19
20Passavant, G. (2013, October). Manufacturer Data on Ethanol Measurements in the SHED. Memorandum to the
docket.
4-177
-------
Chapter 4: Proposed Vehicle Emissions Program
What Commenters Said:
Organization: Volkswagen Group of America, Inc.
Evaporative emissions rules are meant to control fuel-based emissions. The only way to truly
measure fuel-based emissions without the non-fuel emissions confounding the result is to have a
rig. Having a rig-based certification option will eliminate any tedious evap vehicle setup, such as
vehicle background and tire conditioning. Methods can be developed to measure rig-type
emissions In-Use, easing the Agency fears of complex and burdensome testing. VWGoA
specifically requests EPA to continue to optionally accept the ARB PZEV rig procedure until
2025MY. W/ also specifically requests a mid-term review of the Tier 3 evap regulations and
procedures to be completed before the expiration of the rig option currently proposed for 2019.
Our Response: We do not agree with the commenter. The major shortcoming of the rig test is
that it cannot be conducted in any confirmatory testing or in IUVP testing without disassembling
the fuel/evaporative system to remove it from the vehicle. There would be no way to ascertain
the representativeness of the test results relative to what would have occurred when the
fuel/evaporative control system was installed in the vehicle. In response to the request for more
time to use the SHED rig test option, EPA has extended the time we will accept SHED rig tests
results from California LEV III option 1 certifications through the 2021 model year.
Test conditions
What Commenters Said:
Organization: Revecorp Inc.
Second, the proposed rule leaves evaporative emissions testing temperatures the same and the
proposed rule even is considering reducing the temperature for the 3-day running loss test.
Revecorp believes that instead of reducing the upper temperature range limit, it should be
increased to be more real-world. Ozone is a problem during high temperatures, so the propose
rule should test vehicles under higher temperature conditions such as at least 100 degrees F.
Revecorp recommends - All testing should be conducted using fuel which is more similar to
industry average in-use fuel, evaporative emission tests should be conducted over a wider, higher
temperature range.... As noted above, all of these changes and test requirements should be
demonstrated at high altitude.
Our Response:
EPA is not removing the 3-day running loss test from the Federal requirements for either
low or high-altitude testing conditions. EPA did not propose to increase the high end of the
temperature cycle and does not believe it is appropriate to take such actions at this time in the
absence of prior proposal.
4-178
-------
Tier 3 Summary and Analysis of Comments
Multi-day diurnals
What Commenters Said:
Organization: Revecorp Inc.
Third, if the proposed rule is to achieve larger evaporative emissions reductions, vehicles should
be designed to capture evaporative emissions for more than two or three days.
Modeling studies such as those conducted by Sam Reddy (presented at the CRC conference in
2011) indicate that significant evaporative emissions come from vehicles sitting for extended
periods of time such as on new and used car lots for sale and at long term parking locations such
as airports. Emissions from these conditions could be reduced significantly by designing
evaporative emissions control systems to be able to contain more evaporative emissions. If
testing was required to prove that evaporative canisters do not "break through" in less than four
days, significant reductions in evaporative emissions could be gained. In addition, requiring
manufacturers to perform the test out to five days and providing EPA hour by hour
measurements of evaporative emissions (60 hour observations) would give EPA information
about vehicles evaporative emission control system and expected in-use durability/performance.
The cost to for vehicle manufacturers to make this change would be relatively small (increasing
canister volume). Revecorp believes this is a better solution than the requirement to add a
supplemental "scrubber" canister, and would achieve the same result or better - at a lower cost.
Revecorp recommends - ... evaporative emissions test should be conducted for five days with
hourly data provided to EPA and the standard being no breakthrough until after 48 hours.
Our Response:
Data available to EPA indicates that some Tier 2 vehicles emit fuel vapors after the third
day of the diurnal. However, many went for several days beyond the 3-day period before
emissions occurred21 The impact of these emissions on the inventory depends on the fraction of
vehicles which are parked for more than three days, the diurnal temperature conditions when
they are parked, the state of vapor load on the canister before it was parked, and the fuel vapor
pressure. Canisters do not breakthrough to an uncontrolled level because diurnal natural back
purge (20-25% efficiency per day) during cooling allows them to reach a steady-state condition.
A larger canister volume with more total gasoline working capacity would help to reduce multi-
day diurnal emissions assuming it was well purged before the extended park. However, this is
not as straightforward as it sounds because the current test procedure mandates that the canister
be loaded to breakthrough before the evaporative emissions test sequence begins. Normally,
purging a canister with a larger volume of carbon or carbon with higher butane working capacity
carbon takes a greater air volume to purge an equivalent amount of vapor from the carbon. In
some cases the additional purge air volume may not be readily available on small to mid size
21 Lindner,!., Sabisch,M, Glinsky,G.,Stewart,J.,St.Denis,M.,Roeschen,J., (2013) Multi-Day Diurnal Testing, ERG
Contract Report to US EPA.
4-179
-------
Chapter 4: Proposed Vehicle Emissions Program
vehicles/engines. Conversely, a scrubber is a low working capacity carbon which loads easily
and purges quickly. The activated carbon canister is at the heart of the vehicles
evaporative/refueling control system and directly impacts compliance for the SHED rig, canister
bleed, hot soak plus diurnal, refueling, and running loss standards. Accordingly, we are not
lengthening the duration of the hot soak plus diurnal evaporative emissions tests in this rule. The
EPA sets test procedures and performance standards to control these emissions. The use of a
larger canister, a carbon with a greater working capacity, or a canister scrubber is a manufacturer
decision based on consideration of cost, technology, and compliance issues.
What Commenters Said:
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
BETP-Fuel/temp profile (FR29896)
EPA indicates that they intend to incorporate the CARB BETP requirements by reference.
If they do so, then they need to add clarification which indicates that Federal fuels/temperatures
can be used in place of the California fuels/temperatures identified in the CARB regulations.
Our Response:
EPA agrees with the commenter and is adding this provision to the regulatory text.
4.3.9. Certification Emission Test Fuels
What Commenters Said:
a. HDGV Test Fuel
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
HDGV Tier 3 Fuel (FR29895)
Industry supports the optional use of Tier 3 test fuel for HDGV testing.
Our Response:
EPA asked for comment on permitting the manufacturers to use Tier 3 test fuel for
evaporative emissions testing if EPA did not adopt the proposed Tier 3 evaporative emision
standards for HDGVs. EPA is adopting the Tier 3 evaporative emission standards for HDGVs so
4-180
-------
Tier 3 Summary and Analysis of Comments
the point is moot. HDGVs certifying to Tier 3 evaporative emission standards can use Tier 3
certification fuel as early as the 2015 model year if they want to earn allowances.
What Commenters Said:
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
HDGV Tier 3 Cert Fuel (FR28985)
If FIDGV evaporative emission or ORVR testing is conducted on Tier 3 fuel, should engine
exhaust testing be required on the Tier 3 fuel? Evaporative and exhaust testing requirements
should be independent in the Tier 3 regulations. Therefore, exhaust testing should not be required
on the same fuel utilized for evaporative or ORVR testing.
Our Response:
Subsequent to the release of the NPRM and receipt of the comments, EPA met with
manufacturers of engine-dynamometer certified heavy-duty gasoline engines and discussed their
comments. These FIDGV engines are not subject to revised exhaust emission standards in the
Tier 3 rule. The manufacturers stated that procuring and storing Tier 2 fuel for exhaust emission
testing for the relatively few HDGE test groups is not preferable long term, and they agreed that
all HDGE test groups should meet the existing exhaust emission standards on Tier 3 fuel by the
2022 model year. Based on the information EPA received from the manufacturers, the final rule
provides that a requirement to use Tier 3 test fuel for exhaust emission testing can phase-in until
the 2022 model year as deemed appropriate by the manufacturers. Also based on feedback from
the manufacturers, the final rule provides that after the 2022 model year manufacturers can
carryover certification on Tier 2 fuel for up to five percent of sales for atypical applications such
a military vehicles.22 Further discussion of these issues appears in the preamble to the final rule.
4.3.10. Application of Evaporative Emission Standards to Fuels Other than Volatile Fuels
What Commenters Said:
Organization: Truck and Engine Manufacturers Association (EMA)
Application of Evaporative Requirements to Gaseous-Fueled Heavy-Duty Vehicles Above
14,000 Pounds GVWR
22 Passavant, G. (September, 2013)."EPA & Ford Meeting on Issues Related to Tier 3 NPPJVT. Memorandum to the
docket.
4-181
-------
Chapter 4: Proposed Vehicle Emissions Program
a. EPA has proposed that Heavy-Duty Vehicles over 14,000 Ibs. GVWR that run on volatile
liquid fuel (such as gasoline or ethanol) or gaseous fuel (such as natural gas or liquid petroleum
gas ("LPG")) meet the evaporative emission standards of 86.008-10 through the 2017 model year
("MY"), and part 86.1813 requirements in MY2018 and beyond. EPA's proposed requirements
violate the CAA's lead time and stability requirements by imposing new standards with less than
four full model years' lead time and at the same time other, already-promulgated, future
standards are set to go into effect, including new GHG emissions requirements in 2017.
b. EPA requests comment on "adjusting the regulations such that evaporative emissions
standards apply only to volatile liquid fuels, which is the approach [EPA has] taken for nonroad
applications (see, for example, 40 CFR 1060.801)." (78 Fed. Reg. at 29899) EMA supports such
an approach and urges EPA to exclude heavy-duty vehicles that run on gaseous fuel from any
evaporative emission requirements.
c. Fuel systems for gaseous fuels such as natural gas (other than LNG) and LPG are not designed
or intended to vent to the atmosphere and, therefore, do not need to, nor should they, be subject
to evaporative standards. LNG systems are designed to vent to the atmosphere, but also are
designed to be able to go substantial periods of time without the need to vent. Moreover, owners
and operators of LNG-fueled vehicles have substantial incentives to minimize consumption of
fuel - a very valuable commodity. For example, managers of fleets with LNG fueled vehicles
monitor LNG fuel systems and vehicle duty cycles to ensure that no fuel is vented, in order to
prevent fuel waste.
EMA urges EPA to exclude heavy-duty vehicles that run on gaseous fuel from any evaporative
emission requirements. At a minimum, EPA should conduct an appropriate cost-benefit analysis
before including natural gas and LPG-fueled vehicles above 14,000 Ibs. GVWR in evaporative
emission standards.
EMA recognizes that its recommendation to exclude gaseous-fueled heavy-duty vehicles from
evaporative emission requirements does not fully harmonize with ARB. If EPA were to conclude
that such requirements were appropriate, EPA, at most, should apply evaporative emission
requirements only to LPG-fueled vehicles, which is consistent with the current California
requirements.
Exclude all gaseous-fueled vehicles over 14,000 Ibs. GVWR from evaporative emission
requirements.
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
Refueling Standards (FR29899)
EPA requests comment on using this rulemaking as the proper context for applying the refueling
standards to vehicles powered by every kind of fuel.
4-182
-------
Tier 3 Summary and Analysis of Comments
We request comment on adjusting the regulations such that evaporative emission standards apply
only to volatile liquid fuels, which is the approach we have taken for nonroad applications.
The Alliance and Global Automakers agree with the EPA proposal to adjust the regulations such
that evaporative emission standards apply only to volatile liquid fuels, as is the approach taken
for non-road applications and has been an allowed approach under Tier 2 regulations. Excluding
diesel fuel from the standards because it is nonvolatile and natural gas and liquefied petroleum
gas because they are not liquid fuels at atmospheric pressure is the right approach and does not
impact evaporative emission stringency and/or controls.
Volatile Liquid Fuels (FR29899)
Industry agrees with EPA that evaporative emission certification should only be required for
vehicles that operate on volatile liquid fuels.
Organization: California Air Resources Board (CARB)
Comment 5 - Evaporative emission requirements for alternative fuels.
The NPRM requests comment on whether the evaporative emission requirements should be
modified in Tier 3 so that the standards only apply to volatile liquid fuels, which would
effectively remove evaporative emission requirements for liquefied petroleum gas and natural
gas. CARB recommends that U.S. EPA retain the current applicability of the evaporative
requirements by continuing to require that liquefied petroleum gas and liquefied natural gas
vehicles certify to evaporative emission standards. CARB currently has an evaporative emission
requirement for liquefied petroleum gas vehicles. Although existing California evaporative
control regulations do not specifically include liquefied natural gas, we believe controlling
evaporative emissions from liquefied natural gas systems is warranted and intend to pursue such
action in the future. Having an evaporative emission requirement for these vehicles assures that
system venting does not occur during normal diurnal conditions, and that other sources of
evaporative emissions are controlled.
Our Response:
EPA proposed that the Tier 3 evaporative emission standards apply to gaseous fueled
heavy-duty vehicles. The existing evaporative emission standards apply to all HDVs as per 40
CFR 86.008-10 and 1816-08 and the proposed standards are contained in 40 CFR 86.1813. EMA
raised concern with regard to lead time. The final rule provides four model years of lead time for
gaseous-fueled vehicles. The applicability of the Tier 3 evaporative emission requirements for
any given gaseous-fueled heavy-duty vehicle evaporative family depends on when it begins
production in the 2019 model year. If it begins MY 2019 production before the fourth
anniversary of the signature date of the Tier 3 final rule, the Tier 3 evaporative emission
requirements would defer to the 2020 model year. If it begins MY 2019 production on or after
the date which is four years after the anniversary date of the signature of the Tier 3 final rule,
then the Tier 3 requirements apply for the 2019 model year. In addition, due to the nature of the
phase-in inherent in the evaporative emissions program, we expect that manufacturers with
4-183
-------
Chapter 4: Proposed Vehicle Emissions Program
diverse product lines will have the lead time necessary to meet the Tier 3 phase-in percentage
requirements while at the same time deferring compliance with the evaporative emission
requirements for gaseous-fueled vehicles until the 2020 model year or later when the percentage
requirement is 80 percent. In this case they could choose to include these vehicles in the
denominator of the percentage calculation in the applicable model year (either 2019 or 2020).
Small manufacturers have until the 2022 model year to meet the Tier 3 evaporative emission
standards.
Existing evaporative emission standards apply to all gaseous fueled vehicles including
natural gas and liquified petroleum gas (LPG) (see 40 CFR 86.008-10 (b)) and 1816-08. EPA
sought comment on applying the evaporative emission standards only to vehicles using volatile
liquid fuels (a fuel that is liquid at atmospheric pressure and has a Reid Vapor Pressure higher
than 2.0 pounds per square inch - gasoline, ethanol, and methanol), but we did not propose it
explicitly. EPA has examined the comments and once again studied the fuel properties and fuel
system characteristics for the vehicles covered by the existing evaporative emission
requirements. Based on this review, we have decided not to change the applicability except to
add that ethanol is covered as a volatile fuel. The requirements are technically feasible and
appropriate for the technologies involved. The Tier 3 evaporative emission standards apply to all
vehicles using volatile fuels (a volatile fuel is a volatile liquid fuel or any fuel that is a gas at
atmospheric pressure — gasoline, methanol, ethanol, natural gas, and LPG are volatile fuels).
EPA has reviewed the evaporative emission requirements relative to the fundamental
elements of the fuel system designs for these vehicles. Based on this review, we are not applying
all the various evaporative and refueling requirements, only those that are appropriate given the
technology. However, we are applying the standards to sealed systems as well as low pressure
systems. Specifically, EPA has decided to continue not to include the 2-day evaporative emission
standards and test procedures for gaseous-fueled vehicles. This is largely because these vehicles
do not use carbon canisters to control evaporative emissions so canister purge does not come into
consideration. Even though the commenters suggested that gaseous fueled vehicles should not
be subject to standards due to the nature of their design, EPA believes the running loss and three
3-day evaporative emissions tests are needed to ensure control of leaks which could arise in these
sealed systems as a result of mirco cracks and poor connections. The behavior and practices of
owners, operators, or fleet managers is not relevant to certification. While the test procedures for
these standards would apply, EPA believes that meeting consensus standards will provide the
appropriate control and is including regulatory provisions to permit manufacturers to certify
based on related data, engineering analysis, and compliance with national consensus standards
such as ANSI, NGV 1-2006. Assuming the use of these compliance flexibilities and elimination
of the 2-day test, compliance costs should be minor.
4.3.11. Refueling Emissions Control and ORVR
What Commenters Said:
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
4-184
-------
Tier 3 Summary and Analysis of Comments
ORVR on Alternate Fueled Vehicles (FR29899)
ORVR requirements should not apply to CNG, LPG, fuel cell or electric vehicles.
Our Response:
For LDVs, section 202(a)(6) requires that refueling standards apply to vehicles of all
fuels. EPA has more discretion for light trucks and heavy-duty vehicles. Existing EPA
regulations apply the refueling emission standards and related requirements to all complete
vehicles up to 10,000 Ibs GVWR. In the NPRM we asked for comment on the future application
of the refueling standards to only volatile liquid fuels (a fuel that is liquid at atmospheric
pressure and has a Reid Vapor Pressure higher than 2.0 pounds per square inch - gasoline.
ethanol, and methanol) or applying them to all volatile fuels (a volatile fuel is a volatile liquid
fuel or any fuel that is a gas at atmospheric pressure — gasoline, methanol, ethanol, natural gas,
and LPG are volatile fuels).
Based on comments and review of the fuel properties and fuel systems EPA is retaining
most of its existing requirements and adding new requirements. First, as discussed further below,
we will no longer require diesel-powered LDTs, MDPVs, or HDVs to certify to the refueling
emission standard. Second, with regard other fuels (not gasoline) EPA is adding requirements
for other volatile fueled vehicles (e.g., gaseous fueled) to meet the refueling emission standard.
EPA agrees in principle that manufacturers and users have incentive to minimize fuel losses.
However, we believe that the potential exists for vapor losses in such systems from micro leaks
during the refueling event and during operation and that without including all volatile fuel
vehicles in the requirement the potential exists for problems to be overlooked either during
design or installation. Thus, for refueling emissions we have decided to maintain the program
structure already applied to evaporative emissions and are applying refueling emission standards
to all complete volatile fuel vehicles including LDTs, MDPVs, and HDVs.
For LDTs up to and including 8,500 Ibs GVWR not already covered by the refueling
emission standard, the standard would apply to the vehicle when it meets the Tier 3 evaporative
emission standards. Since control of evaporative and refueling emissions is technically linked,
we are requiring that when an MDPV or complete HDV less than or equal to 14,000 Ibs GVWR
not already covered by the refueling emission standard meets the evaporative emission standards
its also meets the refueling emission standard.23 As is the case for evaporative emissions for
these alternative-fueled vehicles, the applicability of the Tier 3 refueling emission requirements
for any given family would depend on when it begins production in the 2019 model year. If it
begins production before the fourth anniversary of the signature date of the Tier 3 final rule, the
Tier 3 evaporative emission requirements would defer to the 2020 model year. If it begins
production on or after the date which is four years after the anniversary date of the signature of
the Tier 3 final rule, then the Tier 3 requirements apply for the 2019 model year. The refueling
and evaporative emission standard implementation schedule provisions are structured the same
and vehicles in any evaporative family must meet the Tier 3 evaporative and refueling emission
standards in the same model year.
23 (For vehicles over 8,500 Ibs GVWR the Tier 3 refueling emission standards apply only to complete vehicles. See
the discussion below regarding ORVR for incomplete HDGVs.
4-185
-------
Chapter 4: Proposed Vehicle Emissions Program
However, due to the nature of the phase-in inherent in the evaporative emissions
program, we expect that manufacturers with diverse product lines will have the lead time
necessary to meet the Tier 3 phase-in percentage requirements while at the same time deferring
compliance with the evaporative emission requirements for gaseous-fueled vehicles until the
2020 model year or later when the percentage requirement is 80 percent. In this case they could
choose to include these vehicles in the denominator of the percentage calculation in the
applicable model year (either 2019 or 2020).
For complete HDVs over 14,000 Ibs GVWR, the refueling emission standard does not
apply until the 2022 model year, even if the vehicle meets the evaporative emission standards in
an earlier model year. Small manufacturers have until the 2022 model year to meet the Tier 3
evaporative emission standards
For both LPG and CNG the refueling test procedures and standards apply, but EPA will
deem a system to be in compliance if they certify compliance with relevant national and
international consensus standards (e.g., ANSI and ISO) related to fuel system and fuel system
component integrity and the refueling connector.
What Commenters Said:
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
ORVR Incomplete Class 3 HDGVs (FR29824, FR29894, FR30051, FR30098)
EPA explains that Class 3 (10,001 - 14,000 Ibs GVWR) HDGVs are further sub-categorized into
complete and incomplete vehicles.
We recommend that EPA clarify that the new ORVR requirement is only applicable to complete
vehicles, and excludes all incomplete vehicles >8.5K GVWR.
HDGV Cert (FR29895)
ORVR requirements must exclude incomplete vehicles >8.5K GVWR.
Our Response:
EPA concurs with this comment and we are finalizing the new ORVR requirement only
for complete class 3 heavy-duty vehicles, excluding all incomplete vehicles greater than 8500 Ibs
GVWR. EPA believes that ORVR controls are technically feasible for incomplete HDGVs, but
at this point, the ORVR requirement applies only to complete HDGVs > 8,500 Ibs GVWR
because manufacturers have indicated to EPA that they would have to establish additional
measures to ensure that the steps taken to complete the vehicle by the secondary manufacturer do
not compromise the integrity and safety of the fuel/evaporative control system (including
4-186
-------
Tier 3 Summary and Analysis of Comments
ORVR) and that the ORVR system continues to perform properly with regard to emissions
control. Incomplete vehicles certified to the ORVR requirements may earn ORVR or
evaporative emission standard allowances.
What Commenters Said:
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
ORVR-Reducing gasoline waste on large tanks (FR29895)
For large tanks, the ORVR test generates 50 or more gallons of waste gasoline. Industry would
like to work with EPA to develop methods to reduce this. We recommend adding the following
(or words to this effect) to §86.1816-05:
"For vehicles with fuel tanks larger than 40 gallons, manufacturers may request alternative
ORVR testing to reduce wasted fuel. The request should include data to indicate that testing is
representative of in use refueling."
Our Response:
EPA agrees that an alternative procedure should be available to address this concern.
Most if not all gasoline fuel tanks over 40 gallons are found on HDGVs over 14,000 Ibs GVWR.
In response to this comment, we have included a test procedure approach for ORVR for vehicles
over 14,000 Ibs GVWR which will allow the use of a seal test as currently contained in the
refueling emission test procedure regulations rather than the full SHED test. This gives
manufacturers the opportunity to demonstrate that their test vehicles have the purge strategy and
canister design (e.g., geometry, back pressure, and gasoline working capacity) based on
engineering analysis and comparison to designs used on lighter-weight GVWR HDGVs which
meet the refueling emission standards and have been certified rather than running the full
refueling test procedure.
What Commenters Said:
Organization: Private Citizen
These comments were submitted as testimony at the Philadelphia, PA public hearing on April 24,
2013.
I was also concerned about some of the vapor and evaporation at gas pumps and whether that
could be included in this, or whether in a separate rule. So I was thinking along the lines of
cumulative that you could do, especially when it comes to congressional negotiations that happen
4-187
-------
Chapter 4: Proposed Vehicle Emissions Program
when something gets lowered. Perhaps an all-inclusive thing might get to our goal without
focusing on one small area.
Our Response:
In May of 2012, EPA finalized a rule to permit states which adopted Stage II vapor
recovery requirements to remove such requirements if certain Clean Air Act requirements related
to future compliance with the ozone NAAQSs are met.24 This was adopted because the
Administrator deemed ORVR to be in widespread use and as provided under the Clean Air Act
there was no need to continue redundant control (Stage II and ORVR) of refueling emissions
after ORVR was in widespread use. The Tier 3 rule expands the ORVR requirement further by
adding the requirement that all complete HDGVs over 10,000 Ibs meet the refueling emission
standards. Nonetheless, the analysis prepared by EPA in support of this decision recognized that
full fleet turnover to ORVR would not occur for more than a decade. EPA expects many states to
remove Stage II since the remaining available emission reductions are small relative to the costs
of maintaining the systems. The hydrocarbon reductions from the Tier 3 will help to ensure that
current and future plans for compliance with the ozone NAAQS are not impacted.
What Commenters Said:
Organization: California Air Resources Board (CARB)
Comment 9 - Refueling: Modifying the terms for exemption from the refueling requirement for
diesel vehicles.
CARB does not support U.S. EPA's proposal to withdraw the condition that the fuel tank
temperature not exceed 130°F in order for a diesel vehicle to qualify for an exemption from the
refueling emission requirement absent data showing there would be no emission impact.
Accordingly, the NPRM should cite any studies supporting such an action. While less volatile
than gasoline, diesel fuel will still generate vapors, especially under higher temperatures. If the
original regulatory language is retained, it would likely act to encourage auto manufacturers to
consider heat exposure when designing and locating the fuel tank, which would result in reduced
refueling emissions.
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
ORVR Diesel (FR29899)
If EPA agrees that diesel ORVR is not a concern (for the same reason diesel is not a concern for
Evap.), EPA should exempt diesel vehicles from the ORVR requirement (i.e., eliminate the
standard and requirement for an attestation).
24 See EPA final rule, "Air Quality: Widespread Use for Onboard Refueling Vapor Recovery and Stage II Waiver".
77 FR 28772, May 16, 2012.
4-188
-------
Tier 3 Summary and Analysis of Comments
Organization: Truck and Engine Manufacturers Association (EMA)
1. Application of Refueling Requirements to Diesel-Fueled Vehicles Above 10,000 Pounds
GVWR
EPA proposes in Section 86.1813-17(b)(l) to add a refueling requirement to diesel vehicles
above 10,000 Ibs. GVWR starting with model year 2018. EPA's proposed requirement violates
the CAA's three-year period of stability by imposing new standards near the same time other,
already-promulgated, future standards are set to go into effect, including new GHG emissions
requirements in 2017, and also violates the CAA's four-year lead time requirement.
Moreover, such a requirement for diesel-fueled vehicles is unnecessary and unjustified. Diesel
fuel is a non-volatile fuel, as EPA correctly points out in Section 1037.103(a)(2), and, for this
reason, is excluded from evaporative emissions pursuant to Section 86.016-l(a)(4)(iii).
Furthermore, Section 202(a)(6) of the Clean Air Act, which EPA references to justify the
application of refueling standards to diesel vehicles, specifically applies only to "new light-duty
vehicles." (78 Fed. Reg. at 29899.) As defined in the CAA, light-duty vehicles are those less than
or equal to 6,000 Ibs. GVWR. Thus, diesel-fueled vehicles greater than 6,000 Ibs. GVWR are not
required to be subject to refueling emission standards.
EMA recommends that all diesel applications be excluded from refueling requirements. If EPA
desires to expand refueling requirements to heavy-duty applications over 6,000 Ibs. GVWR, the
Agency must conduct a full assessment and rulemaking to demonstrate why such requirements
should be expanded to diesel vehicles. As the Proposed Rule generally covers vehicles up to
14,000 Ibs. GVWR, in no case should any refueling standard be made applicable to diesel-fueled
vehicles above 14,000 Ibs. GVWR.
- Exclude all diesel applications over 6,000 Ibs. GVWR from refueling requirements
Our Response:
One commenter said that the requirement to include diesels above 6,000 Ibs GVWR in
the program was advisable and others indicated that it was unnecessary. In the NPRM, EPA
indicated that even with turbulent mixing of the dispensed and tank fuel, the very low vapor
pressure of diesel fuel (<0.05 psi) coupled with the vapor shrinkage conditions which would
occur during the refueling test (cooler fuel dispensed into a warmer fuel in the tank which then
condenses vapor in the headspace) make it very unlikely that a diesel vehicle could fail the 0.20
g/gal refueling emission standard. This is reinforced by the simple chemical equilibrium
calculations. Consider for example, Table 7.1.2 of AP-42 which indicates a vapor pressure of
0.022 psi at 100°F for diesel fuel and by interpolation about 9.6 psi for a 9 RVP fuel. A 9 psi
RVP gasoline fuel gives a refueling emission rate of about 5 grams per gallon. So, a simple
calculation 5(0.022/9.6) = .011 g/gal. This is only about 6 percent of the standard. In the
MOVES model used to support the Tier 3 rule we used a zero value for vapor displacement
emissions. Based on these very low values, EPA is removing the requirement for diesel vehicles
4-189
-------
Chapter 4: Proposed Vehicle Emissions Program
to demonstrate or attest to compliance with the refueling emission standard, except for light-duty
vehicles where it is required under the statute.
4.3.12. Other Comments and Issues
What Commenters Said:
Organization: California Air Resources Board (CARB)
Comment 8 - Evaporative emission testing reduced canister loading allowance for pressurized
fuel systems:
The NPRM requests comment on allowing reduced canister loading for a pressurized system.
This load procedure would be proposed by the auto manufacturer to reflect actual diurnal
canister loading using the manufacturer's specific technology. CARB is concerned that this
would reduce the stringency of the test and may result in future vehicles that have inadequate
ability to purge the canister. The current requirement to load a canister to breakthrough is a
critical part of the evaporative emission test, which ensures that enough purge capability is
designed into the vehicle to purge a full canister during an 11-mile FTP drive cycle. While
CARB recognizes that certain engine technologies in the future will have reduced purge
capabilities, there are already a number of strategies available today that could help compensate,
such as heated purge, optimization of purge strategy to maximize use of available engine
vacuum, and purge pumps. Accordingly, CARB recommends that the canister loading procedure
remain as proposed.
Our Response:
The current load to breakthrough requirement for the canister ensures that purge is
adequate to achieve control during in-use operation. Under this requirement, the manufacturers
cannot over size the canister to be sure the test is passed without also providing for adequate
purge to ensure good in-use performance. At least directionally, this requirement creates a
disincentive to increase canister size or total working capacity as these situations often require
more purge air volume to strip the same amount of fuel vapor from the activated carbon in the
canister. EPA agrees that the possibility of lower purge volumes over the FTP might be
compensated for by other technologies but this is not without a cost in dollars or weight. EPA
did not propose a specific change and we received no auto industry comment in support of a
modification. Thus, we are making no change to this requirement for the Tier 3 rule.
Evaporative Emissions Durability Test Fuel
What Commenters Said:
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
4-190
-------
Tier 3 Summary and Analysis of Comments
a. Durability Test Fuel: The Tier 2 certification test gasoline, Indolene, contains no ethanol, yet
the regulations require the gasoline used for evaporative durability mileage accumulation contain
the highest commercially available ethanol concentration allowed by federal law. We understand
the historical need for this provision, since the market fuel and certification fuel contained
significantly different ethanol content. EPA proposes new Tier 3 certification test gasoline
requirements which would require either 15 percent (proposed) or 10 percent (industry
recommended) ethanol. Since permeation evaporative emissions remain relatively consistent
between E10 and El 5, manufacturers should have the option to use either Tier 3 test fuel or fuel
containing the highest commercially available ethanol concentration for mileage accumulation
on evaporative durability programs.
Recommendation: We recommend EPA provide manufacturers the option to use either Tier 3
test fuel or fuel containing the highest commercially available ethanol concentration for mileage
accumulation on evaporative durability programs. This optional allowance could reduce the
burden on manufacturers to maintain separate Tier 3 test fuel and mileage accumulation fuels
without any negative environmental consequences. [EPA-HQ-OAR-2011-0135-4461-A1, p.28]
Evap Dura (FR29897):
Revisions to the durability test fuel are discussed. To reduce the number of required test fuels
EPA should allow the optional use of Tier 3 test fuel for evap durability testing. Evap durability
demonstration testing should not be required to address the differences between the current
durability fuel and the Tier 3 test fuel.
Evap Dura & Market Fuel (FR 29908):
To reduce the number of required test fuels EPA should allow the optional use of Tier 3 test fuel
for evap durability testing. Evap durability demonstration testing should not be required to
address the differences between the current durability fuel and the Tier 3 test fuel.
Also, the language requiring the evap durability service accumulation test fuel to contain the
highest concentration of ethanol commercially available in any state should be modified. The
concentration of ethanol in the evap service accumulation test fuel should be specified to match
the concentration of ethanol in the Tier 3 test fuel.
Organization: Chrysler Group LLC
EPA should clarify the fuel requirement for evaporative emissions durability testing. The
regulations, 40 C.F.R. § 86.1824-08(f), require that the fuel manufacturers select for service
accumulation "contain[] ethanol in, at least, the highest concentration permissible in gasoline
under federal law and that is commercially available in any state in the United States." In the
preamble to and record for the proposed rule, EPA has discussed its interpretation of those
existing requirements, and particularly the meaning of "commercially available." It is clear that
E15, at 0.017 percent of the current market, should not be considered to be "commercially
available" today. EPA should clarify more generally what "commercial availability" means by
objective criteria that industry can readily understand. Therefore, as detailed above, evaporative
4-191
-------
Chapter 4: Proposed Vehicle Emissions Program
emissions durability fuel should represent that which is the predominate fuel in the market, i.e.,
E10.
Recommendation: Chrysler recommends that EPA clarify the meaning of "commercially
available" in the final rule and define it to mean the predominant fuel in the marketplace,
consistent with our recommendation above, and, currently, E10.
Our Response:
EPA did not propose to change the regulations for evaporative emissions durability
service accumulation fuel, which require such fuel to contain "ethanol in, at least, the highest
concentration permissible in gasoline under federal law and that is commercially available in any
state in the United States. Unless otherwise approved by the Administrator, the manufacturer
must determine the appropriate ethanol concentration by selecting the highest legal concentration
commercially available during the calendar year before the one in which the manufacturer begins
its mileage accumulation."(See 40 CFR 86.1824-08(f)). While there are only minor permeation
differences between E10 and E15 for the materials and systems evaluated we believe it is
important that evaporative systems are designed to meet the emission standards over their full
useful life. El5 is the highest ethanol concentration allowed in gasoline under federal law and
this fuel is commercially available. We cannot predict with certainty whether the use of E15 will
become more widespread in the future but with a 15 year useful life for many Tier 3 vehicles it is
prudent that evaporative system components have the durability to perform well when this fuel is
encountered. Most manufacturers use bench aging for their evaporative system components, so
there should be less compliance burden. If El 5 in-use fuel becomes progressively more
available, we would expect that El 5 service accumulation fuel would be used for whole vehicle
evaporative durability programs.
Deterioration Factors
What Commenters Said:
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
Carryover Durability Data: Deterioration Factor (DF) determinations are very resource intensive,
so we would like to use the DFs generated for Tier 2 (or LEV II) on Tier 3 vehicles. This should
yield a higher DF than if manufacturers developed a new DF using Tier 3 vehicles, since both
variation and deterioration tend to go down with emission standards (lower standards = lower
DF). However, as proposed, the regulations would not allow a manufacturer to carry over a DF if
the useful life test point exceeded the new emission standard (300 mg). Since the Tier 2 and LEV
II evaporative standards were higher than the Tier 3 and LEV III evaporative emissions
standards, this provision would prohibit the use of much of the manufacturers' durability data.
It's likely some of the vehicle full useful life tests will exceed the new standards. However, the
purpose of the DF is to establish the change in emissions as the vehicle ages, and using the Tier 2
4-192
-------
Tier 3 Summary and Analysis of Comments
data should be a conservative means of achieving this information. California recognized this
temporary concern and allowed that carryover data from LEV II evaporative emissions
certification would be capped at the LEV II evaporative emission standards. This is a transitional
issue (and a conservative one), since manufacturers will begin conducting durability testing on
the new Tier 3 evaporative emissions standards as the program phases in.
Recommendation: To prevent very expensive and unnecessary testing and harmonize with the
California LEV III, we request that EPA cap carryover durability data evaporative emission at
the Tier 2 evaporative emission standard limits.
Additionally, evap durability carryover should be capped at LEV 2 evaporative emission
standard levels. This would align the Tier 3 regulations with the LEV 3 regulations.
Organization: Chrysler Group LLC
Additionally, evap durability carryover should be capped at LEV 2 evaporative emission
standard levels. This would align the Tier 3 regulations with the LEV 3 regulations.
Our Response:
EPA understands the premise of the request and agrees that in most cases an evaporative
emissions DF based on Tier 2 technology would be likely to be numerically larger than an
evaporative emissions DF based on Tier 3 technology. The allowance requested by the
commenter, is often referred to as line crossing, is not prohibited in the EPA regulation. EPA
regulations at §86.1824-01 permit manufacturers to develop their full-useful life deterioration
factors for evaporative and refueling emission standards based on the use of good engineering
judgment. These DFs are additive in nature, and when added to the "undeteriorated low mileage"
test value the sum must be less than the applicable emission standard or FEL. Manufacturers
usually certify such that this summed value falls below the emission standard or FEL enough to
provide a margin for in-use compliance and to address variability and other uncertainty.
Regulations (at §86.1824-08) require that evaporative emissions durability assessments must
employ gasoline fuel for the entire mileage accumulation period which contains ethanol in, at
least, the highest concentration permissible in gasoline under federal law and that is
commercially available in any state in the United States (currently El5). Thus, EPA is
permitting the use of this data but requires that: (1) the manufacturers use good engineering
judgment in the testing used to develop their deterioration factors and the assessment and
application of this data in developing deterioration factors, (2) the manufacturers use the
evaporative/refueling emissions test fuel as required in the applicable regulations for Tier 3. The
comment states that "It's likely some of the vehicle full useful life tests will exceed the new
standards." Under current EPA provisions it is possible that the data used to develop the DF
could have a measured emission point above the standard, but the addition of the deterioration
factor to the low mileage test result must not result in an exceedance of the Tier 3 emission
standard or the FEL (under ABT).
Miscellaneous
4-193
-------
Chapter 4: Proposed Vehicle Emissions Program
What Commenters Said:
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
UL (FR30072):
Evap and exhaust useful life are associated in §86.1813-17(a)(5); however, they should be
treated independently in the Tier 3 regulations as described in the preamble, page 29886
Our Response:
During the Tier 3 evaporative emission program phase-in years (2017-2022 model years),
the exhaust and evaporative/refueling systems may have different useful life values if both
systems are not yet meeting the Tier 3 evaporative emission standards. However, the effective
operation of the engine and evaporative systems are interdependent and both need to operate
effectively to get the required in-use performance. A different useful life for exhaust and
evaporative system standards is counter to this important design premise.
What Commenters Said:
Organization: California Air Resources Board (CARB)
Comment 4 - Evaporative emission standards specifically for when an auxiliary (non-road)
engine is installed in a motor vehicle.
CARB supports U.S. EPA's proposal to extend evaporative emission requirements to auxiliary
non-road engines installed on motor vehicles. This would provide additional control of
evaporative emissions. Accordingly, CARB intends to propose alignment with this proposal once
the Tier 3 program is finalized.
Our Response:
We acknowledge CARB's supportive comments and plans to propose alignment on this
requirement. The Tier 3 NPRM proposed that any vehicle with an installed auxiliary engine and
fuel system must be tested in the SHED for hot soak and diurnal emissions with the fuel tank and
fuel system installed. The tank should have a 40 percent fill of the appropriate fuel. If the
propulsion engine and the auxiliary engine used the same fuel then they should be of the same
specification. EPA is finalizing this requirement because emissions of these engines are part of
the overall vehicle system. EPA expects that in many cases these engines will already have some
form of evaporative controls as part of the nonroad program, but there is the potential that they
will not be integrated into the vehicle correctly (e.g., placed near a heat source or there is
insufficient purge) or the engine-based controls may not be sufficient to allow the vehicle as a
whole to pass the Tier 3 emission evaporative standards.
4-194
-------
Tier 3 Summary and Analysis of Comments
What Commenters Said:
Organization: Revecorp Inc.
The proposed regulation does not require sufficient in-use compliance testing.
Revecorp believes that in-use compliance testing for tailpipe and evaporative emissions must
also be performed at high altitude. Proving compliance during certification does not prove that
emissions control system designs are durable. For example, Revecorp's firsthand experience
working on various studies and reviewing the literature indicates that there are many common
failures of evaporative emissions control systems in-use. These include:
Evaporative emissions purge valves sticking closed - This leads to a lack of purge of evaporative
emissions, therefore the canister is overwhelmed and emissions escape through the vent.
Canisters are placed in bad locations - This leads the canisters to age prematurely and crack
allowing emissions to escape to the environment as opposed to be drawn into the engine and
burned.
Canisters with bad intake locations (such as at the top of the wheel well) - This leads to the
canister ingesting dirt in-use and then the dirt deactivating the charcoal in the canister. This again
leads to evaporative emissions escaping continuously from the canister. This condition was
found in a pair of identical vehicles recruited in for CRC study E-91, indicating it was a design
flaw which would only be found with in-use testing.
Revecorp's interpretation of the proposed rule is that in-use verification program (IUVP)
requirements are not being strengthened for tailpipe emissions testing and there are no IUVP
requirements for evaporative emission testing at high altitude. As noted previously, if
engineering demonstration is allowed and there is no IUVP requirement for evaporative emission
control systems at altitude, manufacturer deficiencies will never be discovered.
Revecorp recommends - Vehicle manufacturers should be required to prove in-use compliance
for both tailpipe and evaporative emissions, at both sea level and at high altitude.
Our Response:
EPA does not agree with the commenter's assertion regarding the need for IUVP at high
altitude to assure system performance. As reflected in the commenter's experience and our
recently completed study on evaporative system diagnostic trouble codes (DTCs) from OBD
systems interrogations in I/M programs, evaporative systems do not perform flawlessly.
However, purge related codes were found in less than 0.5 percent of all vehicles which had the
evaporative system OBD monitor ready when they entered the I/M lane.25 This is not to suggest
25 Weatherby, M, Sabisch, M, Kishan, S. (February 2014) Analysis of Evaporative On-Board Diagnostic (OBD)
Readiness and DTCs Using I/M Data.
4-195
-------
Chapter 4: Proposed Vehicle Emissions Program
that purge valve failures are acceptable but to note that (1) they are not common and (2) OBD
finds them without IUVP Overall this is a relatively low rate but reliability of the operation of
the purge valve in use deserves further study and consideration.
In the NPRM, we discussed the issue of failed canisters and laid out the idea that a
malfunctioning canister (whether it is from "poisoning" or a blocked purge air valve) should be
monitored as part of OBD. Manufacturers disagreed with EPA on the feasibility monitoring for
canister failures, but EPA believes this is feasible. More information on the frequency of canister
problems in use is needed before we consider adding an OBD requirement to address this
potential concern. EPA is undertaking a study in response to this question.
Furthermore, the commenter indicated that IUVP testing at high altitude would be useful
in finding problems such as those related to the purge valve failures or non-functioning canisters.
EPA notes that IUVP testing at any altitude would find problems if they occurred on the vehicle
tested. Current pass rates for exhaust and evaporative emissions in IUVP exceed 90 percent.
Specifically with regard to evaporative controls, the sample sizes are small and it is not clear that
adding an IUVP requirement for SHED testing at high altitude would be of great benefit.
What Commenters Said:
Organization: Revecorp Inc.
The proposed rule is not protective enough of the health of Americans living at high altitudes and
population growth continues in these areas at a rate that exceeds the average for the US.
Although only approximately 5% of vehicles operate at "high altitude" (as the NPRM notes), the
proposed TIER 3 rule is a national standard. This national standard must protect all human health
- regardless of location or altitude. Revecorp is concerned that the proposed TIER 3 rule
compromises on testing requirements by allowing for "engineering demonstration" of how
evaporative emissions control systems will function at altitude, as opposed to requiring actual
testing. This could lead to vehicles being produced which are not compliant. [EPA-HQ-OAR-
2011-0135-4816-A2p. 2]
Evaporative emissions are especially difficult to control at high altitude and as EPA notes, they
account for 30 to 40% of hydrocarbon emissions including toxics such as benzene.
Unfortunately, evaporative emission control systems perform worse at high altitude due to
reduced atmospheric pressure. In addition, newer technologies, which the vehicle manufacturers
are just now gaining experience, are making it more difficult for these systems to function
effectively. Gasoline direct injection (GDI) and variable valve timing (WT) are becoming more
popular due to fuel economy regulations. Unfortunately, both reduce pumping losses across the
engine and therefore reduce the vacuum available to purge the evaporative emissions canister
completely. Another detrimental trend for evaporative emissions control systems is that many
manufacturers are downsizing vehicle engines to improve fuel economy and making up for the
loss in performance by turbocharging. Turbocharging causes a loss of vacuum, so strategies to
completely purge the canister are becoming more complex. The lower vacuum is a bigger
4-196
-------
Tier 3 Summary and Analysis of Comments
problem at altitude. As a result of the lower barometric pressure, typical vacuum levels are
reduced by 20% at 5,000 feet and more at even higher altitudes. Additionally, the lower fuel flow
required to maintain stoichiometric air fuel ratio at high altitude areas severely limits purge
options. Therefore, more sophisticated strategies need to be developed and should be physically
tested and proven at altitude. In addition, high levels of dust and particulate matter exist at the
semi-arid high altitude areas of the US. The high level of dust and particulate need to be
accounted for in the design of the evaporative control system and again need to be physically
verified as part of an in-use compliance program.
At high altitude, all of the above problems become even more difficult to overcome due to the
reduced atmospheric pressure. Allowing vehicle manufacturers to use "engineering
demonstrations" (showing on paper how the systems "should" work at high altitude) to prove
that new technologies will function at high altitudes could allow evaporative emissions control
systems which do not function properly or are not durable to be put into production. If these
systems are not designed properly, there is no in-use compliance testing for these vehicles to
identify any deficiencies so vehicles could have high evaporative emissions for their lifetime.
The proposed rule also notes that the new canister bleed standard would not apply at high
altitude. Due to the reduced atmospheric pressure, this test should be required to ensure the
system is working properly.
Revecorp recommends - All testing (tailpipe, evaporative and canister bleed test) should be
required to be performed at high altitude to protect the health of all Americans, including those
living at high altitudes.
Our Response:
EPA understands the basis for the commenter's concern but does not agree with the
conclusion. One can always propose that more testing is preferable, but this must be tempered
against a reasonable expectation of the value added and the cost. In the case of evaporative
emissions control EPA does not have information or data to indicate that specific high altitude
testing requirements is needed.
The key factors which govern compliance are vapor generation (ambient and fuel
temperatures and RVP), canister load (vapor generation rates, canister architecture, and activated
carbon characteristics), and purge (air mass and volume).The basic chemical and physical
principles governing these three processes are well understood and have been modeled
extensively. This understanding creates the basis by which engineering analysis of system
designs and data at low altitude can be used to predict compliance with the standard at high
altitude. More specifically, manufacturers understand and characterize how evaporative systems
operate at low altitude and conduct tests to show compliance with the emissions standards. With
knowledge of the system characteristics for the vehicle at low altitude, models, engineering
analysis, and good engineering judgment can be used to predict how they would perform at high
altitude where there may be different conditions. A good example is purge. Manufacturers know
how much purge air volume is needed to prepare the canister for evaporative and refueling vapor
loads and how much volume is generated on that vehicle as a function of available manifold
vacuum pressure. If the calibration changes at altitude and this reduces the purge volume,
4-197
-------
Chapter 4: Proposed Vehicle Emissions Program
adjustments may be needed depending on compliance margins and other system design factors.
Similarly, the working capacity of the activated carbon in the canister depends on the partial
pressure of the evaporative emission constituents such as butane and pentane. The canister
working capacity determined using low altitude testing methods may need to be adjusted to
assure the same working capacity at high altitude due to the effect of the lower atmospheric
pressure on the evaporative emission constituent partial pressures.
This situation is simpler for exhaust emissions since the air/fuel ratio is governed by
feedback controls from the oxygen sensor. Adjustments are made automatically by the onboard
computer.
Nonetheless, in many cases manufacturers do some developmental testing at high altitude
to confirm the relationship between system performance at low and high altitude. They then use
those results as a basis for future analysis and extrapolation of high altitude performance based
on low altitude system characteristics and calibrations.
4.4. Onboard Diagnostics (OBD) Comments
4.4.1. General OBD System Regulation Changes
What Commenters Said:
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
1. Tier 3 contains a number of changes to the EPA OBD regulations. Namely, beginning in 2017,
Tier 3 adopts all of the latest California OBD regulations and adds a requirement to monitor
"distance since 0.02-inch evaporative monitor decision" and report that value to a scan tool. The
adoption of ARB's latest OBD regulations will include both rate-based monitoring and the 0.02-
inch evaporative emissions leak check monitor. We generally support the changes proposed.
2. CARB is currently proposing some changes to its OBD program in response to the LEV III
program exhaust emission standards. We request comment on incorporating these changes into
this rule or other rules in the future. We support ongoing harmonization with the CARB OBD
regulations, especially if CARB makes adjustments to lead time, phase-ins or stringency of
provisions. If EPA adopts the California OBD regulations by reference, it will important for
EPA to update the reference by regulation in a timely manner. For this rulemaking, EPA should
adopt the latest version of the ARB OBD regulations (i.e., as amended 7-Aug-2012).
3. EPA notes, 'We also would generally expect to continue the current practice allowed by EPA
regulations which is for EPA to accept CARB OBD certifications as satisfying EPA
requirements provided that they include at least all of the requirements covered by the EPA
regulations" [78 FR 29905]. We support that EPA continue the current practice to accept
California OBD certification as satisfying EPA requirements, and this provision continues to be
important to ensure OBD certification [EPA-HQ-OAR-2011-0135-4461-A5, p4]
4-198
-------
Tier 3 Summary and Analysis of Comments
4. MY 2017 Implementation of California OBD Requirements
EPA also plans to align its overall OBD regulations with those in California and has proposed
the same MY 2017 start date for all other OBD requirements. The same concerns about applying
OBD requirements in the initial year of the regulation for the Distance Since Evap Monitoring
Decision and CA OBD Leak Detection Monitor at 0.02" apply for the overall requirements. For
this reason, as well as for simplicity, we recommend the same phase-in for the overall OBD
requirements, aligning with the phase-in for the evaporative emission standards to maintain a
single, streamlined phase in for all OBD requirements.
If EPA is unwilling to phase-in the overall California OBD requirements with the Tier 3 leak
emissions and evaporative emissions standards, we recommend that EPA provide an option for
manufacturers to phase in the requirements at a rate of 30/60/100% beginning in MY 2016- a
year earlier than would otherwise be required - while still maintaining the 60/60/80/80/100
starting in MY 2018 for the Distance Since Evap Monitoring Decision and CA OBD Leak
Detection Monitor at 0.02". EPA should also allow the OBD alternative phase-in (using the point
system) for this option.
Recommendation: We recommend that EPA provide a phase-in for the overall California OBD
requirements consistent with the phase-in for the other Tier 3 OBD requirements - Distance
Since Evap Monitoring Decision and CA OBD Leak Detection Monitor at 0.02" - which we
have recommended aligning with the corresponding leak emissions standards (60/60/80/80/100
starting in MY 2018).
Alternatively, EPA could offer a separate phase-in for the overall OBD requirements, not
including the Distance Since Evap Monitoring Decision and CA OBD Leak Detection Monitor at
0.02", of 30/60/100% beginning in MY 2016. EPA should also allow the OBD alternative phase-
in (using the point system) for this option.
Our Response:
EPA proposed to adopt, with a few adjustments, the CARB regulatory requirements
related to OBD II (see California Code of Regulations (CCR) §1968.2 dated May 18, 2010). We
received comment that since our NPRM was issued, CARB was completing an update of its
OBD II regulations and that EPA should adopt these provisions in lieu of the May 18, 2010
provisions. We have reviewed these updates and concur with the commenters, so we are
adopting the provisions officially approved by CARBs Office of Administrative Law on July 31,
2013, with certain adjustments and exceptions discussed in the preamble. As mentioned in the
NPRM and supported by the commenters, EPA expects to continue to work closely with CARB
on future OBD updates and to continue the current practice allowed by EPA regulations which is
for EPA to accept CARB OBD certifications as satisfying EPA requirements provided that they
include at least all of the requirements covered by the EPA regulations.
Several commenters requested a phase-in compliance approach in lieu of a fixed
compliance date, but no specific justification was provided by the commenters and EPA could
not establish a need for this accommodation since the most recent changes to CARB OBDII
4-199
-------
Chapter 4: Proposed Vehicle Emissions Program
regulations (2013) did not meaningfully affect provisions regarding vehicles under 14,000 Ibs
GVWR which have been in place since 2006. LDVs, LDTs, MDPVs and vehicles under 14,000
Ibs GVWR already comply with CARB OBDII requirements and use the CARB certification as
the basis for EPA certification. However, EPA agrees that phase-ins may be appropriate in
situations where there was an explicit Federal requirement that was less rigorous than the
California OBDII requirement (e.g., 0.020" evaporative leak monitor) and when there is a new
Federal requirement (e.g., Distance Since Evap Monitoring Decision). After considering the
comments received, EPA is permitting a limited and minimal phase-in for the 0.020" leak
detection criterion for the OBD evaporative system monitoring requirement. We are permitting
this phase-in, because a few vehicle models still only meet the 0.040" monitoring threshold in
their Federal configuration and complying with the 0.020" CARB OBD II requirement entails
validating performance in high altitude and cold weather regimes not seen in California. Thus,
the 0.020" requirement would be new for those few models currently certified only to the EPA
evaporative leak monitoring requirement. We are, therefore, implementing the following phase-
in provision for the 0.020" leak detection criterion for the OBD evaporative system monitoring
requirement. First, if a vehicle model meets the 0.020"requirement in the 2016 model year it is
not eligible for the phase-in provision. No backsliding is permitted. Second, for manufacturers
with models not meeting the CARB OBDII evaporative system leak monitoring requirement in
the 2016 MY(see CCR 1968.2(e)(4)), they will be permitted to delay product-wide compliance
with the 0.020" leak provision of the evaporative system monitoring requirements until the 2018
model year by engaging in a voluntary early phase-in. This phase-in would begin in the 2016
model year and conclude in the 2018 model year at a 100 percent implementation rate. For
example, a manufacturer could delay attaining 100 percent compliance with the OBD
evaporative system leak monitoring requirement until the 2018 model year by complying in the
2016 model year using a percentage which is at least as large as the delay for the 2017 model
year (e.g., 40% in 2016 MY, 60% in 2017MY, and 100% in 2018MY).
What Commenters Said:
Organization: California Air Resources Board (CARB)
Comment 1 - Version of OBD II Regulation Adopted:
Overall, CARB supports U.S. EPA's incorporation of the OBD II regulation section 1968.2 into
the federal requirements. However, CARB adopted additional changes to the 2010 version of the
OBD II regulation that is referenced in the proposal. Specifically, CARB adopted changes which
became operative on August 7, 2012. The 2012 amendments to the OBD II regulation are
considered minor and include changes to the permanent fault code requirements and in-use
monitor performance requirements to address issues related to hybrid vehicles (including plug-in
hybrid electric vehicles), delays to the required start date for three minor diesel-related monitors,
and an extension to the allowance for particulate matter filter monitors to exclude detection of
specific failure modes. CARB recommends that U.S. EPA adopt the more recent version of the
OBD II regulation to be more closely aligned with California.
Our Response:
4-200
-------
Tier 3 Summary and Analysis of Comments
EPA agrees and is adopting the 2013 version of CARBs OBD II regulations (13 CCR
1968.2), with certain adjustments and exceptions, as discussed in the preamble.
What Commenters Said:
Organization: Appalachian Mountain Club (AMC)
Overall we support these provisions including the adoption of the California Onboard Diagnostic
System (OBD) requirements.
Organization: Michigan Department of Environmental Quality (MDEQ)
The US EPA is proposing to update the OBD regulations to parallel the California Air Resource
Board's current requirements. This shows the requirements can be accomplished while
simultaneously making it easier for car manufacturers to have fewer regulatory requirements to
track and for which to adjust.
Our Response:
EPA agrees and is adopting the 2013 version of CARBs OBD II regulations (13 CCR
1968.2), with certain adjustments and exceptions, as discussed in the preamble. We expect to
continue the current practice allowed by EPA regulations which is for EPA to accept CARB
OBD certifications as satisfying EPA requirements provided that they include at least all of the
requirements covered by the EPA regulations.
4.4.2. Specific Revisions to EPA OBD Regulatory Requirements
What Commenters Said:
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
1. We recommend several additional changes below. These changes will streamline the
requirements and recognize that manufacturers begin at different levels of compliance with the
regulations.
Storing information in Non-Volatile RAM (NVRAM) and clearing codes:
As proposed, §86.1806-17 would require the vehicle to store the last evaporative leak check
result, and the distance (in miles) since that result was determined. Further, the leak check results
(both results and miles since determination) must be scan readable (i.e., readable with a generic
scan tool). We agree with these changes, and the SAE J1979 committee is developing a PID to
store the miles since determination. The regulation would also require that the leak check
results and mileage since determination be retained in system memory even if codes are cleared
or the vehicle battery is disconnected. This would require manufacturers to store the results and
4-201
-------
Chapter 4: Proposed Vehicle Emissions Program
mileage in NVRAM. Currently, vehicles only store permanent fault codes (i.e., when a monitor
fails a component) in NVRAM. Expanding that memory is both costly and time consuming. We
understand that the intent of this requirement is to be able to determine 1) if the monitor has
made a pass/fail determination within the past 750 miles and 2) whether the determination was
either "pass" or "fail." The requirement to store this in NVRAM was designed to prevent a
situation where a vehicle could pass the Leak Emissions Standard test by disconnecting the
battery or clearing the OBD codes. We support this intent, but believe it can be accomplished
much more efficiently. By simply resetting the "distance since determination" PID to maximum
value (40,722 miles) when codes are cleared or the battery is disconnected, the PID would only
reset to zero when an evaporative monitoring pass/fail decision is made. This would prevent a
vehicle from passing the IUVP leak emissions standard test by simply disconnecting the battery
or clearing codes, since the vehicle would not meet the requirements for determination within
750 miles. Moreover, since the counter will only reset after the evap monitor has made a
decision, and since the absence of a DTC will indicate that the monitor passed and the presence
of a DTC will indicate it failed, there is no need to store the pass/fail decision in NVRAM. The
SAE J1979 Committee is currently developing the requirements for the "Distance Traveled Since
Evap Monitoring Decision" PID and will include the requirement to set the PID to maximum
value when the codes are cleared or when the battery is disconnected, and only reset the counter
to zero when the evaporative emissions monitor has made a pass/fail decision.
Recommendation: To implement this proposal, we recommend the following change to §86-
1806(b)(l):
(1) OBD systems must record in computer memory the result of the most recent successfully
completed diagnostic check for a 0.020 inch leak The required data must be able to be used to
determine records include the miles driven since the last check occurred., and-the pass/fail resul^
and if there has not been a completed check since the computer memory was last cleared (e.g.,
from a scan tool command or battery disconnect). The system may be designed to keep data only
from the previous 750 miles of driving. The leak chock results data must be reported in a
standardized format consistent with other data required for the 080 system, scan readable, and
must bo retained in system memory oven if codes arc cleared or the vehicle loses battery power.'
Organization: California Air Resources Board (CARB)
Further, U.S. EPA is proposing vehicles support and report a new standardized data parameter
that will be used to indicate the distance traveled since the OBD evaporative system leak monitor
was last completed successfully, if the system passed or failed during the monitoring event, and
to store this data in non-volatile memory such that it can survive battery disconnects and on-
board computer memory clear events. This proposed requirement is intended to determine
whether off-board leak detection testing needs to be performed on individual vehicles during the
In-Use Verification Testing (IUVP) program.
Concerning the proposal for a new standardized data parameter, CARB has worked with the
SAE J1979 committee to develop an updated standard to achieve this. However, the draft SAE
proposal created to meet the intent does not completely align with the U.S. EPA proposed
regulatory language. Specifically, it does not provide for the data to be stored in non-volatile
memory but instead has the value report a clear indication if the data has been erased and the
4-202
-------
Tier 3 Summary and Analysis of Comments
monitor has not successfully run since then. In this situation, it appears to CARB that this will
satisfy the intent and provide a clear indication that the off-board leak test should be performed
on such a vehicle selected for IUVP testing. Secondly, the draft parameter indicates the mileage
since the last successful test completion but does not directly indicate whether it was a pass or
fail decision. Instead, current fault code status must be used and the absence (or presence) of an
evaporative system leak fault code will confirm whether the monitor last ran and passed (or
failed, respectively). Again, CARB believes this will provide the necessary data for the intended
use in IUVP testing. Accordingly, CARB recommends the proposed language be modified in the
final proposal to eliminate the requirement that the new data be able to survive battery
disconnects and computer memory clear events as well as remove the requirement that the new
parameter directly report whether the last monitoring event was a pass or a fail.
Our Response:
We agree with the commenters. Since the manufacturers normally implement the OBD
requirements using technical guidance from the consensus-based SAE J1979 standard we think it
is important for our requirements to be compatible with that standard provided that it will meet
the EPA objective and that it be done for the lowest possible cost. EPA received further input
from commenters on this issue which slightly revised the suggested language in the comment.
We are adopting provisions consistent with the input from commenters. 6
What Commenters Said:
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
MY 2017 Implementation of "Distance Since Evap Monitoring Decision" As currently written,
the regulations require, beginning in MY 2017, 100 percent implementation of the OBD distance
since evap monitoring decision. Implementing even a relatively straightforward requirement
such as this one, across all vehicles adds burden - in this case, unnecessarily. We do not believe
it was EPA's intent to require a 100 percent phase-in of this requirement in MY 2017. Moreover,
the leak emissions standard does not begin until MY 2018 and then phases in with the Tier 3
evaporative emissions standards (60/60/80/80/100 percent starting in MY 2018), and the phase in
for the two should be aligned.
Recommendation: We recommend the following change to §86-1806(b): '(b) The following
additional provisions apply to vehicles that are also certified to the leak emissions standard of
S86.1813-17(a)(4V
Our Response:
26 Passavant, G. (January, 2014). "Manufacturer Input on Distance Since Last Evaporative Monitoring Decision".
Memorandum to the docket.
4-203
-------
Chapter 4: Proposed Vehicle Emissions Program
EPA concurs with this comment. The "Distance Since Evap Monitoring Decision"
requirement will be phased in on vehicles as they meet the leak standard. This occurs over the
2017-2022 or 2018-2022 model years depending on the compliance option used for the 2017
model year.
What Commenters Said:
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
MY 2017 Implementation of California 0.02" Leak Detection Monitor:
The proposed regulation would also drop the option for federal vehicles to comply with an OBD
leak detection monitor requirement of 0.04" and require all federal vehicles to comply with the
current California OBD requirements, including the 0.02" evaporative leak detection monitor (in
lieu of the federal 0.04" leak detection monitor) and rate based monitoring requirements (which
are not included federally). As proposed, the 0.02" leak detection requirements would be
required for each manufacturer's full federal fleet in MY 2017. While several manufacturers
comply with the California OBD requirements nationwide already, other manufacturers certify a
large portion of their federal fleet to the federal OBD standards. Validating their entire federal
fleet to the 0.02" leak detection monitor will involve significant resources and cannot be
efficiently implemented in the time between the Tier 3 final rule and MY 2017. For example, the
federal vehicles will need to be validated at high-altitude, a requirement that does not exist in the
California program. Rather than a 100 percent requirement in MY 2017, we recommend EPA
provide a phase-in period to allow these manufacturers to more efficiently manage limited
resources needed to make this conversion. The Tier 3 program already involves a number of
different phase-in provisions. Creating another phase-in that is not aligned with other provisions
would add burden on both the manufacturers and the agency in the form of unnecessary new
reporting and tracking compliance requirements. For simplicity, we recommend linking the
phase-in of the 0.02" monitor requirement with the percentage phase-in steps that have already
been proposed for the new evaporative and leak test emission standards. We recognize the
agency might be reluctant to allow this long phase-in of the basic monitor requirement because
there may be incremental air quality benefits associated directly with the change from the 0.04"
to the 0.02" monitor requirement independent of the leak test standard and accompanying OBD
changes (i.e., distance monitor, etc.). However, we believe there will be enough incentive for
manufacturers to make this conversion as soon as possible that a separate, more rapid and more
rigid phase in requirement will be unnecessary. Given that not all manufacturers are in this
situation (i.e., significant use of the 0.04" option) and given there will be a strong incentive to
move as quickly as possible to achieve a full national fleet of vehicles that can be sold anywhere
in the U.S., it seems very unlikely that any manufacturer would take the fullest advantage of the
phase-in schedule. Use of the 0.04" monitor would prevent a vehicle from becoming a "50-state"
vehicle. Under the current Tier 2 regulations, there are multiple differences between the EPA and
California requirements that might cause a manufacturer to opt for separate fleets. But given that
EPA has worked to align the Tier 3 program as much as possible with the California LEV III
program, in the future there will be fewer reasons for a manufacturer to opt for separate fleets,
particularly if the only difference is the OBD leak monitor. Therefore, we don't think EPA needs
to add complexity to the program by creating another phase-in that is not aligned with other
aspects of the Tier 3 program.
4-204
-------
Tier 3 Summary and Analysis of Comments
Recommendation: We recommend that EPA provide a phase-in for the CA OBD leak detection
monitor requirements (0.02" detection) that aligns with the phase-in for the leak emissions
standards (60/60/80/80/100 starting in MY 2018).
Our Response:
There are very few manufacturers who are not already fully compliant with the 0.020"
cumulative diameter threshold for leak diameter monitoring on all of its Federal models, even
though the current Federal requirement is only for a 0.040" cumulative diameter threshold. The
0.020" threshold has been in effect in California for over ten years. EPA sees no need for a full
phase-in as requested, but sees the value in giving those manufacturers who are not yet fully
compliant with the 0.020" threshold some flexibility to integrate the prove-out of their
monitoring hardware and software with other field work. Therefore, we are providing a very
limited phase-in flexibility wherein manufacturers may defer 100 percent compliance for their
gasoline-powered vehicles less than or equal to 14,000 Ibs GVWR until the 2018 model year by
pulling ahead compliance to the 2016 model year for a percentage equal to or greater than that
percentage that they wish to defer to the 2018 model year.
What Commenters Said:
Organization: California Air Resources Board (CARB)
Currently U.S. EPA accepts On-Board Diagnostics (OBD) systems certified to CARB's OBD II
regulation as systems also satisfying the federal OBD regulation. The NPRM proposes to codify
this practice by adoption and incorporation by reference CARB's OBD II regulation (title 13,
California Code of Regulations, section 1968.2) as approved on June 17, 2010, effective starting
in 2017 model year, except for the following provisions:
- sections 1968.2(a), (b), (k) are excluded
- the tampering clause in section 1968.2(d)(1.4) is excluded
- the camshaft/crankshaft alignment monitoring requirement in section 1968.2(e)(15.2.1)(C)
would only apply to vehicles with variable valve timing systems
The proposal also includes a requirement for manufacturers to perform additional demonstration
testing of the evaporative system monitor capability and to report additional standardized data
through the OBD system that would identify how many miles since the evaporative system leak
detection monitor last completed.
Comment 2 - Camshaft/crankcase Alignment Monitoring Requirement:
The proposal includes a specific exclusion for the requirement in the CARB OBD II regulation
section 1968.2(e)(15.2.1)(C) that OBD systems must be able to detect a misalignment between
the crankshaft and camshaft. This is not a technology-forcing requirement and has been in place
for many years in California-certified OBD II systems. CARB believes that a reduction in
diagnostic capability and emission benefit would result due to the exclusion of
camshaft/crankshaft alignment monitors on vehicles without variable valve timing systems,
because alignment malfunctions can routinely occur in-use due to timing belts slipping or being
4-205
-------
Chapter 4: Proposed Vehicle Emissions Program
improperly installed during maintenance involving timing belt replacement. Therefore, CARB
recommends that the proposed regulation language be changed by removing the specific section
that limits camshaft/crankshaft alignment monitoring to only those vehicles equipped with
variable valve timing systems.
Our Response:
This limitation is specifically included in the current EPA OBD requirements and EPA
did not specifically propose to remove it. EPA did not receive input on this issue from any
manufacturer and we did not seek their input on CARB's comment. Thus, while EPA believes
CARS's perspectives have merit, we are not in a position to remove this limitation without input
from the manufacturers and others. It should be noted that the requirement will continue to
apply to vehicles equipped with variable valve timing systems and these are expected to become
more prevalent in response to the 2017-2025 model year GHG and fuel economy standards.
What Commenters Said:
Organization: California Air Resources Board (CARB)
Comment 3 - Evaporative Leak System Monitor:
The NPRM proposes a requirement for manufacturers to phase-in a 0.020 inch leak detection
monitor as CARB's OBD II regulation has required since the 2002 model year. CARB supports
this change to align detection capability nationwide. Further, most manufacturers already extend
this capability nationwide by certifying a single system that meets CARB requirements.
Our Response:
We agree with this commenter. The requirement is feasible and being met by almost all
manufacturers. The requirement will apply to all 2017 and later model year gasoline-powered
highway motor vehicles up to 14,000 Ibs GVWR. The regulations provide a limited flexibility
which will permit manufacturers to defer full compliance to the 2018 model year if they comply
in the 2016 model year with a percentage equal to or greater than the percentage they defer from
the 2017 model year to the 2018 model year.
What Commenters Said:
Organization: Ford Motor Company (Ford)
Adoption of CARB OBD Regulations: EPA has proposed to require that all Federal vehicles
meet California's On-board Diagnostic regulations by the 2017MY. Ford has two specific
concerns with this proposal:
0.020" OBD Leak Monitor Detection Threshold - This proposal would require that all Federal
vehicles adopt California's 0.020" leak detection monitor threshold by the 2017MY. The existing
4-206
-------
Tier 3 Summary and Analysis of Comments
EPA requirement for leak detection is 0.040", but under the Tier 3 proposal, 100% of a
manufacturer's fleet would be required to meet this more stringent requirement in 2017MY.
Although most Ford vehicles have a leak detection monitor capable of meeting this new
requirement, they have not been validated to the 0.020" threshold on a nation-wide basis. In
order to accomplish robust leak detection to the more stringent standard nationally, additional
validation and data collection must be performed that account for more varied environmental and
market fuel conditions encountered in all 50 states (e.g., high-altitude, additional fuel
formulations, etc.). Since OBD assessments are typically performed holistically as part of new
program vehicle development, the processes and engineering resources are not in place to
retroactively reassess existing systems to account for these new conditions. For these reasons,
Ford proposes that EPA allow a phase-in for the 0.020" requirement of sufficient duration to
allow migration to the new standard to be accomplished via new program development, which is
typically on a three-year cycle.
Recommendation: Ford proposes that EPA provide a phase-in for the new 0.020" OBD leak
detection threshold that would allow Ford to utilize its existing, robust new program OBD
validation processes and avoids the need to employ ad-hoc, off-program actions. Ford requests
that such a phase-in start no earlier than 2016MY, and extends for at least three years.
Our Response:
There are very few manufacturers who are not already fully compliant with the 0.020"
cumulative diameter threshold for leak diameter monitoring on all of its Federal models, even
though the current Federal requirement is only for a 0.040" cumulative diameter threshold. The
0.020" threshold has been in effect in California for over ten years. EPA sees no need for a full
phase-in as requested, but sees the value in giving those manufacturers who are not yet fully
compliant with the 0.020" threshold some flexibility to integrate the prove-out of their
monitoring hardware and software with other field work. Therefore, we are providing a very
limited phase-in flexibility wherein manufacturers may defer 100 percent compliance for their
gasoline-powered vehicles less than or equal to 14,000 Ibs GVWR until the 2018 model year by
pulling ahead compliance to the 2016 model year for a percentage equal to or greater than that
percentage that they wish to defer to the 2018 model year.
4.4.3. Provisions for Emergency Vehicles
What Commenters Said:
Organization: Ford Motor Company (Ford)
Emergency Vehicle Provisions:
The CARB OBD regulation required manufacturers to meet unique monitoring requirements to
detect malfunctions causing an air/fuel ratio imbalance between cylinders. By the 2016 model
year, all vehicles certified to the CARB OBD II requirements must meet the final Air/Fuel Ratio
Imbalance Monitor (AFRIM) requirement final emissions threshold of 1.5 times the applicable
standard. We believe there should be an exemption from the AFRIM threshold for emergency
4-207
-------
Chapter 4: Proposed Vehicle Emissions Program
vehicles. We will illustrate this comment with the example of Ford's 2016 and later model year
Police Interceptor package. In order to meet the 1.5 times the standard emissions threshold, the
Police Interceptor would require an added underbody catalyst. This would result in performance
degradation for the Police Interceptor, with a 0-100 MPH time increase that may not meet police
agency expectations. As a result, CARB had agreed to provide Ford with an exemption from the
AFRIM requirement on the Police Interceptor package using a provision in the California
Vehicle Code that allows emergency vehicles to meet Federal emissions requirements:
27156.2 Notwithstanding any other provision of law, any publicly owned authorized emergency
vehicle operated by a peace officer, as defined in Section 830 of the Penal Code, any authorized
emergency vehicle, as defined in Section 165 and used for fighting fires or responding to
emergency fire calls pursuant to paragraph (2) of subdivision (b) or pursuant to subdivision (c) or
(d) of that section, and any publicly owned authorized emergency vehicle used by an emergency
medical technician-paramedic, as defined in Section 1797.84 of the Health and Safety Code, is
exempt from requirements imposed pursuant to California law and the regulations adopted
pursuant thereto for motor vehicle pollution control devices.
Ford had planned to certify the 2017 and subsequent model year Police Interceptor without the
AFREVI monitor since the Federal OBD requirements do not require this monitor. The vehicle
could then be certified and sold in all 50 states. The Police Interceptor is being developed to meet
all other Federal and California emissions and OBD requirements. However, EPA's proposal to
require CARB OBD II on all Federal vehicles by the 2017 model year would force the Police
Interceptor to again have to meet the AFRIM requirement, and thus force those vehicles to be
fitted with an underbody catalyst with the associated vehicle performance reduction for the sole
purpose of keeping emissions below 1.5 times the standard just in case an unlikely cylinder
imbalance malfunction were to occur. We do not think it was EPA's intent to impose OBD
requirements on Federal vehicles that CARB would not require on vehicles sold in California.
Recommendation: With respect to the AFRIM issue specifically, Ford recommends that EPA
provide an exemption from the CARB OBD II AFRIM requirement in 13 CCR 1968.2,
paragraph (e)(6.2.1)(C) through at least the 2019 model year for emergency vehicles. This would
allow Ford the sufficient lead time to implement the necessary design changes need to meet the
CARB OBD II AFRIM requirement without needing to add hardware or other solutions that
reduce vehicle performance levels. More generally, we encourage EPA to adopt language in the
final Tier 3 rules to the effect that EPA will recognize and adopt, with respect to Federal
vehicles, any OBD exemptions that CARB may allow for California vehicles.
Our Response:
EPA has held follow-up discussions on this matter with Ford and CARB staff and has
reviewed the technical data regarding the ability of the law enforcement vehicles to meet the
AFRIM monitoring threshold requirements.27 Adding an underbody catalyst to these two vehicle
configurations just to meet the AFRIM requirements adds cost and more importantly reduces
performance needed in these types of vehicles. These two test groups (comprised of law
27 Passavant, G, (January, 2014). Information Related to CARB AFRIM OBD Requirements for Emergency
Vehicles. Memorandum to the docket.
4-208
-------
Tier 3 Summary and Analysis of Comments
enforcement vehicles) are not required to meet the current AFRIM monitoring requirements and
CARB does not intend to require it in the future. Since Ford expects to have compliant
powertrains for these test groups by the 2020 model year and the only potential adverse
environmental effect is failure to notify the user in the event of an AFRIM imbalance situation,
EPA will defer this requirement for these vehicles through the 2019 model year. The final rule
also incorporates broader provisions that address potential future OBD issues related to
emergency vehicles.
4.4.4. Future Considerations
1. OBD Monitoring of the Performance of Canister Carbon
What Commenters Said:
Organization: Revecorp Inc.
Although tailpipe emissions are significant, Revecorp is most concerned with changes in the
proposed rule related to the control of evaporative emissions. This is because evaporative
emissions account for 30 to 40% of hydrocarbon emissions, including toxics such as benzene.
Research conducted for EPA has shown that vehicles can have failures of the evaporative
emissions control systems (such as deactivation of the canister substrate) that OBDII systems do
not identify. This is in part because sensing of the activity of the canister carbon is not monitored
by OBDII systems, so when it fails, it goes unnoticed. This can lead to vehicles with
uncontrolled evaporative emissions for the life of the vehicle, which is not identified by the
OBDII system (to notify the motorist to get the vehicle repaired) and not caught by in-use
compliance testing because it is not required for evaporative emissions control systems (at least
not at high altitude).
On-board diagnostic systems should be strengthened to improve identification of in-use excess
emissions.
As noted above, without a way to identify failures of the emissions control systems in-use,
vehicles will have high emissions for their lifetimes. In one phase of a study conducted for EPA
on in-use vehicle evaporative emissions (EPA contract #: EP-C-06080, the "Ken Caryl" phase of
the study) 50% to 75% of the vehicles which were identified to have high evaporative emissions
did not have the check engine light illuminated (the excess emissions were not identified by the
OBDII system). This is in part because a major component of the evaporative emissions control
system (the evaporative canister) is not checked by the OBDII system to ensure it is functioning
properly. In some cases, canisters were damaged and evaporative gases were escaping into the
environment. The canister of a vehicle is the single most important component of the evaporative
emissions control system and should be monitored to ensure it is performing properly.
Revecorp recommends - Monitoring of canister performance should be added to the OBDII
requirements.
4-209
-------
Chapter 4: Proposed Vehicle Emissions Program
Our Response:
EPA agrees that the canister performance is critical to the overall ability of the
evaporative system to control vapor emissions. While it is easy to estimate the emissions
increase which occurs if the canister partially or totally fails - and it is quite large- it is more
difficult to determine the frequency and cause of such failures. Purge valve failures leading to
saturated canisters are infrequent (-0.5% )28 based on an initial review of OBD data, but the
frequency of saturated canisters related to other causes such as blocked air inlet purges and
ingestion of liquid is unknown and is informed only by anecdotal information. EPA is now
engaged in further study of this question to help determine the frequency of such occurrences
and whether an OBD requirement is justified.
What Commenters Said:
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
OBD monitoring of the performance of the canister carbon is neither necessary nor practical. The
OBD system currently verifies that the canister is in the system (large and small evaporative leak
test) and that purge is working correctly. Also, if the system is designed properly so that there is
no liquid water or fuel entering the canister, the carbon in modern canisters does not deteriorate
during the life of the vehicle. Therefore, diagnosis of the canister carbon is unwarranted.
Furthermore, diagnosis of the performance of the carbon would be extremely difficult. The
potential diagnostic signals associated with vapor in the fuel today, e.g., (closed loop fueling
feedback) have too much variation to make a reliable diagnostic. Even adding new hardware for
diagnosis e.g., carbon bed temperature sensor, would not provide for a reliable diagnostic.
Our Response:
EPA disagrees with the assessment of the commenter with regard to the technical
capability to monitor canister performance. If it is warranted, there are several sensing schemes
which could be used to monitor the in-use performance of the canister itself. However, as
mentioned above, more information on the frequency and causes of failures is needed, so we are
not putting a requirement in place at this time.
What Commenters Said:
Organization: California Air Resources Board (CARB)
Comment 4 - Evaporative System Activated Carbon Canister Monitor and Purge Monitor:
28 Weatherby, M, Sabisch, M, Kishan, S. (2014) Analysis of Evaporative On-Board Diagnostic (OBD) Readiness
and DTCs Using I/M Data.
4-210
-------
Tier 3 Summary and Analysis of Comments
Though not officially part of U.S. EPA's proposal, U.S. EPA seeks comments about a provision
to require the OBD system to monitor the activated carbon canister, specifically for evidence of
proper performance of the canister such as adequate adsorption and desorption of hydrocarbon
vapors. U.S. EPA believes such performance could be sensed with existing hardware on the
vehicle (e.g., looking for changes in canister carbon bed temperature or presence of vapor in the
fuel going to, the intake manifold after a cold start or refueling event). CARB currently has no
requirement for monitoring the canister performance but given the increasing importance of
keeping evaporative emissions to a minimum, believes the idea merits further exploration and
research before making any decisions on implementing any regulation changes. Important
aspects that would likely need to be investigated include whether monitoring of the activated
carbon canister can indeed be robustly done with the existing hardware on the vehicle.
Our Response: EPA agrees that given the potential emissions impact, this issue merits further
research and assessment. EPA believes that if monitoring is warranted, there are several sensing
schemes which could be used to monitor the in-use performance of the canister itself. However,
as mentioned above, more information on the frequency and causes of failures is needed before
we propose a requirement.
2. OBD Monitoring of Vacuum Pump or Other Assist Hardware
What Commenters Said:
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
If a vacuum pump or other hardware is utilized to assist canister purge, the current OBD
regulations would require that this pump and associated hardware be monitored.
Organization: California Air Resources Board (CARB)
Regarding devices used to assist purge flow, increased usage of such devices in the future is
possible, so further investigation into monitoring of these devices is warranted. However, CARB
believes the OBD regulation already covers monitoring of these devices to some extent, since
they would be considered evaporative system components and consequently required to be
monitored under comprehensive components requirements for proper function.
Our Response:
EPA appreciates these comments and will include such monitoring approaches in future
certification reviews if a future system uses a vacuum pump or other hardware to assist canister
purge.
3. Diagnose Sealed Fuel Systems Down to 0.010" or 0.015"
What Commenters Said:
4-211
-------
Chapter 4: Proposed Vehicle Emissions Program
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
Industry is opposed to OBD detection of leaks smaller than 0.020" in sealed fuel systems. Fuel
system leaks that are less than 0.020" are extremely difficult to detect and are even more difficult
to repair. Attempting to detect leaks smaller than 0.020" would result in a high risk of false MILs
and numerous vehicles that could not be repaired. Ultimately, this would undermine the public's
confidence in OBD performance.
Organization: California Air Resources Board (CARB)
EPA also seeks comments about a provision regarding pressurized and sealed fuel tanks. Based
on its observation of a trend towards fuel tanks with higher operating pressures or tanks sealed to
the atmosphere during normal operation and data showing a leak in these systems will result in
substantially higher emissions relative to traditional low pressure systems. U.S. EPA seeks
comment on the feasibility and cost of requiring OBD leak detection monitors to detect a smaller
diameter orifice on such systems - specifically, something in the magnitude of a 0.010 inch leak
in lieu of a 0.020 inch leak anywhere in the evaporative system. This would apply to vehicles
with a designed in-use operating pressure in excess of 0.36 psi (10 inches of water).
CARB currently has no such requirements but given the increasing importance of keeping
evaporative emissions to a minimum, believes the ideas merit further exploration and research
before making any decisions on implementing any regulation changes. Important aspects that
would likely need to be investigated include the feasibility of the repair service community to
efficiently diagnose and repair detected leaks with a smaller diameter.
Our Response:
EPA understands the concerns regarding detection and repair raised by the commenter.
While there is limited data on emission rates as a function of cumulative orifice diameter, more
information on the frequency of such leaks is needed before pursuing this requirement.
4. Vent Leaking Sealed Fuel Tank System to Canister
What Commenters Said:
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
EPA has requested comment about opening the vent valve on key off if a leak is detected in a
sealed fuel system. This concept will not reduce the emissions from a leak because the flow
through the canister is more restrictive than the flow out of the leak. Therefore, almost all of the
vapor will go out the leak rather than into the canister.
Organization: California Air Resources Board (CARB)
4-212
-------
Tier 3 Summary and Analysis of Comments
For sealed tanks, U.S. EPA seeks comments on an added requirement that any valve that isolates
the fuel tank vapor from the canister be defaulted to an open position at key off (allowing vapors
to vent to the canister) if the OBD monitor detects a leak and illuminates the malfunction
indicator light. Further, for pressurized systems, U.S. EPA is asking for comment on setting the
evaporative system monitor threshold in the 0.010 inch to 0.015 inch range, specifically asking
for the feasibility of this threshold, the effects of the vent valve open requirement on on-board
refueling vapor recovery, and the reparability of leaks below 0.020 inch.
C ARB currently has no such requirements but given the increasing importance of keeping
evaporative emissions to a minimum, believes the ideas merit further exploration and research
before making any decisions on implementing any regulation changes. Important aspects that
would likely need to be investigated include the feasibility of the repair service community to
efficiently diagnose and repair detected leaks with a smaller diameter.
Our Response:
EPA did not make a specific proposal with regard to venting of higher pressure or sealed
fuel tanks with leaks to the activated carbon canister. While sealed or higher pressure fuel tank
system designs are not common today, the reduction of vacuum for purge air in some future
engines/powertrains (e.g., hybrids) creates the possibility that manufacturers will turn designs of
this nature to help reduce vapor loads to the canister and thus allow for lower purge volumes.
Manufacturers should be cognizant of the issue raised by EPA in the NPRM and put emphasis on
system durability. In coordination with CARB, EPA will continue to evaluate trends in fuel
system operating pressures and potential design features leading to leaks.
4.4.5. Heavy-duty Vehicles
What Commenters Said:
Organization: Truck and Engine Manufacturers Association (EMA)
OBD Requirements as Applied to Hybrid Applications:
Current ARB and EPA OBD regulations include provisions requiring hybrid components and
systems to be monitored under "comprehensive component monitoring" requirements. Those
provisions pose a significant barrier to the introduction of hybrid system technology. In fact, the
current OBD provisions will substantially limit the availability of hybrid vehicles nationwide.
The use of hybrids in medium- and heavy-duty commercial applications is an emerging
technology, and the market for that technology is very small relative to overall heavy-duty
applications and to the light-duty passenger car industry. There are multiple medium- and heavy-
duty duty-cycles and applications where hybrids might be practical. Additional time is needed to
develop and commercialize hybrid technologies for those applications, and even more time and
resources are needed to develop proper diagnostics on hybrid components. As things stand,
however, current OBD regulations will result in little or no use of hybrid applications in
California, depriving the state (and any state that has opted in to California's standards) of the
4-213
-------
Chapter 4: Proposed Vehicle Emissions Program
fuel efficiency and other benefits that hybrid technology has to offer. The same is true
nationwide, as EPA's existing hybrid OBD requirements for heavy heavy-duty vehicles and
engines become fully implemented. Engine and vehicle manufacturers are continuing their
efforts to assure that hybrid technology - an important GHG emission-reduction opportunity -
will be available throughout California and the remainder of the U.S. EPA should support those
efforts by adopting (and, as necessary, amending) OBD regulations without hybrid component
monitoring requirements. EPA should take additional time to assess whether a potential
regulatory framework can be developed, with appropriate lead time, that assures the feasible,
cost-effective and reasonable development of hybrid system and engine technologies and
associated diagnostic strategies.
Take additional time to assess the appropriate regulatory framework for cost effective and
feasible development of hybrid technology and diagnostic strategies
Our Response:
EPA acknowledges EMA's comment opposing hybrid component monitoring
requirements. However, no detail regarding the technology issues or problems or systems
integration was provided in the comment. Follow-up discussion with EMA indicated that their
major concern regarding OBD and hybrids was related to vehicles over 14,000 Ibs GVWR.29
After confirming that the Tier 3 rulemaking does not address OBD requirements for vehicles
over 14,000 Ibs GVWR and that appropriate lead time would be provided for new heavy-duty
vehicles equal to or less than 14,000 Ibs GVWR, EMA was satisfied with the way EPA was
implementing the OBD requirements for hybrids for vehicles equal to or less than 14,000 Ibs
GVWR. Thus, we are not deferring adoption of OBD requirements as part of the Tier 3 rule. As
discussed below, the OBD provisions of the Tier 3 rule do not apply to vehicles/engines over
14,000 Ibs GVWR.
14.000 Pounds GVWR and Under:
EPA is proposing to update its OBD regulations to be consistent with ARB's current OBD
requirements ("OBDII"). (78 Fed. Reg. at 29903-29904.) For Heavy-Duty Vehicles (8,501-
14,000 GVWR), those requirements would begin in MY2018. Although adopting the OBDII
regulations for MY2018 does not provide the four full model years' lead-time required by the
CAA, EMA recognizes that EPA already accepts certification with ARE OBDII requirements as
satisfying EPA OBD requirements for vehicles 8,501-14,000 Ibs. GVWR. As a result, EMA
generally supports such alignment with ARB's OBDII regulations, with the significant exception
noted above. EPA should support manufacturers' efforts to assure the cost-effective commercial
use of hybrid technology by adopting the OBDII regulations without hybrid component
monitoring requirements. If EPA were to adopt the OBDII provisions without removing such
requirements, the successful commercialization of hybrid vehicles throughout the nation would
be substantially impeded, if not thwarted altogether.
Adopt the ARB's OBDII regulations without requirements to monitor hybrid components
29 Passavant, G. (February, 2014). "OBD for Hybrid HD Vehicles/Engines". Memorandum to the docket.
4-214
-------
Tier 3 Summary and Analysis of Comments
Our Response:
As discussed above and in the preamble, EPA proposed adoption of CARB 2010 OBD
regulations and based on comments is adopting the version of the CARB OBD regulations
completed in July 2013. With the exception of hybrids EMA did not oppose the 2018 model
year implementation of OBD for HDVs equal to or less than 14,000 Ibs GVWR. EMA's
comment concerning hybrid component monitoring is addressed above. As discussed in the
preamble, to be consistent with the manner in which the Tier 3 exhaust emission standards are
being implemented for the heavy-duty vehicles between 8,501 and 14,000 Ibs GVWR, the OBD
requirements will apply in MY 2019 based on the Job 1 (first production) date for the
vehicle/engine model. If the vehicle/engine model Job 1 date is before the fourth anniversary
date of the signature of the Tier 3 rule the Tier 3 requirements will be mandatory in the 2020
model year, not the 2019 MY. If the Job 1 date is on or after the fourth anniversary of the
signature date of the Tier 3 rule the OBD requirements will apply beginning in MY 2019. The
Tier 3 OBD requirements will apply to all 8,501-14,000 Ib HDVs beginning in MY 2020.
Over 14.000 Pounds GVWR:
In the preamble to the Proposed Rule, EPA indicates that the proposal to adopt the California
OBD regulations also "includes heavy-duty vehicles above 14,000 Ibs. GVWR, though these
vehicles would not need to meet the new requirements related to leak testing" (78 Fed. Reg. 98,
29919). Based on other language in the Proposed Rule, including footnote 313 of the preamble
(which specifically states, "We are not proposing to change the [OBD] requirement for
incompletes and vehicles with a GVWR above 14,000 Ibs.") (78 Fed. Reg. at 29904), EMA
believes the reference on page 29919 of the Federal Register notice is a drafting error. EMA also
believes the language is an error because EPA has not (1) conducted any technical analysis of
applying ARB's heavy-duty OBD regulations; (2) properly notified heavy-duty manufacturers of
any such significant changes to the EPA heavy-duty OBD program; or (3) proposed any other
changes in the rule that would require the application of California's OBD regulations to vehicles
and engines over 14,000 Ibs. GVWR. EPA objects to the finalization of any portion of the
Proposed Rule that would adopt ARE OBD regulations applicable to vehicles and engines over
14,000 Ibs. GVWR. EPA should clarify in the final rule that no changes to EPA's on-board
diagnostics requirements applicable to engines and vehicles over 14,000 Ibs. GVWR are being
proposed or made in conjunction with the Tier 3 Standards (other than OBDII regulations that
may be applicable to optionally chassis-certified HDGVs). Without this vital clarification, the
Proposed Rule will be subject to immediate challenge.
State unequivocally in the final rule that no changes are being made to OBD requirements for
engines or vehicles greater than 14,000 Ibs. GVWR
Our Response:
The OBD provisions in this rulemaking do not apply to vehicles/engines over 14,000 Ibs
GVWR, except that the Tier 3 OBD regulations are applicable to optionally chassis-certified
HDGVs.
4-215
-------
Chapter 4: Proposed Vehicle Emissions Program
4.4.6. Other Issues
What Commenters Said:
Organization: VNG.CO
Reduced OBD Testing Requirements - Today: Current EPA rules allow alternative fuel
conversion SVMs to meet much less stringent OBD requirements than OEMs certifying small
volume test groups, with as few as four tests of major diagnostic monitors (including fuel trim
lean & rich, catalyst deterioration, engine misfire, and oxygen sensor) instead of the many more
typically required for OEM certification.15 As illustrated above, the NPC Future Transportation
Fuels report notes that "certification, OBD, aftertreatment and calibration" are a significant
source of per-vehicle incremental costs for current OEM NGV production. This is particularly
the case for pickup trucks, where it is the single largest cost component at about $4,500 per
vehicle, or nearly 40% of the total incremental cost.16 Harmonizing OEM small vehicle test
group OBD testing procedures with the reduced OBD testing requirements for alternative fuel
converters is consistent with the 'level playing field' provided in most other respects between
these categories of vehicles, and represents a major opportunity to reduce NGV incremental costs
for consumers without sacrificing the efficacy of the certification process.17 Reduced OBD
Testing Requirements - Under Tier 3: The Tier 3 NPRM proposes that EPA "upgrade" its OBD
regulations "to be consistent with the latest CARB regulations." These CARB OBD
requirements are similar to the EPA requirements for full OEM testing and have similarly been
cited as "the most costly and time-consuming requirement in the certification process" for NGV
aftermarket conversions in the California market.19 Compared to the streamlined EPA OBD
testing requirements for SVM converters, which usually require a handful of demonstrations that
can be reviewed in a week, the CARB process currently requires "more monitors to be adjusted,
more tests for each of these monitors to be conducted, lots of documentation on the adjustments
to monitors, and many more test results to be reported," resulting in application packages of 200-
300 pages long that take CARB about 3 months to review.20 This adds considerable expense as
well as time to the certification process, and this barrier to entry has limited the availability of
NGV conversions in California.
For this reason, CARB is working with stakeholders to develop less extensive OBD testing
requirements for small-volume converters that will be aligned with EPA SVM requirements
while still ensuring effective emissions performance. Similar to EPA's rules for SVM converters,
the new CARB rules would require as little as four demonstrations of OBD systems. 21
Consistent with EPA's judgment in allowing reduced OBD test requirements for converters,
CARB notes that this reduced testing burden will still preserve emissions benefits of their
emissions program. EPA should ensure that the Tier 3 rules incorporate these and other changes
that CARB makes to streamline its own regulations.
CARB's proposed rule changes will significantly reduce barriers to entry in California; however,
like the current EPA OBD rules, they apply only to aftermarket converters. In order to
effectively harmonize these rules with existing EPA definitions as well as to provide a level
4-216
-------
Tier 3 Summary and Analysis of Comments
playing field between SVMs and OEM small-volume test groups, the Tier 3 rules should specify
that any reduced aftermarket OBD requirements introduced by CARB will be applicable to OEM
small-volume test groups as well as SVM converters up to the current 15,000 vehicle thresholds
for each.
Our Response:
EPA met with the commenter in October 2013 to get a better understanding of the issues
raised in their comments.30 EPA proposed that the new OBD provisions not apply to the small
businesses and SVMs (as defined in the rule) until the 2022 model year. Provisions for OEM
small volume test groups were not included. We understand the commenter's view that OBD is
potentially more costly for small volume test groups and could be a deterrent to a manufacturer
decision to develop these configurations. EPA is willing to consider reduced OBD requirements
for small volume test groups of OEMs (perhaps on a transitional or interim basis). However,
EPA cannot before the fact commit to applying the changes under consideration in California.
Commenters' footnotes:
15 Environmental Protection Agency. "Clean Alternative Fuel Vehicle and Engine Conversions Final
Rule: Webinar for Conversion Manufacturers." 27 April 2011.
http://www.epa.gov/otaq/consumer/fuels/altfuels/conversions-fr-webinar.pdf
16 National Petroleum Council. "Future Transportation Fuels: Natural Gas Analysis." 28 May 2013. pp.
53-58. http://www.npc.org/FTF-report-080112/Chapter 14-Natural Gas-052813.pdf
17 Existing EPA regulations (40 CFR 86.1806-01) already afford flexibility in OBD testing requirements
and allow EPA to reduce them if requested to do so by the applicant, particularly in the case of alternative
fuel vehicles. EPA should encourage both SVM converters and OEMs certifying NGVs in small-volume
test groups to request and utilize these reduced OBD requirements.
18 Environmental Protection Agency. "Control of Air Pollution From Motor Vehicles: Tier 3 Motor
Vehicle Emission and Fuel Standards; Proposed Rule." Federal Register, 21 May 2013, p. 29903.
19 Ward, Peter F, on behalf of Alternative Fuels Advocates LLC. Letter to Annette Hebert, Chief Mobile
Source Operations Division, California Air Resources Board. 22 Feb 2013.
20 Carmichael, Tim, on behalf of California Natural Gas Vehicle Coalition. Letter to Annette Hebert,
Chief Mobile Source Operations Division, California Air Resources Board. 8 Mar 2013.
21 California Air Resources Board. "Public Workshop: Proposed Amendments to the Alternative Fuel
Conversion Certification Procedures for New and In-Use Vehicles and Engine." 1 May 2013.
http://www.arb.ca.gov/msprog/onroad/altfuelconv/5-l-2013_Alt%20Fuels%20Wkshp%20V8.pdf
4.4.7 Requirements for Alternative Fuel Vehicle Converters
What Commenters Said:
Organization: National Propane Gas Association (NPGA)
EPA is seeking changes to its regulations that address fuel specifications for gasoline and vehicle
emission standards for certain types of vehicles. The changes would essentially harmonize
EPA's regulations with those of the California Air Resources Board's (CARB) Low Emission
30 Passavant, G. (October, 2013). EPA and VNG.CO Meeting on Tier 3 NPRM. Memorandum to the docket.
4-217
-------
Chapter 4: Proposed Vehicle Emissions Program
Vehicle III (LEV III) program. Implementation would be phased-in over a period of time
beginning from the year 2017 and running thru 2025.
From a broader perspective, NPGA disagrees with the basic premise that EPA should seek to
harmonize its regulations with the CARB LEV III program. NPGA does agree with EPA's stated
intention of allowing manufacturers to be able to sell their product in all 50 states. However,
NPGA argues that the reverse should occur, i.e. CARB should harmonize its regulations with the
federal regulations.
CARB's regulations are typically more stringent than EPA's, and it's unclear as to whether all of
their changes over the years have met the level of administrative approval required by EPA.
What is clear, though, is the higher costs to certify to CARB's regulations, which have served as
a major deterrent for alternative fuel converters to make a business case for participation in the
California market. EPA has even acknowledged that there is a significant cost burden to comply
with CARB's OBD requirements, for example. As a result, can an economic justification be
made (by CARB or EPA within this NPRM) for the emissions benefits lost due to this cost
avoidance?
NPGA believes there is no demonstrated benefit (costs or otherwise) to using CARB's
regulations over EPA's regulations. CARB's regulations pertaining to alternative fuels have not
been updated since the 1990s. While NPGA understands they are working on changes to the
regulations, those changes remain in development and have yet to be finalized. NPGA simply
believes that EPA's alternative fuel regulations are easier to manage from a cost/benefit
perspective compared to CARB's.
In 2011, NPGA lauded EPA for promulgating a series of changes that streamlined the
regulations for certifying alternative-fueled vehicles. This action was a culmination of years of
effort on the part of both the alternative-fuel industry and EPA in recognition that such changes
would increase the availability of clean-burning vehicles, and, thus, have the effect of reducing
pollution. These changes have had an overall positive impact on the industry and
environmentally, as well, since that time.
NPGA would caution EPA to be mindful of changes to its emission regulations that would result
in more conventional fueled vehicles put into operation than alternative fueled vehicles (as has
occurred in California). Similarly, any changes put in place should not offset the gains in
pollution reduction from EPA's 2011 rulemaking that has resulted in more alternative-fueled
vehicle certifications.
Our Response:
EPA acknowledges the NGPA comment regarding the value and wisdom of harmonized
programs and their view that CARB should harmonize with EPA. We are sensitive to this
concern, but the overwhelming sense of the commenters was that an EPA Tier 3 program
harmonized with the CARB LEV III program was superior to separate requirements whenever
possible. EPA agrees with those commenters who supported a harmonized program to the
greatest degree possible. However, there are a few places where the programs are not fully
4-218
-------
Tier 3 Summary and Analysis of Comments
harmonized and the OBD requirements for alternative fuel converters are one of these situations.
The thrust of the commenters concerns related to OBD seemed to be that EPA was adopting
CARBs requirements including those for OBD for alternative fuel converters. This is not the
case. EPA adopted small alternative fuel converter provisions in 2011 and the converters will
still be able to meet the OBD requirements using the provisions of 40 CFR 85 Subpart F. We are
aware of the work ongoing at CARB, but it is not yet final.31 As CARB updates its OBD
regulations in the future EPA will consider these changes and propose to adopt them, if
appropriate.
4.5. Emissions Test Fuel ("Cert Fuel")
4.5.1. Proposed Changes to Gasoline Emissions Test Fuel
4.5.1.1. Ethanol Content of Emissions Test Fuel
What Commenters Said:
Organization: Advanced Biofuels Association (ABFA)
ABFA is an advocate for "Technology Neutrality" throughout the regulatory system and believes
that biofuels should be allowed to compete freely in the market. Most of the fuel and fuel
additive regulatory system at US EPA Office of Transportation Air Quality (OTAQ) over the
years has been written to address the market of the first generation biofuel, ethanol, which has
been the only cost effective widely available biofuel. Ethanol has been and will continue to be a
valuable commercial product for the nation's fuel supply and energy independence. However,
new biofuel molecules are beginning to enter the market for blending with gasoline. New
biofuels under development by ABFA member companies and others provide multiple
commercial and societal benefits such as high energy density, low RVP and improved emissions
parameters. In addition, many of these fuels would not require any new infrastructure
investments to reach the market and grow into a significant contributor to the nation's energy
independence.
EPA requests comment on proposing to change the certification test fuel to El5 and supports this
by stating that the requirements in the Renewable Fuel Standard (RFS-2) will reflect El 5 as the
fuel in the US marketplace in the next 10-15 years. (78 Federal Register 98, 29909). This
statement is driving towards a specific molecule and is picking winners and losers in the
marketplace. The RFS-2 requires that the US fuel supply contain 36 billion gallons of renewable
fuel by 2022 which can be achieved by drop in molecules with more energy density. When the
RFS-2 was created in 2007, fuel consumption in the US was rising, the NHSTA efficiency
standards had not been promulgated, and the US government had great hope for the consumer
acceptance of E85. Since 2007, fuel consumption in the US has decreased significantly and will
continue to decrease based on CAFE standards and other factors. E85 consumption reached 38.6
million gallons in 2011 and consumption has remained flat since then. This change to the EPA
31 See http://www.arb.ca.gov/regact/2013/altfuel2013/altfuel2013.htm downloaded on February 19. 2014 for more
information.
4-219
-------
Chapter 4: Proposed Vehicle Emissions Program
certification test fuel will not achieve the RFS-2 requirements. US EPA creating a technology
neutral position and allowing the marketplace to pick the winners is the correct approach. This is
working today, through the higher RIN values driving the market to work with more energy
dense fuel to fill the based D6 RIN pool.
In addition, concern for the implications of this rule with respect to vehicle certification, ABFA's
concerns with proposed changes in the certification test fuel extend particularly to the use of this
same fuel as the baseline for Part 79 fuel and fuel additive registration.
EPA requests comment on a plan to allow the use of El 0 as the certification test fuel with a
transition scheme to E15 in the future. (78 Federal Register 98, 29910) would create confusion in
the marketplace by changing the base fuel for Part 79 registration. Advanced biofuel companies
are working to gain the data set to provide to US EPA to register their fuels for on road use. The
change in base fuel will add extra costs, through further toxicity testing and research to comply
with Tier I and Tier II requirements, without adding value to the companies or environmental
benefits. The base fuel should remain a non additive fuel without oxygen to allow for the least
regulatory burden for todays and future fuel/engine company complying with the US EPA
regulatory requirements.
Many companies have already invested significant sums in research and development, facility
citing, engineering design, or equipment purchase. Of particular concern to ABFA is what signal
granting a specific molecule for certification fuel will have on the RFS and would have on the
plans to begin breaking ground on a number of advanced biofuels plants. Several ABFA member
companies, including Dynamic Fuels, Neste Oil, KiOR, Gevo, Lanzatech, UOP, Solazyme and
others have already made significant capital investments to build these production facilities that
are up and running at a commercial scale today. Our regulations need to stay technology neutral
to provide a marketplace that has not pre-picked the winners and losers.
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
Automakers support updating certification test fuels in Tier 3 regulations, as California did in its
recent LEV III rulemaking.
Automakers support updating and expanding Federal Certification Test Fuels (hereafter "test
fuels"). We advocate that EPA's final rule implement a Tier 3 E10 9 psi RVP test fuel as the
gasoline test fuel (18) rather than an El5 test fuel. There are only about 20 U.S. retail outlets for
El5 to date, and there are numerous other uncertainties affecting El5 fuel use. Trying to be
"forward looking" by assuming significant national market penetration of E15 is too speculative
a basis to require that El 5 be the federal test fuel for 2017 and beyond.
EPA should make the final Tier 3 gasoline test fuel E10 9 psi RVP.
Organization: Mercedes-Benz USA, LLC on behalf of Daimler AG
4-220
-------
Tier 3 Summary and Analysis of Comments
The primary factors which directly impact the Mercedes-Benz GHG Compliance Plan are sulfur
content in market fuel, and ethanol content and octane level in certification and market fuel.
Organization: American Coalition for Ethanol (ACE)
Ethanol content - We strongly support EPA's proposal that emissions test fuel, which is also
commonly referred to as 'certification fuel,' needs to be updated to account for the present-day
reality that the most popular in-use fuel in the U.S. (E10) contains ethanol, and under the RFS,
higher blends of ethanol, including but not limited to El 5, will become the rule and not the
exception.
Indolene, the current certification fuel, is not and never has been useful in the practical sense
because motorists never used it in vehicles. Updating the federal emission test fuel to better
match today's in-use fuel and to reflect the reality that the RFS will call for an ever-increasing
level of ethanol in gasoline is the appropriate step to take. We strongly support El 5 as the new
certification fuel, and suggest that if a phased-in process is chosen, it be done in such a way that
it provides incentives for manufacturers that chose to optimize for El 5 at an earlier date.
What is critical is that the certification fuel no longer act as an implicit impediment to higher
ethanol blends. We expect ethanol blends to increase from El5 to even higher level blends to
capture the economic value and clean air benefits associated with these higher-level blends. As a
result, it is critical that EPA institute a certification fuel regimen that is easily adaptable to higher
level blends.
Organization: American Fuel & Petrochemical Manufacturers (AFPM)
The AFPM supports a new certification fuel. Indolene should be replaced with E10, not E15.
Almost all gasoline today is E10. It is speculative to expect that E15 will be the predominant
fuel in 2017. The current certification fuel in California is E10, and one of the stated goals of
this proposal is harmonization with California standards.
Organization: American Honda Motor Co., Inc.
Testing on El 5:
EPA is proposing that El 5 become the standard test fuel based upon, for example, anticipated
market acceptance, the continued implementation of RFS, and the adoption of this standard by
the retailing industry. While it is true that EPA has approved El5 for use in newer vehicles, this
highly controversial decision has seen numerous detractors. Furthermore, the RFS could be
implemented in any number of ways, including: a) a higher percentage of E85 in the market,
combined with E10, or b) a higher percentage of drop-in biofuels in the market, combined with
E10, and/or higher E85 sales. Test fuel should represent the market, not forecast the market.
Additionally, Test fuel should be harmonized as much as possible with California - one of the
overriding goals of the regulation. We believe it is premature and inappropriate to require as the
test fuel a market fuel which is, for all intents and purposes, not in the market today. We
recognize that E10 is the standard fuel in the market today, and believe it would be reasonable
and appropriate to make E10 the standard fuel for Tier 3. We also recommend that EPA adopt an
4-221
-------
Chapter 4: Proposed Vehicle Emissions Program
approach to handle the period in which market fuels are in transition. For example, if and when
El 5 becomes more than 1/3rd (one-third) of all gasoline sales, then EPA will adjust the Tier 3
rules and adopt the fuel with appropriate phase-in testing requirements. Perhaps EPA is over-
correcting for a previous, perceived error. E10 was a common market fuel available at the time
of Tier 2, and yet EO remained the test fuel. It is counter to EPA's prior practice to require El 5,
an anticipated - but in no way certain - market fuel, to become its new, de facto standard.
Organization: American Lung Association
Update Certification fuel to in-use fuel. In separate decisions announced in October of 2010 and
January in 2011, EPA granted a waiver request to major manufacturers of ethanol to increase the
allowable limit of ethanol in gasoline to 15 percent starting for vehicle model years 2001 and
after. Although EPA has increased the permissible amount of ethanol in gasoline to 15 percent,
this fuel is still not widely available in the marketplace. Most gasoline sold today contains up to
10 percent ethanol by volume (E10). We believe that the certification fuel should match the fuel
being sold in the market. Based on current gasoline sales, E10 should be the certification fuel.
We urge EPA to adopt an approach that gives the agency the flexibility to update and match the
certification fuel with the current market fuel without further rulemaking. Under such an
approach, perhaps a triggering event such as the suggested 30 percent market share of gasoline
sold with fifteen percent ethanol (El5), could prompt EPA to change the certification fuel. Two
model years of lead time for such a switch should be sufficient time for the auto manufacturers to
accommodate any such change. EPA should continue to have flexibility to make modifications to
certification fuel specifications as appropriate.
Organization: American Motorcyclist Association (AMA)
Changing the certification fuel to El5 or E30 is at odds with the 22 million motorcycles and all-
terrain vehicles currently in use, not to mention the legacy fleet of cars, boats, lawnmowers,
generators and hundreds of millions of small engines in commerce today. None, of these vehicles
and engines is designed to operate on fuel with more than 10 percent ethanol.
Automobile and motorcycle manufacturers must certify that on-highway vehicles produced will
meet applicable U.S . Environmental Protection Agency and National Highway Traffic Safety
Administration emissions, fuel economy and safety requirements prior to selling vehicles. The
fuel vehicles must use for this requirement is called the certification fuel. The current
certification fuel is EO - that is, fuel that has no ethanol content.
The current certification fuel should not be changed to reflect 'forward-looking' assumptions
about what 'could become a major gasoline blend over the next 10-15 years,' such as El5 or
E30. These changes would be contrary to both the letter and spirit of the law.
Currently, the risks of increasing El 5 in the marketplace will negatively impact every American.
Since the EPA used only one test to determine whether El 5 is safe for vehicles before granting a
waiver, the AMA urges the agency to allow for an independent scientific study of E15. We also
request that motorcycles and ATVs be included in the study.
4-222
-------
Tier 3 Summary and Analysis of Comments
The AMA has expressed concerns about El5 being mistakenly used and damaging engines in
motorcycles and ATVs, and about the continued availability of gasoline that has no ethanol, or
gasoline with only a 10 percent blend, that is safe for use in motorcycles and ATVs.
Organization: American Petroleum Institute (API) and the Association of Fuel &
Petrochemical Manufacturers (AFPM)
Certification Fuel Recommendations:
EPA proposes to change the ethanol content of the gasoline certification test fuel from 0% (EO)
to 15% (El 5) by volume. The certification fuel should be representative of a gasoline/ethanol
blend that is currently prevalent in the market place; that fuel is one which contains 10% ethanol
(E10) by volume.
EPA does not state its legal basis for designating test fuels, and incorrectly asserts sweeping
discretion under the statute. EPA makes the forward-looking prediction that the Renewable Fuels
Standard will lead to an expansion of El 5, despite the limited commercial availability of the fuel
today. EPA needs to set certification fuel standards that reflect current driving conditions, and
the Agency would be in violation of the Clean Air Act if it specifies an ethanol content above
10% for gasoline certification fuel.
Emissions certification fuel:
EPA proposes to change the gasoline test fuel for light-duty and heavy-duty vehicles from EO
fuel to E15 fuel. 78 Fed. Reg. at 29825. EPA does not state its legal basis for designating test
fuels, and instead asserts in a conclusory way that "we believe we have discretion under the
statute to transition from EO to E15 test fuel...." Id. at 29910.
EPA presents essentially no factual basis for this proposed change to El 5. According to EPA,
"[i]n-use gasoline has changed considerably since EPA's fuel specifications for emissions testing
of light- and heavy-duty gasoline vehicles were last set and first revised." Id. at 29908. EPA
predicts that the second iteration of the federal renewable fuels standard program ("RFS2") will
lead to further changes in in-use fuel, including "expansion of the number of retailers that offer
E15." Id. In response to changes in in-use fuel that EPA alleges have already occurred, as well as
"forward-looking" predictions about the ethanol and sulfur content of future in-use fuels, EPA
proposes to update the gasoline test fuel provisions of 40 C.F.R. § 1065.710. Id. EPA proposes to
make this change despite its admission that "El 5 is only commercially available at a limited
number of fuel retailers" at present. Id. at 29909. EPA's prediction about increased El 5 fuel
availability in the future is premised on "instability in crude oil pricing and growing RFS2
renewable fuel requirements." Id.
EPA seeks comment on its proposed approach of changing the test fuel from EO to El 5,
including the "forward-looking nature" of this proposal. Id. at 29910. Additionally, EPA seeks
comment on potential alternative approaches, including designation of E10 as the test fuel and
later "transition[] to El 5 as the market further transitions to El 5 in use." Id.
EPA should specify E10 as the certification fuel. The CAA requires EPA "to insure that vehicles
are tested under circumstances which reflect the actual current driving conditions under which
4-223
-------
Chapter 4: Proposed Vehicle Emissions Program
motor vehicles are used, including conditions relating to fuel, temperature, acceleration, and
altitude." CAA § 206(h). Accordingly, test fuels must "reflect current driving conditions." Id.
(emphasis added).
In keeping with this clear statutory requirement, we agree with EPA that the certification fuel
should be switched to an ethanol containing blend because ethanol blends are the most prevalent
type of gasoline currently in the market place. However, we disagree with the proposed selection
of El 5. As was stated in the RIA, most gasoline in the United States contains 10% ethanol by
volume. The certification fuel should represent the most common grade of fuel sold, which is
E10.
We disagree with EPA's proposal to select a fuel that is "forward looking with respect to the
maximum gasoline ethanol concentration Tier 3 vehicles could expect to encounter." Id. This
creates two fundamental legal problems.
First, establishing a "forward looking" test fuel violates the statute because a "forward looking"
fuel such as El5 does not accurately or reasonably reflect the "current" fuels used by affected
vehicles. The plain language of the statute does not permit EPA to substitute the phrase "forward
looking" for "current." EPA's own analysis shows that E10 is the most prevalent ethanol blend
in the market today. Thus, E10 is the only ethanol blend that may be specified as a test fuel at
this time.
Second, even if the statute could be construed (arguendo) as authorizing EPA to set a "forward
looking" test fuel, EPA has not put forward adequate factual justification to do so in this case.
EPA's proposed "forward looking" E15 test fuel is based on the assertion that E15 "could
become a major gasoline blend over the next 10-15 years." Id. at 29909. However, there are no
data or analyses in the proposal or underlying record that support this prediction. Absent such
factual support, adopting El5 as a test fuel would be arbitrary and capricious.
Notably, even if the Agency attempted to assemble factual justification for a "forward looking"
El 5 test fuel, it could not do so. Given the lack of announced El 5 compatible vehicles,
automobile manufacturer warranty statements, lack of refueling infrastructure and the 15 to 18
year timeframe to turn over the vehicle fleet, E10 will continue to reflect "current driving
conditions" over the timeframe under consideration by EPA.
In light of these problems, the only legally-viable course would be to select E10 as the
certification fuel and transition to El 5 if and when El 5 become the most prevalent fuel in the
market. We support a market review at some point in the future after 2017 to gauge El 5 usage
and growth projections. This review could coincide with the technology review for the CAFE
standards. The certification fuel should not switch to El5 until it becomes the dominant fuel in
the marketplace.
Lastly, if EPA ultimately decides to switch to El 5, that switch must be accomplished through
notice and comment rulemaking. EPA suggests in the proposal that such a switch might be
accomplished automatically in the future by establishing "a "trigger point" (e.g., 30 percent of
gasoline is El5) in the Tier 3 final rule to prompt an automatic move to El5 after a certain period
4-224
-------
Tier 3 Summary and Analysis of Comments
of time, e.g., two or three years." Id. at 29910. EPA alternatively suggests that it "could simply
set a future date (e.g., 2020) with sufficient time for transitioning to El5 test fuel." Id. Neither of
these approaches is legally viable.
Switching to El5 based on a 30% trigger would require a future factual determination that the
30% trigger has been exceeded. Such a determination constitutes "factual data" that must be set
out in a proposed rule before EPA may take a final action based on those data. CAA § 307(d)(3).
Switching to El 5 at a fixed point in the future based on a current prediction of when El 5 will
become the prevalent fuel is problematic because, as explained above, EPA has not set out
sufficient data or analyses to justify a prediction as to when El 5 might become the most
prevalent gasoline. Thus, the only viable way to establish El5 as a test fuel is to do so through
notice and comment rulemaking at the point when El5 become the prevalent ethanol blend in the
market.
EPA proposes to increase ethanol content from zero to 15 volume percent. EPA believes that this
level is forward-looking with respect to the maximum gasoline ethanol content Tier 3 vehicles
could expect to encounter. Concerns with vehicle fuel system compatibility still need to be
addressed in advance of its introduction into commerce. This level of ethanol is also not aligned
with CARB LEV III test fuel which is set at 10 volume percent. To ensure the most effective
evaporative emissions control system for in-use operation we would suggest that EPA considers
setting the ethanol content at 10 volume percent and then adjusting the RVP requirement of the
certification fuel to account for the allowed 1 psi waiver. Only until vehicle manufacturers all
warrant their vehicles for E15, and E15 represents a dominant portion of the overall market
(including California) would it would be appropriate for EPA to consider shifting to an El 5
certification fuel while ensuring no loss of effectiveness of the emission control systems.
EPA has failed to explain the legal basis for its test fuel proposal. EPA asserts that the primary
legal authorities for the proposed rule are CAA §§ 202, 206, and 211. 78 Fed. Reg. at 29828-29.
EPA rulemaking pursuant to these provisions is subject to the requirements of CAA § 307(d).
See § 307(d)(l)(E), and (K). Pursuant to § 307(d)(3), EPA is required to include in the proposal
the "statement of its basis and purpose" for the action, which must include "the major legal
interpretations ... underlying the proposed rule."
Yet, EPA does not include any discussion in the proposal of the legal basis for the proposed test
fuel provisions. For example, there is no explanation of what section or sections of the statute
authorize it to designate a new test fuel. Similarly, there is no discussion of the scope and extent
of the Agency's authority to establish test fuels and specify particular parameters that such fuels
must meet. The Agency simply asserts that "we believe we have discretion under the statute to
transition from EO to E15 test fuel." 78 Fed. Reg. at 29910.
Because EPA has not provided any explanation of the statutory authority on which it relies for
the proposed test fuel provisions, API has no opportunity to comment on that key issue and any
rule promulgated would be contrary to the Clean Air Act and unlawful. To try and resolve this
problem, EPA must re-propose the rule to provide an opportunity to comment on the Agency's
legal basis for designating a new gasoline test fuel.
4-225
-------
Chapter 4: Proposed Vehicle Emissions Program
EPA has failed to consider fully the CAA § 21 l(f) implications of its test fuel proposal. Section
21 l(f) limits the fuels and fuel additives that a manufacturer may lawfully "introduce into
commerce, or [] increase the concentration in use of." Pursuant to § 21 l(f)(l) and (2), only those
fuels and fuel additives that are "substantially similar to any fuel or fuel additive utilized in the
certification of any model year 1975, or subsequent model year, vehicle or engine under section
[206 of the CAA]" may enter commerce for general use in light duty vehicles, or for use by any
person in any motor vehicles, manufactured model year 1974 or later. (Emphasis added.) EPA
may grant a § 21 l(f)(4) waiver from these commercial limitations if the requested waiver "will
not cause or contribute to a failure of any emission control device or system (over the useful life
of the motor vehicle, motor vehicle engine, nonroad engine or nonroad vehicle in which such
device or system is used) to achieve compliance by the vehicle or engine with the emission
standards with respect to which it has been certified under pursuant to sections [206 and
As EPA explains in the preamble to the proposed rule, pursuant to § 21 l(f)(4), EPA granted a
partial waiver for use of El 5 by light-duty vehicles model year 2007 and later, and then extended
the waiver to include model year 2001-2006 light-duty vehicles. 78 Fed. Reg. at 29909 n. 320
(citing 75 Fed. Reg. 68094 (Nov. 4, 2010) and 76 Fed. Reg. 4662 (Jan. 26, 201 1)); see also id. at
2991 1. EPA also concluded at the time that E10 was not a certification fuel for purposes of
determining whether mid-level blends could be put into commerce under the authority of §
21 l(f)(l) (under a "substantially similar" determination) rather than pursuant to a waiver issued
under § 21 l(f)(4). 75 Fed. Reg. at 68143. In issuing those partial waivers, EPA placed conditions
designed to, among other things, minimize potential misfueling. EPA complemented those
conditions through a later rule, known as the "El 5 Misfueling Mitigation Measures Rule," which
included misfueling prohibition, fuel pump labeling, PTDs, and ongoing implementation survey
requirements as a "direct and efficient way to further reduce the potential for misfueling and the
emission increases that would result from misfueling." 76 Fed. Reg. 44406, 4441 1 (July 25,
201 1) (emphasis added); see also 78 Fed. Reg. at 2991 1. EPA stated that these additional
requirements were directed to "El 5 that is introduced into commerce in accordance with the
partial waivers" and thereby operated "collectively and in tandem with the partial waiver
conditions [to] maximize the likelihood that El 5 is used only in motor vehicles covered by the
partial waivers and minimize the potential for emissions increases that might otherwise occur."
Id. (emphasis added).
EPA proposes to establish El 5 as a certification fuel for purposes of implementing the Tier 3
standards. This raises two legal issues that EPA has failed to address in the proposal: (1) will
establishing El 5 as a certification fuel authorize El 5 to be put into commerce pursuant to §
21 l(f)(l); and (2) if so, what effect does this have on the previously-issued El 5 partial waivers
and corresponding misfueling mitigation rule? EPA's failure to address these key questions
violates its rulemaking obligations under § 307(d)(3) and renders the proposed rule arbitrary and
capricious due to the Agency's failure to identify and address key policy and legal implications
of designating El 5 as a certification rule.
On the question of whether establishing E15 as a certification fuel authorizes E15 to be put into
commerce pursuant to § 21 l(f)(l), language in the preamble seems to suggest that in EPA's
view, once EPA designates El 5 as a test fuel under § 206, El 5 could be introduced into
4-226
-------
Tier 3 Summary and Analysis of Comments
commerce only for the vehicle type(s) for which El 5 were specifically authorized for use as a
test fuel. For example, manufacturers could introduce into commerce El5 for use into heavy-
duty vehicles without a § 21 l(f)(4) waiver "[i]f. . . new heavy-duty gasoline vehicles or engines
begin testing on E15 for certification." 78 Fed. Reg. at 29911. Under this implied legal analysis,
commenters are left to guess about many key aspects of this important issue. For example, does
EPA believe this is how the statute must be interpreted? If not, what other interpretations did
EPA consider and why were alternative interpretations rejected?
For example, some might suggest that, under Section § 21 l(f)(l), if E15 were designated a new §
206 test fuel, El5 could be introduced into commerce for use in any vehicle without a § 21 l(f)(4)
waiver. This view might be supported by the assertion that Section 21 l(f)(l) does not appear to
place any limitation on the vehicle types for which a fuel can be introduced into commerce once
that fuel is designated as a test fuel under § 206. Therefore, assuming arguendo that El 5 were
designated as a test fuel only for light-duty vehicles, it could still be introduced into commerce as
a fuel for use in any vehicle under § 21 l(f)(l). We would view such an approach as not
authorized by the Clean Air Act and therefore unlawful, yet, EPA does not explain how the
sweeping language of § 21 l(f)(l) can or should be construed as being as highly constrained as it
appears to suggest. Thus, EPA's treatment of this key issue falls far short of the Agency's §
307(d)(3) obligation to clearly set out "the major legal interpretations and policy considerations
underlying the proposed rule."
As importantly, EPA also essentially failed to explore the obvious and important implications of
introducing E15 into commerce under § 21 l(f)(l) with regard to the previously-issued E15
partial waivers and misfueling mitigation rule. For example, if E15 may be put into commerce
under § 21 l(f)(l), do the previously-issued E15 partial waivers continue to have relevance,
meaning, or legal applicability? Would the waivers essentially be rendered moot (if E15 may be
introduced into commerce under § 211(f)(l) by virtue of E15 becoming a certification fuel), and
does EPA have authority to limit the types of engines/vehicles that may use El 5, as it did under
§ 21 l(f)(4) by issuing the so-called "partial waivers"? Similarly, does the misfueling mitigation
rule have any force or effect, given that it was issued for purposes of facilitating implementation
of the El 5 partial waivers? If it does remain in effect, is the misfueling mitigation rule adequate
given that it was designed to be implemented in conjunction with the misfueling mitigation
measures required to be implemented as a condition of using the E15 partial waivers? These are
just a few examples of the many important questions that arise by virtue of designating E15 as a
certification fuel.
We certainly reserve our ability to challenge each and every one of these issues and the most that
EPA says on these important issues is in its discussion of the potential use of El 5 in heavy-duty
vehicles, where the Agency notes that the potential for misfueling in heavy-duty vehicles would
be addressed in a future action. 78 Fed. Reg. at 29911. Thus, EPA's treatment of this key issue
falls far short of the Agency's § 307(d)(3) obligation to clearly set out "the major legal
interpretations and policy considerations underlying the proposed rule." As a result, EPA must
re-propose the rule and set forth a clear interpretation of §§ 206, 21 l(f)(l), and 21 l(f)(4). EPA
must explain what effect the proposed rule would have on the El 5 partial waivers and misfueling
mitigation rule and propose provisions to fill any potential regulatory gaps that are created by
designating E15 a certification fuel. And, EPA must also address the major policy considerations
4-227
-------
Chapter 4: Proposed Vehicle Emissions Program
that flow from these key legal issues. EPA's proceeding with a final test fuel rule for El5 at this
time would clearly be unlawful under the Clean Air Act. We seriously question whether El 5 can
be lawfully made a test fuel under current circumstances for the reasons we have stated but
certainly the Agency must do a far more thorough job of legal analysis in a re-proposal before it
could seriously propose El5 as a test fuel.
The EPA should set a new certification fuel as E10, not El5. It will be likely the predominant
fuel in the marketplace.
Also on the certifying in a higher octane and higher ethanol content fuel, EPA should not go
ahead of the requisite research and collaboration necessary to set new fuel specifications. And
I'm out of time, so thank you for doing ppms.
Organization: Appalachian Mountain Club (AMC)
EPA proposes to update the federal emission test fuel to better reflect real world fuel. We
generally support this concept and believe EPA should replace Indolene with the currently
available ethanol blend E10. Higher ethanol blends have yet to be widely used in the marketplace
and can impact greenhouse gas and fuel economy standards. EPA should use one consistent test
fuel and consider the impacts across all of these regulations.
Organization: BMW of North America, LLC
Harmonization of certification fuel is particularly crucial as with the new standards and
increasing product complexity and regulation stringency manufacturers have to bear increased
test burden. The proposed Tier 3 standard requires a different certification fuel than the
certification fuel adopted by CARS's LEV III program. In the interest of a 50-state certification
and to avoid double testing, BMW proposes that EPA adopt an E10 fuel as required by CARB
and allow in-use testing with the harmonized fuel.
Organization: BP Products North America Inc.
Establish E10 as Emissions Test Fuel- BP supports a change in EPA's federal emissions test fuel
from Indolene to a fuel representative of the predominant, in-use gasoline. Recognizing the huge
uncertainties underlying ElS's potential market penetration and the predominance of E10 in the
market today, BP recommends that EPA adopt E10 as the test fuel for the foreseeable future.
If/when El 5 were to become the predominant, in-use gasoline, then and only then, should EPA
adopt El5 as the test fuel.
Organization: California Air Resources Board (CARB)
CARB staff disagrees with U.S. EPA's decision to propose 15 percent by volume ethanol
gasoline (El5) as the Tier 3 certification test fuel.
El 5 is currently used in a handful of fueling stations nationwide and its further penetration into
the marketplace is uncertain. Given the practical considerations inherent in providing a refueling
4-228
-------
Tier 3 Summary and Analysis of Comments
infrastructure for El5, it may not proliferate very rapidly into the marketplace. On the other
hand, E10 has been tested thoroughly as a viable fuel for all vehicles currently on the road and,
as a result, will likely be widely used for the foreseeable future. However, should El 5 constitute
a significant fraction of the federal commercial fuel market at some future date, it would be
appropriate at that time for U.S. EPA to reconsider the composition of Tier 3 certification fuel.
Organization: Chevron Products Company
We are opposed to the designation of E15 as a certification fuel given that it is not representative
of the fuel in wide use in the market and it is not compatible with most of the vehicles in the
existing fleet.
The proposal to shift the certification fuel to El 5 ignores the many difficulties in marketing
E10+ blends to the existing vehicle fleet. CRC has produced compelling data that demonstrate
the incompatibility of E15 with millions of 2001+ vehicles. In addition, E15 remains illegal in
California and will remain so for the foreseeable future. It is not clear when, if ever, El 5 will be
a practical product option. The potential for El0 and its higher RVP to continue to be the
predominant gasoline used by Tier 3 vehicles creates a disconnection between certification
requirements and real in-use emissions. Therefore, a more prudent path for EPA would be to
adopt an El 0 certification fuel at this time, deferring an E15 fuel until market conditions justify
the change.
Organization: Chrysler Group LLC
Gasoline Emissions Test Fuel:
- Regulations that specify a fuel for testing purposes should match the fuel predominantly sold in
the U.S., which we believe should be at least 70% prevalent in the marketplace.
- The certification test fuel for gasoline-fueled vehicles should be an E10, 9.0 psi RVP test fuel,
rather than an El5 certification test fuel.
- The exhaust and evaporative durability test fuel should also be representative of the
predominate fuel in the market, which is E10.
- EPA should adopt a mechanism to accurately determine the point in the future at which El 5
becomes 70% prevalent in the marketplace, building in appropriate lead time.
Definition of Commercially Available Fuel and Resultant Federal Test Fuel Requirements:
Chrysler believes that test fuels should match the predominant fuel in the marketplace. Matching
test fuel to marketplace fuel will ensure vehicles deliver the performance and environmental
benefits in the real world as they were certified. Chrysler thus believes that test fuel used in
certification testing and also in durability mileage accumulation should be representative of the
predominant fuel in the market. This includes the ethanol content of the gasoline.
Accordingly, Chrysler supports updated Tier 3 test fuel requirements for gasoline-fueled vehicles
to match test fuel to fuel that is predominate in the marketplace, but objects, however, to EPA's
proposal to require use of 15 percent ethanol by volume (El 5) as the test fuel for certification
testing beginning Model Year 2017. Requiring use of E15 would conflict with the objective to
match test fuel to in-use fuel, because El5 is not the predominate fuel in the market. In addition,
4-229
-------
Chapter 4: Proposed Vehicle Emissions Program
Chrysler is concerned that EPA has not defined the term "commercially available," as used in the
fuel requirements for evaporative emissions durability testing. This concern is broadened where
EPA, in a similar context, uses the phrases "commercial gasoline that will be generally available
through retail outlets", and "national average in-use fuels" to describe fuels for exhaust or
evaporative durability service accumulation. Further, EPA, in seeking comment on the
appropriateness of the alternative test fuel provisions at 40 C.F.R. § 1065.701(c), which also
refers to fuels that would be "readily available nationwide" and implies this to mean
"commercially available." Chrysler finds these terms add considerable uncertainty when
determining appropriate certification and durability test fuel requirements, and proposes that
EPA adopt a single definition to represent the predominate fuel in the market that would apply to
all test fuels.
To the degree EPA seeks to interpret El 5 as (1) "commercially available", (2) a "national
average in-use fuel", or (3) "readily available nationwide', Chrysler believes that E15, which is
not currently available to the vast majority of American vehicle owners, is neither "commercially
available", "national average in-use fuel", nor "readily available nationwide". In order for EPA
to require that any specific fuel blend, such as El5, be used in certification testing or durability
testing, that fuel blend must be representative of the predominate fuel in the market.
As EPA acknowledges in the proposed rule, El 5 fuel is not representative of the currently
prevailing gasoline in the marketplace. Instead, E10 is the predominate gasoline in the market.
Moreover, there is no basis to believe that E15 will displace E10 as the predominate market
gasoline in the foreseeable future. Although Chrysler recognizes EPA's attempts to be "forward
looking" in selecting an appropriate long-term test fuel, the Agency's unsupported assumptions
of significant market penetration of E15 are too speculative at this point to support specifying the
use of El 5 as the federal test fuel for Model Year 2017 and beyond.
Chrysler believes that EPA should define one term to describe the fuel that is predominate in the
marketplace and therefore appropriate for fuels used in regulatory testing. In general, the
meaning of the term "commercially available" in a legal or regulatory context is highly
dependent on the context in and purpose for which it is used. For example, an item might be
considered "commercially available" in certain contexts—where the objective is simply to
ensure that the product can be obtained—if it can be purchased from an online provider, even if
there is only one such provider and even if such provider is located in a geographically distant
location. In the context of EPA's fuel requirements for durability testing, however, it makes little
sense to define "commercially available" in this fashion. Rather, in the durability testing context,
the term "commercially available" is intended to ensure that testing is representative of real
world in-use conditions — i.e., to ensure that manufacturers are conducting durability testing
with a type of fuel that is representative of fuel that is actually being used in the marketplace.
This ensures that durability in testing is likely to assure durability of cars in actual use.
Obviously, gasoline cannot be ordered on the Internet or from a distant location. Rather, the
ability to obtain the fuel has to involve ready access nearby to the routes of cars throughout the
country.
EPA appears to agree with this interpretation of the term "commercially available" in the
preamble to the proposed rule:
4-230
-------
Tier 3 Summary and Analysis of Comments
First, under the alternative test fuels provisions in 40 C.F.R. § 1065.701(c), a manufacturer must
show that its proposed alternative test fuel is "commercially available." EPA states in the
preamble to the proposed rule that, pursuant to this provision, manufacturers can "petition the
Administrator for approval of a higher octane, higher ethanol content test fuel if they could
demonstrate that such a fuel. . . would be readily available nationwide. . . ." In other words, in
the preamble, EPA interprets the term "commercially available" in the alternative test fuel
context — with the same goal of using a fuel representative of real world conditions — to mean
"readily available nationwide." Chrysler supports this interpretation.
Second, EPA states in the preamble to the proposed rule regarding the requirement in 40 C.F.R.
§ 86.1823-08(d) that manufacturers use fuel with the highest ethanol content "commercially
available" for purposes of evaporative emissions durability testing (7). With respect to
this requirement, EPA states that "as El5 in-use fuel becomes progressively more available, we
would expect that El 5 service accumulation fuel would be used for whole vehicle evaporative
durability programs." Again, EPA indicates that the term "commercially available" excludes
fuels, such as El5, that are currently available only at select service stations in limited states.
Even with this interpretation that "commercially available" means "readily available
nationwide," Chrysler believes that EPA must also specify the threshold for which a fuel is
"readily available nationwide" in order to ensure that the industry has a clear understanding of
what is required. In that light, the Onboard Refueling Vapor Recovery (ORVR) rules may
provide a useful analogy for determining when market conditions have changed such that El 5
fuel may be deemed "commercially available" or "readily available nationwide." Under the
Clean Air Act, EPA has authority to waive certain requirements of the Stage II vapor recovery
program when EPA finds that ORVR systems are in "widespread use" in the highway vehicle
fleet. Because Stage II vapor recovery systems on gasoline pumps and ORVR systems on
highway vehicles perform the same function (i.e., capture ambient vapors emitted during
refueling), the reasoning goes, there would be little environmental benefit gained by requiring
Stage II vapor recovery systems on gasoline pumps if ORVR systems on highway vehicles are
already in "widespread use." On May 9, 2012, EPA determined that ORVR was in "widespread
use" and therefore waived the requirement that certain current and former ozone nonattainment
areas implement Stage II vapor recovery systems on gasoline pumps. In reaching this
determination, EPA relied in part on its marketplace finding that "75 percent of gasoline will be
dispensed to ORVR-equipped vehicles by April 2012." Therefore, a "widespread use" threshold
of 70% to express "commercial availability" or "readily available nationwide" appears to be a
reasonable threshold to determine commercial availability of gasoline for the purpose of defining
test fuels, and is consistent with our proposal to have such test fuels be the predominate fuel sold
in the marketplace.
El5, which is not available except at twenty service stations across the United States, is not
representative of the predominant market fuel in the U.S., nor is it commercially available for
virtually all cars in the United States today. It is unclear when EPA might consider El 5 to
become "commercially available" such that it must be used for service accumulation during
durability testing. EPA should clarify that the term "commercially available" should mean
"readily available nationwide," consistent with EPA's apparent interpretation of the term in the
preamble to the proposed rule, and thus representative of the predominant fuel in the
4-231
-------
Chapter 4: Proposed Vehicle Emissions Program
marketplace. Today, only approximately twenty of the more than 120,000 retail outlets in the
United States, or 0.017%, offer E15 fuel (13). Even E85, which is generally still not considered
to be widely available in the marketplace, is available at approximately 2,350 service stations, or
one hundred times as many stations as El 5 (14). In the context of evaporative testing
requirements to test on "highest ethanol content commercially available", E85 does not meet the
test of availability. Not only is El 5 not the prevailing market fuel, it should not be considered
"commercially available," since only a tiny fraction of motorists can get it. It is unreasonable to
consider a fuel "commercially available" if that fuel is only available at select service stations in
a limited number of states. Accordingly, it is inappropriate for EPA to specify El 5 as the test
fuel for purposes of any of the motor vehicle testing requirements.
Recommendation: Chrysler recommends that EPA define the fuel requirements for certification
and evaporative emissions durability testing to mean the predominant fuel in the market, which
we believe should be at least 70% of the fuel sold in the U.S. marketplace. We also recommend
that EPA adopt a mechanism to determine, based on objective criteria and sufficient lead time,
the point in the future at which the predominate fuel in the marketplace reaches that threshold.
EPA proposes to adopt an El5 certification test fuel requirement for gasoline-fueled vehicles
(15). However, El 5 is not representative of the predominate fuel in the marketplace, which is
E10. In fact, E15 is not even "commercially available" to the vast majority of American vehicle
owners. As such, it is unreasonable and inappropriate to specify El5 as the certification test fuel.
Chrysler supports EPA's goal of specifying test fuel that best matches gasoline that the American
public's vehicles actually use (16). Chrysler also supports EPA's goal of continuing to monitor
the relationship between test fuel and in-use fuel so that testing is undertaken with fuel that
represents real-world conditions. E10 is the fuel that the American public's vehicles use, because
it is the predominate fuel in the market; as such, EPA should designate E10, not E15, as the
proper certification test fuel. Further, matching the ethanol content of certification test fuel to
fuel used in the real world is important because the combustion of higher ethanol content
gasoline (for example, of El 5 versus E10, and both as compared to Indolene) tends to decrease
volatile organic compound ("VOC") and carbon monoxide ("CO") emissions, but tends to
increase oxides of nitrogen ("NOX") emissions (17). Accordingly, adopting a higher ethanol
content gasoline for certification testing is not necessarily "conservative," because increasing the
ethanol content of gasoline affects pollutants differently.
Organization: Consumers Union;Consumers Union (comment campaign)
While Consumers Union supports EPA's intention to match the certification fuel with the fuel in
the market, we recommend an alternative plan in case the market for El 5 does not overcome the
significant barriers to its development and materialize by 2017. Even if E10 continues to be the
dominant fuel in the market, testing vehicles for fuel economy purposes at El 5 would be more
accurate than the current EO certification fuel. Adjustment factors can always be used to account
for predicted differences (as they are now). However, uniform adjustment factors after the fact
are never as precise as using the market fuel to begin with; automakers that maximize efficiency
for E15 may distort fuel economy expectations of consumers using E10 at the pump. Therefore,
4-232
-------
Tier 3 Summary and Analysis of Comments
we would recommend that E10 be the default fuel, unless El5 comes to dominate the market, at
which point a reasonable lead time could be provided to switch the certification fuel again.
Organization: Countrymark; Small Business Refiners (SBRs)
Adoption of a 15 percent ethanol by volume (El 5) certification test fuel for gasoline-fueled
vehicles.
EPA recommends matching today's in-use gasoline federal emission testing fuel standard by
replacing 'Indolene' (EO) fuel with 15% ethanol blend (El 5). Justification for the change is to
better match today's in-use gasoline.
According to EIA data, only eight fueling stations across three states were selling El5 by the end
of 2012. As a comparison, more than 159,000 fueling stations were in service at the end of 2012.
EIA and DOE both believe that El 5 penetration into the market is not likely to occur in
significant volume.
In addition, the change to El 5 is not supported by the oil industry, automotive industry, or
consumer groups such as AAA because El 5 brings several new problems to the market place. A
few significant problems are listed below:
- A higher blending ratio of ethanol is more corrosive, potentially resulting in engine damage or
a shorter engine and fuel system operating life.
- Containment failures cause greater environmental damage because ethanol promotes mixing of
gasoline and water; which would not readily occur without ethanol being present.
- Ethanol cannot be transported via pipeline like other transportation fuels. Ethanol must be
shipped by rail or truck; resulting in additional investments for unloading and storage capacity.
- Transporting ethanol by rail or truck increases greenhouse gas emissions, reducing the claimed
benefit of using renewable fuels.
- Blending El 5 instead of E10 reduces the amount of butane that can be blended into gasoline.
This directly reduces the refinery's profitability and restricts flexibility in our operations because
we need to find another disposition for butane instead of blending with gasoline.
Organization: Countrymark
CountryMark is not interested in producing El 5 and recommend that the fuel standard be
changed to a 10% (E10) blend of ethanol to represent today's in-use gasoline based on fuel
demand across the country.
CountryMark started blending ethanol at ten percent long before being obligated under the RFS
because it made economic sense. We have not sold CBOB (Conventional Blend stock for
Oxygenate Blending) or RBOB (Reformulated Blend stock for Oxygenate Blending) to support
E15 since it was approved as a fuel option. CountryMark is not planning to incorporate E15 as a
major volume product into our portfolio of offerings to our customers. We do not see sufficient
demand to justify changing operations at our refinery to meet El 5 specifications.
Organization: Small Business Refiners (SBRs)
4-233
-------
Chapter 4: Proposed Vehicle Emissions Program
The SBRs are opposed to a requirement to produce E1S and recommend that the fuel standard be
changed to a 10% (E10) blend of ethanol to represent today's in-use gasoline based on fuel
demand across the country.
Organization: E.I. du Pont de Nemours and Company
Of particular interest to DuPont is EPA's proposal to set E-15 as certification fuel. Renewable
fuels have a bright future in the U.S. and the Tier 3 proposed rule includes a number of
provisions that can support making renewable fuels more widely available and economically
competitive. Expanding renewable fuels will ultimately drive reductions in air pollutants that are
the target of the Tier 3 rule.
Setting E-15 as certification fuel is also the baseline for evaluating the health effects of the fuel
and associated fuel additives, a mandatory part of the Part 79 registration process. DuPont
suggests that EPA recognize this in the final rule and address how this fact affects existing and
future registrations.
First, DuPont supports EPA's proposal to set E-15 as emissions test fuel for light-duty cars and
trucks as well as heavy-duty gasoline vehicles. EPA must create the requisite incentives for auto
manufacturers to pursue advanced engine technologies. Setting E-15 as test fuel is a step in the
right direction.
EPA is requesting comment on alternative approaches to implementing E-15 as test fuel. Two of
the options include requiring E-15 to be available on the market before transit!oning to E-15 as
test fuel. Instead, DuPont suggests that EPA set a specified future date for transit!oning to E-15
test fuel. This approach would ensure E-15 as test fuel providing certainty for fuel producers and
auto manufacturers, and provide the requisite incentives to automakers to advance engine
technologies.
Organization: ExxonMobil
If the Agency decides to proceed with this rulemaking, we recommend that EPA: specifies E10
as the new vehicle certification fuel.
Organization: General Motors LLC (GM)
The basic foundation of today's Federal Certification Test Fuels (hereafter "test fuels") has been
in place and relatively unchanged since 1975. GM supports updating and expanding the test fuel
to be aligned with current predominate market fuel. We advocate that EPA's final Tier 3 rule
specify an E10 9 psi RVP test fuel gasoline rather than the proposed El5 test fuel that does not
represent the market. There are only about 20 U.S. retail outlets selling El5 to date and most of
these do not offer El 5 during the summer. Because there are many uncertainties about the further
expansion of El 5, EPA's "forward looking" approach assumes significant national market
penetration of El 5 that is too speculative for something as important as the Federal test fuel for
2017 and beyond.
4-234
-------
Tier 3 Summary and Analysis of Comments
As mentioned above, transit!oning to an El 5 test fuel is too speculative for something as
important as the Federal test fuel for 2017 and beyond. If there needs to be a provision to
recognize El 5 in the future, then GM suggests a two-step approach.
Organization: Private Citizen
However, I do not support the new proposal for El5 gasoline since this fuel has not yet been
adequately tested on non-flex fuel gasoline engines, or proven safe for motorcycles or other
small engines. Besides, we already have 10% ethanol (E10) which is enough to help reduce
tailpipe emissions without creating excessive problems due to ethanol's incompatibility with fuel
system materials and designs in older vehicles. Ethanol-blend fuels reduce fuel economy (miles
per gallon) due to lower energy content, and they often cause damage to fuel systems... doing
this in the name of cleaner air just does not make sense to me. There's got to be a better way to
deal with the ethanol issue, and I urge you to please oppose the introduction of El 5 without
adequate testing.
Organization: Growth Energy
Growth Energy Strongly Supports an El5 Certification Fuel. Growth Energy saw the need for
additional market space for the higher biofuel volumes called for under the RFS and
subsequently filed the Green Jobs waiver for El 5 with the Agency in 2009. In January 2011, the
Agency approved the waiver for 2001 and newer light duty vehicles. While the critics have done
everything in their power to prevent higher ethanol blends in the marketplace, EPA has
appropriately proposed that all 2017 and newer vehicles should be certified on E15 - clearly the
fuel of the near future. Nearly the entire American vehicle fleet fuels on E10 today rather than
EO. Now that EPA has approved E15 and the U.S. Supreme Court has denied cert on the critics'
lawsuits to prevent the implementation of the El 5 waiver, it only makes sense that the
automakers and the American public move forward with what EPA views as a major
transportation gasoline blend in the near term. With the approach of the RFS blend wall, it only
makes sense that we will begin to see more El5 in the marketplace. Additionally, more than 70
percent of the vehicles on the road have been approved by EPA for use with El 5 (model years
2001 and newer), and two major automobile manufacturers, Ford and General Motors, are
already warrantying their newer vehicles for El5 - GM for model years 2012 and 2013 and Ford
for model year 2013 (Oil Price Information Service, October 2, 2012 "Ford and GM Okay E15
Blends for Newer Vehicles").
Organization: INEOS Bio
EPA requests comment on various ways to transition the emissions test fuel, including: starting
with a transition to E10 with a transition to E15 as that fuel becomes more widely available on
the market; transit!oning to El5 after a few years to allow time for this greater market
availability or by setting a date certain by which the transition should be made to help drive El 5
market availability; or, allowing "vehicle manufacturers to request approval for an alternative
certification fuel, such as a high-octane 30 percent ethanol by volume (E30) for vehicles they
might design or optimize to use such a fuel" so that it may "help manufacturers that wish to raise
4-235
-------
Chapter 4: Proposed Vehicle Emissions Program
compression ratios to improve vehicle efficiency, as a step towards complying with the 2017 and
later light-duty greenhouse gas and CAFE standards."
INEOS Bio commends EPA's efforts to transition the certification test fuel for emissions from
light duty cars and trucks and heavy duty gasoline vehicles ("the emissions test fuel") to better
reflect the current and future in-use fuel as it begins to contain greater volumes of biofuels
resulting from the continued development and commercialization of biofuels and increased RFS
volumes in the market.
INEOS Bio believes that EPA should work to set the emissions test fuel to promote the highest
achievable level of ethanol and octane in the U.S. fuel supply. Encouraging the highest
achievable level of biofuels would help meet EPA's overall goal in this rulemaking to address
the impacts of motor vehicles and fuels on air quality and public health. (11) For instance,
cellulosic ethanol from waste in comparison to gasoline contains no sulfur and has over 100%
reduced GHGs. In addition, ethanol combusts without producing air toxics, which are the main
source of particulate matter. Blending ethanol in gasoline also reduces the need for unhealthy
detergent additives which are mandated to reduce the formation of engine deposits from gasoline
that increase exhaust emissions and result in the loss of fuel economy and performance. These
benefits of ethanol only rise with higher blends, so INEOS Bio encourages EPA to issue a final
rule that would encourage and drive the highest achievable blend which we feel in this given
market would be El5.
Organization: International Council on Clean Transportation (ICCT)
While the ICCT supports using a more representative fuel for certification testing, El5 is not
representative of in-use fuel. El 5 can cause damage if it is used in small engines or in legacy
vehicles. El 5 is also specific to ethanol, which encourages the use of food feedstocks instead of
more environmentally friendly feedstocks. Finally, El 5 provides significant evaporate cooling,
which manufacturers could exploit to generate higher fuel economy on the tests than the vehicles
actually experience in use. The ICCT recommends that the test fuel use E10.
The ICCT is concerned about the proposed revision to use El5 for the certification test fuel.
While the ICCT supports using a more representative fuel for certification testing, El5 is not
representative of in-use fuel. E10 is representative of current in-use fuels and should be used for
emission and fuel economy testing.
In addition to violating the principle of using representative fuels, the ICCT has a number of
serious concerns with El 5: El 5 can cause damage if it is used in small engines or in legacy
vehicles; it is specific to ethanol (as opposed to drop-in biofuel pathways), which encourages the
use of food feedstocks such as maize instead of more environmentally friendly feedstocks; and it
provides significant evaporative cooling, which manufacturers could exploit to generate higher
fuel economy on the tests than the vehicles would actually experience in use.
The problems and potential damage if El 5 is used in small engines and legacy vehicles has been
well documented by Honda and other vehicle manufacturers and will not be repeated here.
However, our concerns in this area are exacerbated by the lack of systems to provide proper fuel
4-236
-------
Tier 3 Summary and Analysis of Comments
and prevent misfueling in-use. For example, EPA has not proposed a system to separate the use
of E15 for newer vehicles and E10 for older vehicles. A system where E15 is used for regular
fuel and E10 for premium fuel would encourage misfueling of small engines and legacy vehicles,
as customers choose El 5 just because it is cheaper. If E10 is used for regular fuel and El 5 is
used for premium fuel, this would discourage the use of E15 in vehicles that could safety use it
and would make it impossible for older vehicles requiring premium to be refueled properly.
Thus, for the refueling system to work properly, it appears that service stations must provide
separate pumps for both E10 and E15 for both regular and premium. This is not likely to happen
unless EPA requires it. Until EPA addresses the refueling situation with regulations, refueling
will almost certainly be marked by confusion and misfueling.
The ICCT is also concerned that El 5 will encourage biofuels made from food crops, instead of
advanced biofuels. It is not currently cost effective to make ethanol out of cellulose, thus simply
increasing the ethanol blend wall is effectively a mandate for more food-based biofuels that can
easily be turned into ethanol. Several pathways for producing advanced biofuel from feedstocks
such as cellulose will likely be able to deliver drop-in fuels. As drop-in fuels have a higher value
in the market than ethanol, it is also possible that cellulose can be more profitably turned into
drop-in fuels. El5 would work against this by incentivizing ethanol, not drop-in fuels. Insofar as
wider use of El 5 supports increased use of maize and sugarcane ethanol rather than driving
investment into advanced drop-in fuel pathways, the net result will be that El5 would tend to
increase food prices (and hence worldwide food insecurity), decrease biodiversity due to land use
change, create yet another roadblock for advanced, lower carbon, biofuels and deliver less
carbon reductions than would be available from commercializing biofuels from cellulosic wastes
and energy crops.
The ICCT's third concern is specific to fuel economy testing. 15% ethanol content provides
significant evaporative cooling in the cylinder. This would allow manufacturers to advance
ignition timing or make other modifications to improve fuel economy on the test cycles—and
which would not be likely to be achieved in-use.
The advantages of even E10 were demonstrated in a test program to maximize engine power on a
variety of fuels. Grassroots Motorsports (December 2012) tested a Mazda Miata on the following
fuels, using a standalone computer to tune the vehicle for each fuel on a Dynapack dynamometer
to make the most power: The charge cooling effect of E10 boosted performance on 93-octane
fuel by about 1.5% and virtually matched the performance of 100 and 105-octane race gasoline.
GRM couldn't redesign the engine and they were only able to revise spark timing, camshaft
timing, and air/fuel ratio at WOT. The higher charge cooling with El5 would allow
manufacturers to do even more optimizing—and they could also optimize the design of the
engine over all operation.
It is important that the fuel used for certification and fuel economy testing be representative of
in-use fuel. However, that fuel is currently E10, which is what should be used for EPA's testing.
E15 is not currently representative of in-use fuel and its use for EPA testing has several major
problems, which should be avoided.
Organization: Marathon Petroleum Company LP (MFC)
4-237
-------
Chapter 4: Proposed Vehicle Emissions Program
EPA must also change the proposed certification fuel to a 10% ethanol blend with the 1 psi
waiver that is more representative of current and future transportation gasoline.
EPA proposes to change the gasoline test fuel for light-duty and heavy-duty vehicles from EO
fuel to E15 fuel. 78 Fed. Reg. at 29825. EPA does not state its legal basis for designating test
fuels, and instead asserts in a conclusory way that "we believe we have discretion under the
statute to transition from EO to El5 test fuel...." Id. at 29910.
EPA presents essentially no factual basis for this proposed change to E15. According to EPA,
"[i]n-use gasoline has changed considerably since EPA's fuel specifications for emissions testing
of light- and heavy-duty gasoline vehicles were last set and first revised." Id. at 29908. EPA
predicts that the second iteration of the federal renewable fuels standard program ("RFS2") will
lead to further changes in in-use fuel, including "expansion of the number of retailers that offer
E15." Id. In response to changes in in-use fuel that EPA alleges have already occurred, as well as
"forward-looking" predictions about the ethanol and sulfur content of future in-use fuels, EPA
proposes to update the gasoline test fuel provisions of 40 C.F.R. § 1065.710. Id. EPA proposes to
make this change despite its admission that "El 5 is only commercially available at a limited
number of fuel retailers" at present. Id. at 29909. EPA's prediction about increased El 5 fuel
availability in the future is premised on "instability in crude oil pricing and growing RFS2
renewable fuel requirements." Id.
EPA seeks comment on its proposed approach of changing the test fuel from EO to El 5,
including the "forward-looking nature" of this proposal. Id. at 29910. Additionally, EPA seeks
comment on potential alternative approaches, including designation of El 0 as the test fuel and
later "transition to El5 as the market further transitions to El5 in use." Id.
EPA should specify E10 as the certification fuel. The CAA requires EPA "to insure that vehicles
are tested under circumstances which reflect the actual current driving conditions under which
motor vehicles are used, including conditions relating to fuel, temperature, acceleration, and
altitude." CAA § 206(h). Accordingly, test fuels must "reflect current driving conditions." Id.
(emphasis added).
In keeping with this clear statutory requirement, we agree with EPA that the certification fuel
should be switched to an ethanol containing blend because ethanol blends are the most prevalent
type of gasoline currently in the market place. However, we disagree with the proposed selection
of El 5. As was stated in the RIA, most gasoline in the United States contains 10% ethanol by
volume. The certification fuel should represent the most common grade of fuel sold, which is
E10. Likewise, the octane rating should coincide with the dominant grade which is 87 AKI
regular unleaded. The only exception should be for engines that require premium unleaded fuel
as stipulated in the vehicle owners' manual.
We disagree with EPA's proposal to select a fuel that is "forward looking with respect to the
maximum gasoline ethanol concentration Tier 3 vehicles could expect to encounter." Id. This
creates two fundamental legal problems.
4-238
-------
Tier 3 Summary and Analysis of Comments
First, establishing a "forward looking" test fuel violates the statute because a "forward looking"
fuel such as El5 does not accurately or reasonably reflect the "current" fuels used by affected
vehicles. The plain language of the statute does not permit EPA to substitute the phrase "forward
looking" for "current". EPA's own analysis shows that E10 is the most prevalent ethanol blend
in the market today. Thus, E10 is the only ethanol blend that may be specified as a test fuel at
this time.
Second, even if the statute could be construed (arguendo) as authorizing EPA to set a "forward
looking" test fuel, EPA has not put forward adequate factual justification to do so in this case.
EPA's proposed "forward looking" E15 test fuel is based on the assertion that E15 "could
become a major gasoline blend over the next 10-15 years." Id. at 29909. However, there are no
data or analyses in the proposal or underlying record that support this prediction. Absent such
factual support, adopting El5 as a test fuel would be arbitrary and capricious.
Notably, even if the Agency attempted to assemble factual justification for a "forward looking"
El 5 test fuel, it could not do so. Given the lack of announced El 5 compatible vehicles,
automobile manufacturer warranty statements, lack of refueling infrastructure and the 15 to 18
year timeframe to turn over the vehicle fleet, E10 will continue to reflect "current driving
conditions" over the timeframe under consideration by EPA.
In light of these problems, the only legally-viable course would be to select E10 as the
certification fuel and transition to El 5 if and when El 5 become the most prevalent fuel in the
market. We support a market review at some point in the future after 2017 to gauge El 5 usage
and growth projections. This review could coincide with the technology review for the CAFE
standards. The certification fuel should not switch to El5 until it becomes the dominant fuel in
the marketplace.
Lastly, if EPA ultimately decides to switch to El 5, that switch must be accomplished through
notice and comment rulemaking. EPA suggests in the proposal that such a switch might be
accomplished automatically in the future by establishing "a "trigger point" (e.g., 30 percent of
gasoline is El5) in the Tier 3 final rule to prompt an automatic move to El5 after a certain period
of time, e.g., two or three years." Id. at 29910 EPA alternatively suggests that it "could simply
set a future date (e.g., 2020) with sufficient time for transitioning to El5 test fuel." Id. Neither of
these approaches is legally viable.
Switching to El5 based on a 30% trigger would require a future factual determination that the
30% trigger has been exceeded. Such a determination constitutes "factual data" that must be set
out in a proposed rule before EPA may take a final action based on those data. CAA § 307(d)(3).
Switching to E15 at a fixed point in the future based on a current prediction of when E15 will
become the prevalent fuel is problematic because, as explained above, EPA has not set out
sufficient data or analyses to justify a prediction as to when El 5 might become the most
prevalent gasoline. Thus, the only viable way to establish El5 as a test fuel is to do so through
notice and comment rulemaking at the point when and if El 5 becomes the dominant ethanol
blend in the market.
4-239
-------
Chapter 4: Proposed Vehicle Emissions Program
EPA proposes to increase ethanol content from zero to 15 volume percent. EPA believes that this
level is forward-looking with respect to the maximum gasoline ethanol content Tier 3 vehicles
could expect to encounter. Concerns with vehicle fuel system compatibility still need to be
addressed in advance of its introduction into commerce. This level of ethanol is also not aligned
with CARB LEV III test fuel which is set at 10 volume percent. To ensure the most effective
evaporative emissions control system for in-use operation we would suggest that EPA considers
setting the ethanol content at 10 volume percent and then adjusting the RVP requirement of the
certification fuel to account for the allowed 1 psi waiver. Once the manufacturers all warrant
their vehicles for E15, and E15 represents a dominant portion of the overall market (including
California) then it would be appropriate for EPA to consider shifting to an El 5 certification fuel
while ensuring no loss of effectiveness of the emission control systems.
Legal justification for new test fuel - EPA has failed to explain the legal basis for its test fuel
proposal. EPA asserts that the primary legal authorities for the proposed rule are CAA §§ 202,
206, and 211. 78 Fed. Reg. at 29828-29. EPA rulemaking pursuant to these provisions is subject
to the requirements of CAA § 307(d). See § 307(d)(l)(E), and (K). Pursuant to § 307(d)(3), EPA
is required to include in the proposal the "statement of its basis and purpose" for the action,
which must include "the major legal interpretations ... underlying the proposed rule."
Yet, EPA does not include any discussion in the proposal of the legal basis for the proposed test
fuel provisions. For example, there is no explanation of what section or sections of the statute
authorize it to designate a new test fuel. Similarly, there is no discussion of the scope and extent
of the Agency's authority to establish test fuels and specify particular parameters that such fuels
must meet. The Agency simply asserts that "we believe we have discretion under the statute to
transition from EO to E15 test fuel." 78 Fed. Reg. at 29910.
Because EPA has not provided any explanation of the statutory authority on which it relies for
the proposed test fuel provisions, MFC has no opportunity to comment on that key issue and any
rule promulgated would be contrary to the Clean Air Act and unlawful. To try to resolve this
problem, EPA must re-propose the rule to provide an opportunity to comment on the Agency's
legal basis for designating a new gasoline test fuel.
EPA has failed to consider fully the CAA § 21 l(f) implications of its test fuel proposal. Section
21 l(f) limits the fuels and fuel additives that a manufacturer may lawfully "introduce into
commerce, or [] increase the concentration in use of." Pursuant to § 21 l(f)(l) and (2), only those
fuels and fuel additives that are "substantially similar to any fuel or fuel additive utilized in the
certification of any model year 1975, or subsequent model year, vehicle or engine under section
[206 of the CAA]" may enter commerce for general use in light duty vehicles, or for use by any
person in any motor vehicles, manufactured model year 1974 or later. (Emphasis added.) EPA
may grant a § 21 l(f)(4) waiver from these commercial limitations if the requested waiver "will
not cause or contribute to a failure of any emission control device or system (over the useful life
of the motor vehicle, motor vehicle engine, nonroad engine or nonroad vehicle in which such
device or system is used) to achieve compliance by the vehicle or engine with the emission
standards with respect to which it has been certified under pursuant to sections [206 and
4-240
-------
Tier 3 Summary and Analysis of Comments
As EPA explains in the preamble to the proposed rule, pursuant to § 21 l(f)(4), EPA granted a
partial waiver for use of El 5 by light-duty vehicles model year 2007 and later, and then extended
the waiver to include model year 2001-2006 light-duty vehicles. 78 Fed. Reg. at 29909 n. 320
(citing 75 Fed. Reg. 68094 (Nov. 4, 2010) and 76 Fed. Reg. 4662 (Jan. 26, 2011)); see also id. at
29911. EPA also concluded at the time that E10 was not a certification fuel for purposes of
determining whether mid-level blends could be put into commerce under the authority of §
21 l(f)(l) (under a "substantially similar" determination) rather than pursuant to a waiver issued
under § 21 l(f)(4). 75 Fed. Reg. at 68143. In issuing those partial waivers, EPA placed conditions
designed to, among other things, minimize potential misfueling. EPA complemented those
conditions through a later rule, known as the "El 5 Misfueling Mitigation Measures Rule," which
included misfueling prohibition, fuel pump labeling, PTDs, and ongoing implementation survey
requirements as a "direct and efficient way to further reduce the potential for misfueling and the
emission increases that would result from misfueling." 76 Fed. Reg. 44406, 44411 (July 25,
2011) (emphasis added); see also 78 Fed. Reg. at 29911. EPA stated that these additional
requirements were directed to "El 5 that is introduced into commerce in accordance with the
partial waivers" and thereby operated "collectively and in tandem with the partial waiver
conditions [to] maximize the likelihood that El5 is used only in motor vehicles covered by the
partial waivers and minimize the potential for emissions increases that might otherwise occur."
Id. (emphasis added).
EPA proposes to establish El 5 as a certification fuel for purposes of implementing the Tier 3
standards. This raises two legal issues that EPA has failed to address in the proposal: (1) will
establishing El5 as a certification fuel authorize El5 to be put into commerce pursuant to §
21 l(f)(l); and (2) if so, what effect does this have on the previously-issued El 5 partial waivers
and corresponding misfueling mitigation rule? EPA's failure to address these key questions
violates its rulemaking obligations under § 307(d)(3) and renders the proposed rule arbitrary and
capricious due to the Agency's failure to identify and address key policy and legal implications
of designating El 5 as a certification rule.
On the question of whether establishing El 5 as a certification fuel authorizes El 5 to be put into
commerce pursuant to § 21 l(f)(l), language in the preamble seems to suggest that in EPA's
view, once EPA designates El5 as a test fuel under § 206, El5 could be introduced into
commerce only for the vehicle type(s) for which El 5 were specifically authorized for use as a
test fuel. For example, manufacturers could introduce into commerce El5 for use into heavy-
duty vehicles without a § 21 l(f)(4) waiver "[i]f. . . new heavy-duty gasoline vehicles or engines
begin testing on E15 for certification." 78 Fed. Reg. at 29911. Under this implied legal analysis,
commenters are left to guess about many key aspects of this important issue. For example, does
EPA believe this is how the statute must be interpreted? If not, what other interpretations did
EPA consider and why were alternative interpretations rejected?
For example, some might suggest that, under Section § 21 l(f)(l), if E15 were designated a new §
206 test fuel, El 5 could be introduced into commerce for use in any vehicle without a § 21 l(f)(4)
waiver. This view might be supported by the assertion that Section 21 l(f)(l) does not appear to
place any limitation on the vehicle types for which a fuel can be introduced into commerce once
that fuel is designated as a test fuel under § 206. Therefore, assuming arguendo that El5 were
designated as a test fuel only for light-duty vehicles, it could still be introduced into commerce as
4-241
-------
Chapter 4: Proposed Vehicle Emissions Program
a fuel for use in any vehicle under § 21 l(f)(l). We would view such an approach as not
authorized by the Clean Air Act and therefore unlawful, yet, EPA does not explain how the
sweeping language of § 21 l(f)(l) can or should be construed as being as highly constrained as it
appears to suggest. Thus, EPA's treatment of this key issue falls far short of the Agency's §
307(d)(3) obligation to clearly set out "the major legal interpretations and policy considerations
underlying the proposed rule."
As importantly, EPA also essentially failed to explore the obvious and important implications of
introducing El5 into commerce under § 21 l(f)(l) with regard to the previously-issued El5
partial waivers and misfueling mitigation rule. For example, if E15 may be put into commerce
under § 21 l(f)(l), do the previously-issued E15 partial waivers continue to have relevance,
meaning, or legal applicability? Would the waivers essentially be rendered moot (if El 5 may be
introduced into commerce under § 211(f)(l) by virtue of E15 becoming a certification fuel), and
does EPA have authority to limit the types of engines/vehicles that may use El 5, as it did under
§ 21 l(f)(4) by issuing the so-called "partial waivers?" Similarly, does the misfueling mitigation
rule have any force or effect, given that it was issued for purposes of facilitating implementation
of the E15 partial waivers? If it does remain in effect, is the misfueling mitigation rule adequate
given that it was designed to be implemented in conjunction with the misfueling mitigation
measures required to be implemented as a condition of using the E15 partial waivers? These are
just a few examples of the many important questions that arise by virtue of designating E15 as a
certification fuel.
We certainly reserve our ability to challenge each and every one of these issues and the most that
EPA says on these important issues is in its discussion of the potential use of El 5 in heavy-duty
vehicles, where the Agency notes that the potential for misfueling in heavy-duty vehicles would
be addressed in a future action. 78 Fed. Reg. at 29911. Thus, EPA's treatment of this key issue
falls far short of the Agency's § 307(d)(3) obligation to clearly set out "the major legal
interpretations and policy considerations underlying the proposed rule." As a result, EPA must
re-propose the rule and set forth a clear interpretation of §§ 206, 21 l(f)(l), and 21 l(f)(4). EPA
must explain what effect the proposed rule would have on the El 5 partial waivers and misfueling
mitigation rule and propose provisions to fill any potential regulatory gaps that are created by
designating E15 a certification fuel. And, EPA must also address the major policy considerations
that flow from these key legal issues. EPA's proceeding with a final test fuel rule for El5 at this
time would clearly be unlawful under the Clean Air Act. We seriously question whether El 5 can
be lawfully made a test fuel under current circumstances for the reasons we have stated but
certainly the Agency must do a far more thorough job of legal analysis in a re-proposal before it
could seriously propose El5 as a test fuel.
Organization: Maryland Department of the Environment
When reviewing the use of ethanol in the certification fuel, we would recommend that EPA, in
the final rule, specify E10 as the test fuel. This recommendation is based on data at this time that
shows that El 5 is only a small segment of the U.S. fuels market, it currently is not use in
Maryland, and it is questionable at this point if this fuel will see wide spread automotive use. In
addition, in the past MDE has voiced its concern over the use of El 5 based on its increased
volatility which leads to increased ozone emissions particularly in the summer months.
4-242
-------
Tier 3 Summary and Analysis of Comments
Organization: American Energy Alliance (comment campaign)
One overlooked part of this proposed rule is that it also changes the "vehicle certification fuel" to
El5 (gasoline that is 15 percent ethanol). This is a stealthy way for EPA to get much more wide-
spread adoption of El 5 fuel at gas stations, a mandate that threatens engines and forces
Americans to use a less efficient fuel.
Organization: National Association of Clean Air Agencies (NACAA)
Use of Ethanol in Certification Fuel — NACAA supports the concept of ensuring that fuels used
for certification purposes should more closely reflect real-world fuels. Given the prevalence of
E10 (gasoline containing 10 percent ethanol by volume) in the marketplace, we encourage EPA
to replace the currently used indolene with E10 as an emissions certification fuel. We are
concerned that EPA has proposed to require that El5 be used as certification fuel. At this time,
E15 is only a small segment of the U.S. fuels market and the prospect of it becoming a widely
used automotive fuel in the future is questionable. Further, use of El 5 in the existing vehicle
fleet will have adverse impacts on vehicle emissions (e.g., aldehydes orNOx). Therefore, we
recommend that in the final Tier 3 rule EPA specify El0 as a test fuel and continue to monitor
the use of E15 in the marketplace. Should use of E15 become significant, the agency can then
consider the pros and cons of revising test fuel specifications to include E15.
Organization: National Association of Convenience Stores (NACS) and Society of
Independent Gasoline Marketers of America (SIGMA)
The Proposal would make El5 the default test fuel for most non-flex-fuel vehicles ("FFVs"). At
the present time, E10 is the predominant fuel in the market. It appears unlikely that El5 will gain
significant market share in the immediate future. NACS and SIGMA do not oppose making El 5
the test fuel provided that vehicles certified on El5 do not experience significant deterioration in
their performance when they run on E10. If such performance problems do occur, the Agency
must consider delaying designating E15 as the default test fuel, and, instead, make E10 the test
fuel until El 5 is more widely available and utilized.
By developing specifications for ethanol-based fuels, and requiring other potential fuels of the
future to petition the Agency for approval as a test fuel, the Proposal could entrench ethanol's
current role in the fuel supply, and impose artificial impediments to new fuels coming to market.
Certification Fuels:
The Proposal would update the specifications of the certification test fuel with which vehicles
demonstrate compliance with emissions standards. For non-FFVs, the Proposal moves away
from "Indolene" ("EO") to an El5 test fuel. In addition, the Proposal would for the first time
impose detailed specifications for the E85 emissions test fuel used for FFV certification. Finally,
the Proposal would allow vehicle manufacturers to request approval for an alternative
certification fuel for vehicles they might design or optimize for use on such a fuel.
Non-FFVs:
4-243
-------
Chapter 4: Proposed Vehicle Emissions Program
The Proposal designates El5 as the test fuel for most non-FFVs to demonstrate compliance with
emissions standards. NACS and SIGMA members recognize and appreciate that the Agency is
trying to be "forward-looking" in designating E15 as the test fuel. While E10 blends are
relatively standard across the country, much higher percentages of ethanol—including El 5—will
need to be consumed to satisfy the RFS's increasing consumption targets. If El 5 is going to be
the predominant fuel in the future, the market needs adequate notice so it can, among other
things, adjust its infrastructure accordingly.
Of course, the Proposal designates El 5 as the default test fuel before that fuel has made any
substantial inroads into the marketplace. At the present time, E10 is the predominant fuel, and
there are few indications that this will change in the immediate future. The Proposal would
nonetheless require vehicle manufactures to calibrate their vehicles to meet the proposed Tier 3
standards on fuel containing 15 percent ethanol by volume. NACS and SIGMA do not oppose
this provided that vehicles certified on El5 do not experience significant deterioration in their
performance when they run on E10. If such performance problems do occur, the Agency must
consider delaying designating E15 as the default test fuel, and, instead, make E10 the test fuel
until El5 is more widely available and utilized.
The retail market is not ready to accommodate sufficient volumes of higher fuel blends to satisfy
the RFS requirements at the present time. From retailers' perspective, a primary concern is
infrastructure compatibility. By law, all equipment used to store motor fuel—including
underground storage tanks ("USTs") and dispensers (gas pumps) must be certified by a
nationally recognized testing laboratory.4 If a retailer fails to use appropriately certified
equipment, that retailer may also be violating tank insurance policies, state tank fund program
requirements, bank loan covenants, and other local regulations. In addition, the retailer could be
found negligent per se based solely on the fact that his fuel dispensing equipment is not certified.
Currently, there is essentially only one organization that certifies such equipment - Underwriters
Laboratories ("UL").
Prior to 2010, UL had not listed a single dispenser as compatible with any fuel containing more
than 10% ethanol. This means that any dispenser in the market prior to 2010 is not legally
permitted to sell E15, E85, or anything else above 10% ethanol. As a practical matter, this means
that a significant number of retailers wishing to sell blends >E10 must replace their dispensers, at
an average cost of $20,000 per dispenser. For some retailers who operate specific models of
dispensers, there are UL-approved retrofit kits now available for approximately $4,000, but the
number of units that can utilize these kits is uncertain. Further, a substantial majority of UST
systems may have to be replaced as well. Once a retailer begins to replace underground
equipment, the cost can escalate rapidly and can easily exceed $100,000 per location. (Many of
these units are manufactured to be compatible with high concentrations of ethanol, but are not
listed as such by UL.)
For a significant number of fuel retailers across the United States, these costs preclude them from
upgrading their infrastructure. Unless Congress or the Agency develops a mechanism by which
current equipment is permitted to store and dispense higher fuel blends lawfully, retailers will
remain reluctant to market E15. That the RFS mandates will continue to escalate annually does
4-244
-------
Tier 3 Summary and Analysis of Comments
not change this fact. Designating El 5 as the default test fuel for most motor vehicles on the road
will also not change this fact.
Further, and perhaps more importantly, until consumers indicate their desire to purchase El5,
very few retailers will consider offering the new fuel even if their equipment is legally approved
to do so. Since the approval of E15, there has been no discernible effort to generate consumer
demand and encourage the availability of this product in the market. Quite the contrary, most
information readily available to consumers discourages the use of E15.
Therefore, if designating El 5 as the default test fuel for non-FFVs could impair vehicles' ability
to run on E10, NACS and SIGMA urge the Agency to make E10 the default test fuel to better
reflect market realities. This does not preclude the Agency from transit!oning to El 5 if legal and
logistical barriers are removed. This could include a market review in the next several years and,
if warranted, further regulatory action to implement the change from E10 to E15.
Organization: National Automobile Dealers Association (NADA)
EPA should not approve El 5 as a test fuel given its lack of presence in the marketplace and
should not mandate a test fuel above 9 psi RVP.
To enable Tier 3 tailpipe standard compliance, EPA should establish a 10 ppm gasoline sulfur
average with a 25 ppm cap. To enable Tier 3 evaporative emissions compliance, a Reid Vapor
Pressure (RVP) cap should be set at 9 psi for a maximum E10 gasohol test fuel.
Organization: National Corn Growers Association (NCGA)
We are supportive of the EPA's approach toward establishing E15 as the new certification fuel
for 2017 and later non-flexible fueled light duty vehicles and the lower sulfur specifications for
in-use gasoline.
EPA has proposed changing the certification test fuel from one containing no ethanol to El 5
with a volatility of 9 psi and an AKI of 87.
Separate from the Tier 3 rule, EPA approved the use of El 5 in 2001 and later on-road motor
vehicles (5). It would not make sense to certify new motor vehicles on EO, or even E10, when
EPA has approved the use of E15 in these vehicles. Nearly all vehicles in the United States are
currently fueled with E10, and the RFS volume requirements will increase the renewable fuel
content in the United States beyond today's levels. Therefore, it is important that the 2017 and
later model year vehicle fleet be certified on the expected future ethanol content rather than the
historic ethanol content.
We believe El 5 certification test fuel should be introduced for the 2017 vehicle model year as
EPA has proposed. In view of the long history of ethanol use as E6, E10 and E85, the
understanding of materials compatibility and engine and emission impacts of ethanol blends in
the literature for 30 years, it seems that a phase-in schedule might not be required to design and
calibrate for E15. EPA discussed potential delays in the introduction of E15 test fuel to allow
4-245
-------
Chapter 4: Proposed Vehicle Emissions Program
auto manufacturers to make calibration adjustments. While phase-in schedules or "trigger points"
may be necessary and appropriate in some situations, we believe it would reduce complexity and
improve market certainty to require the introduction of El 5 test fuel for the 2017 model year as
EPA has proposed.
A specific model year requirement such as 2017 provides a 3-year lead time for vehicle design
and calibration changes. This would also provide important certainty for fuel providers,
marketers and distributors that a trigger point approach would not provide. Perhaps more
importantly, implementation in the 2017 model year would eliminate the need for an interim E10
test fuel and the related complexity and potential vehicle modifications associated with a second
test fuel transition.
Organization: National Marine Manufacturers Association (NMMA)
NMMA's focus is on the EPA's request for comment regarding changing the automotive and
light duty truck certification fuel to El5, and perhaps higher, and its effect on boats and marine
engines due to potential misfueling and continued availability of suitable fuel.
The NMMA is not opposed to the use of ethanol as an additive in gasoline. Our members have
been designing their engines and fuel systems to be compatible with E10 since the early 1980's.
Our very serious concern is that the 12.1 million recreational boats currently registered the
United States, and those boats currently being manufactured, have not been designed to be
compatible with gasoline that has ethanol content greater than 10%. Recreational marine fuel
systems are not unique in this regard. The overwhelming majority of non-road engines, from
chainsaws to weed trimmers to lawn mowers, operate similarly to recreational marine engines
with open loop systems including a carburetor that is set at the factory and designed to be—and
required by EPA to be—tamper proof.
In addition to the well-documented physical evidence of ethanol's damaging effects on marine
and other non-road engines, NMMA's concerns are also based on the physical properties of
ethanol in gasoline. Gasoline is a mixture of many hydrocarbon compounds that consist mainly
of hydrogen and carbon. Ethanol contains hydrogen and carbon, but it also contains oxygen. The
exact air-fuel ratio needed for complete combustion is called the "stoichiometric air-to-fuel
ratio." This ratio is about 14.7 to 1 on a weight basis for gasoline that does not contain any
ethanol. When more ethanol is added to gasoline, less air is required for complete combustion
because oxygen is already contained in the ethanol. For example, for E10 the stoichiometric air-
to-fuel ratio is 14 to 14.1 pounds of air per pound of fuel. The stoichiometric air-to-fuel ratio for
straight ethanol is 9 to 1, so as the proportion of ethanol in gasoline increases so must the air-to-
fuel ratio decrease. To deliver the required power for a given operating condition, engines are
designed to consume enough air and fuel to generate the required energy. The marine engine is
designed and calibrated to anticipate a specific fuel-to-air ratio and nothing different. Because
ethanol blended fuels require more fuel for the same amount of air to achieve stoichiometric
conditions, the fuel system must adapt by introducing more fuel. If additional fuel is not
introduced to compensate for the ethanol, the resulting mixture has less fuel than needed and the
engine experiences a condition known as "enleanment."
4-246
-------
Tier 3 Summary and Analysis of Comments
Enleanment can lead to a variety of performance problems. For example, the combustion and
exhaust gas temperatures will be higher, engine starting may be harder, and the engine speed
control may become inaccurate. The increased combustion and gas temperatures resulting from
lean operation can result in severe damage to pistons, head gaskets, catalysts and emission
related components, which in turn may result in the failure of the engine and increased exhaust
emissions.
A series of engine evaluations conducted by Mercury Marine and Volvo Penta under direction of
the Department of Energy (DOE) National Renewable Energy Laboratory (NREL) concluded
that ethanol content in gasoline at 15% by volume will severely damage marine engines and
cause them to exceed the EPA emission standards (1). There is no remedy for this engine
damage and emission standard exceedances and it is not limited to recreational marine engines.
These same issues apply to other product categories including outdoor power equipment, heavy
duty engines, and snowmobiles. If E15 becomes the primary fuel in the US marketplace,
misfueling will occur, EPA and California emission standards will be violated, engines will be
damaged, and the American consumer will be stuck paying the bill for a misguided US
government policy that needs to be corrected now.
EPA requests comment on proposing to change the certification test fuel to El 5 and supports this
by stating that the requirements in the Renewable Fuel Standard (RFS-2) will result in El5
becoming the predominant fuel in the US marketplace in the next lO^lS years (78 Fed. Reg. at
29,909.) The RFS-2 requires that the US fuel supply contain 36 billion gallons of renewable fuel
by 2022. When the RFS-2 was created in 2007, fuel consumption in the US was rising, the
NHSTA motor vehicle fuel efficiency standards had not been promulgated, and the US
government had great hope for the consumer acceptance of E85. Since 2007, fuel consumption in
the US has decreased significantly with expectations of its continued fall. E85 consumption
reached a high of 38.6 million gallons in 2011 and consumption has remained flat since then.
The E85 contribution to achieving the requirements in the RFS-2 is insignificant.
El 5 can never achieve the requirements of RFS-2. This proposed change to the EPA certification
test fuel will not even come close to achieving the RFS-2 requirements. EPA would need to
mandate E20, E25 or even E30 for all engines immediately to meet 36 billion gallons in nine
years. Such action would not be logical or responsible. Thus, EPA is attempting to create a
piecemeal approach to achieving a federal requirement that cannot be achieved without great
cost to vehicle and engine owners. The RFS-2 is a broken government mandate that may have
made sense in 2007, but does not work today. NMMA recommends that the certification fuel
remain at E10 and that EPA formally request that Congress and the Administration amend the
Renewable Fuel Standard to reflect America's realistic motor fuel resources for 2013 and
beyond.
EPA also requests comment on a plan to allow the use of E10 as the certification test fuel with a
transition scheme to E15 in the future (78 Fed. Reg. at 29,910). This would create far more
confusion for consumers than an immediate mandate to E30. At this time, gasoline retailers that
choose to sell El 5 must clearly label the gasoline pump with the following specific language:
4-247
-------
Chapter 4: Proposed Vehicle Emissions Program
If El 5 becomes the certification test fuel for automobiles and light trucks, it will soon be the
primary fuel in the marketplace. Being the primary fuel in the marketplace, marine engine
manufacturers will have to follow suit and design engines to operate on El 5. However, their
open loop engines require a fuel to air ratio for optimum combustion that is limited and far less
able to tolerate variation in fuel oxygen content. In other words, an engine designed for 2.5%
oxygen will not operate efficiently or meet the emission standards with a fuel containing 5%
oxygen.
NMMA is also concerned with consumer labeling. The current label (see above) clearly states
that if the consumer uses El 5 in any engine other than a 2001 and newer passenger vehicles, it is
a violation of federal law. If El 5 is the only available fuel in the marketplace, breaking this law
seems inevitable. Is this violation of federal law a misdemeanor or a felony? When does the label
come off; when does it stay on? When does it apply, when doesn't it apply? NMMA is
concerned that this proposal shows no consideration for the consumers who own the millions of
lawn mowers, garden equipment, farm machinery, snowmobiles, and marine engines that will be
damaged and no longer covered under manufacturer warranties. This automobile and light truck
proposal will start the process of moving the nation's fuel supply to El5 and destroying marine
and other non-road engines in the process.
Organization: Natural Resources Defense Council (NRDC)
The gasoline emissions test fuel should be modified to reflect typical fuel on the market, which is
gasoline blended with 10 percent by volume of ethanol (E10).
Gasoline Emissions Test Fuel Should be Modified to Reflect Typical Fuel on the Market (E10).
NRDC agrees with EPA's objective to make the emission test fuel closer to the fuel widely
available in the market, and we support increasing the ethanol volume to 10 percent but not 15
percent as proposed by EPA. Today the share of El 5 in the gasoline market is small and the
likely share during the implementation of the Tier 3 rule with model year 2017 and beyond is
highly uncertain. It is conceivable that instead of reaching El 5 as the majority fuel in the market,
renewable drop-in fuels could have a significant market penetration. If drop-in fuels such as
biobutanol capture a large share of the market, EPA would have to evaluate the need for
subsequent changes in the test fuel to match the marketplace.
EPA should establish a market penetration level above 10 percent for ethanol and/or drop-in
fuels that would trigger a regulatory proceeding to establish new test fuel parameters. In
establishing the trigger volume, EPA should consider the rate at which ethanol or other
renewable fuel volumes are increasing in the market, the amount of time required to complete a
rulemaking and the amount of time required for manufacturers to adopt the new fuel in their test
procedures. The goal should be to set the trigger such that any changes to the test fuel could be
adopted by the manufacturers when the new test fuel would be representative of the majority of
the fuel on the market.
Organization: New York State Department of Environmental Conservation
4-248
-------
Tier 3 Summary and Analysis of Comments
We have long urged EPA to adopt emissions certification gasoline specifications that are
representative of actual commercial gasoline. We are pleased with EPA's steps in this direction.
However, nearly all commercial gasoline contains nominally 10 volume percent ethanol (El 0).
We urge EPA to adopt El 0 as the standard certification gasoline ethanol content, not the E 15
standard proposed.
Certification fuels should be comparable to commercial fuels. We have long urged EPA to
change certification gasoline standards to more closely reflect commercial gasoline. Currently,
the two have little in common. Current certification gasoline contains no ethanol, and regulations
allow unrealistically low amounts of aromatics and olefms.
The use of certification gasoline that reflects the properties of commercial gasoline rather than a
'super-clean' certification gasoline provides additional assurance that the air quality benefits
promised by Tier 3 are realized. We support EPA's steps in this direction.
10. Commercial gasoline overwhelmingly contains nominally 10 volume percent ethanol. This is
the ethanol content that should be specified for certification gasoline.
The Department opposes EPA's proposal to require 15 volume percent ethanol in certification
gasoline (El5). This is no more reflective of commercial gasoline than today's requirement for
zero ethanol. Allowing El 5 in certification gasoline violates the principle of testing and
certifying vehicles on a fuel that reflects the properties of commercial gasoline. El 5 is not a
significant commercial fuel now, and there is no assurance that El5 will become a significant
commercial fuel in the future.
Unless and until El 5 becomes the dominant commercial gasoline blend, it should not be the
certification gasoline blend.
Organization: Northeast States for Coordinated Air Use Management (NESCAUM)
EPA should ensure that certification fuels accurately represent fuels used in the real world. In our
view, the proposed requirement that certification fuel contain 15% ethanol by volume (El 5) is
inappropriate at this time. Given that this fuel represents a tiny fraction of the present-day fuels
market in the U.S., that its future prospects as a mainstream automotive fuel are highly uncertain,
and that its use in the existing fleet of vehicles will have non-trivial effects on vehicle emissions,
we urge EPA to refrain from specifying El 5 as a test fuel at this time. We suggest that EPA
continue to monitor the use of ethanol as a blended transportation fuel. If and when El 5 comes to
represent a significant share of the national market, EPA should reassess its implications and the
potential benefits and drawbacks to revising the test fuel specifications.
Organization: Ozone Transport Commission (OTC)
In the past, EPA has relied upon multiple fuels that are not necessarily available at public
gasoline dispensing facilities for use in testing vehicles to meet certification standards. OTC is
encouraged by EPA's decision to have one national real world test fuel. This will ensure that the
cars on the road are indeed meeting the emission standards for which they were tested. However,
4-249
-------
Chapter 4: Proposed Vehicle Emissions Program
OTC suggests that EPA rely on current, rather than future, in-use fuel, in particular in regards to
ethanol concentrations. Currently, El 5 (15% ethanol) is available in only six states having only a
small market share nationally due to concerns about its effects on automobiles and infrastructure
at gasoline dispensing facilities. Furthermore, California requires E10 (10% ethanol) as a test
fuel, which would lead to a discrepancy between vehicles tested to meet California vehicle
standards and those tested to meet EPA standards. OTC calls on EPA to require E10 test fuel to
maintain nationally harmonized testing standards, and only move to El5 at a point in when E15's
market share becomes the predominant fuel blend.
Organization: PBF Energy Inc.
PBF supports the EPA in the concept that the proposed vehicle emission test fuels should match
the properties of fuels found typically in the market place. The EPA should revise the proposed
rule to require that vehicle manufacturers certify emissions with E10 test fuel until such time as
E15 is actually widely used by consumers as- opposed to a projected to be used by consumers.
As an alternative, we recommend that the EPA defer to using the DOE/EIA projection of El 5
usage and timeline of prevalence as the inception date for allowing El 5 to be used in new
vehicle emission certification.
Organization: Pennsylvania Department of Environmental Protection (DEP)
The EPA is correct to develop a certification test fuel for vehicles subject to Tier 3 standards that
is more representative of the fuel that is used in the nation. DEP has urged EPA to change the
certification fuels to represent a real-world fuel on other proposed rulemakings. DEP supports
EPA moving to a fuel that is more representative of the fuel available to the nation's motorists.
In the future, EPA should make timely changes to the certification fuel as relevant changes occur
to fuels available in the marketplace.
The EPA should not use a certification test fuel that contains 15 volume percent ethanol before
that fuel is in widespread use in the nation. DEP questions whether there is sufficient ethanol
supply to meet the 15 volume percent threshold when Tier 3 goes into effect. The ethanol
industry reports that 20 of 211 biofuel plants are currently idled and many more are not
producing at capacity. If the current drought in the Midwest continues into the summer of 2013
(which the National Oceanic and Atmospheric Administration predicts), some ethanol plants will
likely be shut down permanently. Current annual biofuel production is at least 5 billion gallons
per year short of demand predicted in the EPA's proposed rule. The certification fuel should
include 10 volume percent ethanol until over 50 percent of the fuel supplied to the nation's
motorists meets the 15 volume percent ethanol requirement. In addition, CARB will be using 10
volume percent ethanol in their certification fuel. The federal fuel certification requirements
should be harmonized with CARB's requirements. Importantly, as previously stated, ethanol
increases NOX emissions and the increase to 15 volume percent of ethanol in gasoline seems to
be counter to the desire to achieve NOX reductions through the use of Tier 3 gasoline.
Organization: Phillips 66 Company
4-250
-------
Tier 3 Summary and Analysis of Comments
Certification fuel - We oppose El5 as the certification fuel. Phillips 66 does not support the
proposed change in the certification test fuel to El 5. The certification fuel should be reflective of
the predominant fuel in the marketplace that the vehicle will be operated on. EPA states "if El 5
were not to enter the gasoline pool in significant quantities, it may be more appropriate to require
that vehicles be calibrated for and tested on E10". We believe that the certification fuel should be
E10 and not El5. EPA asked for comment on an approach that would set the test fuel at E10 but
potentially transition to El5 in the future, either through a review and rulemaking process or by
some "trigger". Should EPA determine in the future that another change in the certification fuel
is warranted, we believe that EPA should go through a rulemaking process. We would not
support an automatic "trigger" without opportunity to review and comment through a rulemaking
process.
Organization: POET, LLC
The Tier 3 Proposed Rule appropriately recognizes that ethanol blending is of central importance
to our nation's transportation fuel supply and accomplishing key environmental and energy
security goals. To achieve these goals and maintain a robust domestic transportation fuel supply,
it is critical that EPA design its regulations to facilitate the production, blending, and use of
ethanol as a transportation fuel.
Ethanol already provides about 10% of the fuel for our nation's cars and trucks. Indeed, most
fuel available today contains 10% ethanol in every gallon distributed at the pump. EPA has
recently approved blends of 15% ethanol (El5) for over 230 million vehicles on the road today
(those that are model year 2001 and later). Additionally, over 14 million flexible-fueled vehicles
(FFVs) are on the road that can run on gasoline and blends containing up to 85% ethanol.
Ethanol is also inherently less toxic than petroleum fuels. Clean-burning ethanol is essentially
comprised of alcohol with a small amount of denaturant to render it unsuitable for drinking.
Ethanol use enables key air emission reductions of paniculate matter, benzene, sulfur, and other
toxic constituents as compared to gasoline. In fact, ethanol itself contains no benzene, generally
no sulfur, and overall has significantly lower air emissions than gasoline on a toxicity-weighted
basis.
The continued use of E10 and the increased use of E15 are critical to our nation's energy supply
and environmental goals.
To summarize, POET strongly supports two key certification fuel provisions in the Proposed
Rule. In particular, POET supports both the 'E15' base certification fuel, as well as an optional
E30 certification fuel. It is also imperative that RVP limits must not be a barrier to the use of
higher ethanol blends.
POET supports EPA's proposed E15-based certification fuel.
EPA proposes a new certification fuel that would be applicable to conventional gasoline
vehicles, including those that run on ethanol blends up to E15. This new certification fuel is
proposed to include an ethanol content of 15%.
4-251
-------
Chapter 4: Proposed Vehicle Emissions Program
POET strongly supports a new certification fuel based on an ethanol content of 15%. POET
agrees with EPA that such a fuel is a 'better match' for fuels in use today (which generally have
an ethanol content of at least 10%) than the current certification fuel that contains no ethanol.
Moreover, such a certification fuel is appropriate considering the increased use of El 5 over the
next few years. EPA is correct in pursuing a 'forward-looking El5 test fuel for light- and heavy-
duty gasoline vehicles,' and an El5 certification fuel better reflects the near future. EPA notes
that 'In-use fuel is projected to continue to change with the implementation of the RFS2
program' including with 'the expansion of the number of retailers that offer E15.' POET agrees.
The RFS2 obligations mandate increasing amounts of renewable fuel, the majority of which is
likely to be ethanol. Furthermore, the percentage of gasoline use from vehicles on the road that
are capable of using E15 (including model years 2001 and newer) is greater than 70% and is
growing as new vehicles replace aging ones.
An E15 certification fuel is preferable to an E10 certification fuel because an E10 certification
fuel may become obsolete in the timeframe covered by this rulemaking. Moreover, EPA has
sought to have vehicles tested with the highest available ethanol content that reflects actual
driving conditions; for conventional 'gasoline' vehicles, this would be an ethanol content of
15%.
Moreover, without enabling El 5 use, EPA would be frustrating the will of Congress in enacting
the RFS, and EPA must adhere to regulatory requirements that it act to 'ensure' that RFS targets
are met. Congress has mandated that EPA 'promulgate regulations to ensure that gasoline'
contains the applicable volumes specified in the RFS. An El5 certification fuel is necessary for
the RFS targets to be met in the short term, by allowing for the deployment of vehicles that can
best make use of (are optimized for) this fuel.
El5 should also become the prevalent fuel in the marketplace for compatible vehicles (model
year 2001 and newer) given ethanol's inherent economic advantages as a blendstock and octane
enhancer.
Organization: Renewable Fuels Association (RFA)
RFA supports the proposal to establish El 5 as the certification test fuel beginning in 2017. RFA
supports EPA making much needed modifications to the certification fuel used by auto
manufacturers in certifying all new vehicles. Significant changes in the composition of the in-use
gasoline available to consumers have taken place since the initial new vehicle certification
requirement circa 1974.
RFA strongly supports moving to a certification fuel that contains the highest level of ethanol
that is likely to be in broad commercial use in the next 5-10 years. In light of EPA's recent
approval of E15 for use in light-duty automobiles built in 2001 or later, and the recent
introduction of E15 at retail gasoline stations in several states, we agree that E15 is the proper
new certification fuel beginning in 2017. The agency should be mindful, however, that RFS2
requirements and the increasing desire for higher octane fuels to maximize engine efficiency are
likely to drive average ethanol content above 15% over the course of the next 10 years.
4-252
-------
Tier 3 Summary and Analysis of Comments
Looking at certification fuel another way, EPA could align the emissions certification fuel with
the diurnal emissions test fuel containing ethanol in at least the highest concentration permissible
in gasoline under federal law and that is commercially available. Further, EPA should include an
incentive for auto manufacturers who phase in the certification of new vehicles with the new
certification fuel ahead of the prescribed schedule for implementation. Early adoption of the new
certification fuel will accelerate the capability of the national car pool with higher ethanol
blended fuels.
Organization: Shell Oil Products for Shell and Motiva
EPA Should Change the Vehicle Certification Fuel to E10. EPA is proposing to change the
vehicle certification fuel to E15. We agree it is appropriate for EPA to change the vehicle
certification fuel from the current EO, since over 90% of all gasoline in the United States is now
blended with ethanol, but believe that EPA should change it to El0 rather than El5. As required
by section 202(h) of the Clean Air Act, EPA's regulations should require that the certification
fuel reflect the fuel that is generally available in the market. Section 202(h) required EPA to
update its regulations to ensure that "vehicles are tested under circumstances which reflect the
actual current driving conditions under which motor vehicles are used, including conditions
relating to fuel, temperature, acceleration, and altitude." As this relates to fuel, it clearly calls for
EPA to use E10, not E15 as the certification fuel.
It is unlikely that El5 use will expand to the degree that EPA suggests in the proposal. EPA fails
to consider that notwithstanding its approval of El 5, the automobile manufacturers do not
support El 5 use in the current fleet and El 5 is not compatible with the infrastructure at the
majority of retail gasoline stations. Studies by the Coordinating Research Council also indicate
that El 5 can damage even 2001 and newer vehicles.
Organization: Sierra Club, Clean Air Watch, Respiratory Health Association
As EPA notes in the proposed standards, in-use gasoline has changed significantly in previous
decades. With the implementation of the Renewable Fuel Standard, most gasoline sold contains
up to 10% ethanol by volume (E10). In the proposed rule, EPA proposes moving from Indolene
to an El5 test fuel. Additionally, EPA is proposing to allow manufacturers to request approval
for alternative certification fuel, such as E30.
However, ethanol blends higher than El0 are not widely available in the market today. Though
EPA recently increased the "blend wall" to El 5, that fuel has not been widely adopted yet, and
significant concerns have been raised about the effect of higher ethanol blends on older vehicles
and small engines. By proposing El5 as a test fuel and blends such as E30 as an alternative test
fuel, EPA is moving ahead of what is happening in the marketplace.
Instead of proposing El 5 as a certification fuel, EPA should instead adopt a certification fuel that
is closer to real-world conditions, such as E10. It is critical that this test fuel requirement be
harmonized with test fuel requirements for other vehicle standards, particularly light-duty
greenhouse gas and fuel economy standards, as increased ethanol blends affect fuel economy.
4-253
-------
Chapter 4: Proposed Vehicle Emissions Program
Though test fuel conditions may need to be updated in the future, EPA should strive to match
real-world fuel conditions.
Organization: Truck and Engine Manufacturers Association (EMA)
EPA has proposed changing on-highway certification test fuel to El 5 with a 9 psi RVP. (78 Fed.
Reg. at 29825.) In addition, EPA requests comment on the potential to utilize E10 rather than the
proposed El5 certification test fuel with a transition scheme to El5 at a future time. (78 Fed.
Reg. at 29914.) Finally, EPA suggests that if it were to adopt an E10 certification test fuel, that
fuel would have a maximum of 10 psi RVP rather than the 9 psi RVP maximum proposed for
E15 certification test fuel. (78 Fed. Reg. at 29895.)
EPA's proposed changes in certification test fuel for heavy-duty vehicles up to 14,000 Ibs.
GVWR would constitute a substantial change from current certification test fuel requirements.
EPA has not provided support in the Proposed Rule for any such change. EPA acknowledged in
the Proposed Rule that its previous Tier 2 regulations do not apply to Heavy- Duty Vehicles
(8,501-14,000 Ibs. GVWR). Yet, EPA now has proposed that such vehicles be subject not only to
more stringent emission standards, but also to a change in certification test fuel. In fact, EPA
previously determined that El5 certification fuel would not be appropriate or acceptable for use
in Heavy-Duty Vehicles (i.e., vehicles over 8,500 Ibs. GVWR) (see reference below). To make
such a change in certification test fuel for Heavy-Duty Vehicles, EPA must first fully review,
analyze and make a new determination that El5 is acceptable for Heavy-Duty Vehicles, which it
has failed to do. EMA does not support such a change.
EMA also strongly objects to any attempt to create a transition scheme where products would be
certified on E10 test fuel for the near term and transit!oned to certification utilizing an El 5 test
fuel at an unspecified future date. In addition, EPA should not adjust the maximum RVP based
on ethanol content, as such a change would have the effect of increasing the stringency of the
new standards.
EMA recognizes that current Heavy-Duty Vehicles are compatible with E10 marketplace fuels.
As such, if EPA makes any change to certification test fuels for Heavy-Duty Vehicles, EPA, at
most, should adopt E10 certification test fuel with a maximum of 9 psi RVP for Tier 3 Heavy-
Duty Vehicles and the engines used in them.
Adopt E10 certification test fuel with a 9psi RVP for light-duty and medium-duty vehicles
subject to the Proposed Rule, with future rulemakings to make similar changes to the test fuels
for other heavy-duty and nonroad spark-ignition engines
Organization: Union of Concerned Scientists (UCS)
EPA should adopt a certification fuel that accurately reflect the fuels used in the marketplace
now and in the timeframe of the vehicles' useful lifetime. The certification fuel should be chosen
to accurately reflect the fuels used in the marketplace now and in the timeframe of the vehicles'
useful lifetime. Today the fuel most commonly in use is E10, so this is an appropriate choice for
a certification fuel. But the fuel market is changing, and is expected to continue to change, and
4-254
-------
Tier 3 Summary and Analysis of Comments
EPA should ensure that the choice of certification fuel is, to the extent possible, technology
neutral, and does not present a barrier to the use of the higher levels of biofuel called for in the
Renewable Fuels Standard. Technical guidance from automakers, fuel producers and other
experts will be required to assess the best course forward, and to determine the pros and cons of
El5, E85, midlevel blends such as E30, or drop-in biofuels as a compliance pathway for the
RFS. We are not in position at this point to judge the outcome of these considerations in detail.
However, we urge that in performing this assessment EPA use realistic volume targets for the
RFS, particularly in the timeframe from 2016 to 2022. We have provided comments in an RFS
rulemaking explaining that the original RFS2 schedule targeting 36 billion gallons in 2022 is no
longer realistic, because of the limited availability of cellulosic biofuels (11). More plausible
timeframes for meeting the full 36 billion gallon target of the RFS are closer to 2030 than 2022,
and because of this there is time for an orderly transition to higher biofuel blends as the
cellulosic biofuel industry scales up. Any introduction of new fuel blends into commerce should
be conducted in a manner that will protect consumers from inadvertently damaging their
vehicles.
Organization: United Steelworkers Union (USW)
One concern our union does have with the proposed rule is the EPA's test fuel specification.
EPA has indicated a desire to move from "Indolene" (EO) to an El5 test fuel. While updating the
federal emissions test fuel to better match today's in-use gasoline is an appropriate approach,
specific issues regarding the "blend wall" and El5 still need to be addressed. USW would
recommend that test fuels meet what is the predominant fuel available at fuel stations in the US.
Our Response:
In the NPRM, EPA proposed that the gasoline emissions test fuel be changed from EO to
E15 as a forward-looking position based on indications following the 2011 E15 waiver decision
that the market would move in that direction. Since the time when we developed the proposal,
several relevant factors have led EPA to reconsider that position, including minimal proliferation
on a national scale of stations offering El 5. The most recent surveys of the market show that
E10 now comprises nearly 100% of in-use gasoline, with very small amounts of EO and El 5
being sold in limited areas where there is specific interest.32 Adopting an El5 test fuel would
also complicate the regulatory streamlining goal of long-term harmonization of the Tier 3 vehicle
emission regulations with California's LEVIII program (which uses E10 for emissions testing).
We received comments supporting use of E10 as emissions test fuel from the automotive
and oil industries, as well as states, NGOs, and private citizens citing the fact that this was most
representative of current market conditions. Other stakeholders involved in fuel marketing and
distribution cited significant infrastructure cost and liability concerns in making El 5 widely
available at existing stations. Some commenters expressed the concern that inclusion of ethanol
in test fuel would indicate an improper EPA policy preference for ethanol over other biofuels.
Ethanol industry commenters generally supported El5 certification fuel as proposed, but
provided no specific timeline on which this blend level would become representative of in-use
fuel.
32 More detail on fuel survey data is available in Chapter 3 of the Regulatory Impact Analysis.
4-255
-------
Chapter 4: Proposed Vehicle Emissions Program
Based on our review of the data and comments and under our authority in section 206 of
the Clean Air Act, specifically including section 206(d), EPA has determined that E10 most
appropriately reflects in-use gasoline around the country today and into the foreseeable future,
and thus we are finalizing E10 as the ethanol blend level for gasoline emissions test fuel for Tier
3 light-duty and heavy-duty gasoline vehicles. (Test fuel for motorcycles and nonroad engines is
not changing at this time.) We are not at this time adopting any specific timeline or criteria for
considering a gasoline emissions test fuel based on a higher-level ethanol blend. Instead we will
continue to monitor the in-use gasoline supply, including the qualities of El 5 in-use, and based
on such review may initiate rulemaking action to revise the specifications for emissions test fuel.
At this time it would not appear necessary to take further action to consider market triggers to
transition automatically to El5 as an emissions test fuel as opposed to through a new rulemaking
process. Rather than speculate about what in-use fuel qualities might be in the future, we prefer
to see how El5 properties develop as it expands into the market.
Some commenters expressed concern about implications for future test programs, fuel
waiver approvals, or Part 79 fuel registrations resulting from a change in test fuel properties.
While we recognize these interactions may occur, we believe the change in emission test fuel is
warranted to ensure that the certification process more appropriately reflects in-use conditions
Some comments included in this section also pertain to the RVP level of emission test
fuel, a topic to which we respond in Chapter 4.5.1.2 of this document.
Some comments included in this section also pertain to evaporative durability or service
accumulation fuel, a topic to which we respond in Chapter 4.3 of this document.
Some comments included in this section also pertain to allowances for alternative
emission test fuel, a topic to which we respond in Chapter 4.5.1.5 of this document.
4.5.1.2. Emission Test Fuel RVP
What Commenters Said:
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
Automakers support updating certification test fuels in Tier 3 regulations, as California did in its
recent LEV III rulemaking. EPA has asked for comment on adopting certification gasoline
containing 10 percent ethanol with a 10 psi Reid Vapor Pressure (RVP). We have further
discussion on the ethanol content in our fuels comments below; however, we cannot and do not
support Tier 3 if EPA adopts a certification gasoline with an RVP of 10 psi. The proposed
evaporative emission standards are already extremely challenging to meet across all the required
vehicle sizes based on the 9 psi fuel proposed by EPA. An RVP of 10 psi would significantly
increase the stringency of these standards, and we are not confident that the evaporative
standards could be met with such fuel. Moreover, adopting a certification gasoline with an RVP
of 10 psi would eliminate several years of work by the EPA, ARB, and automakers to harmonize
4-256
-------
Tier 3 Summary and Analysis of Comments
the Tier 3 and LEV III requirements and make them equal in stringency. While it might be
possible to make test procedure changes for an equivalent stringency on the higher RVP, such
changes would require years of additional work by EPA, ARB, and automakers and, even then,
would be more likely to result in confusion and uncertainty than in equivalence. Automakers
support a Tier 3 test fuel with 10 percent ethanol and RVP of 9 psi.
While EPA has used typical or average values for a number of the specific properties for the
proposed Tier 3 test fuel, the Agency has made exceptions to its market representative approach
for sulfur and RVP. In the case of sulfur, EPA is proposing more stringent market gasoline
average sulfur limits, so the test fuel sulfur levels appropriately reflect this change. However, in
the case of test fuel RVP, EPA is setting the level to reflect the maximum allowable content
rather than representative market levels. Although 10 psi RVP fuels are in use in the market as a
result of the statutory 1 pound waiver, a 9 psi RVP limit for both an E10 test fuel and
summertime market gasoline is a better outcome for advanced vehicle technologies and for the
environment, as detailed below.
As noted above in Part I, the auto industry supports harmonization of Federal Tier 3 and
California LEV III vehicle emission programs, to allow use of one test fuel and test regime to
certify for both Tier 3 and LEV III purposes. Toward that end, while there are differences
between these standards, we continue to support permanent mutual reciprocity to allow OEMS to
use one test fuel and its related testing regime per test vehicle, either LEV III or Federal Tier 3,
for purposes of both Federal and California certification. Such reciprocity is appropriate for all
test fuels, not just gasoline.
U.S. gasoline is now supplied as E10 in virtually all markets. The summer RVP for this fuel
varies from 7 to 10 psi, depending on local regulations and marketers. California recently
finalized new El0 7 psi RVP certification fuel (LEV III), based on the market fuel prevalent in
California. As noted above, the Alliance and Global Automakers recommend that EPA set the
Tier 3 test fuel ethanol level at 10% (E10), which represents over 90% of U.S. gasoline.
Likewise, we deem it critical that EPA set the test fuel Reid Vapor Pressure (RVP) at no greater
than 9 psi, which is toward the high end of the allowed range, but which appeared as the 75th
percentile of the U.S. samples in the 2012 Alliance North American Fuels Survey (excluding
California).
We understand that an El0 9 psi RVP cert fuel would be acceptable to California for purposes of
mutual reciprocity certification. Transit!oning to an El 5 test fuel incorporates speculative
assumptions that are not justified for the foreseeable future as a basis for such a critical
regulatory decision.
EPA cannot and should not try to compensate by imposing an RVP for the certification test fuel
that results in more stringent emission tests for vehicles. This is not a legitimate replacement for
fulfilling its duty under the Congressional mandate to regulate market fuel RVP. EPA should use
9 psi RVP test fuel in a final Tier 3 rule, until it has had the opportunity to complete its
evaluation of RVP regarding market fuel, and if necessary, then review RVP of test fuel.
4-257
-------
Chapter 4: Proposed Vehicle Emissions Program
Tier 3 Test Fuel Gasoline Should Be 9 psi RVP. Vapor pressure is the most important property
for engine start-up performance. At higher ambient temperatures, the fuel can vaporize
prematurely prior to reaching the injectors, disrupting the fuel flow to the engine and creating
"vapor lock." As defined in ASTM D4814, vapor lock conditions arise "if too much vapor is
formed, the fuel flow to the engine can be decreased, resulting in loss of power, rough engine
operation, or engine stoppage." During cold ambient temperatures, fuels with vapor pressures
that are too low can prevent the engine from starting at all or have poor warm-up performance.
Thus, the vapor pressure and distillation profile of gasoline are controlled based on seasonal
ambient temperatures so that fuel vaporizes easily (higher vapor pressure) in cold weather to
assist in engine starting, and vaporizes less easily (lower vapor pressure) in warm weather to
prevent vapor lock and reduce evaporative emissions.
ASTM D4814, Standard Specification for Automotive Spark-Ignition Engine Fuel, specifies for
Class A fuel a 9 psi maximum for the continental U.S. during the summer low-VOC (Volatile
Organic Compound) RVP blending season. The 10 psi market gasoline in some conventional
markets is largely due to Congressional action. Enacted in part to increase the volume of
renewable fuels in the U.S. marketplace, the Energy Policy Act of 2005 and the Energy
Independence and Security Act of 2007 have greatly increased the penetration of E10 in the
market. Under section 21 l(h) 4 of the CAA, these E10 blends are allowed an additional 1 psi of
vapor pressure relative to conventional gasoline or other ethanol blends.
EPA states in the NPRM (26): While the volatility (i.e., RVP) of CARB's E10 test fuel is 7.0 psi
to be representative of in-use gasoline in California during summer months, conventional E10 in
the rest of the country is currently around 10 psi. Thus, should we finalize E10 instead of E15, in
the absence of any standard to reduce the in-use RVP of E10 to 9.0 psi or lower, we would also
have to consider raising the RVP of certification test fuel to 10 psi to reflect the RVP level of the
current in-use fuel.
EPA should not raise the test fuel RVP to 10 psi. Inspection of the 2010-2012 Alliance of
Automobile Manufacturers Summer fuel surveys for US States (excluding Alaska, Hawaii and
California data) in Figure 1 [Figure 1 can be found on p. 55 of Docket number EPA-HQ-OAR-
2011-0135-4461-Al.pdf] show that the 50th percentile of samples is an 8 psi fuel while 10 psi is
only found at the 95th percentile of samples. 9 psi would be approximately the 75th percentile of
samples. Thus a 9 psi E10 fuel would be a realistic market fuel representation when considering
an E10 certification fuel.
But for the one pound waiver, E10 would be 9 psi or lower, which it is in many parts of the
country. As discussed further below, for both vehicle performance and environmental benefits,
we advocate that EPA reduce market fuel RVP to 9 psi (including 8 psi with a 1 pound waiver to
9 psi for E10). Therefore, it would be consistent to use a 9 psi test fuel. We disagree that there is
any technical basis or policy reason for imposing a 10 psi test fuel. Requiring the use of a 10 psi
test fuel would have the effect of making the Tier 3 evaporative emissions tests more stringent
for vehicles than anticipated, more stringent than California, and would disrupt optimal
harmonization efforts.
EPA should make the final Tier 3 gasoline test fuel E10 9 psi RVP.
4-258
-------
Tier 3 Summary and Analysis of Comments
HDGV and 10 PSI: Industry opposes 10 psi RVP test fuel for use in evaporative emission
testing. The use of a 10 psi test fuel would impact the stringency of the evaporative emission
standards and prevent a single national harmonized emissions program
E10 & 10 PSI (FR29910): Industry opposes a 10 PSI test fuel. This impacts the stringency of the
Tier 3 evap standards and does not harmonize with the ARB LEV III program.
Organization: American Coalition for Ethanol (ACE)
Volatility/Reid Vapor Pressure (RVP) — EPA is proposing to change the Federal certification
test fuel specification to a 9 RVP gasoline with 15% ethanol, and while we could support such a
change, it again seems inconsistent with past practices and with EPA's goal of having test fuel
more accurately reflect the fuels available in the marketplace. The highest volume fuel in the
U.S. is 10 Ib. RVP E10. [EPA-HQ-OAR-2011-0135-4313-A2 p. 3]
Organization: American Lung Association
We believe that if the Reid Vapor Pressure (RVP) of E10 in current in-use fuel is 10 psi, then the
certification fuel should also be 10 psi.
Organization: American Petroleum Institute (API) and the Association of Fuel &
Petrochemical Manufacturers (AFPM)
The test fuel vapor pressure should be set at 10 psi dry vapor pressure to reflect the 1 psi waiver
afforded to fuels containing 9% to 10% ethanol found in the Clean Air Act.
Organization: California Air Resources Board (CARB)
RVP 10 fuel implications for evaporative emission testing: The NPRM requests comment for
alternatively requiring E10 with Reid Vapor Pressure (RVP) of 10 psi rather than El5 with an
RVP of 9 psi for certification emissions testing. Regardless of which fuel U.S. EPA finalizes for
Tier 3, CARB strongly recommends that U.S. EPA maintain an equivalent level of stringency for
evaporative emission testing with LEV III evaporative emission standards. If equivalency
between Tier 3 and LEV III evaporative requirements is maintained, CARB intends to accept test
data from vehicles tested using Tier 3 certification fuel and temperatures to meet the LEV III
evaporative emission requirements. See also our more detailed comments on Tier 3 certification
fuel on pages 21-22 of this document.
Comment 1 — Gasoline Certification Test Fuel: CARB recommends that the U.S. EPA
reconsider the use of E15 as the Tier 3 certification fuel. Instead, CARB recommends a
certification fuel with 10 percent volume ethanol (E10) and a 9.0 pounds per square inch (psi)
RVP. An E10 certification fuel would be consistent with the current federal commercial fuel. In
addition, the 9.0 psi RVP would allow the Tier 3 program and California's LEV III program to
be consistent with the certification fuel reciprocity precedent set between the Tier 2 program and
California's LEV II program. The evaporative emission difference between U.S. EPA's 9.0 psi
RVP Tier 2 certification fuel and California's LEV II 7.0 psi RVP certification fuel was offset by
4-259
-------
Chapter 4: Proposed Vehicle Emissions Program
U.S. EPA's use of a lower diurnal temperature range. California's LEV III certification fuel has
approximately a 7.0 psi RVP and CARB staff recommends that U.S. EPA maintain a 9.0 psi
RVP fuel such that the evaporative emission offsets between the California LEV III certification
fuel and the Tier 3 certification fuel can be maintained.
Organization: Chrysler Group LLC
The certification test fuel for gasoline-fueled vehicles should be an E10, 9.0 psi RVP test fuel,
rather than an El5 certification test fuel.
Reid Vapor Pressure of Certification Test Fuel: Chrysler recommends that EPA set the
certification test fuel RVP at 9.0 psi rather than the 10.0 psi proposed by EPA for E10 fuel. RVP
is a measure of the volatility of gasoline. Because gasoline evaporates more readily the higher
the RVP, use of a higher RVP fuel for evaporative emissions testing imposes a more difficult
compliance burden on automobile manufacturers (18). Indolene, the current certification test
fuel, has an RVP of 9.0 psi. EPA proposes to retain this 9.0 psi RVP if E15 is adopted as the
certification test fuel. However, EPA proposes to raise the RVP from 9.0 psi to 10.0 psi should it
instead adopt E10 as the certification test fuel. EPA states that raising the RVP to 10.0 psi for
E10 fuel is necessary in order to be representative of conventional E10 gasoline. [EPA-HQ-
OAR-2011-0135-4326-A2, p. 7]
EPA's reasoning that it is necessary to raise the RVP for E10 certification test fuel is flawed
(21). Although it is true that conventional E10 gasoline can have an RVP of up to 10.0 psi during
summertime months, it is not the case that E10 gasoline with an RVP of 10.0 psi is
representative of in-use fuel. Rather, 9.0 psi fuel is most representative of in-use fuels. First,
many states require use of reformulated gasoline, which has a lower RVP than conventional E10
gasoline in the 7.0 to 8.0 psi range. In fact, according to EPA, reformulated gasoline is currently
used in 17 states and the District of Columbia, and accounts for approximately 30 percent of
gasoline sold in the United States (22). In addition, although conventional E10 gasoline can as a
statutory matter have an RVP of up to 10.0 psi, some conventional E10 gasoline has an RVP that
is lower than 10.0 psi. Thus, as explained in comments provided by the Alliance of Automobile
Manufacturers, 9.0 psi fuel represents the 75th percentile RVP level of U.S. gasoline in
summertime months based on samples collected as part of the 2012 Alliance North American
Fuels Survey (excluding California, Alaska, and Hawaii, which have even lower RVP levels)
(23). Accordingly, 9.0 psi fuel is actually more representative of in-use fuel than 10.0 psi fuel
and thus is more appropriate for use as the certification test fuel. [EPA-HQ-OAR-2011-0135-
4326-A2, pp. 7-8]
Organization: Ford Motor Company (Ford)
EPA has proposed a new Tier 3 certification test fuel containing 15% ethanol (El5) with a Reid
Vapor Pressure (RVP) of 9psi. While the percentage of ethanol in the proposed fuel does not
match that of the 10% ethanol (E10) fuel adopted by California in LEV III, the proposed vapor
pressure ensures equivalent stringency with respect to evaporative emissions when this fuel is
used with the temperatures specified in EPA evaporative test procedures. However, in the NPRM
EPA also explored the ramifications of instead proposing an E10 test fuel, indicating that such a
4-260
-------
Tier 3 Summary and Analysis of Comments
change could result in an RVP other than 9 psi. Ford strongly opposes any increase in the RVP
of the test fuel since this would require either heavily modified test procedures or higher
standards in order to maintain equivalent stringency to CA LEV III.
Recommendation: Ford recommends that EPA adopt a 9 psi test fuel to maintain equivalent
stringency with CA LEV III and to ensure that the Tier 3 evaporative emissions standards are
achievable. [EPA-HQ-OAR-201 l-0135-4349-A2,p.3]
Organization: General Motors LLC (GM)
Although not proposed by EPA, throughout the Tier 3 NPRM, EPA discusses the possibility of a
10 psi (pounds per square inch) Reid Vapor Pressure (RVP) test fuel for evaporative emission
testing. GM is strongly opposed to the use of 10 psi test fuel for evaporative emission testing.
GM has vast experience conducting evaporative emission testing with both CARB 7 psi and EPA
9 psi test fuels. The Tier 3 regulations were intended to harmonize with the CARB LEV III
regulations and to propagate zero evaporative emission vehicle technologies across the nation.
GM vehicle designs are predicated on the understood equivalent performance between the EPA
and CARB test procedures and test fuels. Moving to a 10 psi test fuel would decouple this basic
design premise, destroy the reciprocity agreement between EPA and CARB and would represent
a significant stringency increase for vehicles complying with the Tier 3 requirements. Simply
put, Tier 3 would not be harmonized with LEV III, and the goal of a single national program
would be undermined.
While EPA has used typical or average values for a number of the specific properties to define
their proposed test fuel, the Agency made exceptions to its market representative approach for
sulfur and RVP. In the case of sulfur, EPA is proposing to impose more stringent market
gasoline average sulfur limits, so the test fuel sulfur levels appropriately reflect this change.
However, in the case of test fuel RVP, EPA is setting the level to reflect the maximum allowable
rather than representative market levels. Although 10 psi RVP fuels are in use in the market as a
result of the statutory 1 pound waiver, a 9 psi RVP limit for both an E10 test fuel and
summertime market gasoline is a better approach for advanced vehicle technologies and for the
environment, as detailed below. [EPA-HQ-OAR-2011-0135-4288-A1, p. 11 ]
U.S. gasoline is supplied with E10 in virtually all markets. The summer RVP for this fuel varies
from 7 to 10 psi, depending on local regulations and markets. California recently finalized new
E10 7 psi RVP certification fuel (LEV III) based on the market fuel prevalent in California. GM
recommends that EPA set the Tier 3 test fuel ethanol level at 10% (E10) which represents over
90% of U.S. gasoline. [EPA-HQ-OAR-2011-0135-4288-A1, p.12]
Likewise, we deem it critical that EPA set the test fuel RVP at no greater than 9 psi, which is
toward the high end of the allowed range, appearing in the 75th percentile of the U.S. samples in
the 2012 Alliance North American Fuels Survey (excluding California).8 This is discussed
further in Section XV. [EPA-HQ-OAR-2011-0135-4288-A1, p.12]
4-261
-------
Chapter 4: Proposed Vehicle Emissions Program
In conversations with members of CARB, an E10 9 psi RVP certification fuel would be
acceptable to California for purposes of mutual reciprocity certification. [EPA-HQ-OAR-2011-
0135-4288-Al,p.l2]
Tier 3 Gasoline Certification Test Fuel Should Be 9 psi RVP: Vapor pressure of gasoline is the
most important property for engine start-up performance. At higher ambient temperatures, the
fuel can vaporize prematurely prior to reaching the injectors, disrupting the fuel flow to the
engine and creating "vapor lock". As defined in ASTM D4814, vapor lock conditions arise "if
too much vapor is formed, the fuel flow to the engine can be decreased, resulting in loss of
power, rough engine operation, or engine stoppage." During cold ambient temperatures, fuels
with vapor pressures that are too low can prevent the engine from starting at all or cause it to
have poor warm-up performance. Thus, the vapor pressure and distillation profile of gasoline is
controlled based on seasonal ambient temperatures so that fuel vaporizes easily (higher vapor
pressure) in cold weather to assist in engine starting and vaporizes less easily (lower vapor
pressure) in warm weather so as to prevent vapor lock and reduce evaporative emissions. [EPA-
HQ-OAR-2011-0135-4288-A1, p. 15]
ASTM D4814, Standard Specification for Automotive Spark-Ignition Engine Fuel, specifies for
Class A fuel a 9 psi maximum for the continental U.S. during the summer low-VOC (Volatile
Organic Compound) RVP blending season. The 10 psi market gasoline that EPA sees in some
conventional markets is largely due to Congressional action. Enacted, in part, to increase the
volume of renewable fuels in the U.S. marketplace, the Energy Policy Act of 2005 and the
Energy Independence and Security Act of 2007 have greatly increased the penetration of El 0 in
the market. Under section 21 l(h) 4 of the Clean Air Act, these E10 blends are allowed an
additional 1 psi of vapor pressure relative to gasoline or other ethanol blends. [EPA-HQ-OAR-
2011-0135-4288-Al,p.l5]
EPA states in the NPRM, "While the volatility (i.e., RVP) of CARB's E10 test fuel is 7.0 psi to
be representative of in-use gasoline in California during summer months, conventional E10 in
the rest of the country is currently around 10 psi. Thus, should we finalize E10 instead of E15, in
the absence of any standard to reduce the in-use RVP of E10 to 9.0 psi or lower, we would also
have to consider raising the RVP of certification test fuel to 10 psi to reflect the RVP level of the
current in-use fuel." [EPA-HQ-OAR-2011-0135-4288-A1, p.15]
We disagree that there is any technical basis or policy reason for imposing a 10 psi test fuel.
Requiring the use of a 10 psi test fuel would have the effect of making the Tier 3 evaporative
emissions tests far more stringent for vehicles than anticipated, and far more stringent than
California LEV III, completely undermining the harmonization efforts. Further, making
certification fuel have a vapor pressure of 10 psi would have the effect of making this high vapor
pressure summer gasoline permanent. [EPA-HQ-OAR-2011-0135-4288-A1, p. 16]
EPA cannot and should not try to compensate for the widespread use of ethanol by imposing an
RVP for the certification test fuel that results in more stringent emission tests for vehicles. This
is not a legitimate replacement for fulfilling its congressionally mandated duty to regulate market
fuel RVP. EPA should use 9 psi RVP test fuel in the final Tier 3 rule. [EPA-HQ-OAR-2011-
0135-4288-Al,p.l7]
4-262
-------
Tier 3 Summary and Analysis of Comments
Inspection of the 2010-2012 Alliance of Automobile Manufacturers summer fuel surveys for
U.S. States (excluding Alaska, Hawaii and California data) in Figure 1 show that the 50th
percentile of samples is an 8 psi fuel while 10 psi is the 95th percentile of samples. 9 psi would
be approximately the 75th percentile of samples. Thus, a 9 psi E10 fuel would be a realistic
market fuel representation when considering an E10 certification fuel. [[See Docket Number
EPA-HQ-OAR-2011-0135-4288-A1, p. 16 for Figure 1.]]
Organization: Marathon Petroleum Company LP (MFC)
The test fuel vapor pressure should be set at 10 psi dry vapor pressure to reflect the 1 psi waiver
afforded to fuels containing 9% to 10% ethanol found in the Clean Air Act.
Organization: Michigan Department of Environmental Quality (MDEQ)
The MDEQ, Air Quality Division is supportive of a change to the emission test fuel
specifications to E-10 as long as this revision is accompanied by a Reid Vapor Pressure (RVP) of
9.0 pounds per square inch (psi). As an ozone compliance strategy, many of the large
metropolitan areas throughout the country are required to use reformulated gasoline or fuels with
lower RVP limits than the 10.0 psi claimed to reflect fuels in conventional use. The test fuel
should be reflective of what is widely used in the marketplace, especially large urban areas
where there is high traffic volume.
Organization: Mitsubishi Motors R&D America, Inc (MRDA)
EPA requested comments on a proposal for Federal E10, 10 psi RVP certification fuel.
Additionally, Tier 3 evaporative emissions standards are proposed to align with LEV III
standards. Though there are differences in the Federal and California test procedure temperature
profiles to account for the climatic differences, proper specification of Federal and California
certification fuels - specifically RVP - makes the results equivalent. It is commonly known that
9 psi RVP 'Federal' fuel using Federal evaporative test procedures yields similar test results as 7
psi RVP California fuel using California procedures. Therefore, the use of a 10 psi RVP federal
certification fuel will no longer allow for harmonized evaporative emissions standards. Clearly,
this issue is caused by the statutory one pound waiver that allows market availability of 10 psi
RVP fuel as a temporary regulatory mechanism. Therefore, creating a long-term certification fuel
should not be based on a temporary waiver. [EPA-HQ-OAR-2011-0135-4281-A1, p.2]
Mitsubishi Motors emphasizes that permanent reciprocity of certification test fuel with the LEV
III program should not include a fuel with an RVP higher than 9 psi RVP. Federal certification
fuel higher than 9 psi RVP would result in significant challenges in meeting the Tier 3
evaporative emission standards. The lack of harmonization in this area leads to an increased
development burden upon automotive manufacturers because they need to account for possible
higher vehicle engineering costs, redundant-testing, and increased costs due to additional test
fuel stock. We request that EPA focus on incorporating a permanent reciprocity regime with
Federal E10, 9 psi RVP fuel and federal test procedures into the Tier 3 Final Rule in order to
follow through on its goal to harmonize with LEV III. Doing so would follow the
4-263
-------
Chapter 4: Proposed Vehicle Emissions Program
Administration's call for 'smart regulations' by eliminating areas of duplicative and unnecessary
compliance requirements. [EPA-HQ-OAR-2011-0135-4281-A1, p.2]
EPA should not consider a federal certification test fuel higher than 9 psi RVP. [EPA-HQ-OAR-
2011-0135-4281-Al,p.2]
Organization: National Automobile Dealers Association (NADA)
EPA should not approve El 5 as a test fuel given its lack of presence in the marketplace and
should not mandate a test fuel above 9 psi RVP.
To enable Tier 3 evaporative emissions compliance, a Reid Vapor Pressure (RVP) cap should be
set at 9 psi for a maximum E10 gasohol test fuel.
Organization: National Corn Growers Association (NCGA)
We concur with EPA's proposed 9 psi vapor pressure specification for an El5 test fuel. The
current vapor pressure specification for Tier 2 vehicles is also 9 psi. However, we recommend
that in-use ethanol blends between El 1-50 be allowed a 1 psi waiver, as applied to E10. Our
rationale is further discussed within the In-Use Fuel section of these comments. We do not
believe that a 1 psi waiver for El 1-E50 necessitates a vapor pressure specification greater than 9
psi for E15 certification fuel, any more than E10 (which has a 1 psi waiver) does for the current
certification gasoline. Reformulated Gasoline (RFG) areas and non-RFG areas without a waiver
currently have less than 9 psi fuel in the summer; vapor pressure greater than 9 psi only occurs in
non-RFG areas with a waiver. EPA did not propose further volatility controls for in-use fuel as a
part of the Tier 3 fuel requirements. If EPA were concerned about the emission increases of Tier
3 (or even Tier 2) vehicles certified on 9 psi fuel and operated on 10 psi in the summer, it would
have done so as a part of its fuel proposal. [EPA-HQ-OAR-2011-0135-4285-A1, p. 4]
EPA points out in the Tier 3 proposal that vapor generation in motor vehicles using 10 psi fuel in
the summer is 25% higher than with 9 psi fuel. However, this does not mean that vehicle
emissions are increased by 25% because modern vehicle evaporative emission control systems
are designed to capture emissions for a 3-day park at high temperatures. The Tier 3 evaporative
emission standards represent a very high level of evaporative emission control, whether the
vehicles are operated on 9 psi fuels or in some areas of the country, 10 psi. [EPA-HQ-OAR-
2011-0135-4285-Al,p. 4]
Organization: National Marine Manufacturers Association (NMMA)
EPA requests comment regarding the option of allowing the use of CARB E10 certification test
fuel through model year 2019 with a Reid Vapor Pressure (RVP) of 10 psi (78 Federal Register
29895). EPA currently provides a 1 psi waiver for E10, but this waiver will not apply to E15.
NMMA recommends that the 1 psi waiver also be eliminated for E10. EPA finalized a rule in
2010 that requires boat builders to meet stringent new evaporative emission requirements.
California is currently in the process of finalizing a rule. In order to meet these new emission
4-264
-------
Tier 3 Summary and Analysis of Comments
standards, marine engine manufacturers need a combination of new technology and a fuel with a
minimum evaporation rate. [EPA-HQ-OAR-2011-0135-4294-A1, p. 4]
Organization: New York State Department of Environmental Conservation
Certification gasoline Dry Vapor Pressure Equivalent should reflect worst case commercial
summer conventional gasoline. Currently certification gasoline has a nominal Reid Vapor
Pressure (RVP) of 9 psi. EPA's proposed certification gasoline standard has a nominal Dry
Vapor Pressure Equivalent (DVPE) of 9 psi. These are the same except for a change in test
method (and acronym). While a 9 psi standard is appropriate for EO or El5 gasoline, it is not
appropriate for a certification gasoline standard based on current commercial practice.
In much of the country conventional summer gasoline DVPE is nominally 10 psi through a
combination of Clean Air Act Section 21 l(h)(l)'s summer gasoline maximum RVP of 9 psi and
the 1 psi waiver for E10 contained in Section 21 l(h)(4). As long as this remains the case, it is the
worst case for evaporative emissions, and is what evaporative emission control systems should
be designed for and certified on.
Organization: Phillips 66 Company
The test fuel specifications for an E10 test fuel should align with the ASTM gasoline
specification, D4814. EPA stated that they reviewed refinery test data as one input to set
specifications for various test fuel properties. Currently, all conventional gasoline is tested and
reported to EPA on a clear basis so would not include the effect of the blended ethanol that is
added downstream. A few notes on selected properties: [EPA-HQ-OAR-2011-0135-4463-A1, p.
7]
- Vapor pressure should be set at 10 psi which accounts for the 1 psi vapor pressure waiver
allowed for E10 blends throughout most of the United States [EPA-HQ-OAR-2011-013 5-4463-
Al,p.7]
Organization: Revecorp Inc.
First, the proposed TIER 3 rule states that testing for evaporative emissions will be conducted on
fuel which is lower RVP than the current certification fuel. Unfortunately, at a time which EPA
is proposing decreasing the RVP of the certification fuel may, in-use vehicles are being exposed
to higher RVP fuel. For example, it is possible that because E85 is being sold and more flexible
fuel/E85 capable vehicles are in the fleet that a vehicle could end up with a high RVP fuel in the
tank. For instance, if a motorist had than the certification fuel is a good possibility. For example,
if an E85 vehicle is refueled with 19 gallons of E10 fuel when it had one gallon of E85 in the
tank, the ethanol content of the resulting fuel would be 13.75%, near the highest RVP of any
gasoline/ethanol blend, and with an RPV much higher than the new proposed certification fuel.
Vehicle testing should be performed with fuel which is representative of the fuel which vehicles
are expected to see under the most severe in-use conditions.
Organization: Truck and Engine Manufacturers Association (EMA)
4-265
-------
Chapter 4: Proposed Vehicle Emissions Program
Adopt E10 certification test fuel with a 9psi RVP for light-duty and medium-duty vehicles
subject to the Proposed Rule, with future rulemakings to make similar changes to the test fuels
for other heavy-duty and nonroad spark-ignition engines
Organization: Volkswagen Group of America, Inc.
VW strongly supports capping both market fuel and testing fuel at 9 psi RVP. This allows
current vehicles to operate their evaporative emissions controls more efficiently. An immediate
real-world air quality benefit will be soon in non-attainment areas that typically see 10 psi RVP
summertime fuels. A 10 psi RVP test fuel will unnecessarily increase the stringency of the Tier 3
evap rule. Since CARB does not view a 10 RVP E10 test fuel as equivalent to their gasoline test
fuel for LEVIII, a 10 RVP fuel walks further away from a harmonized National Program.
Our Response:
In addition to our proposal to change the ethanol level of gasoline emission test fuel, we
raised the question of which volatility level (or RVP) would be most appropriate. The Clean Air
Act Section 21 l(h)(l) sets a national limit on summertime RVP in northern conventional
gasoline areas of 9.0 psi to control ozone pollution. However, Congress included a waiver
allowance in Section 21 l(h)(4) granting an additional 1 psi RVP to 10% ethanol blends, meaning
that E10 could have an RVP of up to 10 psi in conventional gasoline areas unless specifically
prohibited by state or local rules. (Under Section 21 l(h)(4), El5 is not covered by the waiver
and thus is restricted to 9 psi nationwide.)
Some commenters, including some automakers, recommended leaving the RVP of
emissions test fuel at 9 psi for reasons including the fact that raising the specification to 10 psi
would increase the stringency of the proposed evaporative emission standards significantly. We
agree that the resulting increased vapor generation rates during the refueling test would increase
emissions (by about 10 percent and during the hot soak, diurnal, canister bleed, and running loss
tests by as much as 25 percent in total). While the likely increase in canister volume in response
to higher certification fuel RVP would not be difficult for automakers to accommodate in most
cases, there are additional uncertainties regarding cost and feasibility of strategies for removing
the larger vapor loads from the canister during vehicle operation (vapor "purging"). Some
vehicles have adequate engine vacuum available to accomplish the increased vapor purge, while
others may require new or innovative approaches to increase purge volume or efficiency
(discussed in more detail in preamble Section IV.C.3).
Several other commenters, including some NGOs and environmental groups, supported
setting certification gasoline RVP to 10 psi to be representative of the worst-case volatility
vehicles may see in the market, making the test procedure more stringent than in the proposed
program and likely bringing further evaporative emission reductions in-use.
Raising the certification test fuel RVP to 10 psi would also impact the equivalency of
CARB and EPA evaporative emission test procedures. EPA and CARB have maintained a
reciprocity agreement regarding test fuel RVP and test temperatures for hot soak plus diurnal and
running loss emissions since the mid 1990's. (California requires the use of 7 psi RVP test fuel,
which, in conjunction with higher test temperatures, produces equivalent results to the federal
4-266
-------
Tier 3 Summary and Analysis of Comments
test procedures using 9 psi fuel for purposes of evaporative control system design.) CARB has
accepted ORVR emission test results using Federal test fuels and test temperatures since the
1998 model year. If we were to adopt 10 psi test fuel, we would likely need to develop and
adopt new test procedure adjustments in order to maintain the equivalency of CARB and EPA
evaporative procedures (and allow reciprocal acceptance of test data generated under either
agency's program).
A review of 2011 gasoline batch data submitted to EPA shows that just under half of
summertime gasoline was conventional gasoline at 10 psi RVP. An additional third was
reformulated gasoline at approximately 7 psi, with the remainder having intermediate RVPs
under local volatility control programs. The volume-weighted average RVP of this data is
approximately 8.7 psi. Thus, approximately half of summertime gasoline falls above 9 psi and
half falls below, suggesting an emissions test gasoline with RVP of 9 psi aligns well with in-use
fuel nationwide. In addition, virtually all of the areas that have elevated summertime ozone
levels where excess evaporative VOC emissions would be of greatest concern already control in-
use gasoline RVP to levels less than 9 psi. Furthermore, under CAA Section 21 l(a)(5),
governors can request that the 1 psi waiver for El 0 not apply in their state if it causes an
emissions increase that contributes to air pollution. Any state exercising this authority would
have in-use E10 RVP levels limited to 9 psi.
After considering these technical and policy issues in the context of the available data and
comments received, we conclude that the most appropriate approach is to maintain the RVP at 9
psi for Tier 3 emissions test fuel.
Some comments included in this section also pertain to the ethanol level of emission test
fuel, a topic to which we respond in Chapter 4.5.1.1 of this document.
Some comments included in this section also pertain to the RVP of in-use gasoline, a
topic to which we respond in Chapter 4.5.1.6 of this document.
Some comments included in this section also pertain to properties of emission test fuel
for flexible fuel vehicles (FFVs), a topic to which we respond in Chapter 4.5.2 of this document.
4.5.1.3. Test Fuel Octane
What Commenters Said:
Organization: American Coalition for Ethanol (ACE)
Octane — EPA recommends reducing test fuel octane to around 87 (R+M)/2, to be
representative of in-use fuel, and plans to continue to allow manufacturers to test vehicles on
premium-grade gasoline should vehicles require it. EPA also plans to continue to allow
exceptions for lower-volatility fuel for high altitude testing. There has never been a requirement
that manufacturers prove that premium would actually be used by the operator of a premium-
only vehicle if that certification method is chosen. The high-altitude exception has been offered
4-267
-------
Chapter 4: Proposed Vehicle Emissions Program
although most vehicles will rarely have that kind of fuel available and if ever operate at high
altitude. The current test fuel is not available commercially in any gas stations.
EPA asks whether there should be a maximum octane of gasoline used in certification. ACE
conversely suggests that there should be a minimum octane for the base fuel, so that refiners
cannot continue to control ethanol in the marketplace by simply creating lower-octane, lower-
quality blendstocks or 'BOBs' each time ethanol percentage is increased. This refiner practice
prevents automakers from using of ethanol's higher octane to make cleaner vehicles. While EPA
has been unfairly charged with 'picking winners and losers' through its rules, failing to maintain
a minimum octane for base fuels would effectively allow the oil industry to 'pick winners and
losers,' by allowing refiners to retain full control of octane offered by ethanol and other
additives, by creating BOBs that swallow all of that octane value before it can reach vehicles and
consumers. [EPA-HQ-OAR-2011-0135-4313-A2 p. 2-3]
EPA must maintain high standards for octane, especially because future GHG emissions
reductions proposed by the Agency on automakers are very likely to lead to advanced vehicle
technologies (turbocharging, downsizing, increasing compression ratios, etc) that require and
indeed optimize on clean and high-octane qualities that fuel ethanol delivers. Failing to require
higher octane or to protect minimum octane allows choices to be dictated by the petroleum
industry.
To update certification fuel with high-octane ethanol but propose to reduce - or even simply
maintain - overall octane seems inconsistent.
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
Automakers support EPA's proposed option for use of premium grade octane gasoline as test
fuel, consistent with California LEV III.
Organization: American Petroleum Institute (API) and the Association of Fuel &
Petrochemical Manufacturers (AFPM)
EPA proposes to change the measurement approach from RON to AKI and lower the octane
requirement to be consistent with regular unleaded fuel at 87.0 (R+M)/2. EPA also proposes to
allow manufacturers to test on a fuel with a minimum octane rating of 91 (R+M)/2 for those
vehicles where operation on high-octane gasoline is required by the manufacturer. EPA is
seeking comment on the need for limiting the maximum octane of gasoline used in the
certification of premium-required engines and vehicles. As we will note in other comments, EPA
needs to limit the range of certification fuels to those that are readily found in the marketplace.
As such, a high-octane fuel specification should be limited to 91 (R+M)/2. While higher octane
fuels exist in the marketplace, they are not available on a nation-wide basis and several
significant regions offer 91 (R+M)/2 premium gasolines as the highest level available. These
regions include California, Arizona and Nevada where a significant number of premium vehicles
are sold by the manufacturers. Increasing the minimum octane of premium to satisfy these
vehicles and support their certification would add significantly to the cost of this regulation.
4-268
-------
Tier 3 Summary and Analysis of Comments
Another related issue is the requirements that would need to be put on the vehicles to insure that
only premium fuel is used in these vehicles. Will there be a requirement for special nozzles? Will
the "Check Engine" light illuminate? This special accommodation makes no sense unless EPA
has some way of insuring that this fuel is used in the field if it is required for emissions or other
regulatory performance attributes (e.g., CAFE compliance). Finally, if a consumer fills a
"premium" vehicle with regular, does that constitute misfueling?
The octane level for high altitude test fuels is not mentioned in the text, but rather shown in
Table IV-21, where EPA is proposing that the octane be set at 87.0 minimum (R+M)/2. Market
fuels in high altitude regions currently meet an 85 minimum (R+M)/2; we question the need to
raise the octane for the high altitude certification fuel if EPA is indeed looking "to better match
today's in-use fuel". Again, increasing the minimum octane to satisfy these vehicles and support
their high altitude certification would add significantly to the cost of this regulation.
Organization: Biotechnology Industry Organization (BIO)
BIO urges EPA against lowering octane as it allows for higher biofuel content in the emissions
test fuel in order to further help meet its air quality and public health goals, as well as to help
incentivize automakers and help them meet their CAFE requirements for model vehicles 2017
and beyond. Instead, BIO suggests EPA issue a final Tier 3 rule that maximizes investment in
biofuels, while also encouraging the highest achievable level of octane in transportation fuel.
Ethanol and other alcohol fuels including butanol, for instance, have higher octane ratings than
neat fuel. Higher octane enables higher compression and more efficient combustion engines. It
could actually boost fuel economy in future car models, according to a 2009 report by Sandia
National Labs, if automakers concentrate on developing smaller engines with higher
compression and turbocharging. For this reason, encouraging higher octane could help
incentivize automakers to make cars that run on cleaner fuel.
High octane blends including ethanol offer a means for auto manufacturers to meet higher
mileage CAFE standards. In fact, during EPA's recent public hearing on Tier 3, Mercedes Benz
publically announced its support for the Tier 3 proposed emissions test fuel and urged adoption
of higher blends including E25. And last year, the Alliance for Automobile manufacturers filed
comments to EPA stressing the need to transition to higher blends.
Organization: Chevron Products Company
Similar to CARB, EPA proposes to provide for both 87 and 91 anti-knock index (AKI)
certification fuels. This provides a base certification fuel that more closely matches AKI of the
currently predominant in-use fuel, while providing a mechanism for vehicle manufacturers to
utilize the benefits of a higher octane fuel if they so choose. This plan provides a good balance
between OEM flexibility and recognition of customer choice. However, the specification for the
'Premium' certification fuel should include a maximum AKI limit in order to ensure certification
on a similar octane to that which may be encountered in-use.
Organization: Growth Energy
4-269
-------
Chapter 4: Proposed Vehicle Emissions Program
Growth Energy also recognizes ethanol's important octane contribution to the nation's fuel
supply. As it relates to the proposed El 5 certification fuel, we feel that it is crucial that any
additional octane achieved by adding ethanol should not be lost by making lower-octane gasoline
blendstocks (CBOB and RBOB). As such, Growth Energy believes that a higher octane level
certification fuel would be more reflective of in-use octane levels of E15 fuels. Specifically, we
recommend a base octane level of 88.5 - 90 [(R+M)/2J.
Organization: INEOS Bio
INEOS Bio urges EPA against lowering octane as it allows for higher biofuel content in the
emissions test fuel in order to further help meet its air quality and public health goals, as well as
to help incentivize automakers and help them meet their CAFE requirements for model vehicles
2017 and beyond. Instead, INEOS Bio suggests EPA issue a final Tier 3 rule that encourages the
highest achievable level of octane in transportation fuel.
Ethanol has a high octane rating and is cleaner burning and more cost effective than the
petroleum based octane equivalent. Higher octane enables higher compression and more efficient
combustion engines. It could actually boost fuel economy in future car models, according to a
2009 report by Sandia National Labs, if automakers concentrate on developing smaller engines
with higher compression and turbocharging. For this reason, encouraging higher octane could
help incentivize automakers to make cars that run on cleaner fuel.
High octane blends of ethanol offer a means for auto manufacturers to meet higher mileage
CAFE standards. In fact, during EPA's recent public hearing on Tier 3, Mercedes Benz
publically announced its support for the Tier 3 proposed emissions test fuel and urged adoption
of higher blends including E25. And last year, the Alliance for Automobile manufacturers filed
comments to EPA stressing the need to transition to higher blends. A higher octane Tier 3 fuel
that includes >E15 is needed in order for these engines to meet the new CAFE standards for
2017 and beyond.
Organization: Marathon Petroleum Company LP (MFC)
EPA proposes to change the measurement approach from RON to AKI and lower the octane
requirement to be consistent with regular unleaded fuel at 87.0 (R+M)/2. EPA also proposes to
allow manufacturers to test on a fuel with a minimum octane rating of 91 (R+M)/2 for those
vehicles where operation on high-octane gasoline is required by the manufacturer. EPA is
seeking comment on the need for limiting the maximum octane of gasoline used in the
certification of premium-required engines and vehicles. As we will note in other comments, EPA
needs to limit the range of certification fuels to those that are readily found in the marketplace.
As such, a high-octane fuel specification should be limited to 91 (R+M)/2. While higher octane
fuels exist in the marketplace, they are not available on a nation-wide basis and several
significant regions offer 91 (R+M)/2 premium gasolines as the highest level available. These
regions include California, Arizona and Nevada where a significant number of premium vehicles
are sold by the manufacturers. Increasing the minimum octane of premium to satisfy these
vehicles and support their certification would add significantly to the cost of this regulation.
Another related issue is the requirements that would need to be put on the vehicles to insure that
4-270
-------
Tier 3 Summary and Analysis of Comments
only premium fuel is used in these vehicles. Will there be a requirement for special nozzles? Will
the "Check Engine" light illuminate? This special accommodation makes no sense unless EPA
has some way of insuring that this fuel is used in the field if it is required for emissions or other
regulatory performance attributes (e.g. CAFE compliance). Finally, if a consumer fills a
"premium" vehicle with regular, does that constitute misfueling?
The octane level for high altitude test fuels is not mentioned in the text, but rather shown in
Table IV-21, where EPA is proposing that the octane be set at 87.0 minimum (R+M)/2. Market
fuels in high altitude regions are currently meeting an 85 minimum (R+M)/2 and so we question
the need to raise the octane for the high altitude certification fuel if EPA is indeed looking "to
better match today's in-use fuel". Again, increasing the minimum octane to satisfy these vehicles
and support their high altitude certification would add significantly to the cost of this regulation.
Organization: National Automobile Dealers Association (NADA)
Lastly, a minimum 87 AKI octane rating should be specified.
Organization: National Corn Growers Association (NCGA)
The second principle is that ethanol blends above E10 should be splash blended to increase
octane above the level of 87 Anti-Knock Index (AKI). This will provide higher-octane fuels (not
higher priced "premium" grade fuels) to be widely used by the motoring public and by the
automakers in meeting the 2017 and Later Model Year Light-Duty Vehicle GHG and Corporate
Average Fuel Economy (CAFE) Standards.
Ethanol has a high octane value which has allowed gasoline refiners to supply gasoline
blendstock for blending with ethanol at levels significantly below 87 AKI, typically 84, such that
the finished gasoline meets the 87 octane minimum requirement. Virtually all El5 sold at retail
today is made using the same 84 AKI octane base gasoline, resulting in a higher blended octane
in the range of 88.5 to 90 AKI. EPA has proposed an 87 to 88.4 octane range for Tier 3 test fuel.
We believe the El 5 certification fuel octane should reflect the octane of retail El 5, ensuing that
the octane value of ethanol will continue to be used to raise the octane of finished gasoline above
87 AKI. Our recommendation is that EPA establishes an octane range of 88.5 to 90 AKI for E15.
This would also be consistent with an incremental improvement in fuel economy and GHG
emission performance of modern vehicle electronic emission control systems to this increase in
octane.
Organization: POET, LLC
POET supports a base certification fuel octane level above 88.4. As part of proposing an updated
certification fuel (based on El 5), EPA proposes lowering the specified based gasoline octane to
87.0-88.4 (R+M)/2 'to be representative of in-use fuel, i.e., regular-grade gasoline.' (26) POET
suggests that an octane of 88.5-90 (approximately 1.5 points higher) would more likely reflect
in-use octane levels for an El5 certification fuel, and an even higher octane range may be
appropriate. Indeed, as vehicle manufacturers move toward downsized and turbocharged
4-271
-------
Chapter 4: Proposed Vehicle Emissions Program
engines, octane levels of in-use fuels are likely to increase. As noted above, a recent MathPro
refinery study found that increasing the fuel octane to 92 AKI from 88 AKI would result in
significant cost savings.
Our Response:
Since more than 80% of the gasoline consumed in the U.S. is regular grade E10 gasoline
with an AKI of approximately 87, we proposed changing the octane rating of emission test fuel
to 87-88.4 AKI. This is also consistent with CARB's LEV III gasoline test fuel specifications.^
Manufacturers can continue to use high-octane gasoline (with a minimum octane rating of 91
AKI) for testing of premium-required[2] vehicles and engines as well as for testing unrelated to
exhaust emissions. (Comments related to the expected AKI levels of E15 in-use are no longer
relevant since we are finalizing E10 test fuel.)
Regarding establishing an upper limit for the octane range of premium test fuel, we have
decided to mirror CARB's approach of no upper limit (California is an area where lower-octane
premium fuels are likely to be found in-use). Further, manufacturers will design their premium
fuel required vehicles for the minimum octane market premium fuel which as discussed by
commenters is 91 AKI and any additional octane will not impact the emission performance of
the vehicle. Because of this, we do not feel it is necessary to constrain the certification test fuel
upper limit which would likely add additional cost to the certification test fuel.
An 85 AKI test fuel was not specified for high altitude regular gasoline because modern
spark ignition vehicles are designed and optimized for octane levels around the typical regular
grade of 87 AKI, and this is typically stated in the owner's manual. (In the case of mid-grade or
premium-required vehicles, the higher octane specification is for all altitudes.) While these
modern vehicles can compensate for octane levels less than the manufacturer's design intent, this
typically results in loss of performance and can cause an increase in GHG and other emissions.
Thus, it did not seem necessary or prudent to specify an additional unique test fuel with lower
octane specifications for emission testing at high altitude when manufacturers recommend or
require a minimum of 87 AKI or higher for the majority of their modern vehicles.
Some comments we received on the octane specification for emissions test fuel focused
on the need for in-use fuels to have higher octane and/or ethanol content. Those commenters
appear to have misunderstood the purpose of the change in emission test fuel specifications.
EPA is using its Clean Air Act section 206 authority to establish requirements for emission
certification testing, including specifications of the emissions test fuels with which vehicles
demonstrate compliance with emissions standards, in order to better reflect the ethanol content
and other properties of in-use fuels. While we are taking action to limit sulfur content in in-use
gasoline under authority in Clean Air Act sections 21 l(c)(l) and (2), we did not propose other
changes to in-use fuels. Any action related to changing the octane level of in-use gasoline is
outside the scope of this rulemaking.
[1] LEV III test procedures, including a description of test fuel, can be found at 13 CCR 1961.2.
^ Premium-required defined at §1065.710(d).
4-272
-------
Tier 3 Summary and Analysis of Comments
Some comments included in this section also pertain to the octane level of in-use fuel, a
topic to which we respond in Chapter 4.5.1.6 of this document.
4.5.1.4. Other Test Fuel Specifications
What Commenters Said:
Organization: American Coalition for Ethanol (ACE)
Aromatics — We support reductions in aromatics such as benzene and note that EPA
acknowledges in the NPRM that has ethanol blending has increased due to the RFS, harmful
levels of aromatics in gasoline have declined by nearly twenty percent over the last decade. El 5
and other high-level ethanol blends will continue to help reduce aromatics further. Further, there
is a growing body of evidence that shows that mid-level ethanol blends (i.e., E30) result in even
greater reductions in aromatics and resulting improvements in air quality and improved health
for U.S. citizens. We know that the Urban Air Initiative is submitting a detailed set of comments
on this particular point that deserve careful consideration by EPA.
Organization: Growth Energy
With the addition of ethanol, it may also be necessary to lower aromatics, benzene and specific
distillation temperature limits. ASTM Standard D4814 uses a T50 distillation temperature range
of 150 - 190 degrees, while EPA proposed a T50 distillation temperature range of 170-190
degrees. EPA should lower the lower end of the temperature range to be consistent with the
ASTM standard. Growth Energy also agrees that additional ethanol use can, and should, lower
total aromatics, and that should be reflected in the El 5 certification fuel accordingly.
Organization: American Petroleum Institute (API) and the Association of Fuel &
Petrochemical Manufacturers (AFPM)
In addition to ethanol content, EPA proposes to make changes to several other certification fuel
characteristics. API and AFPM recommend consistency between certification fuel standards and
the industry recognized standard for gasoline developed by ASTM International: ASTM D4814.
API and AFPM provide detailed comments relating to certification fuel octane, distillation
temperatures, sulfur, aromatics, olefins, aromatics distribution, and the test methods used by
laboratories to determine these characteristics.
The specifications of the test fuel should match up with the specifications set by ASTM in
D4814. To match up to ASTM D4814 the following changes should be made:
The T50 minimum distillation temperature should be 150 F for E10. The proposed T50
temperature of 170 to 190 F is too high given the boiling point depression caused by the
formation of an azeotrope between the hydrocarbons and the ethanol. Addition of the ethanol
depresses the T50 point by 20 to 350 F.
4-273
-------
Chapter 4: Proposed Vehicle Emissions Program
EPA proposes to adjust the gasoline distillation temperatures to better reflect today's in-use
gasoline / E10. We question EPA's approach to determining market fuel quality levels as they
"relied heavily on the AAM North American Fuel Survey trends" which are limited in scope and
confounded by lumping all fuel grades in to one analysis. EPA also used 2009 refinery
compliance data to assist in setting specifications. Using more recent data, and to be consistent
with ASTM specifications, we believe EPA should revisit the distillation properties with
suggested T10 in the range of 110°F to 130°F and more importantly a relaxation of the minimum
TSOto 150°F.
EPA is proposing to lower the sulfur content of the test fuel to 8-11 ppm to be consistent with the
proposed Tier 3 standards. We cover comments on acceptability in the technical justification
section of these comments (section LA).
EPA is proposing to reduce the aromatics content to better match today's in-use gasoline/El0.
As stated before, EPA is using limited AAM data and historic refinery compliance data to
support their conclusion on market fuel quality. In addition, Figure 3-4 in the RIA shows refinery
batch data for olefm levels and not aromatics. Based on our own data analysis, average aromatics
content for El0 blends during the past two summer seasons was 25 volume percent.
Similar to aromatics, average olefm content for El0 blends during the past two summer seasons
was 9 volume percent.
EPA is proposing to include a distribution of aromatics in the certification fuel to ensure that it is
more representative of in-use gasoline. We do not have recent data on composition distributions
for aromatics in market fuel but would note the sum of the distributed aromatics maximum is less
than the "Total Aromatic Hydrocarbons" maximum (24.4 versus 24.5).
EPA proposes to update some of the gasoline test procedures which we encourage them to do,
noting that some of the procedures may not be appropriate for E15 test fuels (e.g. D525 does not
include El 5). We also suggest EPA use the latest versions of the following tests: D2699, D2700,
D5191, D86, D5453, D3237, D130, D381, and D512 as these all have 2012 releases.
From an overall standpoint, we would recommend that EPA consider moving to an E10 regular
unleaded certification fuel that adequately describes the current in-use fuel quality while also
ensuring enough severity that Tier 3 emissions standards provide substantive emissions changes
versus Tier 2. To that end we would suggest the following key certification fuel properties:
Ethanol Content 9.8 to 10.2 volume %
Octane 87 to 88.4 (R+M)/2
DVPE 9.7 to 10.2 psi
T10 110°Ftol30°F
T50 150°Ftol70°F
T90 310°Fto330°F
FBP 380°Fto420°F
Aromatics 21.5 to 26.5 volume %
Olefms 6 to 12 volume %
4-274
-------
Tier 3 Summary and Analysis of Comments
Finally, EPA should indicate that the certification test fuel properties should in no way be
interpreted as limiting for in-use fuels. Rather the narrow range of fuel properties provides
consistency for EPA and industry when evaluating results from standard emissions tests and
other certification test programs.
Organization: Chevron Products Company
The proposed specifications that define the distribution of aromatics in the certification fuel
appear to be overdone. This level of detail will encourage certification fuel blenders to blend
near-pure components rather than more typical refinery streams. EPA and others have identified
examples of test programs in which fuel blends that used atypical refinery streams to blend test
fuels produced atypical emissions results. The proposal should be dramatically simplified, at a
minimum.
Organization: Chrysler Group LLC
EPA should adopt low sulfur market, refiner and retailer gasoline fuel.
Organization: Mercedes-Benz USA, LLC on behalf of Daimler AG
The primary factors which directly impact the Mercedes-Benz GHG Compliance Plan are sulfur
content in market fuel, and ethanol content and octane level in certification and market fuel.
Organization: INEOS Bio
We applaud the EPA for efforts in the Tier 3 proposal to reduce sulfur to lOppm, change the
emissions test fuel to El5 and encourage flexibility to allow for even higher blends, however
there were indications that the EPA intended to address the issue of aromatics and higher
minimum octane requirements, but this does not appear to be the case. As a cellulosic ethanol
producer, INEOS Bio would like to further highlight the benefits of higher ethanol blends and
how they can compliment and even improve upon the proposed Tier 3 requirements.
Organization: Marathon Petroleum Company LP (MFC)
Test fuel specifications should be consistent with ASTM D4814. The specifications of the test
fuel should match up with the specifications set by ASTM in D4814. To match up to ASTM
D4814 the following changes should be made:
The T50 minimum distillation temperature should be 150 F for E10. The proposed T50
temperature of 170 to 190 F is too high given the boiling point depression caused by the
formation of an azeotrope between the hydrocarbons and the ethanol. Addition of the ethanol
depresses the T50 point by 20 to 35 F.
4-275
-------
Chapter 4: Proposed Vehicle Emissions Program
EPA proposes to adjust the gasoline distillation temperatures to better reflect today's in-use
gasoline / E10. We question EPA's approach to determining market fuel quality levels as they
"relied heavily on the AAM North American Fuel Survey trends" which are limited in scope and
confounded by lumping all fuel grades in to one analysis. EPA also used 2009 refinery
compliance data to assist in setting specifications. Using more recent data, and to be consistent
with ASTM specifications, we believe EPA should revisit the distillation properties with
suggested T10 in the range of 110°F to 130°F and more importantly a relaxation of the minimum
TSOto 150°F.
EPA is proposing to lower the sulfur content of the test fuel to 8-11 ppm to be consistent with the
proposed Tier 3 standards. We cover comments on acceptability in the technical justification
section of these comments (section LA).
EPA is proposing to reduce the aromatics content to better match today's in-use gasoline/El0.
As stated before EPA is using limited AAM data and historic refinery compliance data to support
their conclusion on market fuel quality. In addition, Figure 3-4 in the RIA shows refinery batch
data for olefm levels and not aromatics. Based on our own data analysis, average aromatics
content for El0 blends during the past two summer seasons was 25 volume percent.
Similar to aromatics, average olefm content for El0 blends during the past two summer seasons
was 9 volume percent.
EPA is proposing to include a distribution of aromatics in the certification fuel to ensure that it is
more representative of in-use gasoline. We do not have recent data on composition distributions
for aromatics in market fuel but would note the sum of the distributed aromatics maximum is less
than the "Total Aromatic Hydrocarbons" maximum (24.4 versus 24.5).
EPA proposes to update some of the gasoline test procedures which we encourage them to do,
noting that some of the procedures may not be appropriate for E15 test fuels (e.g. D525 does
not include E15). We also suggest EPA use the latest versions of the following tests: D2699,
D2700, D5191, D86, D5453, D3237, D130, D381, and D512 as these all have 2012 releases.
From an overall standpoint, we would recommend that EPA consider moving to an El0 regular
unleaded certification fuel that adequately describes the current in-use fuel quality while also
ensuring enough severity that Tier 3 emissions standards provide substantive emissions changes
versus Tier 2. To that end we would suggest the following key certification fuel properties:
Ethanol Content 9.8 to 10.2 volume %
Octane 87 to 88.4 (R+M)/2
DVPE9.7to 10.2 psi
T10 110°Ftol30°F
T50 150°Ftol70°F
T90310°Fto330°F
FBP380°Fto420°F
Aromatics 21.5 to 26.5 volume %
Olefms 6 to 12 volume %
4-276
-------
Tier 3 Summary and Analysis of Comments
Finally, EPA should indicate that the certification test fuel properties should in no way be
interpreted as limiting for in-use fuels. Rather the narrow range of fuel properties provides
consistency for EPA and industry when evaluating results from standard emissions tests and
other certification test programs.
Organization: New York State Department of Environmental Conservation
In addition to setting the ethanol content of certification gasoline at 10 volume percent, EPA
should specify a T50 range consistent with the norms reported on E10 batch reports. We note
that, except for T50, the certification gasoline distillation curve proposed by EPA lines up quite
well with data from New York's 2011 and 2012 gasoline sampling program for summer
conventional gasoline.
Minimum requirements for certification gasoline aromatics and olefms content, and a realistic
distribution of aromatics should be promulgated as proposed.
Current federal certification gasoline standards have no minima for olefms and aromatics. This
allows gasoline with unrealistically 'clean' properties to be used for vehicle certification and
testing. The Department strongly supports EPA's proposal to specify ranges (minimum and
maximum) for aromatics and olefms that are based on actual gasoline production batch reports.
Equally important is EPA's proposal to require a realistic distribution of aromatics by carbon
number.
These changes will ensure that vehicle development, certification, and compliance testing is
conducted on a fuel that reflects the composition of commercial gasoline. Although we can't
quantify emissions benefits, we expect that this will lead to more effective emissions control
systems and real world air quality benefits.
Organization: Phillips 66 Company
The test fuel specifications for an E10 test fuel should align with the ASTM gasoline
specification, D4814. EPA stated that they reviewed refinery test data as one input to set
specifications for various test fuel properties. Currently, all conventional gasoline is tested and
reported to EPA on a clear basis so would not include the effect of the blended ethanol that is
added downstream. A few notes on selected properties:
The T50 minimum should be 150° F. T50 specifications for clear gasoline or subgrades is 170° F
while the minimum specification for ethanol blended gasoline is 150° F, accounting for the
boiling point depression experienced with ethanol blends.
API and AFPM have included additional detailed comments on other properties based on more
recent data and refining experience with gasoline blends, testing, and specifications. We support
the comments offered by API and AFPM.
Organization: POET, LLC
4-277
-------
Chapter 4: Proposed Vehicle Emissions Program
POET also suggests that aromatics, benzene, and certain distillation temperature limits may
warrant lowering from what EPA has proposed, to reflect a 15% ethanol content. For instance,
EPA explains that 'Additional ethanol blending to produce El5 is expected to result in even
greater aromatics reductions.' POET agrees that increased ethanol use, including via El5, can
lower both total aromatics and benzene, and that lower certification fuel aromatics levels are
appropriate. Both ethanol and aromatics are a source of octane, but ethanol provides a more cost-
effective and less toxic source of octane.
Organization: Renewable Fuels Association (RFA)
Further, RFA suggests the EPA recognize the T50 temperature allowed for ethanol blended
gasoline of 150°F in the ASTM International "Standard Specification for Gasoline for use in
Spark Ignition engines." RFA strongly recommends EPA align the certification fuels for non-
road engines as soon as possible with on road engine certification fuel. This ensures a uniform
capability of spark ignition engines thereby streamlining the fuel options in the retail
marketplace.
Organization: National Automobile Dealers Association (NADA)
To enable Tier 3 tailpipe standard compliance, EPA should establish a 10 ppm gasoline sulfur
average with a 25 ppm cap. To enable Tier 3 evaporative emissions compliance, a Reid Vapor
Pressure (RVP) cap should be set at 9 psi for a maximum E10 gasohol test fuel.
Our Response:
We are adjusting gasoline distillation temperatures to reflect in-use E10 gasoline to the
extent practical. This includes adjustments to our proposed T10, T90 and FBP ranges based on
review of updated Alliance of Automobile Manufacturer (AAM) fuel surveys and refinery batch
data. Regarding T50, these data show a range of approximately 150°F to 220°F in in-use
gasoline. Adopting a wide specification range for test fuel may have undesirable effects on
consistency of results between facilities and over time. Therefore, we have chosen a range of
190-210°F to represent both conventional and reformulated gasoline (as well as maintaining
some overlap with CARB's specification of 205-215°F).
We are lowering the sulfur content of test fuel to 8-11 ppm to be consistent with the Tier
3 in-use gasoline sulfur standards. Sulfur level has a significant effect on emissions so it is
important to adjust the test fuel specification to match in-use fuel.
We are setting a benzene test fuel specification of 0.5-0.7 volume percent to represent in-
use fuel under the MSAT2 regulations. The MSAT2 standards, which took effect January 1,
2011, limit the gasoline pool to 0.62 volume percent benzene on average.33
We are lowering the range of aromatics content in the test fuel to better match today's in-
use E10 gasoline, and narrowing the range to limit variability of results. Data from recent
gasoline batch data as well as AAM surveys support a specification of 22-26 volume percent. In
33 72 FR 8434 (February 26, 2007).
4-278
-------
Tier 3 Summary and Analysis of Comments
addition to total aromatics and benzene, the updated test fuel requirements place boundaries on
the distribution of aromatics by carbon number (i.e., prescribed volume percent ranges for each
of C7, C8, C9, and C10+ hydrocarbons). There is evidence that the heaviest aromatics in
gasoline contribute disproportionately to PM emissions, so compliance with emission standards
should be demonstrated on fuel with a composition representative of in-use gasoline.
Some commenters raised concerns about specific ranges for distribution of aromatics
resulting in use of pure compounds in test gasoline, which would not be representative of in-use
fuels. On the contrary, historically, emissions test gasoline has often been found to have been
produced using high proportions of single components (e.g., toluene, iso-octane), and therefore
having additional targets for distribution of aromatics may actually increase the number of
blending components required to meet the specification.
We are adjusting the olefins specification to a range of 4-10 volume percent to better
match in-use E10 gasoline.
We are setting a range of 8-11 ppm sulfur in test fuel to mirror the expected in-use range
after phase-in of the in-use sulfur standard of 10 ppm average.
We are adding distillation residue, total content of oxygenates other than ethanol, copper
corrosion, solvent-washed gum, and oxidation stability specifications to better control other
performance properties of test fuel. These specifications are consistent with ASTM's D4814
gasoline specifications and CARB's LEV III test fuel requirements.
We are updating some of the gasoline test methods previously specified in §86.113 with
more appropriate, easier to use, or more precise test methods for ethanol-blended gasoline. Key
changes include replacement of ASTM D323 with ASTM D5191 for measuring vapor pressure;
replacement of ASTM D1319 with ASTM D5769 for measuring aromatics and benzene; and
replacement of ASTM D1266 with three alternative ASTM test methods (D2622, D5453 or
D7039) for measuring sulfur. The most current versions of these standards were specified where
possible and appropriate. Additional response to comments on test methods can be found in
Chapter 6.1.1.7 of this document.
Some commenters raised concerns about over-reliance on AAM fuel surveys, as well as
lumping premium and regular grade fuel data together, when performing our analyses. We have
generally performed parallel analyses using both refinery batch as well as AAM survey data
where possible, and have focused on regular grade fuel where we believed that inclusion of
premium would bias the results.
More information on data and analyses related to test fuel specifications can be found in
Chapters oftheRIA.
4.5.1.5. Alternative Test Fuel Option / E30
What Commenters Said:
4-279
-------
Chapter 4: Proposed Vehicle Emissions Program
Organization: Advanced Biofuels USA (ABFUSA)
As part of the recently proposed Tier 3 motor vehicle fuel and emission regulations, EPA
included a very forward looking idea that could bring higher octane, higher ethanol gasoline to
the marketplace.
However, Advanced Biofuels USA recommends the following changes to assure consumers get
maximum fuel economy and climate change mitigation benefits from the EPA proposed higher
octane, higher ethanol gasoline.
• Advance Biofuels USA fully supports EPA's recognition of the importance of higher
octane, higher ethanol (in the range of E30) gasoline as a cost-effective way to allow
manufactures to maximize the efficiency of smaller, more efficient engines that utilize
high combustion pressures to meet 2022 EPA CO2 vehicle emission standards.
• To provide a smooth path to making this "higher octane, higher ethanol content gasoline"
available nationwide EPA Tier 3 regulations should not require individual vehicle
manufacturers to certify the availability of this fuel. Instead, EPA should use their
authority under section 211 of the Clean Air Act to provide for the commercial
availability of this "higher octane, higher ethanol content gasoline."
• In using the Clean Air Act authority to assure that "higher octane, higher ethanol content
gasoline" is available nationwide EPA should implement a reasonable phase-in schedule
tied to manufacturer production plans required to meet 2017 and later EPA GHG
requirements. This schedule should be based on the "vehicles would not operate
appropriately on other available fuels, and such a fuel would result in equivalent
emissions performance," information.
• EPA should allow vehicle manufacturers that certify new vehicles with the "higher
octane, higher ethanol content gasoline" to also certify that those vehicle are able to also
operate on existing E10 or E15 fuels. These vehicles would be called "E30 capable."
• By building up the number of these "E30 capable" vehicles that could get the same
mileage with a lower cost fuel, the demand for E30 would increase. This demand would
create a nationwide E30 infrastructure that would then allow for the marketing of "E30
Optimized" Vehicles designed to provide the fuel economy and GHG reductions
necessary to meet 2022 CO2 reduction standards.
• Since a higher proportion of lower cost ethanol is used to produce E30, the "higher
octane, higher ethanol content gasoline" would probably not be priced above current 87
octane E10 regular. Therefore, it should not be referred to as "premium" fuel in final
regulations or for purposes of marketing but rather should be labeled differently. For
example, "E30 capable/E30 optimized regular" fuel.
• EPA should also provide flexibility in the Tier 3 regulations so that other renewable,
negative GHG (as compared to petroleum) octane additives could be substituted for 30%
ethanol when they are commercially available. In order to qualify, these additives should
have to meet the applicable EPA regulations for fuel composition, aromatic content, and
certification testing. This flexibility would promote competition in the biofuel
marketplace that would result in the most sustainable low GHG solutions possible.
By adopting a flexible, market-based Tier 3 higher octane, higher ethanol content gasoline
program, EPA would send a clear Demand Certainty Signal to markets and fuel providers. This
4-280
-------
Tier 3 Summary and Analysis of Comments
clear signal will provide the currently missing certainty to the financial markets that will be
needed to provide the capital for total-biomass ethanol and other advanced biofuel plants that
will be needed to provide the "higher octane, higher ethanol content."
This clear market signal would be the most effective way to meet EPA's stated goal to, "provide
a market incentive to increase ethanol use beyond E10 and enhance the environmental
performance of ethanol as a transportation fuel by using it to enable more fuel efficient engines."
EPA has the Authority under Section 211 of the Clean Air Act to Implement Nationwide
Marketing of "Higher Octane, Higher Ethanol Content Gasoline".
As stated in this proposed rulemaking on pages 32-33, "We are proposing to adopt gasoline
sulfur controls pursuant to our authority under section 21 l(c)(l) of the CAA. This section allows
EPA to establish a fuel control if at least one of the following two criteria is met: (1) The
emission products of the fuel cause or contribute to air pollution which may reasonably be
anticipated to endanger public health or welfare; or (2) the emission products of the fuel will
impair to a significant degree the performance of any emissions control device or system which
is either in general use or which the Administrator finds has been developed to a point where in a
reasonable time it will be in general use were the fuel control to be adopted." Using the first
criteria, the establishment of a higher fuel octane gasoline standard (fuel control) would reduce
CO2 emissions from direct injection/turbocharged/high compression engines below what they
would be with lower octane, lower ethanol content gasoline. This reduction in CO2 emissions,
which have previously been determined by EPA to endanger public health and welfare as climate
change inducing Greenhouse Gas (GHG) emissions, would come from two effects:
1. These engines would run at higher efficiency with the higher octane fuel meaning less
fuel/mile would be used resulting in less CO2 emissions/mile.
2. The increase from 10% to approximately 30% ethanol would decrease life-cycle
GHG emissions when compared to the petroleum gasoline components it would
replace. This decrease would become especially significant as proportion of total
biomass (i.e. cellulosic) ethanol increases in accordance with the previously
Congressional enacted Renewable Fuel Standard (RFS).
It making this determination to require higher octane, higher ethanol content gasoline, it is also
important to recognize that these interrelated engine design features (the Ford Ecoboost is an
example) are being implemented for the sole purpose of meeting the 2022 EPA CO2/DOT mpg
standards while also providing enough power to compensate for the extra weight requirements of
DOT safety standards, and are not being introduced for high performance reasons. Therefore,
engines using this integrated package of design features are as much emission control devices as
the catalytic converters that caused EPA to issue lead elimination and sulfur reduction fuel
control regulations to ensure lower HC, NOX, and CO emissions.
Organization: Algenol Biofuels Inc.
Algenol appreciates EPA's effort - as it proposes to strengthen the Tier 3 Vehicle Emission and
Fuel Standards— to accomplish that objective in a way that maximizes the role which can be
4-281
-------
Chapter 4: Proposed Vehicle Emissions Program
played by biofuels blending. We believe EPA correctly seeks to implement the emissions
reduction and biofuels blending programs in the most mutually reinforcing way.
In particular, we wish to endorse and encourage the Agency's proposal, in part IV.D. of the
proposed Tier 3 rule, to allow vehicle manufacturers to specify an appropriate alternative test
fuel, such as an E30 fuel, (a) for vehicles designed to be operated with higher octane, higher
ethanol fuel,, optimized to run on an E30 or higher blend, (b) and/or for flex-fuel vehicles
(FFVs). If properly constructed, this option will help to increase the demand for biofuels and
facilitate their use by the driving public. We understand that Mercedes and other major auto
manufacturers have expressed serious interest in this opportunity.
Auto makers and the driving public could benefit from having such fuel available and used in
designated vehicle models because it would enable the manufacturers to raise compression ratios
and improve fuel efficiency in such vehicles as a step towards complying with 2017 and later
greenhouse gas and CAFE fuel economy standards. The public would also benefit, economically
and environmentally, from such an incentive for increased ethanol use, especially insofar as the
ethanol involved is from a low-cost source such as Algenol.
EPA proposes, in Part IV.D.l of the proposed Tier 3 rule, to allow manufacturers to petition for
use of such a mid-level blend in government-required emission certification tests for designated
vehicles if the manufacturers demonstrate that (1) such fuel would be available nationwide, (2)
such fuel would be used by operators of such vehicles, (3) the vehicles would not operate
appropriately if other fuels were used, and (4) the fuel would result in equivalent (or better)
emissions performance to that attainable with other fuels.
We believe that the first condition above is beyond what can realistically be expected for auto
manufacturers to demonstrate. Instead, EPA needs to take regulatory action, under its existing
statutory authorities, to mandate that the E30-type fuel be available in adequate quantities by a
date certain. One precedent is EPA's action in the 1970s to mandate the availability of unleaded
gasoline to match the introduction of vehicle models with catalytic converters to prevent
"poisoning" of those catalysts which greatly reduced automotive emissions below prior level.
(The writer supervised that action while at the Agency.)
We also urge that EPA give careful consideration to any reasonable requests that auto
manufacturers may make regarding the 2d - 4th Agency-proposed criteria listed above. The 3rd
and 4th criteria seem generally appropriate, depending on how they are actually interpreted and
applied, but the second criterion - that drivers will use the fuel — may need to be modified
somewhat or at least applied with some flexibility. For example, this condition probably should
not be construed to mean there is no conceivable way that some perverse drivers might not
circumvent the intent of the program even though they would risk damaging the vehicle and
possibly paying more for the inappropriate fuel they used.
As a company that will be well along into commercial production of ethanol by the time an E30
type option would go into effect, Algenol has a strong and legitimate interest in urging that the
conditions for allowing use of E30 type fuels for vehicle testing be realistic to meet so that the
statutory objective of expeditiously expanding use of ethanol can be meaningfully met. This
4-282
-------
Tier 3 Summary and Analysis of Comments
Algenol interest is consistent with the public interest which gave rise to the Congressional
directive that EPA act to maximize biofuels use in vehicles insofar as this reasonably supports its
other mandates such as minimizing vehicle emissions.
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
Recommendations concerning Optional Alternative Test Fuels. Vehicle manufacturers
understand there could be a need for new test fuels in the future, such as higher octane, higher
ethanol blends, depending on how market fuels develop and vehicle technology needs evolve.
For example, there are potential benefits from such blends, assuming the octane grades of the
underlying gasoline blend-stocks do not decline. Higher octane/higher ethanol content can
improve vehicle efficiency and lower emissions. Automakers concur that EPA needs to replace
the flawed existing regulations applicable to use of alternative test fuels [40 C.F.R. Sec.
1065.701(c)], discussed further below. We look forward to working with the Agency and other
stakeholders to develop better options for use of alternative test fuels.
One pathway to improved vehicle efficiency and lower GHG emissions is to increase the engine
compression ratio (CR). However, an increased compression ratio may result in the engine then
requiring a fuel with a higher octane number (ON), a measure of the fuel's resistance to auto-
ignition that causes engine knock. Since ethanol has a higher ON than gasoline, a mid-level
gasoline-ethanol blend (prepared by blending denatured ethanol into an E10) will have a higher
octane number than the E10, allowing the engine to be designed with a higher CR. Ethanol's
high heat of vaporization, which produces an in-cylinder cooling effect, and its high octane
number compared to that of gasoline, can increase engine efficiency. The resultant efficiency
gains can also enable a downsizing of the engine to reduce weight, thereby further reducing fuel
consumption and CO2 emissions.
The Coordinating Research Council (CRC) conducted studies investigating the body of literature
on the effects of fuel octane quality and ethanol on engine and vehicle fuel economy and
emissions. Additionally, Ford conducted a study comparing the effects of E10, E20, and E30 on
fuel economy and CO2 emissions from a turbocharged direct-injection engine. This engine was
tested with CRs of 10:1 (current production) and 11.9:1. The selected E20 fuel operated in an
engine with 11.9:1 CR gave similar knock performance to the E10 fuel operated in the same
engine set to 10:1 CR. Similarly, the E30 fuel run at 11.9:1 CR resulted in knock performance
equivalent to E20 at 10:1 CR.
In a vehicle simulation study, the E20 at 11.9:1 CR provided a 5% reduction in CO2 emissions
compared to the El0 at 10:1 CR, and the fuel economy (MPG) was about the same, even though
the energy content of fuel was lower as the ethanol content was increased. To provide the desired
high knock resistance, the higher-ethanol fuels should be formulated by using a blend-stock that
retains the octane quality of the current blend-stock used for regular-grade E10.
While higher ethanol, higher octane fuels can be useful in all types of engines to varying degrees,
they are of particular benefit to direct-injection (DI) engines. Ethanol has a much greater heat of
4-283
-------
Chapter 4: Proposed Vehicle Emissions Program
vaporization when compared to gasoline. Thus, in a port fuel injected (PFI) engine, the chemical
octane rating increase can be realized, but since much of the fuel is inducted into the cylinder
after it has vaporized, the cooling effect is not captured. In a DI engine, however, the fuel is
introduced directly into the cylinder in liquid form and its vaporization in the cylinder results in a
larger cooling effect compared to the PFI case, thereby reducing the octane number requirement
of the engine. However, advanced design for either type of engine can benefit from higher octane
fuels.
EPA Should Replace Flawed Regulations Governing Use of Alternative Test Fuels. Automakers
agree that alternative test fuel provisions in the current regulations at 40 C.F.R. §1065.701(c) are
flawed and cannot work in practice. For dedicated vehicles, the regulation requires that the new
fuels are commercially available in the marketplace, and it appears to presume the fuels would be
widely available without any vehicles being available to use them, which underscores the
unresolved "chicken and egg" dilemma.
In addition, automakers would have to show that the vehicles will not run or be durable on
standard test fuel. Alternatively, automakers could show that the vehicles would not meet
emissions limits on standard test fuel, any of which would likely make them virtually unsellable.
Finally, Sec. 701(c) does not address F Factors that would have to be determined for these
alternative fuel types. Rather than imposing unrealistic regulatory pre-conditions, EPA needs to
replace them. Were an OEM (Original Equipment Manufacturer) to petition to test using El5 test
fuel under these provisions, meeting the criteria would be very challenging. Automakers would
be happy to work with the Agency to develop a workable process and criteria for access to
alternative test fuels.
Approval of a fuel for certification test purposes does not change the CAA requirements that a
new fuel be independently approved by EPA as substantially similar (or a waiver provided) with
full notice and comment rulemaking before use as market fuel (other than for FFV use for up to
E85). Likewise, mechanisms to avoid mis-fueling in vehicles or equipment not designed for a
particular fuel use would also have to be addressed.
EPA's Approach to Use of Premium Octane Grade Test Fuel is Sound:
Automakers support consistent reliance on communications in the Vehicle Owner Manual
regarding vehicle care and fueling recommendations, and terms of warranty protection. Owner
Manuals requiring (as opposed to recommending) use of Premium Octane Grade fuel are the best
means to determine which vehicles should be allowed to test on it. Also, EPA's approach
provides critical consistency with California's requirements.
EPA is seeking comment on the need for limiting the maximum octane of gasoline used in the
certification of premium-required engines and vehicles. Automakers agree that there is no need
to limit the maximum octane of gasoline used in the certification of premium-required engines
and vehicles. Doing so would be an obstacle to the development of highly efficient optimized
vehicles.
4-284
-------
Tier 3 Summary and Analysis of Comments
EPA should finalize its provision for testing on premium grade octane fuel as written.
Organization: Clean Fuels Development Coalition (CFDC)
That notwithstanding, we respectfully draw your attention to several key points that warrant
repeating. First is the issue of an E30 certification fuel and providing this option to automakers.
Many have indicated their support for high octane fuels and recognize ethanol's outstanding
octane properties. For those automakers that choose to do so allowing them a pathway to E30
certification is another key step in expanding ethanol markets. For those who find this an
attractive option, a poison pill can easily be viewed as the requirement that automakers must
demonstrate the commercial availability of such fuel. We believe that is unnecessary and quite
impractical. The successful CAFE credits afforded to automakers for example in the production
of flex-fuel vehicles, was achieved completely independent of fuel supplies and automakers
responded to market developments and regulatory signals by making more FFVs available. We
believe this would certainly be the case with E30 capable or optimized vehicles and if not
corrected in this Rule would be a huge impediment to them. Similarly, infrastructure would
follow as more of these vehicles were on the road.
Organization: American Coalition for Ethanol (ACE)
Alternative Certification Fuel (such as high-octane E30):
To the contrary, EPA suggests auto manufacturers can chose to certify on higher ethanol blends
only if they 'could demonstrate that such a fuel would be used by the operator and would be
readily available nationwide.' ACE believes such a requirement gives control of fueling
decisions to the oil industry, which has historically only allowed new blends of ethanol when
they have been required to do so by law. Oil company contracts banned E10 when it came to the
marketplace, effectively restricted E85 to only a small percentage of stations, and currently
prevent branded stations from marketing El 5. The oil industry would likely continue to limit
El 5 and place an outright ban on E30 if it believed it could keep car manufacturers from
building cars that can run on those fuels that compete with their petroleum fuels.
ACE appreciates that EPA proposes to allow automakers who wish to increase compression
ratios and make other advanced technology improvements to engines in order to comply with the
2017 and later GHG/CAFE standards, to specify an alternative high-octane test fuel such as E30.
As such, we strongly recommend that EPA establish an E30 (or comparable midlevel blend of
ethanol and gasoline) certification fuel for model year 2017 and beyond, with incentives offered
to encourage automakers to make that conversion earlier. As mentioned earlier, ACE strongly
objects to EPA's suggestion that approval of such a certification fuel is contingent upon
automakers' demonstrating commercialization of the fuel.
That suggested requirement is undoubtedly based on E85 history and the lack of E85 fueling
stations. ACE believes that to be an inaccurate comparison. Automakers did not build E85
vehicles. They built Flex-Fuel Vehicles. As such, since FFVs could also use unleaded gas or
E10, station owners already had fueling infrastructure to provide fuel for FFVs. Buying new
tanks, lines, and dispensers to offer fuel to vehicles a station could already serve only made sense
at times of large price advantage, and later, when marketers learned how to use E85 and blender
4-285
-------
Chapter 4: Proposed Vehicle Emissions Program
pumps to recover some of the ethanol blending margin that refiners were taking through pre-
blended E10. Had automakers built E85 vehicles, station owners would have had to decide
whether to concede that market to competitors or offer E85 for a competitive edge.
In an effort to set aside the infrastructure discussion, while ACE prefers a higher octane fuel with
a minimum ethanol content requirement such as E30 for use as a certification fuel, as an
alternative, we feel so strongly about ethanol's octane value that we could support a transitional
certification fuel that specifies only the octane of the fuel. Existing fuel infrastructure could be
used to sell a 95 octane consumer fuel product, for example, so the burden of infrastructure
development would not fall on ethanol. We believe other properties of ethanol will make octane
from ethanol more attractive to automakers, and ACE believes in a truly competitive
marketplace, ethanol's low-cost will present profit opportunities for station owners that far
outweigh costs to upgrade, while still allowing late adapting fuel stations an option for fueling
new vehicles with high octane gasoline.
In either case, it is imperative that automakers let fuel station owners and consumers know well
in advance what kind of fuel — either by octane or by ethanol content, or both - they plan to use
in the future, so that both have time to adjust for that fuel.
A recent New York Times article1 reports that an alternative certification fuel such as E30 enjoys
widespread support from technical automotive experts. Mr. William H. Woebkenberg, senior
engineer for fuels policy in the U.S. at Mercedes-Benz, says in that article than an E30 blend in
an engine designed to use it would be attractive to consumers with 'ridiculous power and good
fuel economy,' and that those car owners would seek out the fuel.
As EPA balances the administration of Tier 3, CAFE-GHG, and the RFS; adoption of higher
octane blends such as E30 will be critical to achieving the desired impact of each policy. As
such, EPA should more directly facilitate higher octane blends like E30 by independently
approving a cert fuel for these types of higher octane blends.
Organization: California Air Resources Board (CARB)
C ARE staff also supports the proposed requirements for midlevel ethanol blends, as well as
promoting the power plant design advantages of an E30 blend.
Organization: Chevron Products Company
Per the request for comment on the appropriateness of the alternative test fuel provisions at
§1065.701(c), the regulatory language already in place appears to be adequate given that it: 1)
requires EPA approval for the specifics of the test fuel; and 2) requires that there be evidence of
commercial availability. The existing language presents no barriers to the certification of a
higher octane, higher ethanol content test fuel and there is no need to more precisely specify the
makeup of potential alternative fuels. The current language allows for innovation in vehicle and
fuel technology. A more prescriptive approach which dictates the nature of such fuels would
only create barriers to such innovation. It should be noted that the example of a higher octane,
4-286
-------
Tier 3 Summary and Analysis of Comments
higher ethanol fuel to be used in dedicated vehicles is in fact an alternative fuel rather than
gasoline and should be treated as such.
Organization: American Council on Renewable Energy (ACORE)
Furthermore, I believe there are several additional critical points that should be considered before
EPA finalizes Tier 3 regulations, and propose actions that should be taken as soon as possible. I
make these proposals based on my experience as the Director of the Office Regional Operations
at the Federal Energy Administration (FEA). There I learned the hard way that oil companies,
during our national transportation fuels crises in the mid-1970s, illegally took action to gain
corporate power and position at the expense of the American people. The FEA took many of the
oil companies to court and won fines large enough to fund the energy offices in all 50 states for
about two decades. During this critical period of fuel shortage, they were still able to boost
profits, eliminate the lower levels of their competition (some independent fuel suppliers and most
of the "mom and pops" that were involved in their own communities) to further concentrate their
dominance of the transportation fuels market.
From then on, I made a personal commitment to find an alternative to oil as the primary source
of transportation fuels. With the help of EPA Administrator Russell Train and C. Boyden Gray,
Council to Vice President Bush, I was able to help pioneer the ethanol industry starting in 1975
and the biodiesel industry in 1980.
It is critically important that the United States Congress and the Administration take into full
consideration critical factors impacting America's biofuels program. Congress should not be
swayed by the powerful and widespread exaggerations berating ethanol orchestrated by the oil
industry and their colleagues, who seek lower corn prices. The animal feeders and those formerly
enjoying the benefits of subsidized corn through lower meat prices, at the taxpayers' expense,
want a return to the earlier lower corn prices that were below free market costs. This would lead
to the U.S. being forced to return to price subsidies that would disrupt free market forces to the
disadvantage of corn farmers worldwide; and, at the expense of the U.S. taxpayers.
The ethanol industry has eliminated the need for corn subsidies in the U.S., resulting in free
market prices worldwide for both corn and ethanol. This was due to the major contributions by
the ethanol industry and the wisdom of the Congress and the Administration to terminate the
VETEC that was subsidizing ethanol.
Additionally, this rhetoric, focused on discounting critical factors such as resource availability,
technological and scientific advances, the U.S commitment to keep the world's oil lanes open for
commerce, true climate change factors, and the value of additional externalities, swayed the
political process so that these factors are no longer principal determinants in establishing market
share for various transportation fuels. These conditions have been overpowered by political clout
influenced by misinformation and there are two very recent examples of this. The first is when
Harley Davidson motorcycle riders circling the Capitol building and visiting congressional
offices to persuade votes against E-15. EPA regulations do not allow the use of this fuel in
motorcycles, and motorcycles have easy access to ethanol free gasoline. The second entails
restaurant associations demonstrating on the Hill and in the media against the RFS claiming that
4-287
-------
Chapter 4: Proposed Vehicle Emissions Program
corn-based ethanol leads to an increase in meat and dairy prices. It is well established that oil
costs have a greater impact on these prices than the cost of corn, and the availability of distillers
grains replacing a good percentage of the feed for animals is further offsetting in reducing the
cost of meat. Then there is the issue that meat, particularly fatty meats, is losing popularity;
already leading to less corn in some animal diets. These factors and other mentioned externalities
including sustainability, climate change, short and long range economics, and job creation — not
misleading rhetoric must be used in the making process. One can assume the connection to the
bikers and the restaurateurs to the oil industry are not hypothetical.
The oil industry was using similar spin tactics almost a century ago to promote dangerous lead in
gasoline to attain needed octane for advanced engines. They defeated cleaner-burning ethanol
preferred by many, including Henry Ford. They were able to knowingly allow the introduction of
lead into the blood of millions of Americans for almost a half century at enormous expense to
human health, including very significant reductions in the IQs of kids living in traffic-congested
cities until it was phased out and replaced by methyl tertiary butyl ether (MTBE). This chemical
is a known ground water contaminant; again, large oil conglomerates restrained the use of 113
octane ethanol while advancing that contaminant. When MTBE was phased out because of
ground water contamination and other health problems, their preference shifted to aromatics for
most of their octane, leaving it up to free market forces, the marketplace and legislation (the
Renewable Fuel Standard—RFS) to incorporate ethanol into gasoline at the 10% level. The oil
industry, the public, and the environment benefited from the RFS because of its lower financial
cost as an octane enhancer and the other previously mentioned externalities. However, oil began
to lose market share. About that time, EPA was considering authorization of the additional use of
ethanol (E-15), and it was becoming apparent that the auto industry was looking into 93+ octane
fuels to permit the transition to direct injected, high compression/turbo charged engines. The oil
industry knew this combination could open the way for higher blends of ethanol and increasing
numbers of flex-fuel vehicles (FFV) on the road, serviced by blender pumps. The market share
for gasoline was being seriously threatened - totally unacceptable to the oil industry.
They turned to their proven power base: money, political clout, and disinformation. Their
decade-long campaign has been overwhelmingly successful throughout society, especially in the
Congress that is well attuned to: public opinion and their support; successes of U.S. oil and gas
production; reduced oil import dependence; and, of course, endless anti-ethanol rhetoric. This all
leads to good news for fossil fuels and bad news for ethanol.
Also, allow me to share with you my reaction to the President's speech on climate change.
Thankfully, he mentioned a commitment to biofuels, but without proposed action except for
support for the Renewable Fuels Standard and biofuels in general. I believe he has a stronger
commitment to ethanol, but was restrained by the threatening climate orchestrated by the oil
industry and their compatriots. That is the key reason for my concern. The President of the
United States, his key staff members, Congress, and the public have, for quite some time, been
pressured by the oil industry and their colleagues influenced by the involved animal feeding
industries. This powerful force has set the stage for their victory. Non corn-based biofuels in
general still has good support. The corn-based ethanol industry had a strong support base while
pulling the industry to today's point of providing 10% of the gasoline mix while contributing
billions of dollars annually to the nation's economy. In recent years, however, it has essentially
4-288
-------
Tier 3 Summary and Analysis of Comments
lost most of its support across the board in a variety of ways. This is all good news for the fossil
transportation fuels, and bad news for biofuels, particularly the pioneering corn-based ethanol. I
believe President Obama would like to fix this inequity; but political and public conditions signal
otherwise. However, EPA has the authority and the reasons to come to the support of the
President, the nation and the environment.
Without EPA action, all of this signals a possible demise of the ethanol industry unless EPA
embraces it's legally imposed responsibilities, the compelling signals from the marketplace, and
its mandate to protect the environment (including the consequences of climate change and public
health). Comments by the Coalition/Urban Air Initiative (EFC/UAI) provide all the evidence
needed for EPA to immediately move to correct "oversights" and delays in their actions over the
past 35 years, since the forced phase-down of lead in gasoline. The agency should not be delayed
further by current political and public opinion, nor mislead by a massive and prolonged anti-
ethanol campaign.
To further reinforce the EFC/UAI comments and my recommendation, attached is a copy of my
article published in the Physicians for Social Responsibility: Environmental Health Policy
Institute that reinforces my comments on the unacceptable negative health effects of excessive
aromatics in gasoline. It also touches on the positive environmental benefits of corn and it
conversion in a biofuels.
EPA must carry out their responsibilities and opportunities under the law to protect public health,
the environment, and the best interest of the nation. It should fully support the President's
commitment to biofuels and the essential need to reduce carbon emissions, and not continue to
bow to the powerful influence of those committed to the continuation of oil's dominance in the
transportation sector.
In summary, here are the benefits to the nation, using existing authorities, to reduce aromatic
levels in gasoline, boost levels of renewable alcohols, incentivize the production of FFV, and
encourage funding support at the state and federal levels for the inclusion of blender pumps at
refueling stations:
Reduce oil dependency
Reduce gasoline prices
Give drivers cleaner fuel options
Reduce carbon emissions (Heavy carbon emissions in gasoline are 40% greater than diesel fuel
due to the heavy carbon nature of ultra fine particulates (UFP)). This is in addition to the reduced
carbon footprint of ethanol production and use, both of which are steadily improving. This is not
the case with oil based gasoline that is not sustainable and whose carbon footprint will only rise
as oil gets harder to obtain while getting heavier and dirtier
Reduce health-related costs by tens of billions of dollars annually as verified by studies at
Harvard and elsewhere and also highlighted in the EFC/UAI comments and my enclosed article.
4-289
-------
Chapter 4: Proposed Vehicle Emissions Program
Permit the auto industries to transition up to 93 octane gasoline as a standard fuel thereby
optimizing engine performance through direct injected and higher compression/turbocharged
engines. (Many high-powered new cars use 93 octane fuels)
Permit the attainment of CAFE standards by 2025 - 54.5 MPG
Reduce greenhouse gas emissions and put America on the path to sustainability
Provide drivers choices based on costs and desires in terms of sustainability and carbon
footprints
Catch up with Brazil in terms of the availability of FFV and fuel choices. See the chart on FFV
growth in Brazil.
Maintain the U.S. as a world leader in biofuels. If we are governed by the preferred policies of
big oil, we could fall behind rapidly as indicated by the following:
In France, Amyris and Total announced a successful demonstration flight at the Paris Air Show
of its renewable jet fuel made from Amyris Biofene, and ultimately, from plant sugars. The
Airbus A321 aircraft powered by two Snecma CFM56 jet engines flew from Toulouse to Paris
with a blend of renewable jet fuel produced by Amyris and Total. This demonstration flight was
in support of the French Initiative for Future Aviation Fuels, which seeks to produce and
commercialize alternative, renewable, and sustainable aviation fuels in France in the coming
years.
We are now at a critical junction internationally and domestically in terms of the future of
transportation fuels in the United States. EPA has the legal authority to take the positive actions
outlined above. Doing so would ensure a bright future for the nation in respect to biofuels. If the
EPA allows itself to be driven by the financial and political clout of non-renewable
transportation fuel, biofuels will likely be seriously weakened in the U.S. This is an advantage
the oil industry, Brazil, Europe, and other regions including China will gain. The EPA's failure
to adequately support FFVs and reductions in the authorized level of aromatics in gasoline has
already placed us on the downhill slide.
It is not to too late for the EPA to take the steps they are authorized to implement, and are
morally compelled to do so. While updating of their pertinent models and the imputing of good
and sensible data will fully justify the recommendations cited above, timing is critical. There is
already fully enough data compelling EPA to act now, and they must do so in the best interest of
the United States and world's population, as well as the environment.
America has made great strides in reducing emissions and improving the efficiency of our
transportation system and technologies. But simply utilizing less petroleum, while essentially
relying on it as the single transportation fuel source is not sufficient to meet the energy,
economic and environmental security we need. What would happen if the Iranians decided to
close down the Strait of Hormuz? Despite expanded oil production in our country, the price of
oil, determined on the global market, would immediately jump to $150-200 per barrel, severely
4-290
-------
Tier 3 Summary and Analysis of Comments
impacting our economy. There are also many plausible terrorist and natural disaster scenarios
that would produce a similar impact.
Every barrel, every percentage of oil that we can displace as America's primary fuel for
transportation with clean, domestic biofuels mitigates against the potential negative impact of an
oil price spike on our government, military, overall economy and individual consumers.
Therefore, efforts to improve vehicle emissions and reduce sulfur content must include the
utilization of greater amounts of clean, domestic biofuel.
Higher amounts of biofuel in our gasoline have significant economic benefits. In 2011, an
ethanol blend of 10% reduced gasoline prices by $1.09. At the same time, over 200 ethanol
plants and refineries supported economic growth across America's heartland. As a whole, the
ethanol industry employs over 400,000 Americans across the country, especially in rural regions
of the country, and is doing so during a time where many Americans need steady employment
and good jobs.
The economic benefits of incorporating higher amounts of ethanol into our gasoline are
significant, but so are the positive environmental impacts at the local, regional and global level.
Higher octane/ higher ethanol content gasoline will reduce the harmful air emissions that lead to
an estimated 50,000 premature deaths a year. Ethanol is the cleanest and most affordable source
of octane on the market today, displacing toxic aromatics such as benzene and toluene. In
addition, ethanol reduces greenhouse gas emissions by 40-50% when compared directly to
gasoline.
Organization: A 2nd Opinion Inc.
As an informed consumer I am very concerned about the use of gasoline containing up to 15%
ethanol (E15). My owner's manuals approve the use of gasoline containing up to 10 percent
ethanol (E10) and imply that if I use other fuels my warranties are not valid. I know the
Environmental Protection Agency (EPA) thinks El5 will not harm cars produced since 2001. I
also know that the Coordinating Research Council (CRC) has tested cars and fuel system
components and concluded that millions of cars could have fuel system failures if they use El 5.
No one will know whether the EPA or the CRC is right for several years. But as a senior citizen
consumer, I do not want to take the risk. Therefore I am proposing as a compromise, that you ask
the Renewable Fuels Association (RFA) to administer a Federal Auto Repair Trust. Here is an
outline of how it would work:
Each ethanol producer/importer would volunteer to deposit one cent for every gallon of fuel
grade ethanol produced in or imported into the United States in the trust which RFA should
volunteer to administer.
Owners of gasoline powered vehicles that are not certified as flex fuel (E85) that experience fuel
system failures after using El 5 can recover their out of pocket repair costs from the trust. The
trustee may use trust funds equal to %5 of the disbursed repair costs to cover administration
costs.
4-291
-------
Chapter 4: Proposed Vehicle Emissions Program
Effective with the 2014 model year, all gasoline powered vehicles sold to U.S. consumers must
be certified to perform on high ethanol content gasoline.
Ten years after all gasoline powered new cars sold in the United States are required to be
compatible with El5 or five years after the last claim was paid whichever is later, the RFA
would be required to disburse the unused FART to the ethanol producers/importers who made
the FART deposits. If EPA is right, the depositors get all their FART deposits back. If CRC is
right each depositor would get a pro rata share of the residual FART. (Individual FART
deposit*(Total FART deposits minus 105%FART costs)/Total FART deposits))
I selected the acronym to get your attention and to have a little fun. I hope it did brighten your
day. But, I am serious about wanting environmental justice concerning these repair costs. The
people should not pay for repairs to fuel system damage caused by the ethanol industry. Nor
should the refining industry. The industry that profits from El 5 should provide environmental
justice and 0.15 cent's per gallon gasoline is cheap.
EPA should encourage RFA to jump on this opportunity as a good compromise.
Organization: American Lung Association
EPA outlines a process whereby a manufacturer could design vehicles to operate on higher
octane and higher ethanol content gasoline, i.e. E30 or higher. We support the approach that
allows for a petition for certification on such fuels if demonstrated that such fuels will be readily
available nationwide, will be used by the vehicle operators, vehicles would not operate
appropriately on other available fuels, and such a fuel would result in equivalent emission
performance. All confirmatory testing should be conducted on fuel that matches the certification
fuel.
Organization: American Petroleum Institute (API) and the Association of Fuel &
Petrochemical Manufacturers (AFPM)
EPA states that its goal is to reduce the number of certification fuels that manufacturers would
need to use to test their vehicle fleet, and yet the Agency proposes to allow vehicle
manufacturers to request approval for alternative certification fuels such as those which contain
high ethanol content. EPA should not finalize rules allowing vehicle manufacturers to certify on
various alternative blends, as it could have a significant impact on market dynamics, such as a
proliferation of boutique fuels that do not fit with the existing fungible fuel system or service
stations storage tank configurations.
Proliferation of Test Fuels: Part of EPA's goal in modifying the certification fuel for Tier 3 was
to reduce the number of certification fuels manufacturers would need to use to test their vehicle
fleet. This goal seems to be lost as EPA is proposing to "allow vehicle manufacturers to request
approval for an alternative certification fuel such as a high-octane 30 percent ethanol by volume
(E30) blend for vehicles they might design or optimize for use on such a fuel. This could help
manufacturers that wish to raise compression ratios to improve vehicle efficiency, as a step
toward complying with the 2017 and later light-duty greenhouse gas and CAFE standards (2017
4-292
-------
Tier 3 Summary and Analysis of Comments
LD GHG). This in turn could help provide a market incentive to increase ethanol use beyond
E10 by overcoming the disincentive of lower fuel economy associated with increasing ethanol
concentrations in fuel, and enhance the environmental performance of ethanol as a transportation
fuel by using it to enable more fuel efficient engines." Allowing each manufacturer the option to
request approval of an alternative certification fuel could have significant impact on market
dynamics. One can imagine a proliferation of boutique fuels necessary to support each individual
manufacturer so that their vehicles can actually deliver the fuel efficiency as tested. This
approach does not fit with the extensive, fungible fuel system that currently exists nor does it
account for the limited fuel options available at service stations due to underground storage tank
configuration and space availability. How does EPA expect to steward such a program and who
would be responsible for determining the cost, well-to-wheel emissions, etc? One could image an
engine that is optimized on an alternative certification fuel but would not deliver anywhere near
the efficiency on in-use fuels so regarding alternative fuel availability, what would be the market
threshold for EPA to consider allowing a manufacturer to use an alternative certification fuel?
Organization: E.I. du Pont de Nemours and Company
The second provision that could encourage larger volumes of renewable fuels involves allowing
vehicle manufacturers to request approval for an alternative certification fuel such as high-octane
E-30. This provision would give vehicle manufacturers the requisite flexibility to improve
vehicle efficiency for purposes of complying with 2017 and later CAFE standards. DuPont
supports allowing vehicle manufacturers to request alternative certification fuels. This policy
could promote higher octane fuels coming to market with lower cost as compared to their high
octane petroleum fuel counterparts. High octane petroleum fuels are more expensive than
gasoline with an 87 rating, but higher biofuel blends with higher octanes should cost less per
gallon of regular gasoline.
The Tier 3 rule objectives go hand in hand with CAFE standards and would create the flexibility
that automobile manufacturers need to comply with both CAFE and the Tier 3 rule. The purpose
of CAFE standards is to reduce energy consumption that will help address our country's
dependence on imported oil, save consumers money at the pump, and reduce emissions of
greenhouse gases that contribute to global climate change. Higher ethanol blends will help make
significant contributions to these objectives.
Opponents of the Tier 3 rule have argued that "proposed fuel specifications should be based on
sound science in order to accommodate vehicle technologies that automakers will commercialize
in significant volumes within the same timeframe required of fuel providers." They fault EPA for
not identifying "automotive technologies that would be utilized to comply with increased CAFE
standards or a Tier 3 standard that would benefit from lower sulfur gasoline."
These arguments seek to solidify the current obstacles that prevent both higher blends of biofuels
and advanced vehicle technologies. Without the availability of higher ethanol blends on the
market, automobile manufacturers are hesitant to develop engines that can accept the fuel.
Likewise, without advanced engines, market demand for advanced fuels is lacking. DuPont
believes that the Tier 3 rule is an opportunity to make progress towards overcoming these
obstacles.
4-293
-------
Chapter 4: Proposed Vehicle Emissions Program
The second provision that could encourage larger volumes of renewable fuel involves allowing
vehicle manufacturers to request approval for alternative certification fuel, such as E-30. And
that would give vehicle manufacturers the requisite flexibility to improve vehicle efficiency for
purpose of complying with the 2017 rules.
We support allowing vehicle manufacturers to request approval for alternative certification fuels,
higher octane fuels, because ethanol cannot be used as an octane booster. High octane petroleum
fuels are more expensive than gasoline with an 87 rating. But higher ethanol blends with higher
octane should cost less per gallon than regular gasoline.
Organization: Biotechnology Industry Organization (BIO)
Biofuels and Their Supportive Policies Are Important to the U.S. Fuel Supply; The federal RFS,
enacted in 2005 and updated in 2007, is an important tool in achieving the objectives of energy
independence and pollution reduction. The RFS is the single most important federal policy
driving investment and commercialization of conventional and advanced biofuels. Biofuel
production under the RFS has already displaced nearly 10 percent of gasoline consumption and
will account for more than 20 percent of U.S. transportation fuel by 2022. Biofuel production
under the RFS reduced the need for imported oil by more than 462 million barrels in 2012.
Investment in biofuels, largely spurred by the RFS, has led to the development of facilities like
INEOS Bio's in Vero Beach, Florida, and KiOR's in Columbus, Mississippi, which represent
several hundred million dollars of investment in the United States and are poised to begin
production of the next generation of renewable fuel from non-food feedstocks this year. Dozens
more advanced biofuel projects are planned or under construction, as highlighted in the attached
Appendix I, illustrating the visible success the RFS has had in driving development of highly
skilled, well-paying jobs in rural America. Biofuel production under the RFS has led to the
employment of 380,000 Americans. And, 800,000 employment opportunities could be created by
2022.
BIO firmly believes that the limits to market access for biofuels commonly referred to
collectively as the "blend wall" represent a series of barriers contrived by obligated parties to
prevent biofuels from gaining access to the marketplace. Multiple avenues exist for blending
additional volumes of biofuel into the nation's fuel supply. For instance, as the proposed rule
recognizes, El 5 blends are approved and ready for use, and production of flex fuel vehicles
continues to increase. These options, combined with the introduction of new "drop-in" fuel
molecules, provide a suite of opportunities for the growth of the entire biofuels industry and RFS
compliance.
The main obstacle to this growth and compliance is the dilatory tactics of obligated parties to
pursue the options available to them. Obligated parties have had over five years to begin
establishing the infrastructure necessary to distribute RFS-mandated biofuel volumes, but have
taken few steps to do so. EPA should therefore resist all efforts by obligated parties to postpone
updating the U.S. certification fuel for emissions from light duty cars and trucks and heavy duty
gasoline vehicles, or to reduce RFS obligations based on blend wall claims. Any concession by
4-294
-------
Tier 3 Summary and Analysis of Comments
EPA to accommodate these assertions regarding the blend wall will only serve to embolden
obligated parties in their effort to resist compliance with the Clean Air Act.
Instead, as it is able through the proposed rule and other administrative actions, the Agency
should encourage the development of biofuels - and the engines and infrastructure to support
them. This development can readily grow the supply of biofuels in the market and overcome the
blend wall by allowing Renewable Identification Numbers (RINs) to truly reflect their market
value. It will also help drive the market and encourage retailers to adopt new infrastructure as
reflected in Appendix I.
EPA should also seek to identify opportunities to grow biofuel markets, including for drop-in
biofuels. Reconsideration of the gasoline base fuel would enable engine manufacturers to
optimize beneficial characteristics of biofuels in engine design, while expedited approval of new
molecules would provide obligated parties with additional options for compliance not subject to
blending limitations.
BIO believes the final Tier 3 rule has the potential to continue the progress of the biofuels
industry and help alleviate the "blend wall" as described above. BIO commends EPA's efforts to
transition the certification test fuel for emissions from light duty cars and trucks and heavy duty
gasoline vehicles ("the emissions test fuel") to better reflect the current and future in-use fuel as
it begins to contain greater volumes of biofuels resulting from the continued development and
commercialization of biofuels and increased RFS volumes in the market.
BIO believes that EPA should work to finalize its Tier 3 rulemaking and set the emissions test
fuel to encourage further investment and adoption of biofuels, including advanced drop-in
biofuels, ethanol and other fuel molecules. Encouraging biofuels, including higher blends of
ethanol and drop-in biofuels would help meet EPA's overall goal in this rulemaking to address
the impacts of motor vehicles and fuels on air quality and public health. For instance, advanced
drop-in biofuels have the same molecular make-up of traditional petroleum-based fuels, but they
contain little or no sulfur and have significantly reduced GHGs. In addition, ethanol combusts
without producing air toxics, which are the main source of particulate matter. Blending ethanol
in gasoline also reduces the need for unhealthy detergent additives which are mandated to reduce
the formation of engine deposits from gasoline that increase exhaust emissions and result in the
loss of fuel economy and performance. These benefits of biofuels only rise with higher blends.
BIO encourages EPA to issue a final rule that maximizes investment and adoption of all biofuels,
including higher blends and drop-ins.
EPA requests comment on various ways to transition the emissions test fuel, including: starting
with a transition to E10 with a transition to E15 as that fuel becomes more widely available on
the market; transitioning to El5 after a few years to allow time for this greater market
availability or by setting a date certain by which the transition should be made to help drive El 5
market availability; or, allowing "vehicle manufacturers to request approval for an alternative
certification fuel, such as a high-octane 30 percent ethanol by volume (E30) for vehicles they
might design or optimize to use such a fuel" so that it may "help manufacturers that wish to raise
compression ratios to improve vehicle efficiency, as a step towards complying with the 2017 and
later light-duty greenhouse gas and CAFE standards."
4-295
-------
Chapter 4: Proposed Vehicle Emissions Program
As BIO describes above, we urge EPA to issue a final Tier 3 rule that maximizes investment and
adoption of biofuels. We agree with EPA's assertion that E10 is widely available and in use in
today's market, and that E15 and E85 are the most likely near term biofuels that have the
potential to be widely available. We believe the final rule should allow for and encourage
ongoing and greater use of E85 and El 5, as well as the transition to higher blends of ethanol, and
drop-in biofuels.
To this end, BIO urges EPA to issue a final rule that recognizes that new biofuel molecules are
beginning to enter the market for blending with gasoline in order to help achieve the goals of the
RFS. As such, we urge EPA to ensure that any transition in the certification fuel not create
unintended barriers and burdensome costs to companies working to register these new biofuel
molecules under Part 79 fuel and fuel additive registration.
As EPA suggests, one way to accomplish this goal may be to allow vehicle manufacturers to
request approval for an alternative certification fuel such as a high-octane E30 blend for vehicles
they might design or optimize for use on such a fuel. BIO is aligned with EPA's assertion in the
proposed rule that "[t]his could help manufacturers that wish to raise compression ratios to
improve vehicle efficiency, as a step toward complying with the 2017 and later light-duty
greenhouse gas and CAFE standards...[t]his in turn could help provide market incentive to
increase ethanol use beyond E10 by overcoming the disincentive of lower fuel economy
associated with increasing ethanol concentrations in fuel, and enhance the environmental
performance of ethanol as a transportation fuel by using it to enable more fuel efficient engines."
Preliminary results from Oak Ridge National Laboratory and the University of Wisconsin
suggest that moderate biofuel blends increase the benefits of the use of Reactivity Controlled
Compression Ignition (15). Therefore in order to further help manufacturers develop engines
which can utilize biofuels which deliver reductions in greenhouse gas emissions; BIO
recommends EPA sets the "R-factor," which may be considered as the efficiency with which the
vehicle engine adapts to fuel variations, at a value of 1.0. By doing so, EPA will ensure the final
rule does not discourage manufacturers from developing engines that can utilize new cleaner
burning biofuels in order to achieve the overall goals of the proposed rule.
BIO believes the final Tier 3 rule has the potential to compliment other transportation related
policies, including the RFS, which promote the continued development, commercialization and
use of biofuels. We commend EPA's efforts to transition the emissions test fuel to help promote
the goals of the RFS. BIO believes that EPA should work to set the emissions test fuel to
maximize investment and adoption of all biofuels, including higher ethanol blends and drop-ins,
while also maximizing the level of octane in the U.S. fuel supply. This action would help EPA
meet one of its stated goals of this rulemaking to address the impacts of motor vehicles and fuels
on air quality and public health.
To this end, we believe the final rule should allow for and encourage ongoing and greater use of
E85 and El 5, as well as the transition to higher blends of ethanol and drop-in biofuels.
BIO also supports EPA's proposal to allow vehicle manufacturers to request approval for an
alternative certification fuel such as a high-octane E30 blend for vehicles they might design or
4-296
-------
Tier 3 Summary and Analysis of Comments
optimize for use on such a fuel. As described above, we agree with EPA that this provision could
help auto manufacturers meet their CAFE requirements.
BIO and its members look forward to working with EPA and the affected parties of this
rulemaking to ensure implementation of a final Tier 3 rule that promotes the continued
development of the biofuels industry as expressed in these comments, while complimenting
other transportation regulatory requirements, including under the RFS and CAFE.
Organization: Energy Future Coalition and Urban Air Initiative
Together, we file these comments in response to EPA's questions in the Proposed Rule about
whether to approve "an alternative certification fuel such as a high-octane 30 percent ethanol by
volume (E30) blend" to enable more efficient engine design. Our comments will show not only
the advisability of approving a mid-level ethanol blend certification fuel, but also the need to
facilitate its use in order to comply with EPA's obligation to regulate mobile source air toxics
(MSATs or "hazardous air pollutants"). We hope EPA will take advantage of this opportunity to
usher in a new era of cleaner, higher-efficiency fuel, improving the lives of millions of American
citizens and protecting both public health and the environment.
The Proposed Rule will deliver substantial economic, health, and environmental benefits to the
nation; however, it inadequately considers the opportunity to enlarge those benefits and save tens
of billions of dollars annually by facilitating a gradual shift in the content of motor vehicle fuel
for light-duty vehicles. A mid-level ethanol blend fuel, created by splash-blending ethanol with a
gasoline blendstock that meets current standards, would maximize benefits to air quality and
public health, automotive performance, and consumer costs:
As another example, the Proposed Rule notes, "as the ethanol level increases, the volatility
increase caused by blending ethanol with gasoline begins to decline, such that at E30 there is
only about a 0.5-psi RVP increase." Indeed, a recent study by a team from Ford Motor Company
and AVL Powertrain Engineering, Inc. concluded that RVPs match that of base gasoline at
ethanol concentrations of 30%. This has positive implications for the cost and ease of
compliance with the Proposed Rule.
- Automotive performance: As the Proposed Rule notes, the use of a 30% ethanol blend "could
help manufacturers that wish to raise compression ratios to improve vehicle efficiency, as a step
toward complying with the 2017 and later light-duty greenhouse gas and CAFE standards." The
Ford/AVL team reviewed 10 properties of ethanol blends and concluded that "a mid-level
ethanol blend (greater than E20 and less than E40) appears to be attractive as a long-term future
fuel for the US, especially if used in vehicles optimized for such a fuel." Despite the lower
energy content of ethanol, in an optimized vehicle with an E30 blend, vehicle fuel economy and
range can be maintained. An E30 blend in an engine designed to use that fuel would have
"ridiculous power and good fuel economy," one senior automotive engineer said.
- Consumer costs: Numerous economic analyses suggest that the replacement of gasoline with
ethanol reduces the cost of gasoline. A recent study at Louisiana State University, for example,
found that each additional billion gallons of ethanol reduces gasoline prices as much as $0.06 per
4-297
-------
Chapter 4: Proposed Vehicle Emissions Program
gallon. Considering U.S. ethanol production in 2010 was more than 13 billion gallons, this would
suggest savings of up to $0.78 per gallon of gasoline. Since Americans consume more than 130
billion gallons of gasoline annually, this amounts to more than $100 billion a year in savings.
Increased ethanol use would further reduce gasoline consumption, increase those savings, and
lessen the nation's vulnerability to economic disruption due to fluctuations in oil prices.
For all these reasons, we commend EPA for requesting comment on whether it should approve
"an alternative certification fuel such as a high-octane 30 percent ethanol by volume (E30) blend
for vehicles [automobile manufacturers] might design or optimize for use on such a fuel." Our
answer to this is an emphatic 'yes.' But in order to achieve this environmentally sound and
technologically progressive result, EPA must remove certain hurdles erected by the current
regulatory regime, described below in Part II.
EPA Leadership Is Needed on Fuel Content for Certification and Consumer Use. In its Proposed
Rule, EPA requests comment on whether the Agency should approve "an alternative certification
fuel such as a high-octane 30 percent ethanol by volume (E30) blend for vehicles [automobile
manufacturers] might design or optimize for use on such a fuel." Our answer to this is an
emphatic 'yes.' But in order to achieve this environmentally sound and technologically
progressive result, EPA must remove certain hurdles erected by the current regulatory regime.
Under current EPA regulations, automobile manufacturers must show that a fuel is
"commercially available" before the Agency will approve it for emissions and fuel economy
certification. In its proposed Tier 3 Rule, EPA states that the fuel must "be readily available
nationwide" to meet this standard and that the onus is on automobile manufacturers to make this
showing. However, EPA also asks for comment on whether this standard is appropriate. In the
case of a new mid-level ethanol blend, we believe that this standard is not appropriate.
Automobile manufacturers are not in a position to ensure that a fuel is available to consumers
nationwide. The market will not produce and distribute a fuel until there are vehicles that run on
it, and automobile manufacturers will not mass-produce vehicles optimized to run on a new fuel
without assurance that these vehicles can be certified on that fuel, and that the fuel will be
available to consumers. Moreover, Congress has prohibited the introduction of commercial fuel
that is not already "substantially similar" to an existing certification fuel, so even if automobile
manufacturers were capable of creating a market for a new fuel by themselves, they would not be
allowed to do so until that fuel, or one like it, had already been approved as a certification fuel.
The current regulatory regime thus imposes a catch-22 on the industry: The fuel necessary for
the next generation of motor vehicle engines cannot be introduced into the market until it is
approved as a certification fuel, and it cannot be approved as a certification fuel until it is
available in the market. EPA should take the initiative to solve this collective action problem by
(1) approving for new motor vehicle certification a high-octane, mid-level ethanol blend, subject
to the environmentally relevant parameters discussed below, (2) ensuring that such fuel can be
made commercially available, and (3) removing the regulatory disincentives that currently inhibit
the development of efficient vehicles optimized or dedicated to operate on this fuel.
A market shift from current light-duty motor vehicle fuel to a mid-level ethanol blend would
result in numerous benefits to the environment, to American drivers, and to the public at large.
Increasing the ethanol content of fuel would reduce the United States' dependence on foreign oil
4-298
-------
Tier 3 Summary and Analysis of Comments
and achieve corresponding benefits to national security. A mid-level ethanol blend would
provide desirable gains in clean octane, enabling better engine performance and efficiency.
Finally, and of primary significance for the purpose of these comments, adding ethanol to fuel
would allow EPA to satisfy its statutory obligation to regulate air toxics in motor vehicle fuel by
reducing aromatic hydrocarbons that naturally occur in gasoline and are added to motor vehicle
fuel as an octane enhancer. Reducing these harmful additives by adding octane-rich (and less
expensive) ethanol would substantially reduce the environmental impact of fuel emissions with
direct health benefits and lower pollution control costs to the public, especially in urban areas.
EPA Should Approve a Splash-Blended Mid-level Ethanol Blend Certification Fuel. Requiring
automobile manufacturers to demonstrate that a fuel is "commercially available" as a
precondition to EPA's approval of it as a certification fuel stifles technological progress and
blocks environmental benefits, as described above. By seeking comment on this requirement,
EPA opens the door to the kind of Agency leadership on air toxics that Congress contemplated in
section 202(1). EPA should make the first move in this direction by approving a splash-blended
mid-level ethanol blend as an optional certification fuel. In due course, EPA should require all
new motor vehicles to be certified on this fuel and then phase down the aromatics content of fuel
in response to its obligation to reduce MSATs. In other words, the Agency should over time
make a mid-level ethanol blend the standard, rather than an alternative, certification fuel.
It is important, however, to ensure that the quality of the gasoline portion of the fuel is not
permitted to deteriorate before blending with ethanol, thereby reducing the environmental
benefits of aromatics reduction. In its notice of proposed rulemaking, EPA asks for comment on
whether the Agency needs "to specify more precisely the makeup of [a higher octane, higher
ethanol content] fuel" in terms of "ethanol content, as well as other fuel parameters." The
answer is yes. Since ethanol contains only one molecule, with unvarying chemical properties, it
is critical that, when regulating fuel blends for environmental purposes, EPA specify certain
parameters of the gasoline "blendstock for oxygenated blending" or "BOB," that is, the gasoline
portion of the fuel to which ethanol will be added. This is crucial because the gasoline
blendstock is the only portion of the fuel that has variable characteristics.
As noted above, the components of motor vehicle fuel evaporate at different temperatures, and
the components that evaporate at high temperatures have the worst environmental and health
effects.
Because ethanol, unlike gasoline blends, contains just one molecule, it has a single, relatively
low, boiling point. Ethanol's T50 is the same as its T90—78.4 degrees Celsius—this is the
temperature at which every molecule of ethanol would evaporate at sea level. Standard gasoline,
on the other hand, is capped at a T50 of 121 degrees Celsius and a T90 of 190 degrees Celsius by
ASTM, a non-governmental standards-setting organization. (The ASTM limit is a ceiling, not a
floor; refiners are free to create fuel blends with lower T90 values (and T50 values as low as 77
degrees Celsius), and ethanol makes this possible.) In other words, standard gasoline contains
much higher distillates than ethanol and is therefore more conducive to tailpipe pollution. Mixing
ethanol directly with current blendstock for oxygenate blending ("splash blending") reduces the
distillation temperatures of the finished fuel proportionally, thereby reducing tailpipe pollution.
4-299
-------
Chapter 4: Proposed Vehicle Emissions Program
The Agency's EPAct model, referred to in the proposed rule only with respect to CO2 emissions,
purports to measure the effects of various fuel blends on many forms of pollution. However,
EPAct does not model splash blending of ethanol with standard gasoline blendstock, but rather
models what would happen if refiners were allowed to alter the gasoline portion of the fuel by
adding dangerous levels of high-distillate aromatics before blending with ethanol. In other
words, EPAct models the pollution effects of ethanol blends containing high-distillate gasoline
formulations that are themselves not permitted to be sold as consumer fuel.
In the real world, EPA should not allow refiners to make gasoline blendstocks more dangerous
before mixing them with ethanol. If only the final fuel needs to maintain certain parameters such
as octane and T50, T90, and endpoint distillation temperatures, refiners could make gasoline
blendstocks dirtier and lower octane, thereby canceling out the performance and environmental
benefits of ethanol. EPA should not permit this to happen. As the Agency noted with regard to
E51-83, "only blendstocks (including butane and NGL) that meet [certain] specifications would
ensure the final blend would always meet the standards." We agree with the Agency that in
blending ethanol into motor vehicle fuel, the use of previously approved blendstocks and
"gasoline that has previously been demonstrated to comply with applicable EPA specifications . .
. prevents inappropriate blending components from being used in the production" of the resulting
ethanol blend. Maintaining current blendstock standards is also necessary to maximize the
engine performance benefits of a mid-level ethanol blend. Thus, EPA should mandate that any
mid-level ethanol blend approved for market and certification fuel be made up of (1) a certain
percentage of ethanol by volume, and (2) a corresponding volume percentage of gasoline which
itself meets all relevant EPA and ASTM standards. In other words, EPA should approve a
"splash-blended" mid-level ethanol fuel.
EPA Should Remove Regulatory Disincentives that Currently Impede the Production of Vehicles
that Run on a Mid-Level Ethanol Blend. In the proposed rule, EPA requests comment on how the
Agency might "provide an incentive for, or remove obstacles to, the development of highly
efficient vehicles optimized for use on higher level ethanol blends." EPA also requests comment
on whether an adjustment is warranted in the compliance calculation for light-duty GHG
standards. These two questions should be viewed through the same lens. EPA should remove
current regulatory disincentives to the production of vehicles designed to run on a mid-level
ethanol blend, while paving the way for a national transition to this fuel by incentivizing
increased production of FFVs. These incentives can be achieved without relaxing GHG
emissions standards, simply by allowing auto manufacturers to account for the life-cycle GHG
benefits of ethanol in their GHG compliance calculation.
In order to begin to phase in a cleaner, higher-ethanol fuel that significantly reduces air toxics
and other dangerous pollutants, while at the same time helping manufacturers design more
efficient engines to reduce GHG pollution, EPA should allow all vehicles capable of running on
a mid-level ethanol blend (dedicated, optimized, or FFV) to be certified for purposes of GHG
emissions and fuel economy on the new high-octane, mid-level ethanol blend.
Organization: Volkswagen Group of America, Inc.
Other Fuel Properties
4-300
-------
Tier 3 Summary and Analysis of Comments
VW supports working with the Agencies on a higher octane/higher ethanol certification fuel that
can be optionally used by the light duty fleet. Vehicles specifically designed for known fuels can
take advantage of efficiencies available. VW sees value in ensuring the specially-designed
vehicles have access to an analogous market fuel. EPA must then ensure that vehicles not
designed for this fuel do not attempt to use it, i.e., gas station pump labeling must assist
customers in choosing the proper fuel for their vehicle. VW supports a 9RVP E 15 and 9 RVP
E30 created by splash blending the appropriate level of ethanol into gasoline.
Organization: Ford Motor Company (Ford)
We also commend EPA on their forward-looking approach for future opportunities for increased
vehicle efficiencies through the request for comments on a higher octane rated alternative
certification fuel. Treating fuels and vehicles as a system is vitally important as fuel quality
improvements are needed to enable the advance technologies that will be needed to meet
requirements of the ONP as well as Tier 3. Thus, further modifications to the market fuel quality
parameters (vapor pressure, elimination of the use of metallic additives, increased detergency
requirements, increased octane rating for gasoline, and increased cetane rating for diesel) are
necessary to allow for a complete integration of vehicles and fuels as a system.
Alternative Certification Fuel: In the NPRM, EPA proposes to allow vehicle manufacturers to
request approval for an alternative certification fuel such as a high-octane 30 percent ethanol by
volume (E30) blend for vehicles designed or optimized for use on such a fuel. EPA requests
comments on an alternative test fuel that is a "higher octane, higher ethanol content gasoline".
While we support the initiative for future collaboration between original equipment
manufacturers (OEMs), government agencies and other stakeholders to maximize vehicle
efficiency in tandem with use of renewable fuels, we believe that there is still further
development needed on the higher ethanol/high octane certification fuel option. Additional
information and recommendations concerning this proposal can be found in Part II (Fuels
section) of this document.
Recommendation: Ford strongly recommends that EPA pursue regulations and other measures to
facilitate the introduction of higher octane rating market fuels, which offer the potential for the
introduction of more efficient vehicles. Progress on this issue will be a key parameter for
consideration in the "One National Program" mid-term evaluation process, as well as for all
phases (current and future) of the medium and heavy-duty GHG and fuel economy standards.
Alternative Certification Fuel Opportunity: As noted above, Ford supports the development and
introduction of an intermediate level blend fuel (E16-E50), with a minimum octane rating of 91
anti-knock index (AKI) that increases proportionally as ethanol is splash-blended on top of the
base Tier 3 gasoline emission test fuel. The development of such a fuel would enable the first
steps to the development of a new generation of highly efficient internal combustion engine
vehicles. We look forward to future collaboration with the EPA on this item. We recognize that a
number of issues will need to be resolved in order to move this concept forward, but there is
concurrence that it is important to work through the details to allow more efficient vehicles to be
made available to American consumers.
4-301
-------
Chapter 4: Proposed Vehicle Emissions Program
In response to the request for comment, Ford provides the following observations and
recommendations as outlined in the following section. Literature has shown that increased
ethanol in gasoline paired appropriately with the resultant commensurate increase in octane
rating of the final blend leads to future opportunities for increased engine efficiency and
therefore, reduced GHG emissions. The addition of ethanol to gasoline blends has many
potential benefits, which have been extensively studied 10,11,12,13,14,15 (see Appendix A).
Ethanol is a renewable fuel which displaces petroleum and has been shown to reduce wells-to-
wheels CO2. Ethanol can be produced domestically, therefore offering the potential to support
domestic energy goals. Advanced engine designs operating on ethanol (with increased octane
rating) have a slightly higher thermal efficiency and produce fewer tank-to-wheels CO2 than
engines operating on straight gasoline. If octane rating is allowed to increase with the added
ethanol, there is opportunity to optimize the engine for further improvements in engine efficiency
(see Appendix). In the U.S., the Energy Independence and Security Act of 2007 (EISA 2007)
mandates 36 billion gallons of renewable fuels in use by 2022. Since starched-based ethanol is
limited to 15 billion gallons in EISA 2007, excess ethanol has been exported to other fuel ethanol
consuming countries. Ethanol has a lower vapor pressure than gasoline; however, the addition of
ethanol at a concentration of about 10%-15% by volume increases the blended vapor pressure by
approximately 1 psi. At levels above this range of ethanol, the vapor pressure of the resulting
mixture begins to decrease. The decrease in vapor pressure allows for decreased evaporative
emissions, however, too low vapor pressure can result in engine starting issues in colder
climates.
The higher octane number (109 Research Octane Number [RON] for ethanol versus today's
regular gasoline 91 RON) and higher heat of vaporization associated with gasoline-ethanol
blends can improve engine efficiency through engine redesign and use of higher compression
ratios. However, the oxygen content in ethanol can result in enleanment for engines with
uncalibrated control modules due to the change in stoichiometric air/fuel ratio (A/F). For
example, assuming EO has a stoichiometric A/F of 14.6, this A/F then becomes 14 for E10, 9.9
for E85 and 9.0 for E100. In addition, ethanol's energy density is less than that of gasoline, with
El 00 having a net heating value (NHV) of 21 MJ per liter of fuel whereas gasoline has a NHV of
33 MJ per liter of fuel. The difference in energy density results in reduced volumetric fuel
economy of ethanol-gasoline blends versus pure gasoline in non-optimized engines. Even though
ethanol's higher octane rating can improve the thermal efficiency of the engine at knock limited
operating conditions, the net result is degraded volumetric fuel economy and this is most
pronounced in E85 blends, which has about 27% less energy per gallon than gasoline.
However, increased octane rating from increased ethanol content has the potential to allow for
fuel economy, performance and emissions improvements through more efficient engine designs.
Raising the minimum octane rating requirement would allow manufacturers to design engines
with greater thermal efficiency through higher compression ratios and/or smaller displacement
turbo-charged engines. Current engines in the fleet could also benefit from the higher octane
rating through more aggressive spark timing during certain driving conditions. High compression
ratio engines are already found in Europe, where 95 RON fuel is typically available in contrast to
the lower 87 AKI (approximately 91 RON) fuel that is most common in the US. The introduction
of higher octane rated/intermediate level ethanol blend fuel would allow for a faster introduction
of more efficient vehicle designs from Europe with lower CO2 emissions and increased
4-302
-------
Tier 3 Summary and Analysis of Comments
efficiency that are designed to operate on 95 RON (-91 AKI) market fuel without the need for
significant design changes as for products destined for the U.S. market.
Of course, there needs to be a long-term strategy for the introduction of such a fuel, and a
number of phase-in issues must be addressed in the meantime. Protection grade fuel must remain
available for those consumers with legacy products as the new alternative fuel is phased in. The
current flexible-fuel vehicle (FFV) fleet can be used in the interim as the transition from lower
(E10) to higher ethanol level blends occurs across the United States. The timing and details
regarding the introduction of optimized vehicles dedicated to the use of the new fuel will need to
be addressed as it becomes important to promote flexibility in this transitional timeframe.
Considering the added complexity in the production of vehicles capable of greater than E10 or
FFVs, continued incentives to build such vehicles will likely be necessary. EPA should consider
an extension of the current FFV credit policy (F-factor) to allow credits to be generated for
vehicles with the capability of operating with intermediate ethanol blend fuels. Considering the
complexity and transitional nature of the entire scenario, such burden of proof with respect to
fuel use should not be placed upon either the vehicle manufacturer or vehicle owner.
Alternatively, such credits could be allocated based on national ethanol use beyond that which
can be allocated to an E10 marketplace and then distributed to manufacturers based on sales
volumes of capable vehicles.
The inevitable quandary of sufficient fuel infrastructure must be also addressed. As observed in
the literature cited in this document (see Appendix), sufficient nationwide availability of the
next-generation blend must be assured as future fuels and engine technologies should progress
synergistically, with a consistent, open dialogue between stakeholders and policymakers.
The next-generation fuel must also be favorably valued by consumers in order to incentivize
consistent use throughout the market. As stated above, FFVs offer an opportunity for current
technology vehicles to bridge the gap until the new fuel is available in sufficient quantities
throughout the market. Finally, to assist in the market introduction of the next-generation fuel to
consumers, fuel certification must be addressed and aligned at a federal level. Ford recommends
the use of market fuel specifications that are developed through ASTM working groups (ASTM
D5798, ASTM D4806) as well as subsequent successful working groups.
Recommendation: Ford strongly recommends that EPA pursue regulations and other measures to
facilitate the introduction of higher octane rated market fuels, which offer the potential for the
introduction of more efficient vehicles. Progress on this issue will be a key parameter for
consideration in the "One National Program" mid-term evaluation process, 16 as well as for all
phases (current and future) of the medium and heavy-duty GHG and fuel economy standards.
Organization: General Motors LLC (GM)
EPA Should Replace Flawed Regulations Governing Use of Alternative Test Fuels. GM agrees
that alternative test fuel provisions in the current regulations at 40 CFR 1065.701(c) are flawed
and cannot work in practice. For dedicated vehicles, the regulation requires that the new fuels be
commercially available in the marketplace; in addition, the regulation appears to presume the
4-303
-------
Chapter 4: Proposed Vehicle Emissions Program
fuels would be widely available without any vehicles being available to use them, which
underscores the unresolved "chicken and egg" dilemma.
In addition, manufacturers would have to show that the vehicles will not run or be durable on
standard test fuel. Alternatively, manufacturers could show that the vehicles would not meet
emissions limits on standard test fuel (i.e., measured emissions would be "substantially
unrepresentative"), all of which would likely make them practically unsellable.
Were an OEM to petition to test using El 5 test fuel under these provisions, meeting them would
be very challenging. Rather than imposing unrealistic regulatory pre-conditions, EPA needs to
replace them. GM would be happy to work with the agency to develop a workable process and
criteria for access to alternative test fuels.
Within the rulemaking, EPA puts forward the concept of producing "dedicated E30 vehicles or
FFVs optimized to run on E30 or higher alcohol blends." While there is no way to couple a FFV
and an optimized vehicle together, GM supports the future of higher octane and higher ethanol
content in order to provide a pathway to improved vehicle efficiency and lower GHG emissions.
However, the use of an R-factor less than 1.0 discourages the voluntary use of any fuel with
energy content less than that of the 1975 emissions certification fuel.
Organization: Mercedes-Benz USA, LLC on behalf of Daimler AG
The synergy of increased efficiency, afforded by high octane and the reduced carbon intensity
mid-blend ethanol can offer, along with gasoline direct-injection and other technologies, can
provide remarkable reductions in CO2 output. Mercedes-Benz endorses the concept of a higher
octane, mid-blend ethanol certification fuel such as E25, as suggested in the Tier 3 proposed
rulemaking, and requests consideration of the aforementioned corporate annual CO2 burden
reduction proposals.
Organization: National Corn Growers Association (NCGA)
The first principle is that the RFS volume of biofuels in the future should be met in order to
preserve the 150 million metric tons of annual CO2 equivalent emission reductions attributed to
the RFS 2 program by EPA. These emission reductions are directly dependent on the production
of biofuels and not on the fuel economy of the light-duty on-road motor vehicle fleet. The
volume of biofuels utilized is in turn dependent on having vehicles that are flexible-fuel capable
and those certified on high octane mid-level blends plus a retailer infrastructure offering energy-
competitive pricing. The Tier 3 proposal thus has a direct bearing on the availability of vehicles
to implement the RFS.
Further, we support the recognition that a high octane mid-level ethanol blend (MLEB) would
help automakers achieve the 2017 and later GHG and CAFE standards.
High Octane E30 Alternative Certification Test Fuel: EPA suggested the concept of producing
"dedicated E30 vehicles or FFVs optimized to run on E30 or higher alcohol blends." NCGA
supports fuels with higher octane and higher ethanol amounts that provide a pathway to
4-304
-------
Tier 3 Summary and Analysis of Comments
improved vehicle efficiency and lower GHG emissions. It is important to establish the
composition of a new fuel early to facilitate design and optimization. NCGA recommends two
options for the base gasoline blendstock for oxygenate blending (BOB) and several options for
ethanol content. For base gasoline or BOB, the options are conventional gasoline BOB or
CBOB, an 88 research octane (RON) hydrocarbon that becomes regular E10 when blended with
10 percent ethanol, and a high octane "premium" gasoline BOB or PBOB, a 91-93 RON
hydrocarbon that becomes premium E10 when blended with 10% ethanol. Although EPA
mentions 30 percent ethanol as a candidate blend, other ethanol blend rates such as E25 or E35
may represent a more optimum blend for vehicle and engine design.
NCGA does not believe that the alternative test fuel provisions at §1065.701(c) are practical
since they assume that a new fuel would become widely available prior to the availability of
vehicles designed to use it. Further, a specification for the new fuel is needed ahead of time as a
guide for fuel providers and auto manufacturers.
Were EPA to determine that some sort of "actual use" demonstration of the high octane high
ethanol fuel is required, NCGA would look to recent EPA guidance on this subject. EPA, in its
Draft Guidance for E85 Flexible Fuel Vehicle Weighting Factor for Model Year 2016-2019
Vehicles, stated that the F-Factor (proportion of actual use) "...must be based on the projected
use of the alternative fuel over the life of the ... vehicles." If the determination of the amount of
alternative fuel used requires the vehicles to reach full useful life first, then it is clear that EPA
will have to offer guidance on what this F-F actor value would be because of the uncertainty of
this value.
EPA has the authority to ensure the availability of this new fuel formulation as the vehicles
designed to use it become available if it believes such a fuel would enable meeting the 2017 and
Later Model Year CAFE and GHG Emission regulation and the RFS regulation (8). Consistent
with EPA's expressed belief that improved engine efficiency and reduced GHG emissions would
result from a high octane MLEB and engines designed for its use, we recommend that EPA
consider requiring the nationwide availability of such a high octane fuel.
Perhaps a more workable alternative would be to require a phase-in of all gasoline fueled light
duty vehicles as FFVs beginning in 2017 (50% in 2017 and 100% in 2018). This approach would
clearly communicate EPA's support for the GHG reduction benefits of the RFS program, and
would clear the way for FFVs to become the "bridge" to a new generation of vehicles designed
and optimized to run on a higher octane MLEB. Requiring 2017 and later gasoline vehicles to be
FFVs would remove any limits on ethanol usage, and would provide a clear incentive to service
station owners to offer more gasoline/ethanol fuel choices through the use of blender pumps.
One of the key impediments to additional biofuel use is the number of vehicles with flexible fuel
capability to use blends above El 5. With a current penetration of about 18 million vehicles,
FFVs are still significantly less than 10% of the vehicle fleet. E85 is sold in approximately 3,000
service stations, less than 10 percent of the 165,000 stations, not nearly enough to support
increased ethanol consumption volumes under the RFS. It is our belief that if EPA required 2017
and later vehicles to be certified as FFVs, automakers would more seriously consider petitioning
EPA for a high octane MLEB. A preferred MLEB for both consumers and automakers would
then likely emerge from this process. An FFV requirement would also eliminate any non-FFV
4-305
-------
Chapter 4: Proposed Vehicle Emissions Program
"misfueling" concerns with the 2017 and later fleet. Finally, requiring gasoline vehicles in 2017
and later model years to be FFVs would significantly enhance energy security in the United
States by encouraging the use of renewable, domestically produced fuels.
At least two major automakers have reported research programs evaluating advanced technology
gasoline direct engines with high octane MLEBs.
Organization: Natural Resources Defense Council (NRDC)
Alternative, high-octane certification fuels must be sustainably produced and be the in-use fuel
for vehicles certified on the fuel to receive greenhouse gas or fuel economy credits.
Alternative, High-Octane Certification Fuels Must be Sustainably Produced and be the In-Use
Fuel for Certified Vehicles to Receive GHG or Fuel Economy Credits. EPA proposes to allow
automakers to certify vehicles on an alternative high-octane renewable-blend fuel. NRDC
supports automaker efforts to reduce carbon pollution and increase fuel economy by optimizing
their operation for clean fuels. However, an alternative, renewable certification fuel must be
sustainably produced and available as the regular in-use fuel. NRDC opposes the adoption of
E30 as a certification fuel in this rulemaking because today there is no clear timeline in which
E30 made from sustainably-produced ethanol could be available on a national scale. Establishing
a process to certify vehicles on E30 today would signal support for greater production of ethanol
from food crops such as corn, which have significant environmental and economic risks.
If a renewable, high-octane fuel can be produced on a large scale, it must be widely available.
For an automaker to receive GHG emissions and fuel economy ratings based on the alternative
certification fuel there must be certainty that the certified vehicle is being primarily operated on
that fuel during use by the vehicle consumers.
Organization: POET, LLC
EPA should finalize an optional E30 certification fuel, without unnecessary gating criteria. EPA
proposes 'to allow vehicle manufacturers to request approval for an alternative certification fuel
such as a high-octane 30 percent ethanol by volume (E30) blend for vehicles they might design
or optimize for use on such a fuel.'
An optional E30 certification fuel should be allowed and finalized as soon as possible. POET
agrees with EPA that an E30 certification fuel 'could help manufacturers that wish to raise
compression ratios to improve vehicle efficiency, as a step toward complying with the 2017 and
later light-duty greenhouse gas and CAFE standards,' which 'in turn could help provide a market
incentive to increase ethanol use beyond E10 by overcoming the disincentive of lower fuel
economy associated with increasing ethanol concentrations in fuel, and enhance the
environmental performance of ethanol as a transportation fuel by using it to enable more fuel
efficient engines.'
4-306
-------
Tier 3 Summary and Analysis of Comments
An E30 certification fuel promotes the development of high-octane, clean-running vehicles that
reduce toxic air pollutants and other health-based pollutants such as PM. The use of MLEBs also
provides a clear pathway to meeting RFS mandates for biofuels use.
However, EPA also references certain 'gating' criteria to be able to use an E30/MLEB
certification fuel, including requirements to '[1] demonstrate that such a fuel would be used by
the operator and would be readily available nationwide, [2] vehicles would not operate
appropriately on other available fuels, and [3] such a fuel would result in equivalent emissions
performance.'
Each of the three 'gating criteria' regarding the use of an E30/MLEB certification fuel is
unnecessary and should be avoided.
First, a mandate to demonstrate that 'such a fuel would be used by the operator and would be
readily available nationwide' is unworkable, unnecessary, contrary to EPA's treatment of other
fuels and vehicles, and counter to EPA's stated purpose of the Tier 3 rule. Practically, requiring
that a fuel be 'used by the operator' is unworkable because an auto manufacturer cannot be
expected to make a driver's fueling choice for a vehicle that by design is intended to offer 'fuel
flexibility.' Economics and preferences will dictate which available fuel a driver will select.
Furthermore, this criterion is unnecessary because ethanol, and therefore gasoline blends
containing ethanol, are less expensive and are projected to remain so for the foreseeable future
according to the Energy Information Administration 2013 Annual Energy Outlook. So EPA
should be comfortable that MLEBs will in fact be used in E30-optimized vehicles.
A mandate that a fuel be 'readily available nationwide' is contrary to EPA's treatment of other
fuels and vehicles. For example, manufacturers are not required to demonstrate that 'fuel' for
electric vehicles or fuel cell vehicles will be available nationwide to obtain certain benefits under
EPA's CAFE/light-duty vehicle rule. In fact, electric and fuel cell vehicles do not have a
nationwide refueling availability. Just as automakers are not punished for the limited availability
of electric and fuel cell vehicle 'refueling' stations, which is largely out of the automaker's
control, automakers should not be punished for pursuing vehicle technology that leads to lower
emissions but similarly relies on the availability of fuel, which is provided by another party.
Furthermore, a requirement that a fuel is 'readily available nationwide' is more punitive than,
and inconsistent with, existing provisions for using an alternative certification fuel in existing 40
CFR § 1065.701(c), which merely requires that the fuel in question must only be 'commercially
available.' E10 is available nationwide, but gasoline refiners and producers have sought to limit
the availability of El 5 and higher-level ethanol blends.
Additionally, a mandate that a fuel be 'readily available nationwide' is counter to EPA's stated
purpose of the Tier 3 rule. This requirement is counterproductive because it exacerbates the
'chicken-or-egg' problem that alternative fuels already face: in order to successfully enter the
marketplace, there must be both sufficient alternative fuel and vehicles available to use the fuel.
EPA should do all that it can to promote and expand the use of MLEBs because they can
effectively reduce particulate matter and other toxic air emissions covered by the Proposed Rule.
Unnecessary gating criteria frustrate these emission reduction goals.
4-307
-------
Chapter 4: Proposed Vehicle Emissions Program
A gating criteria mandate to demonstrate that 'vehicles would not operate appropriately on other
available fuels' is inappropriate as it would defeat the purpose of creating 'flex-fuel' vehicles.
FFVs are specifically designed to be able to operate on multiple fuels. Gating criteria should not
unnecessarily limit the flexibility that these vehicles are designed to provide.
Finally, a gating criteria mandate that 'such a fuel would result in equivalent emissions
performance' is unnecessary. EPA already requires that vehicles achieve specific emissions
limits for pollutants (and some emission limits may be specifically tailored for individual fuel
types while maintaining a high overall level of environmental protection).
To address concerns regarding emissions on E30/MLEB-optimized vehicles when using 'other
available fuels' and 'equivalent emissions performance' (i.e., gating criteria 2 and 3, above), an
E30/MLEB-optimized vehicle could also be tested on the El 5 certification fuel, for the purposes
of demonstrating compliance with minimum emission standards. A similar approach is currently
used for 'E85' flex-fuel vehicles where they are also currently tested on EO.
If an E30/MLEB-optimized vehicle is not designed to be run on HLEBs, manufacturers should
be allowed the option of using labeling to clarify that HLEBs should not be used in such
vehicles. Such a labeling is similar to that used for 'premium-only' gasoline vehicles today, and
should be sufficient to address any concerns regarding an E30 vehicle running on HLEBs.
Two alternative certification fuels for FFVs should be available depending on what blend-range
the FFV is optimized for. Within the Ethanol Fuel Family, a manufacturer should be able to use
one certification fuel for vehicles designed to be run on HLEBs (E51-E85) and should be able to
opt to use the E30 certification fuel for vehicles optimized for E30/MLEB blends. POET
believes that these two certification fuels would be representative of in-use fuels for which
MELB and HLEBs would be designed to operate and, pursuant to Clean Air Act Section 206,
reflect 'actual current driving conditions under which motor vehicles are used.'
Regarding vehicles optimized for E30/MLEBs, POET supports EPA's proposal for an optional
E30 certification fuel, as discussed above. POET believes that an E30 certification fuel would
reflect actual, in-use fuels for which vehicles using MLEBs and the 'actual current driving
conditions under which motor vehicles are used' pursuant to CAA Section 206. Furthermore, an
E30 certification fuel would be 'within the range of fuel mixtures' that are 'expected to be found
in use.'
In fact, ethanol cost-effectively provides a high-octane, clean-burning fuel that is important for
meeting recently-tightened corporate average fuel economy (CAFE) and greenhouse gas
emissions standards promulgated by EPA. The Proposed Rule notes that MLEBs could 'help
manufacturers that wish to raise compression ratios to improve vehicle efficiency, as a step
toward complying with the 2017 and later light-duty greenhouse gas and CAFE standards.' 78
Fed. Reg. at 29,825.
Ethanol also provides other important energy security and economic benefits. Significant hikes
in oil prices immediately preceded 10 of the last 11 U.S. recessions. Ethanol today makes up
4-308
-------
Tier 3 Summary and Analysis of Comments
approximately 10% of the gasoline market, expanding our fuel supply and adding diversity,
which dampens the impact of fuel supply disruptions. Ethanol also supports over 365,000 U.S.
jobs.
Looking ahead, as more stringent emissions standards come into play, automakers will need
higher-octane fuel to improve engine efficiency. Automobile engineers have found that 'a mid-
level ethanol blend (greater than E20 and less than E40) appears to be attractive as a long-term
future fuel for the U.S., especially if used in vehicles optimized for such a fuel.' Furthermore,
according to a recent MathPro Refinery study, increasing the fuel octane to 92 AKI from 88 AKI
would result in significant cost savings. Ethanol represents the most likely source for achieving
the fuel mix needed.
As noted above, ethanol costs less than gasoline and expands the available domestic fuel supply.
Additionally, improving the fuel distribution infrastructure to accommodate higher ethanol
blends is cost-efficient. A recent Stillwater Associates study estimates the cost of updating pump
infrastructure nationwide to accommodate higher blends such as E30 at a range of 0.0024 cents
per gallon to 0.0056 cents per gallon on a 15-year amortized basis.
Organization: Governors' Biofuels Coalition
Consequently, we commend the EPA for acknowledging the use higher-octane E30 blends. The
Coalition, however, respectfully recommends that EPA make the following changes to its final
Tier 3 rule:
EPA should indeed establish an E30 certification fuel for model year 2017 and beyond. In
answer to the Agency's questions about the "appropriateness" of the current rules concerning
approval of new certification fuels, the burdens of commercializing such a fuel should not be
placed on the automakers as the relevant rule currently requires, since they have no control over
the fuel manufacturing or distribution system. Instead, EPA should follow the successful 1970s
transition from leaded to unleaded gasoline.
Organization: Growth Energy
Demonstration of Ethanol's Emissions Benefits and Growth Energy's Support of a Workable
E30 Certification Fuel: We wholeheartedly agree with EPA's assessment that the proposed
standards "represent a 'systems approach' to reducing vehicle-related exhaust and evaporative
emissions by addressing the vehicle and fuel as a system." Growth Energy made substantive
comments in this regard on the Agency's proposed Greenhouse Gas Emissions and Corporate
Average Fuel Economy Standards for 2017 and Later Model Year Light-Duty Vehicles aimed to
aid the automakers as they design the next generation of vehicles to take advantage of ethanol's
octane and emissions properties, as well as a means to achieve the goals of the RFS. Numerous
studies have shown that as our nation moves to smaller, higher-compression engines, higher
octane fuels using ethanol will be required. These higher-octane fuels using ethanol should not
be more expensive - like today's premium. Later in these comments we present the results of a
study that shows these fuels will likely be lower cost than today's regular. EPA should insist that
4-309
-------
Chapter 4: Proposed Vehicle Emissions Program
refineries and fuel producers not reduce the quality of our nation's fuel and potentially jeopardize
the nation's air quality.
In particular, Growth Energy fully supports the use of E15 emission certification fuel and the
introduction of a high octane, low-emission E30 certification fuel. We also have several concerns
about some of the hurdles to the potential use of an E30 certification fuel. We also continue to
stress the Agency's importance to the deployment of higher ethanol blends into the marketplace
in order to fulfill the goals of improving air quality, reduce greenhouse gas emissions as well as
to achieve the RFS volumes.
Growth Energy Supports a Workable, E30 Certification Fuel: Based on these findings, Growth
Energy is very supportive of the Agency moving forward with an E30 certification fuel for
emissions; however, we have significant concerns about some of the unnecessary hurdles that
automakers would experience in order to use such a fuel based on EPA's proposal.
First, a requirement that the fuel be readily available nationwide is not workable. The distribution
of fuels is largely dependent on some of ethanol's fiercest critics in the oil industry that continue
to stifle the growth of any blends of ethanol above 10 percent. EPA should not hold the
automakers accountable for the oil industry's effort to limit the availability of these fuels, and
instead, should either allow the automakers to immediately certify their vehicles on E30 or
otherwise require that it be made available nationwide.
Secondly, as we have shown, midlevel ethanol blends make significant contributions to our
nation's air quality. As such, EPA should consider using its authority and weight to make E30
and other midlevel ethanol blends more available in the marketplace and should not penalize
automakers for making vehicles capable of using such fuels. By requiring automakers to
"demonstrate that vehicles would not operate appropriately on other available fuels" would run
completely counter to their already well-established FFV production and the intent of making a
Flexible Fuel vehicle.
Growth Energy is also concerned that the R-Factor of 0.6 in the current fuel economy equation
could also prevent automakers from using an E30 certification fuel option because of its impact
on the fuel economy standards. The current factor is based on the use of older engine technology
using ethanol-free gasoline. Based on information contained in the docket for the proposal
(Aaron Butler, "Analysis of the Effects of Changing Fuel Properties on the EPA Fuel Economy
Equation and R-Factor"), a higher R-factor of 0.8 - 0.9 would be more accurate and may
encourage automakers to use an E30 test fuel.
Finally, while we believe it is essential to remove these hurdles in order for automakers to use
E30 as a certification and in-use fuel; we do, however, want to offer the agency the benefit of the
extensive work and research we have done to get an in-use high octane, low-emission fuel into
the national marketplace. Growth Energy commissioned two studies to look at both the refining
cost, and distribution costs of developing a 92 Anti-Knock Index (AKI), E30 gasoline. These
studies found that an E30 high octane blend can be produced for less than the cost of current
gasoline, and that the development of the infrastructure over time is very affordable. Specifically,
a 92 AKI E30 can be produced for between $11.7 billion and $30.8 billion per year less than the
4-310
-------
Tier 3 Summary and Analysis of Comments
cost of current 88 AKI regular gasoline. Additionally, the costs to develop infrastructure at
terminals and gas stations across the country ranged from $0.0024 to $0.0056 per gallon on a 15-
year amortized basis. The complete studies are attached and are briefly summarized here:
MathPro Refinery Study Shows E30 is Less Expensive Than Today's Gasoline: The MathPro
Refinery study [This study is can be found in Docket number EPA-HQ-OAR-2011-0135-4681-
A4.] employed a linear programming (LP) model of the U.S. refining sector to estimate the main
economic effects in the U.S. refining sector (including changes in the refining costs of
transportation fuel) and on consumers of future fuel standards requiring the use of 92 AKI
gasoline in HiTech vehicles. The refinery LP model incorporates the molar concentration
blending method to represent ethanol's blending octane in the various ethanol/hydrocarbon BOB
blends considered. This method produces estimates of the effective blending octane of ethanol in
given ethanol/BOB blends that are functions of the molar fraction of ethanol in the blend, the
octane of the BOB, and the octane of neat ethanol (109 RON/90 MON). After calibrating the
refinery LP model to replicate reported operations of the U.S. refining sector in 2011, MathPro
used the model to assess alternative assumptions on how the new fuel economy standards and
increased ethanol use in the target years would affect the volume of U.S. gasoline exports in
those years. They estimated the effects on refining economics and consumer costs of the various
92 AKI ethanol blend options by comparing the estimated costs returned by the refinery model in
the 92 AKI Study cases with those in the corresponding 88 AKI Reference cases. Finally, they
developed estimates in this manner for each of two assumed ethanol price scenarios: Low
ethanol price: energy parity (e7K BTU) with the wholesale price of 88 AKI E10, and High
ethanol price: volumetric parity (eVgal) with the wholesale price of 88 AKI E10. MathPro
determined that the impact on the cost of transportation fuel was that the 92 AKI gasoline would
be between $11.7 billion per year (ethanol at volumetric parity) and $30.8 billion per year
(ethanol at energy parity) less than the current AKI 88 regular gasoline.
Stillwater Associates Marketing and Distribution Study Shows E30 Infrastructure is Very
Affordable: Stillwater Associates examined the marketing and distribution costs of widespread
distribution of E30 fuel to 2017 and later motor vehicles (and FFVs). Marketing and distribution
costs consist of bulk transport of additional ethanol, modifications to terminals that store and
blend ethanol with gasoline, and modifications to gasoline stations. Stillwater modeled three
cases. In the first case, it was assumed that E30 would be blended at the terminal and delivered to
the station. Both terminals and stations would incur costs because they are not currently capable
of delivering E30 except for a limited number of Midwest localities. The cost of this case is
estimated at $3.5 billion, an average of $37 thousand per participating service station
or$.0024/gallon on a 15-year amortized basis. The second case recognizes that gasoline quality
regulations could drive a different solution. This second case looked at a combination of terminal
blending and service station blending. The case recognizes that in regions where Reid Vapor
Pressure (RVP) is controlled by reformulated gasoline regulations or by State Implementation
Plans, current gasoline can be blended with fuel ethanol through a "blender pump" at the service
station to make E30 without requiring a special E30 blendstock. In the Conventional Gasoline
areas where RVP is subject to the 1-psi waiver, the E30 would have to be blended with a
separate gasoline blendstock at the distribution terminal and delivered to the gas station. The cost
of the second case is estimated at $5.8 billion, an average of $62 thousand per participating
service station or $.0040/gallon on a 15-year amortized basis. The third case examined the cost
4-311
-------
Chapter 4: Proposed Vehicle Emissions Program
of doing all the E30 blending at the service station through blender pumps and assumes the RVP
regulatory issues for E30 have been addressed. This option provides the retailer with the greatest
level of flexibility for offering different blend levels, but also has the highest cost. The estimated
cost of this case is $8.1 billion, or $87,000 per participating station. Spread over the 15 year life
of the equipment, the cost of the changes is estimated to be $0.0056 per gallon.
Organization: INEOS Bio
INEOS Bio is helping develop new and innovative ways to help fuel America and the world;
providing cleaner burning cellulosic ethanol that helps reduce greenhouse gas emissions and
provide more sustainable sources of fuel and energy. Achieving our nation's goals of less
dependence of foreign sources of oil and cleaner fuels will require our economy to transition to
sustainable energy resources and higher levels of energy efficiency. Toward this end, federal
policy and regulations—including the proposed rule as well as the federal Renewable Fuel
Standard (RFS) and EPA's consistent support and implementation of that policy—play an
important role in helping to drive the commercialization of these technologies. The importance
of federal policy is particularly critical in the transportation fuel sector. The U.S. transportation
system is overwhelmingly and unsustainably reliant on petroleum fuels. These traditional fuels
are a large component of the U.S. greenhouse gas (GHG) emissions inventory and our
overdependence on foreign sources of energy. Rapid transition to more alternative transportation
fuels is essential to reducing GHG emissions and reducing U.S. reliance on foreign sources of
energy.
INEOS Bio also supports EPA's proposal to allow vehicle manufacturers to request approval for
an alternative certification fuel such as a high-octane E30 blend for vehicles they might design or
optimize for use on such a fuel. INEOS Bio is aligned with EPA's assertion in the proposed rule
that "[t]his could help manufacturers that wish to raise compression ratios to improve vehicle
efficiency, as a step toward complying with the 2017 and later light-duty greenhouse gas and
CAFE standards.
We commend EPA's efforts to transition the emissions test fuel to better reflect the current and
future in-use fuel as it begins to contain greater volumes of biofuels resulting from the continued
development and commercialization of biofuels and increased RFS volumes in the market.
INEOS Bio believes that EPA should work to set the emissions test fuel to promote the highest
achievable level of ethanol and octane in the U.S. fuel supply and thus lower the maximum
aromatics and lowest additive concentration (LAC) content in gasoline accordingly. This action
would help EPA meet one of its stated goals of this rulemaking to address the impacts of motor
vehicles and fuels on air quality and public health without increasing gasoline prices. To this
end, we believe the final rule should allow for ongoing and greater use of E85 and El 5, as well
as the transition to higher blends of ethanol and advanced biofuels.
INEOS Bio also supports EPA's proposal to allow vehicle manufacturers to request approval for
an alternative certification fuel such as a high-octane E30 blend for vehicles they might design or
optimize for use on such a fuel. As described above, we agree with EPA that this provision could
help auto manufacturers meet their CAFE requirements and further displace aromatics content in
gasoline. The benefits of Tier 3 that include lower sulfur, aromatics reductions, combined with
4-312
-------
Tier 3 Summary and Analysis of Comments
higher octane and increased use of renewables are significant, and I urge you to work to finalize
this standard as soon as possible and subsequently include them in any future joint proposals
such as CAFE, and National Program standards.
Organization: International Council on Clean Transportation (ICCT)
The draft proposal requested comments on the use of E30 in certification testing: 'We are
proposing to allow vehicle manufacturers to request approval for an alternative certification fuel
such as a high-octane 30 percent ethanol by volume (E30) blend for vehicles they might design
or optimize for use on such a fuel.'
The ICCT is supportive of all ways to increase efficiency. In particular, the ICCT would strongly
support increasing the minimum required octane for all gasoline.
Despite this, the ICCT could support the E30 provision only if the vehicles actually used E30 the
vast majority of the time in-use. However, this would create major infrastructure difficulties:
how would the E30 infrastructure be developed in advance of offering E30-capable vehicles? It
also has potentially troubling consequences related to food-based high-iLUC ethanol, as
discussed above with respect to E15.
An entirely new infrastructure would be needed for E30, which means a huge chicken-and-egg
problem. Just like with E85, one way to attempt to solve this is to offer credits to E30 vehicles,
whether or not they actually use E30 in-use. The ICCT is extremely concerned that this will
become another loophole, with large credits against the CO2/CAFE standards granted and little
use in the real world, reducing the benefits of the standards. Our concerns are magnified by the
recent EPA Manufacturer Guidance Letter on E85 usage, which proposed to grant flexible-fueled
vehicles a 20% E85 usage rate (F-factor), even though E85 usage in the real world has remained
steady at 1.1% since 1998.
Our infrastructure concerns are exacerbated by the fact that E30 has 10% lower energy content
than gasoline. Certainly some of this can be recaptured with higher efficiency, but it won't be a
10% efficiency improvement. Thus, customers will be able to travel further on gasoline (or E10
or E15) than with E30, unless the vehicle is designed so that it does not run well on E10 or E15.
Which raises another concern with vehicles designed for E30. A recent SAE paper on the
impacts of ethanol blends reviewed what happens when a vehicle designed for E30 is run on 87-
octane gasoline:
'With engine downsizing, the reduction in full load torque with regular grade 91 RON fuel will
be proportional to the amount of downsizing, and can result in unacceptable vehicle performance
attributes for aggressive levels of downsizing.'
This loss of performance on regular fuel just makes the chicken-and-egg problem worse. Engines
that require E30 are going to be a tough sell in the market.
4-313
-------
Chapter 4: Proposed Vehicle Emissions Program
Given the infrastructure concerns, E30 must offer substantial efficiency benefits to justify the
investment in a new infrastructure. Thus, the key question is: What is the efficiency benefit of
running on E30 compared to alternative technologies? The Stein 2013 SAE paper concluded:
'From an engine standpoint, the primary motivation for increasing ethanol content is improved
knock resistance'. However, high EGR rates also offer improved knock resistance. If E30
basically duplicates the benefits of boosted-EGR, it is hard to see how creation of a new
infrastructure can be justified.
Given all of the above, E30 should only be allowed if the manufacturer can demonstrate that the
vehicle will almost always be refueled with E30 in-use. Given the historical abuse of similar
provisions, such as FFV credits and the recent proposal for 20% F-factor for FFVs, the ICCT is
very concerned about the potential to also abuse this provision.
The ICCT is also concerned about the revision to use E-15 for test fuel. This is acceptable for
certification vehicles, but it appears the EPA is also proposing to use E-15 for fuel economy
testing. Fifteen percent ethanol provides significant evaporative cooling in the cylinder, which
would allow manufacturers to make modifications to improve fuel economy in the test cycles,
but which would not necessarily be achieved end use.
Organization: Marathon Petroleum Company LP (MFC)
Proliferation of Test Fuels: Part of EPA's goal in modifying the certification fuel for Tier 3 was
to reduce the number of certification fuels manufacturers would need to use to test their vehicle
fleet. This goal seems to be lost as EPA is proposing to "allow vehicle manufacturers to request
approval for an alternative certification fuel such as a high-octane 30 percent ethanol by volume
(E30) blend for vehicles they might design or optimize for use on such a fuel. This could help
manufacturers that wish to raise compression ratios to improve vehicle efficiency, as a step
toward complying with the 2017 and later light-duty greenhouse gas and CAFE standards (2017
LD GHG). This in turn could help provide a market incentive to increase ethanol use beyond
E10 by overcoming the disincentive of lower fuel economy associated with increasing ethanol
concentrations in fuel, and enhance the environmental performance of ethanol as a transportation
fuel by using it to enable more fuel efficient engines." Allowing each manufacturer the option to
request approval of an alternative certification fuel could have significant impact on market
dynamics. One can imagine a proliferation of boutique fuels necessary to support each individual
manufacturer so that their vehicles can actually deliver the fuel efficiency as tested. This
approach does not fit with the extensive, fungible fuel system that currently exists nor does it
account for the limited fuel options available at service stations due to underground storage tank
configuration and space availability. How does EPA expect to steward such a program and who
would be responsible for determining the cost, well-to-wheel emissions, etc? One could image an
engine that is optimized on an alternative certification fuel but would not deliver anywhere near
the efficiency on in-use fuels so regarding alternative fuel availability, what would be the market
threshold for EPA to consider allowing a manufacturer to use an alternative certification fuel?
Organization: National Association of Convenience Stores (NACS) and Society of
Independent Gasoline Marketers of America (SIGMA)
4-314
-------
Tier 3 Summary and Analysis of Comments
Neither stakeholders nor regulators can predict all future changes in gasoline vehicle
technologies or in-use fuels. The proposal contains specific specifications for high-octane
formulations, but specifies that these fuels would be ethanol-based, such as E30. Other high-
octane fuel formulations would require different fuels to undergo a potentially lengthy and
expensive petition process for approval as test fuels.6 The final rule should not pick winners and
losers in this regard. Ethanol may be the octane of choice at present, but EPA should allow for
and encourage new fuels to come to market rather than impose artificial impediments to
innovation.
Organization: National Marine Manufacturers Association (NMMA)
NMMA testified at the CARB Board hearing requesting that the board also allow for the use of
an alternative California certified advanced biofuel, if available.
EPA requests comment regarding an option that would allow vehicle manufacturers to certify
using a 30 percent by volume ethanol blend (E30) in recognition that E85 has not met EPA's
expectations and to encourage light duty vehicle manufacturers' efforts to continue to determine
what might be the optimal ethanol content. Given the large difference in oxygen content between
E10 and E30 this approach will only cause further confusion and misfueling of marine engines
when E30 reaches the marketplace, as would appear to be inevitable based on EPA's proposal.
This bifurcated approach of allowing E10, E30 and E85 in the marketplace does nothing to
address the root problems associated with ethanol-enhanced fuels: ethanol is not gasoline.
Organization: Renewable Fuels Association (RFA)
We support the proposal allowing vehicle manufacturers to request approval for an alternative
certification fuel.
The RFS2 calls for the continuing increase of ethanol content in gasoline. RFA supports the
continued oversight by EPA on upcoming vehicle and fuel technology (specifically, recent
research that suggests increasing the ethanol content in gasoline commensurate with
improvement in octane can provide a vast improvement in vehicle efficiency, fuel economy and
emissions). The auto industry has requested the use of increased ethanol content with increased
octane in an effort to achieve future requirements for the reduction of greenhouse gas emissions.
(Auto Alliance letter dated October 2013.) Further, the RFA commissioned a study of future
influences and importance of fuel octane on future engine developments by Ricardo. Auto
manufacturers will continue to build vehicles reliant on liquid motor fuels that rely on newly
developed engineering improvements to address converging regulatory requirements for
reductions in greenhouse gas emissions and pollutants. Ricardo concludes that "during the next
decade internal combustion engines will become significantly more efficient due to pressures
from regulatory and consumer preference. Similarly, higher minimum fuel octane number will
facilitate the engine technologies that will boost specific power and engine efficiency. Future
powertrain solutions will have a natural thirst for higher octane fuels."
Extensive research is being completed by Original Equipment Manufacturers and others to
further understand the benefits to increasing ethanol content in the current motor fuel pool.
4-315
-------
Chapter 4: Proposed Vehicle Emissions Program
The most significant benefit of higher ethanol blends is the boost to the fuel blend's octane
rating, nearly a 4 to 7 Research Octane Number increase from an additional 10 to 20% ethanol.
A recent Ford Motor Company publication concludes "higher minimum octane ratings for
regular-grade fuel would enable higher compression ratios in future vehicles and is an
opportunity to provide greater engine efficiency and meet increasingly stringent fuel economy
regulations and expectations. Additionally, the change could benefit all vehicles powered by
spark-ignited engines, including PFI and DI engines, hybrid electric vehicles (HEVs), and plug-
in hybrid vehicles (PHEVs)."
This newly developing evidence suggests that ethanol's contribution to the octane level of the
gasoline pool may only beginning to be realized; additional and greater benefits of using ethanol
to increase gasoline's octane provide benefit beyond optimum performance.
However, we are concerned that EPA's proposed criteria for acceptance of a petition for an
alternative certification fuel may be unrealistic and could discourage automakers from pursuing
such a petition. Specifically, the proposed conditions that an automaker must "...demonstrate
that such a fuel would be used by the operator and would be readily available nationwide," and
that "...vehicles would not operate appropriately on other available fuels" may severely limit the
automakers' ability or desire to utilize the alternative certification fuel petition process.
Organization: Mercedes-Benz USA, LLC on behalf of Daimler AG
Mercedes-Benz acknowledges the EPA's request for comment on the incorporation of a higher
octane, mid-blend certification fuel such as E25 as a key enabler for GHG reduction. Octane is
the single most important property of gasoline when determining engine design. Higher octane
fuels permit higher compression ratios which directly improve efficiency while downsizing
engines also results in greater fuel efficiency. The optimized combination of those two actions
with gasoline direct-injection provides remarkable gains in fuel economy but requires high
octane market fuel - higher octane than is available today. Additionally, higher octane,
traditionally provided through selective petroleum refining and blending, can be further
increased through the addition of ethanol in blends between 20-25 percent by volume. This
powerful fuel enjoys both reduced carbon intensity as well as the renewable aspects of ethanol. A
vehicle equipped with a powertrain which is optimized for a high-octane, mid-blend ethanol fuel
(hereafter referred to as 'Tier 3 Fuel') can simultaneously fulfill what the customer desires -
performance and economy, while reducing the environmental impact. Mercedes-Benz vehicle
offerings include those with E25 capability in various global markets and could be introduced
them to the US market if regulatory and commercial conditions warrant. However, in order to
harmonize with global markets, ethanol blend levels should be limited to 25% v/v as the effort to
perform engine development, validation and certification for blends greater than E25 is quite
substantial and necessary to determine hardware compatibility.
One aspect not reduced is the customer expectation of performance, even in the wake of
regulations mandating improved fuel economy. Thus, another key enabler in reducing
greenhouse gas contributions while maintaining performance is increased octane and gasoline.
Octane is the single most important property in gasoline when determining engine design.
4-316
-------
Tier 3 Summary and Analysis of Comments
Higher octane fuels permit higher compression ratios which directly improve efficiency.
Downsizing engines also results in greater fuel efficiency.
The optimized combination of these two actions with gasoline direct injection provides
remarkable gains in fuel economy, but requires high octane market fuel, higher octane than is
available today. Additionally, higher octane traditionally provided through selective petroleum
refining and blending can also be further increased through the addition of ethanol and blends
greater than 20 percent by volume. This powerful fuel enjoys both reduced carbon intensity as
well as the renewable aspects of ethanol.
A vehicle equipped with a power train, which is optimized for a high octane mid-blend ethanol
fuel can simultaneously fulfill what the customer desires, performance and economy, while
reducing the environmental impact. In fact, Mercedes-Benz vehicle offerings include those with
E-25 capability in various global markets, and can introduce them in the U.S. if regulatory and
commercial conditions warrant.
To summarize, the synergy of mid-blend ethanol and higher octane based fuel in the
proposed rulemaking opens the door to far greater possibilities in greenhouse gas and criteria
emission reduction. Mercedes-Benz fully supports these proposed changes in certification and
market fuel,
Organization: American Motorcyclist Association (AMA)
In its Tier 3, rule, the EPA proposes to change the certification fuel to El5, with an allowance for
manufacturers to petition the EPA for an E30 certification. This, proposal, if promulgated, will
create an environment where this country will have a bifurcated fleet of engines requiring
different types of fuel.
Organization: Truck and Engine Manufacturers Association (EMA)
EPA requests comment regarding an option to certify Tier 3 on-highway vehicles using a 30
percent volume ethanol blend (E30). (78 Fed. Reg. at 29938.) In many respects, a bifurcated
certification test fuel option to certify engines/vehicles utilizing an ARB-aligned (E10)
certification test fuel, or a new alternative E30 certification test fuel, could provide a viable
strategy for both RFS-2 compliance and engine/vehicle compatibility on a long term basis.
Existing light-duty flexible-fuel vehicles would provide a market for E30 fuels until such time as
engines/vehicles designed specifically for E30 would be available in the marketplace. If a
manufacturer of a gasoline-fueled nonroad, motorcycle, heavy-duty on-highway engine
determined there was sufficient customer interest, unique E30-certified products could be
developed and marketed utilizing the alternative E30 certification test fuel.
EMA recommends that EPA work with industry to identify and implement the appropriate
ethanol content for such an alternate certification test fuel.
Our Response:
4-317
-------
Chapter 4: Proposed Vehicle Emissions Program
We received a large number of comments regarding our statement that manufacturers
under certain situations could utilize an existing regulatory provision, 40 CFR 1065.710(c), to
petition the Administrator for approval of an alternative test fuel such as a higher octane, higher
ethanol content gasoline. Many of these comments focused on whether it is appropriate for EPA
to use its authority to encourage or require widespread availability and/or use of such alternative
fuels (e.g., higher octane blends, E30, isobutanol, etc.) These commenters appear to have
misunderstood the purpose of the change in emission test fuel specifications. EPA is using its
Clean Air Act section 206 authority to establish requirements for emission certification testing,
including specifications of the emissions test fuels with which vehicles demonstrate compliance
with emissions standards, in order to better reflect the ethanol content and other properties of in-
use fuels. While we are taking action to limit sulfur content in in-use gasoline under authority in
Clean Air Act sections 21 l(c)(l) and (2), we did not propose other changes to in-use fuels. Any
action related to increasing the ethanol level of in-use fuels is outside the scope of this
rulemaking.
The alterative certification fuel provisions are intended to allow manufacturers to request
approval from EPA to certify vehicles that they design, build, and market to run on fuels other
than those for which we have established certification test fuel requirements. In the past, the
Agency has approved alternative fuels such as natural gas in response to petitions under this
provision.
4.5.1.6. Changes to In-Use Gasoline Other than Sulfur Content
What Commenters Said:
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
Benefits of Lower RVP:
In the summer of 2011, EPA sponsored a testing program that was designed to monitor
evaporative emissions, canister loading profiles, and breakthrough emissions from extended
multiday diurnal testing using 9 and 10 psi RVP fuels on 9 vehicles. In the spring of 2012, the
Alliance and Global Automakers sponsored follow-on testing on a subset of the original vehicles,
using 7 psi RVP E10 fuel in order to provide a more comprehensive set of data. The results of
the study were published in Effects of 7, 9, and 10 psi Vapor Pressure Fuels on Multi-Day
Diurnal Evaporative Emissions of Tier 2 and LEV II Vehicles, © SAE paper 2013-01~'1057,
which was presented at the 2013 SAE World Congress in Detroit, MI. The light-duty vehicles
used in the testing program were either certified to Tier 2 federal standards or the current most
stringent evaporative emission standard in the U.S. - California's Low Emission Vehicle (LEV
II) Partial Zero Emission Vehicle (PZEV) standard. The test procedure for both phases of the
testing program consisted of continuous monitoring over 14 days, which is 11 to 12 days beyond
normal testing procedures, of the federal diurnal cycle for evaporative emissions (72° F to 96°
F). Figure 2.
4-318
-------
Tier 3 Summary and Analysis of Comments
For all five vehicles tested, the 7 psi RVP fuel resulted in the least amount of hydrocarbon slip
emissions from the vehicle's carbon canister over the 14-day period, relative to the 9 and 10 psi
RVP fuels. The data show that the 7 psi RVP fuel resulted in the least amount of evaporative
emissions, with the 9 psi RVP fuel resulting in more evaporative emissions relative to the 7 psi
RVP, and the 10 psi RVP resulting in the most hydrocarbon slip emissions. This demonstrates
that appropriate reductions in fuel vapor pressure lead to reduced evaporative emissions.
Meeting the more stringent evaporative emissions standards of Tier 3 is made possible through
coordination of the "fuels and vehicle as a system" — as consistently emphasized in the Tier 3
NPRM. The reduction in summertime market fuel vapor pressure could benefit the entire car
park from day one. In the NPRM, the EPA notes that "attention is needed to insure better in-use
performance of current evaporative emissions." A reduction in market fuel summertime vapor
pressure would reduce the evaporative emissions performance of in-use vehicles and address
EPA's concern.
Tier 3 Summertime Market Gasoline RVP Cap should be 8 psi (9 psi for E10) (per Federal VOC
Control Season):
This would result in a maximum 9 psi RVP E10 market fuel, including the statutory one pound
waiver. The most effective method of controlling prospective evaporative emissions in
conventional gasoline use areas would be to move promptly to lower the maximum allowable
RVP to compensate for the 1 psi waiver, perhaps to 8 psi. When the statutory mandate for the 1
psi ethanol waiver is added, the fuel would then become 9 psi, as it was before the introduction
of ethanol. Similar to EPA's proposed market sulfur standard, this would bring about an
immediate reduction in summer light duty evaporative emissions without waiting for the fleet to
turn over. In addition, there would be reductions in evaporative emissions from all gasoline
powered equipment, further enhancing the benefits of this approach.
10 psi RVP gasoline hinders States seeking to avoid backsliding out of attainment, or
neighboring states trying to meet State Implementation Plans (SIPS) to reduce criteria pollutant
emissions.
9 psi RVP Market and Test Fuel is Important to Avoid Emission Backsliding: EPA has an
overdue obligation under the Congressional mandate in §209 of EISA (and implemented in
section 21 l(v) of the CAA) which requires EPA to complete an anti-backsliding study, and
promulgate fuel regulations to implement appropriate mitigation measures, if needed, based upon
that study. EPA should complete that study immediately and undertake its mitigation rulemaking
to address market fuel RVP.
Organization: American Coalition for Ethanol (ACE)
Along those same lines, ACE recommends immediately extending the 1 Ib waiver to blends from
El 1 to E50), as each additional percentage above 10% ethanol lowers Reid vapor pressure below
that of nearly all gasoline sold in the United States. Adding ethanol to E10 will lower its
volatility, and much in the same way EPA has suggested its rules might allow biobutanol to be
added under an 'averaging' policy, we believe a similar approach would allow ethanol to be
added to E10 to create compliant fuels with between 11 and 50 percent ethanol. Rather than
4-319
-------
Chapter 4: Proposed Vehicle Emissions Program
singling out El5, E30, or any other higher ethanol blend and requiring it to meet higher standards
than other fuels, we believe that rules should welcome blends that are cleaner than the current
industry standard E10.
Organization: National Marine Manufacturers Association (NMMA)
EPA requests comment regarding the option of allowing the use of CARB E10 certification test
fuel through model year 2019 with a Reid Vapor Pressure (RVP) of 10 psi (78 Federal Register
29895). EPA currently provides a 1 psi waiver for E10, but this waiver will not apply to E15.
NMMA recommends that the 1 psi waiver also be eliminated for E10. EPA finalized a rule in
2010 that requires boat builders to meet stringent new evaporative emission requirements.
California is currently in the process of finalizing a rule. In order to meet these new emission
standards, marine engine manufacturers need a combination of new technology and a fuel with a
minimum evaporation rate.
Organization: General Motors LLC (GM)
But for the one pound waiver, E10 would be 9 psi or lower, which it is in many parts of the
country. As discussed further below, for both vehicle performance and environmental benefits,
we advocate that EPA reduce market fuel E10 RVP to 9 psi (by making the vapor pressure
standard at 8 psi, which, with the E10 one pound waiver, would make E10 9 psi).
Tier 3 Certification Fuel RVP and Marketplace Fuel RVP Should both be 9 RVP during the
Summer VOC Control Season. Benefits of Lower RVP: In the summer of 2011, the EPA
sponsored a testing program that was designed to monitor evaporative emissions, canister
loading profiles, and breakthrough emissions from extended multiday diurnal testing using 9 and
10 psi RVP fuels on 9 vehicles. In the spring of 2012, the Alliance of Automobile Manufacturers
and the Association of Global Automakers sponsored follow-on testing on a subset of the
original vehicles, using 7 psi RVP E10 fuel in order to provide a more comprehensive set of data.
The results of the study were published in Effects of 7, 9, and 10 psi Vapor Pressure Fuels on
Multi-Day Diurnal Evaporative Emissions of Tier 2 and LEV II Vehicles, © SAE paper 2013-
01-1057, which was presented at the 2013 SAE World Congress in Detroit, MI. The light-duty
vehicles used in the testing program were either certified to Tier 2 Federal standards or the
current most stringent evaporative emission standard in the U.S. - California's LEV II Partial
Zero Emission Vehicle (PZEV) standard.
For all five vehicles tested, the 7 psi RVP fuel resulted in the least amount of hydrocarbon slip
emissions from the vehicle's carbon canister over the 14-day period, with the 9 psi RVP fuel
resulting in more evaporative emissions relative to the 7 psi RVP, and the 10 psi RVP resulting
in the most hydrocarbon slip emissions. This demonstrates that appropriate reductions in fuel
vapor pressure lead to reduced evaporative emissions.
Meeting the more stringent evaporative emissions standards of Tier 3 is made possible through
coordination of the "fuels and vehicle as a system" as consistently emphasized in the Tier 3
NPRM. The reduction in summertime market fuel vapor pressure could benefit the entire car
park from day one. In the NPRM, EPA notes that "attention is needed to insure better in-use
4-320
-------
Tier 3 Summary and Analysis of Comments
performance of current evaporative emissions." A reduction in market fuel summertime vapor
pressure would reduce the evaporative emissions of in-use vehicles and address EPA's concern.
Tier 3 Summertime Market Gasoline RVP Cap should be 8 psi (9 psi for E10): The most
effective method of controlling prospective evaporative emissions in conventional gasoline use
areas would be to move promptly to lower the maximum allowable RVP to compensate for the 1
psi waiver, perhaps to 8 psi. When the statutory mandate for the 1 psi ethanol waiver is added,
the fuel would then become 9 psi, as it was before the introduction of ethanol. Similar to EPA's
proposed market sulfur standard, this would bring about an immediate reduction in summer
light-duty evaporative emissions without waiting for the fleet to turn over. In addition, there
would be evaporative emissions reductions from all gasoline powered equipment, further
enhancing the benefits of this approach.
10 psi RVP gasoline hinders states seeking to avoid backsliding out of attainment, or
neighboring states trying to meet State Implementation Plans (SIPS) to reduce criteria pollutant
emissions.
9 psi RVP Market and Test Fuel is Important to Avoid Emission Backsliding: EPA and the U.S.
Congress are concerned about the air quality impacts of the higher RVP E10 blends that have
swept across large portions of the nation. Under the Energy Policy Act of 2005, EPA is
mandated to undertake a "backsliding analysis" consisting of the following components: draft
analysis, final analysis, emissions model, and a permeation effects study including evaporative
emissions. Not content with the direction contained within the 2005 Energy Policy Act,
Congress, within the Energy Independence and Security Act of 2007, wrote a section, 209,
entitled "anti-backsliding". In this section, Congress again directed that "the Administrator shall
complete a study to determine whether the renewable fuel volumes required by this section will
adversely impact air quality as a result of changes in vehicle and engine emissions of air
pollutants regulated under this Act." Further, Congress directed that "Not later than 3 years after
the date of enactment of this subsection, the Administrator shall— '(A) promulgate fuel
regulations to implement appropriate measures to mitigate, to the greatest extent achievable,
considering the results of the study under paragraph (1), any adverse impacts on air quality, as
the result of the renewable volumes required by this section; or (B) make a determination that no
such measures are necessary.'"
EPA has a congressionally mandated duty to complete the anti-backsliding study and to
promulgate fuel regulations to implement appropriate mitigation measures based upon that study.
It is in that mitigation rulemaking that EPA should determine the appropriate RVP of gasoline
(certification and market gasoline). To regulate certification fuel RVP in this Tier 3 rule puts the
cart before the horse. Moreover, EPA is trying to fix a problem it sees with in-use gasoline RVP
that Congress has instructed EPA to study and fix with fuel regulations under its RFS authority.
GM does not believe that increasing certification fuel stringency for an evaporative emissions
test that will phase in over the next seven years qualifies as "to the greatest extent achievable."
EPA must pursue the appropriate mitigation measures through a separate rulemaking that is
informed by the results of its study.
Organization: National Automobile Dealers Association (NADA)
4-321
-------
Chapter 4: Proposed Vehicle Emissions Program
EPA should adopt a nationwide Tier 3 summertime market RVP cap of 8 psi for EO and 9 psi for
E10. In addition to enabling Tier 3 compliance, doing so result in evaporative emission reduction
benefits for the in-use fleet.
Organization: Ford Motor Company (Ford)
Market Fuel Vapor Pressure Must be Controlled Based on Seasonal Ambient
Temperatures: Vapor pressure is the most important property for engine start-up performance. At
higher ambient temperatures, the fuel can vaporize prematurely prior to reaching the injectors,
disrupting the fuel flow to the engine and creating "vapor lock". As defined in ASTM D4814,
vapor lock conditions arise "if too much vapor is formed, the fuel flow to the engine can be
decreased, resulting in loss of power, rough engine operation, or engine stoppage." During cold
ambient temperatures, fuels with vapor pressures that are too low can prevent the engine from
starting at all or have poor warm-up performance. Thus, the vapor pressure and distillation
profile of gasoline must be controlled based on seasonal ambient temperatures so that fuel
vaporizes easily (higher vapor pressure) in cold weather to assist in engine starting and vaporizes
less easily (lower vapor pressure) in warm weather so as to prevent vapor lock and reduce
evaporative emissions. Evaporative emissions of hydrocarbons from the fuel system can occur
not only during vehicle operation, but while the vehicle is at rest during periods of increasing
ambient temperatures. Numerous studies have shown that the vapor pressure has a profound
effect on the amount of hydrocarbon evaporative emissions.
Recommendation: Ford supports the reduction in market fuel vapor pressure as detailed in the
Alliance comments. Ford supports an ambient temperature (seasonally) appropriate 8 psi (9 psi
for E10 due to the Ipsi waiver) vapor pressure maximum for market fuels. Likewise, Ford also
supports a 9 psi maximum Tier 3 certification fuel.
Organization: POET, LLC
RVP limits must not be a barrier to the use of higher ethanol blends. RVP is a measure of
gasoline volatility that EPA regulates to address summer ozone (smog) issues. Under the
statutory framework, CAA Section 21 l(h) mandates that EPA shall limit introducing 'into
commerce gasoline with a Reid Vapor Pressure in excess of 9.0 pounds per square inch (psi)'
and further limit RVP in certain nonattainment areas.
Notably, the RVP provisions in CAA Section 21 l(h) only control 'gasoline.' Because E10 and
El 5 have been considered part of the gasoline family, these ethanol blends have been regulated
by these provisions. However, ethanol alone doesn't cause RVP issues, because ethanol itself has
very low vapor pressure. Furthermore, EPA notes that as 'the ethanol level increases, the
volatility increase caused by blending ethanol with gasoline begins to decline.'
Because ethanol blended with gasoline to make E10 can have a RVP slightly above the
regulatory limit, CAA Section 211(h)(4) provides a 1 psi RVP 'ethanol waiver.' In particular,
section 21 l(h)(4) specifies that 'For fuel blends containing gasoline and 10 percent denatured
anhydrous ethanol, the Reid vapor pressure limitation under this subsection shall be one pound
per square inch (psi) greater than the applicable Reid vapor pressure limitations.'
4-322
-------
Tier 3 Summary and Analysis of Comments
Although ethanol blends greater than 10% generally have lower RVP than E10, such blends may
nonetheless be slightly over the regulatory limit and thus merit the use of the 1 psi waiver. These
ethanol blends over 10% are critical to ensure that the RFS standards are met, and they provide a
host of environmental and energy-security benefits. It would be the height of poor policy design
by EPA to bar cleaner fuels that nonetheless may slightly tick over an RVP threshold when a
waiver is given to fuels with a higher RVP.
If RVP issues are not otherwise addressed, then the current 1 psi waiver should be applied to all
ethanol blends over E10 (i.e., El 1 and above).
Alternatively, the above statutory provisions in CAA Section 21 l(h)(4) granting the 1 psi waiver
to ethanol blends should be read as only requiring a minimum 10% ethanol content. EPA
wrongly rejected this approach whereby the '10 percent' threshold should be construed as
requiring 'at least' 10 percent ethanol. EPA reached this approach by indicating that its
interpretation 'harmonizes' sections 21 l(h)(4) and 21 l(h)(5). Section 21 l(h)(5) requires EPA to
not apply the 1 psi waiver on petition by the Governor of a State that the waiver will 'increase
emissions that contribute to air pollution in any area in the State.'
However, EPA's bar on extending the E10 waiver to E15 is not compelled by the statutory
language. EPA ignores that in subsections 21 l(h)(4) and 21 l(h)(5) the references to 'fuel blends
containing gasoline and 10 percent' ethanol used in both can (and should) be interpreted to mean
gasoline fuel blends containing at least 10% ethanol. Additionally, the statute here references
'blends' in the plural, suggesting that an EPA interpretation focusing only on E10, while
ignoring other blends with at least 10% ethanol, is improper.
Furthermore, EPA's bar on extending the E10 waiver to El5 is inconsistent with prior
rulemakings, and Congress' understanding of these prior rulemakings. Congress based the 1 psi
waiver in section 21 l(h)(4), in part, on EPA's 1987 rulemaking that established national
volatility limits for gasoline but expressly allowed E10 to exceed the general RVP limits by 1
psi. In EPA's 1987 rulemaking, EPA established a minimum ethanol content 'to assure that fuel
containing only trace amounts of alcohol does not qualify' for the waiver. Furthermore, EPA's 1
psi waiver prior to the 1991 amendments expressly applied to fuels containing 'at least 9%
ethanol.'
Additionally, EPA's current approach is inconsistent with legislative history in enacting
21 l(h)(4). In particular, Section 216 of House Bill 3030 (1990), which eventually merged with
Senate Bill 1630 to amend the CAA, made clear the 1 psi waiver was applicable to 'gasoline
containing at least 10 percent ethanol.'
Additionally, EPA's current approach bars the use of an environmentally preferable fuel (El5),
because its improved RVP still merits use of the 1 psi waiver, which is illogical.
Furthermore, EPA glosses over the fact that 21 l(h)(5) was enacted as part of the Energy Policy
Act of 2005 whereby Congress set RFS targets that would mandate the greater use of ethanol
blends. EPA's barring the extension of the 1 psi waiver (1) conflicts with precedent, and (2)
frustrates the flexibility to use different ethanol blends to meet RFS targets enacted in the same
4-323
-------
Chapter 4: Proposed Vehicle Emissions Program
EPAct 2005 legislation, and should be reversed. And EPA can do so by simply changing its
statutory interpretation.
Additionally, subsequent to EPA's 2011 misguided interpretation, blend wall concerns that have
arisen placing a premium on EPA allowing the increased distribution of El 5 into commerce.
Quite simply, EPA should change its interpretation and allow extension of the 1 psi waiver to
blends that contain at least 10% ethanol, as the statute allows. By doing so, EPA would promote
the distribution of E15 and MLEBs and help to 'ensure' that the mandated RFS volumes are met.
Organization: Renewable Fuels Association (RFA)
RFA believes EPA should provide equal RVP treatment for E10 and E15. EPA should be
consistent in its treatment of RVP requirements for all ethanol blends up to 15% by volume.
EPA's initial decision to grant the 1 psi waiver to E10 blends was based on two fundamental
findings: 1) that supplies of low-RVP gasoline blendstock for El0 blending were insufficient,
and 2) that the increased volatility associated with the 1 psi waiver was more than offset by
reduced carbon monoxide and exhaust hydrocarbon emissions from E10. Recent analyses have
shown that the vapor pressure of El 5 is lower than it is for E10. Further, there is evidence that
El5 provides greater reductions in carbon monoxide and exhaust hydrocarbon emissions than
E10. Currently there is insufficient low-RVP gasoline blendstock to accommodate E15 blending
(i.e., without a 1 psi waiver). Thus, the same two findings that led EPA to issue the 1 psi waiver
for E10 also apply to E15. As such, if the 1 psi waiver continues to apply to E10, there is no
logical reason that it should not also be applied to E15.
On the other hand EPA should consider eliminating the 1 psi waiver for E10 provided that RVP
caps are administered consistently across all ethanol blend levels. That is, if EPA decides to
discontinue the 1 psi waiver for E10, it should do so immediately so that E10 and E15 are being
treated consistently in the marketplace with regard to RVP. This is a critical issue for the near
term for El 5 to expand in the marketplace. Discrepancies in the treatment of RVP limits for E10
and El 5 are impeding the introduction of El 5. Consistent treatment of RVP requirements for
ethanol blends up to El 5 will reduce the potential for more "boutique fuels" and maximize
flexibility for refiners and gasoline marketers.
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
A national minimum octane number of 87 AKI without regard to altitude is needed as a Federal
Tier 3 standard for regular gasoline
A uniform, nationwide minimum octane grade of 87 AKI (Antiknock Index) is needed to support
MY 2017 vehicle performance, fuel efficiency, and emission reductions. 85 AKI octane grade is
no longer relevant for high altitude use, and lower octane can cause newer engines to perform
poorly. For some time, all OEMS have designed and engineered vehicles to run on 87 AKI, and
all recommend use of minimum 87 AKI for regular gasoline (without regard to altitude) in
vehicle owner's manuals.
4-324
-------
Tier 3 Summary and Analysis of Comments
Federal action is specifically needed to avoid a patchwork of State octane standards allowing for
use of 85 octane grade gasoline, even in some locations that don't qualify as high altitude, such
as South Dakota where access to >85 AKI gasoline in the western part of the State became
limited and problematic. EPA should prevent backsliding of octane levels in refinery BOB
(blend stock for oxygenate blending) because it leads to sub-grade octane product, and in some
areas also stimulates interest in use of metallic additives to boost octane, which are opposed by
automakers (see below).
The AKI (octane number of gasoline, also referred to as the octane grade or rating, is a measure
of its antiknock performance, or resistance to engine knock (often called spark knock because it
is affected by the spark timing of the engine, among other factors). It is based on an index taken
from an average of two different test outcomes, the RON (Research Octane Number) and the
MON (Motor Octane Number). The higher the AKI octane number, the greater the fuel's
resistance is to knock. Spark knock is an abnormal combustion event that creates an acoustic
wave in the combustion chamber, causing a pinging sound. If the knock is severe enough, it can
damage the engine.
It is widely documented that a gasoline's octane number rating and a vehicle's performance are
directly linked.
The U.S. General Accounting Office (GAO) reported on the importance of consumers
understanding the proper gasoline octane rating number to select for their vehicles. The GAO
stated: "Consumers need to buy gasoline with an octane rating that matches their engines' octane
requirements. Buying gasoline with too little octane can cause engine knock, which can damage
an engine, lower engine efficiency, reduce mileage, and increase emissions'.
Since the mid-1980s, U.S. vehicles have included computer controls to monitor combustion
performance and mitigate auto-ignition (knock) by way of three mechanisms: retarding or
delaying ignition timing, fuel enrichment to cool the in-cylinder end-gas environment, and, in the
case of pressurized air induction equipped vehicles, reducing the level of boost or super-charging
in order to rapidly and momentarily reduce engine output. These mechanisms protect the engine
from acute and potentially severe damage. However, they result in sub-optimal combustion,
lower engine efficiency, and in the case of ignition timing delay, result in higher exhaust
temperatures. Although the sophistication and speed of modern vehicle computer controls has
increased greatly since the 1980s, these three primary mechanisms of knock control are still
widely used.
As vehicle manufacturers strive for increasingly efficient designs, the vehicle hardware and
calibrations have been significantly upgraded, putting modern vehicles closer than ever to spark
knock limited operation. For example, as a result of increased number of transmission gears,
more frequent operation at lower engine speeds for a given vehicle speed puts the engine at a risk
of low speed spark knock. Additionally, exhaust emission reduction catalysts are increasingly
"close coupled" to the exhaust manifold to enable their early light-off and high oxidation
efficiency; however, this makes them more susceptible to temperature spikes (thermal stresses)
from auto-ignition events. And because vehicle sizes and therefore their masses remain relatively
high while engine displacements are quickly dropping, the load factor under which an engine
4-325
-------
Chapter 4: Proposed Vehicle Emissions Program
operates is rapidly increasing. In total, today's vehicles and their computer controls are
constantly monitoring and optimizing to the available gasoline octane number in the vehicle's
fuel tank.
The continued presence of 85 and 86 AKI octane grade gasoline, predominantly in the U.S.
Rocky Mountain States, is limiting vehicle efficiency and performance.
U.S. automobiles have been designed for 87 AKI gasoline since the mid-1970's, when the
octane-enhancing organometallic additive tetra-ethyl lead was phased out. As a result of lead
removal, the predominant gasoline octane number of Regular Unleaded was reduced from 917
92 AKI to 87 AKI and engine compression ratios were reduced from 9.0 - 10.5: 1 to 7.5 - 8.5:1.
By contrast, a review of 2012 Model U.S. Car and Light Truck Specifications and Prices from
the Wards Auto group shows compression ratios have again risen substantially since the 1970's
to 10.5 - 11.5:1 for naturally aspirated engines, while the predominant U.S. Regular Unleaded
gasoline octane rating remains at 87 AKI. [
Recent Vehicle Performance Testing using 85 AKI Gasoline [
NOTE: The information in this section of the comments is from an unpublished Alliance study;
it is Alliance work product and protected under copyright. [
In early 2013, the Alliance completed the test portion of a study to measure the magnitude of
performance effects of operating vehicles on lower octane rating gasoline than for which they
were designed. For this study, two 2012 production vehicles were tested in the General Motors
Milford Proving Grounds "variable altitude" emissions facilities during January and February,
2013. The two vehicles are Regular Unleaded (87 AKI) calibrated vehicles and the tests were
conducted using standard chassis dyno emissions test cycles at Sea Level (0 Ft). The emissions
test cycles were chosen to be moderate to heavy customer driving, the LA92 and US06 cycles,
respectively. During the emissions testing procedures, the vehicle's speed was maintained within
a precise window of operation by the trained driver by opening and closing the throttle (part
throttle).
The two octane ratings for the test fuels were 84.8 AKI and 87.8 AKI as reported on the fuel
supplier's Certificate of Analysis, and all other fuel composition properties were held very close
to each other. Ideally, the AKI rating of the "87 AKI" fuel would have been closer to 87.0, but
based on the very large differences in vehicle performance measured (as fuel economy and
emissions), it is expected that this would have made only a marginal change in results.
The vehicle performance data shown below [Table 4 can be found on p. 84 of Docket number
EPA-HQ-OAR-2011-0135-4461-A1] is the percent difference in each parameter from operating
the vehicles on 87 AKI gasoline compared to operating the vehicle on 85 AKI gasoline. The
results for the two repeat US06 cycles were combined and then compared in an "A-B" (87 - 85)
format. A vehicle data logger recorded the Engine Control Module (ECM) serial port
information and no test drift is anticipated from these signals. While operating on the proper (i.e.,
owner's manual specified) 87 AKI gasoline, the vehicles utilized less throttle, required less load
on the engine, and less fuel enrichment, and where applicable, less boost pressure to maintain the
4-326
-------
Tier 3 Summary and Analysis of Comments
vehicle speed traces of the standardized emissions test cycles. [EPA-HQ-O AR-2011-013 5-4461-
Al, pp. 83-84]
The vehicle emission and fuel economy effects can be seen below [Table 5 can be found on p. 85
of Docket number EPA-HQ-O AR-2011-013 5-4461-A1] and were analyzed in an "A-B-A"
format (85-87-85) to account for any test drift from the catalyst response. Substantial
improvements in fuel economy and emissions performance can be seen while operating on the
proper 87 AKI gasoline during these controlled chassis dyno tests.
Sales of 85 AKI gasoline represent less than 4 % by volume of the total U.S. gasoline
production, but they have an impact on 100% of the vehicle fleet.
A recent study by the Coordinating Research Council (CRC) Performance Committee Octane
Group has determined that it requires about 4-5 RON units to allow an increase of one (1)
number in an engine's compression ratio. Auto manufacturers must protect for the lowest octane
in the market, because they cannot know in which U.S. State their vehicles will be sold and
operated, so the presence of an 85 AKI gasoline represents about 0.4 - 0.5 compression ratio of
inefficiency yet to be utilized. This is not to say that powertrain compression ratios would rise by
0.4 -0.5 units if the U.S. had a national 87 AKI minimum limit, but rather there is an equivalent
amount of efficiency gain still unrealized either in terms of higher compression ratios, less fuel
enrichment, more optimal combustion phasing, or higher boost. Additionally, if 87 AKI was the
national minimum market fuel standard, vehicles already on the road would operate a greater
percentage of time without fuel enrichment.
Prompt federal action is needed under Tier 3, to avoid a patchwork of State octane standards and
proliferation of use of 85 AKI gasoline.
Currently, the most widely referenced gasoline specification in the U.S. is ASTM's D4814,
"Standard Specification for Automotive Spark-Ignition Engine Fuel." ASTM does not specify
mandatory octane minimum limits. Determination of the proper octane number to be sold in the
marketplace is left to the discretion of the fuel marketers. Likewise, the NIST Handbook 130
(Laws and Regulations) does specify the proper octane number naming convention at retail, but
does not specify the actual octane number minimums. Because of this uncertainty, some States
have adopted D4814, some have adopted Handbook 130, and some have written their own
specification. For example, South Dakota currently has a "hybridized" octane standard.
Historically, 87 AKI has been sold throughout the U.S. as Regular Unleaded except in Rocky
Mountain States where a mix of 85 and 86 AKI is sold. Also, recently it has been reported that in
certain regions of the country with an excess of hydrocarbons resulting from new fracking
techniques, it has become profitable to truck in 85 and 86 AKI gasoline to lower elevation
regions of the country. This is going backwards in terms of emissions, engine efficiency, and real
fuel economy, for highly localized refinery and retail savings.
Furthermore, in the event that EPA does not set a national minimum 87 AKI octane market
gasoline standard, gasoline with octane lower than that of federal emission test fuel is not
"substantially similar" to the certification test fuel. By closely regulating the octane of emission
test fuel, EPA has signaled that this is a critical fuel parameter. Under CAA §21 l(f) all gasoline
4-327
-------
Chapter 4: Proposed Vehicle Emissions Program
must be "substantially similar" to fuels used for certification. Such fuel should not be allowed to
remain in the market place since new vehicles will be certified to a higher octane fuel (unless
fuel producers can demonstrate such fuels will not cause or contribute to the failure of any
vehicle/engine to meet emission standards, subject to notice and comment).
In support of Tier 3 and other vehicle needs, EPA should promptly set a federal minimum
standard mandating 87 AKI as the minimum octane grade for regular gasoline, without regard to
altitude.
Organization: Ford Motor Company (Ford)
Elimination of Sub-Regular (<87 AKI) Octane Grades: As stated above, the use of fuel with
higher octane ratings will allow improved performance (increased efficiency) for newer vehicles
with certain advanced technologies. However, the use of fuel with an antiknock or octane rating
lower than the required amount may result in vehicle performance loss or engine damage.
Engines equipped with knock sensors will compensate for the reduced octane rating by retarding
spark timing, but this can result in increased fuel consumption and reduced power and may not
completely prevent the engine from knocking. Although altitude and weather have been shown
to affect a vehicle's antiknock requirement for most pre-1984 vehicles, modern vehicles in
today's fleet require the same antiknock index as stated in their owner's manual, regardless of
ambient temperature or altitude. Despite this requirement, several western states with higher
altitude areas in the U.S. continue to offer 85 AKI instead of 87 AKI for their "regular grade"
market fuels, with the potential for expanded availability. This availability and any further spread
of sub-regular grade market octane rating fuels amounts to "back-sliding" in market fuel quality
and subsequently emissions inventory. This is in direct conflict with the agenda outlined in both
the Tier 3 rules and the ONP rules, which will require increased fuel economy in tandem with
decreased CO2 emissions. In order to meet the increasingly stringent requirements of the ONP,
manufacturers will continue to implement a host of new technologies (cylinder deactivation,
hybridization, continuously variable transmission, 9-speed transmissions, Atkinson engines,
downsized turbocharged direct injection engines) that will require a commensurate increase in
market fuel quality and market fuel octane ratings. EPA must recognize that gasoline octane has
become a critical factor in the development and sale of vehicles with newer advanced
technologies, and this issue needs to be addressed on a nationwide basis. As noted above, the
typical octane rating of U.S. marketplace fuel needs to increase, and the sale of gasoline with
sub-regular octane (less than 87AKI) needs to be eliminated nationwide.
Recommendation: Ford supports the Alliance comments calling for a national minimum octane
number of 87 AKI for regular grade gasoline, without regard to altitude.
Organization: Mercedes-Benz USA, LLC on behalf of Daimler AG
The primary factors which directly impact the Mercedes-Benz GHG Compliance Plan are sulfur
content in market fuel, and ethanol content and octane level in certification and market fuel.
Organization: International Council on Clean Transportation (ICCT)
4-328
-------
Tier 3 Summary and Analysis of Comments
What is really needed is higher octane for regular fuel, regardless of the ethanol content. For
example, the Mazda had to reduce the compression ratio of the Skyactiv engine for the US
market, compared with Europe. Also OEMs simply won't try to sell a mass-market non-luxury
vehicle that requires mid- or high-grade gasoline in the US.
Unfortunately, when fuel providers added ethanol to gasoline in the past, the fuel providers took
out the non-ethanol octane elements - so that, in practice, there has never been a gain in octane
through increased ethanol blends. It is likely that El 5 will be the same—we won't actually get
more octane. Rather than focus on ethanol requirements, the EPA should raise the minimum
octane requirements for all gasoline.
Organization: General Motors LLC (GM)
Gasoline Octane is Important for Emissions - Certification and Market Fuels:
Octane rating of both marketplace gasoline and Tier 3 certification fuels is critical to vehicle
performance, including their exhaust emissions. For this reason, GM specifies the octane
requirement of every vehicle in the vehicle owner's manual. GM fully supports the
Alliance/Global comments on octane.
Organization: National Marine Manufacturers Association (NMMA)
For the past three years, NMMA has been working with the US Department of Energy and
Argonne National Laboratory to evaluate a better alternative to ethanol, both as an oxygenate
and a biofuel. Our research and testing of isobutanol is a result of the problems associated with
increased levels of ethanol being used as a fuel additive. Our research findings include:
- Ethanol raises the Reid Vapor Pressure of gasoline, which at higher levels increases
evaporative emissions. Meeting US EPA and CARB RVP standards for ethanol blends therefore
requires refineries to perform additional processing.
- Ethanol is corrosive and is typically shipped via railcar or truck versus less expensive pipelines.
- Ethanol is splash blended at the end of the gasoline refining process, adding additional storage
and labor cost.
- Ethanol is hygroscopic, meaning it has an affinity for water.
- Ethanol at 10% by volume contains approximately 2.5% oxygen. Increasing ethanol content in
gasoline increases the oxygen level (referred to as enleanment) causing open loop engines to
experience increased combustion temperatures.
Rather than simply oppose all ethanol and biofuels, NMMA has dedicated considerable
resources over the past three years to evaluate isobutanol in order to contribute meaningful data
to steer future biofuels policy and ensure compatibility with millions of non-road engines and
boats.
Recreational marine engines and vessels represent a worst-case environment in which to validate
the reliability and compatibility of advanced biofuels such as isobutanol due in part to the usage
cycles of marine products, open-vented fuel systems, likelihood of introducing saltwater into fuel
systems and open-loop engine operation. Successful demonstration of isobutanol-extended
gasoline in the marine environment may ensure that engine emissions, engine and vessel
4-329
-------
Chapter 4: Proposed Vehicle Emissions Program
performance, engine durability and product safety (fuel system related) could be maintained for
millions of small engines including recreational marine engines and vessels. Moreover,
successful testing in a worst-case marine environment will allow for a more streamlined
acceptance of advanced biofuels outside of on-road vehicles, potentially minimizing the effects
of the current bifurcated fuel system.
NMMA has conducted tests on a variety of marine engines and vessels using 16% isobutanol by
volume, which has similar oxygen content to E10, without the other negative properties of
ethanol identified above. The results of our documented and published research thus far indicate
that isobutanol at 16% by volume yields very similar engine emissions, durability, power and
performance as E10.
This summer, NMMA is conducting tests with Argonne National Laboratory under the direction
of the U.S. Department of Energy on a tri-fuel blend of 5% ethanol, 8% isobutanol and 87%
gasoline to determine the effects on emissions, performance and overall fuel compatibility with
recreational marine engines and vessels. Our preliminary analysis has shown that isobutanol acts
to lower Reid Vapor Pressure, while also increasing the overall biofuel quantity in gasoline.
NMMA strongly urges EPA and DOE to take a leadership role in appealing to Congress and the
President to freeze ethanol at 10%, while neutral party technical evaluation are conducted to
assess alternatives that can be introduced at a higher concentration without the negative effects of
ethanol.
Commenter: Refinery Automation Institute, LLC
The EPA proposal lacks transparency and seems politically motivated, in that uses words to
describe 'implied' changes in Aromatics, Olefins, Vapor Pressure, and Distillation temperatures.
This promotes 'back-door' regulations by 'selling' and unambiguously implying 'harmonization'
with California's CARB RFG3 specs (Attachment II). The impact on costs is misleading in
EPA's Tier 3 proposal.
Our Response:
Some commenters suggested that the 1-psi RVP waiver be extended to all ethanol blend
levels, or be eliminated altogether, either of which would create a more level playing field for
higher ethanol blend s in the market. Some commenters expressed support for raising the
minimum octane of in-use gasoline, or for limiting ethanol content and/or encouraging the use of
other oxygenates (e.g., isobutanol). EPA is using its Clean Air Act section 206 authority to
establish requirements for emission certification testing, including specifications of the emissions
test fuels with which vehicles demonstrate compliance with emissions standards, in order to
better reflect the ethanol content and other properties of in-use fuels. While we are taking action
to limit the sulfur content of in-use gasoline under authority in Clean Air Act sections 21 l(c)(l)
and (2), we did not propose other changes to in-use fuels. Any action related to other changes to
in-use fuels is outside the scope of this rulemaking.
4-330
-------
Tier 3 Summary and Analysis of Comments
Some commenters supported revoking the 1-psi waiver on the basis that large reductions
in evaporative emissions would be expected in the in-use fleet, and that EPA should use
authority given under Clean Air Act section 21 l(v) to accomplish this. The requirements of that
section include completion of the anti-backsliding study, which is still underway. Once the study
finalized, EPA will assess what, if any, regulatory action is warranted.
One commenter expressed concern regarding implied changes in fuel properties and
harmonization with California. The commenter appears to be confusing EPA's actions to adjust
the properties of certification test fuel with potential changes to in-use fuel. The Tier 3
rulemaking is only requiring changes to the sulfur level of in-use gasoline.
4.5.2. Proposed Flexible Fuel Vehicle Test Fuel
What Commenters Said:
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
Automakers agree with EPA that it is important that the fuel used to test FFVs reflect the
composition of actual in-use E85 (30). ASTM has revised specification standard, ASTM D5798-
The most significant driver for the recent revisions to the ASTM E51-83 specifications has been
the inability of terminals to blend E85 that met the requirements of the E85 specification using
current gasoline and BOBs (Blend-stocks for Oxygenate Blending). After significant study, the
ASTM task group decreased the minimum allowable ethanol percentage to 51% as the best way
to enable E51-83 blends with compliant RVPs (Reid Vapor Pressure).
There is currently little data available on actual E51-83 fuel composition in the field. Previous
studies on E68-83 showed that much of the field fuel did not meet the ASTM requirements.
These results drove the re-evaluation and revision of the specification mentioned above. Later
this year, the National Renewable Energy Lab, in conjunction with the Coordinating Research
Council, will survey field E51-83 quality. This is the first survey since the revisions were
implemented to ASTM specification D-5798: Ethanol Fuel Blends for Flexible-Fuel Automotive
Spark Ignition Engines. EPA should wait to finalize specifications until it can take into
consideration the results of the NREL-CRC survey when they become available.
EPA should implement our recommended specifications for E85, CNG and LPG detailed above.
Organization: American Lung Association
For Flexible Fueled Vehicles (FFV) test fuel, we support an approach that includes the standard
certification fuel, E10 until such time as EPA revises to E15 if market conditions warrant, as
discussed above, with additional denatured fuel ethanol to meet the 80-83 volume percent.
Organization: American Petroleum Institute (API) and the Association of Fuel &
Petrochemical Manufacturers (AFPM)
4-331
-------
Chapter 4: Proposed Vehicle Emissions Program
API and AFPM support EPA's proposed approach for FFV test fuel. With regards to the
Flexible Fuel Vehicle Test Fuel, we agree with EPA's proposal to make the fuel by blending the
gasoline emission test fuel base stock with higher levels of ethanol to produce the test fuel and
trim the resulting fuel with normal butane to achieve the necessary vapor pressure.
Organization: General Motors LLC (GM)
GM fully supports the Alliance/Global comments on both E85 and CNG/LPG certification fuels.
Organization: Growth Energy
Growth Energy Supports a Workable Flex Fuel Vehicle (FFV) Certification Fuel: EPA also
proposed a specific "E85" test fuel with an ethanol concentration of E80-E83 that would be
made by blending ethanol (either denatured fuel ethanol or neat ethanol) with the El5 test fuel.
Growth Energy continues to believe that Flexible Fuel vehicles will be essential to use the higher
ethanol blends needed to meet the future volumes of the RFS. As such, it only makes sense to
develop a specific test fuel for FFVs. However, we do not think that the specifications are
representative of E85 found in marketplace, especially since ASTM has published its "Flexible
Fuel" specification.
Growth Energy would suggest that the RVP of the FFV certification fuel should be at 9 psi and
the ethanol content should be somewhere in the range of 70-78 percent (E70 - E78).
Organization: Marathon Petroleum Company LP (MFC)
With regards to the Flexible Fuel Vehicle Test Fuel, we agree with EPA's proposal to make the
fuel by blending the gasoline emission test fuel base stock with higher levels of ethanol to
produce the test fuel and trim the resulting fuel with normal butane to achieve the necessary
vapor pressure.
Organization: National Association of Convenience Stores (NACS) and Society of
Independent Gasoline Marketers of America (SIGMA)
The Proposal would make E80-83 the default test fuel for FFVs. While it necessary to develop
regulatory specifications for E85 test fuel blends, the Proposed Rule does not account for the fact
that certain jurisdictions contain caps on ethanol content in E85 below 80 percent. The final rule
should accommodate fuels that will actually be in the market.
NACS and SIGMA applaud the Agency for developing regulatory specifications for E85 test
fuel blends. The lack of specifications up to this point has caused confusion and inconsistency in
the market. There is concern, however, that the proposed specifications of 80-83 volume percent
does not accommodate fuels that will actually be in the market. Indeed, ASTM specifications for
E85 include a fuel containing between 51-83 volume percent ethanol5. In addition, different
jurisdictions in the country contain various restrictions on ethanol content for E85 fuels. Some of
4-332
-------
Tier 3 Summary and Analysis of Comments
these jurisdictions contain caps below 80 volume percent ethanol. As a practical matter, this
means the Agency is proposing default test fuel that simply will not be utilized in these areas.
It is important that the final rule accommodates fuels that will actually be in the market. The
revised ASTM definition for E85 enhances marketers' ability to provide consumers with a fuel
product that might be offered at a more attractive price and provide greater fuel efficiency than is
currently being experienced with blends containing 80-83 volume percent. This flexibility
should be encouraged. NACS and SIGMA would be happy to work with the Agency to develop
a policy for accomplishing this objective in the final rule.
Organization: National Corn Growers Association (NCGA)
And, while we are generally supportive of establishing new certification fuel requirements for
FFVs, we are concerned that the proposed approach would inhibit future development of FFVs
and the development of dedicated vehicles designed for the exclusive use of an high octane
MLEB. Both FFVs and high octane utilizing vehicles are critical to ensuring that the 150 million
metric tons of GHG emission reductions from the RFS volumes are attained.
We support the development of specifications for FFV certification test fuel. As the RFS
requirements ramp up beginning in 2015 and later, we believe the incentive will increase to
produce more ethanol-containing fuels, and that E85 use in FFVs is likely to increase. Now that
ASTM has published and refined its ASTM 5798 "Flex Fuel" specification, it is reasonable to
provide a more complete and comprehensive specification for the FFV certification test fuel.
EPA proposed the E85 ethanol test fuel concentration be limited to E80-E83 and that this could
be achieved by blending ethanol into El 5 test fuel utilizing denatured fuel ethanol or neat
ethanol. EPA also indicated that the vapor pressure of the test fuel should be between 6.0 and 6.5
psi. However, we are concerned that EPA's specification of E85, at E80-E83 with a vapor
pressure of 6.0 to 6.5 psi, could inhibit the development of FFVs meeting the lowest
NMOG+NOx standard bins of the Tier 3 proposed rule.
We recommend that EPA consider the unique challenges and additional costs of designing FFVs
to comply with Tier 3 Bin 20, 30 and 50 standards, and not create FFV test fuel specifications
that unnecessarily increase the stringency of the emission standards and discourage production of
FFVs. We are aware that the U.S. Department of Energy National Renewable Energy Laboratory
and the Coordinating Research Council are beginning another field survey of E85 to compare the
fuel quality to the most recent ASTM D5798-13 specification (16). We recommend that EPA use
information from this survey to select specifications for FFV test fuel from the data rather than
selecting a specification such as E80-83.
16 - CRC Project no. E-85-3, "E-85/E-85-2/E-85-3 National Surveys of E85/Flex Fuel Quality", project
description available at http://www.crcao.org/news/Mid%20Level%20Ethanol%20program/E-85-
3%20Summary.pdf.
Organization: New York State Department of Environmental Conservation
4-333
-------
Chapter 4: Proposed Vehicle Emissions Program
Higher level ethanol blend fuels for use in Flexible-Fuel Vehicles need to be subject to sufficient
regulation to ensure proper function of vehicle emission control systems.
Current regulation of fuels (other than gasoline) for use in Flexible-Fuel Vehicles (FFVs) is
inadequate. As EPA staff note in a memo to the docket for this rulemaking (2) the commercial
fuel must be substantially similar to a heretofore unspecified certification fuel...
We agree with EPA's decision to promulgate standards for FFV test fuel, and believe that such
standards are overdue.
2 - Memo from Jeff Herzog, Assessment and Standards Division, to Docket EPA-HQ-OAR-2011-0135,
dated April 8, 2013 (docket item number EPA-HQ-OAR-0135-0529).
Organization: POET, LLC
Similarly important is the increased use in FFVs of fuel blends that contain from 51-85% ethanol
(referred to in these comments as high-level ethanol blends or HLEBs).
Regarding an HLEB certification fuel, EPA proposes that the certification fuel ethanol content
must be from 80 to 83 volume percent.' (62) Under Clean Air Act Section 206, test fuels should
reflect 'actual current driving conditions under which motor vehicles are used.' EPA should base
an FILEB certification test fuel that is more likely to be representative of HLEBs on which FFVs
are likely to run. (63) POET suggests that an appropriate FFV/HLEB certification fuel is likely to
have a % ethanol content in the 70s.
Otherwise, POET supports the Proposed Rule discussion of HLEB certification fuel logistics
with regard to how the test fuel is prepared for a particular blend level. In particular, EPA notes
that 'Rather than specify ranges for the other fuel parameters as we have done for gasoline test
fuel... we are proposing that the FFV test fuel would be defined based on the results from
blending the proposed El5 standard gasoline test fuel with ethanol.' (64) The Proposed Rule
would also allow the use of butane to trim RVP. In the alternative, BOB and undenatured ethanol
could be used to prepare the test fuel. (65) POET approves of this approach.
62-Id. at29,912.
63 - By comparison, 40 CFR 86.113-94(d) provides specifications for methanol fuels for flexible fuel
vehicles and states 'Mixtures of petroleum and methanol fuels used for exhaust and evaporative emission
testing and service accumulation for flexible fuel vehicles shall consist of the appropriate petroleum fuels
... and a methanol fuel representative of the fuel expected to be found in use ... and shall be within the
range of fuel mixtures for which the vehicle was designed.' (emphasis added). By the same logic, the
ethanol fuel family could have two certification fuels, and those certification fuels need only be
representative of the range of ethanol blends 'within a range' of mixtures 'expected to be found in use.'
64-Id. at29,912.
65 - Id.
Organization: Renewable Fuels Association (RFA)
EPA's proposed certification fuel for flexible fuel vehicles may discourage FFV production due
to potential difficulty in meeting NMOG+NOx standard bin levels.
4-334
-------
Tier 3 Summary and Analysis of Comments
Our Response:
As discussed in Section IV.F.3 of the preamble to the final rule, EPA is finalizing
specifications for the fuel used in flexible fuel vehicles (FFV) emissions testing including
certification testing (see Section IV.F.3. of the preamble to the Tier 3 final rule). Establishing
these specifications for FFV test fuel will resolve confusion and inconsistency among FFV
manufacturers in carrying out their certification and other testing requirements and help to ensure
that FFV emissions are appropriately controlled over the range of in-use fuels.
The base fuel stock used to formulate FFV emissions test fuel must comply with the
specifications for the standard E10 emissions test fuel as described in Sections IV.F. 1 and IV.F.2
of the preamble to the Tier 3 final rule. This practice avoids the need to specify the ranges for a
number of fuel parameters as we have done for gasoline test fuel and helps to minimize the
number of test fuels that a vehicle manufacturer must store. Denatured fuel ethanol (DFE) that
meets the specifications in the Tier 3 final rule must be blended into this base fuel stock to attain
an ethanol content of 80 to 83 volume percent in the finished test fuel. Commercial grade
normal butane can be added as a volatility trimmer to meet a 6.0 to 6.5 psi RVP specification for
the finished test fuel.34
As an alternative to the use of DFE to formulate FFV test fuel, neat (undenatured) fuel
grade ethanol can be used. As an alternative to using a finished E10 standard gasoline test fuel
to produce FFV test fuel, the gasoline blendstock used by the fuel provider to produce a
compliant E10 test fuel can also be used to produce the FFV test fuel. This would allow ethanol
to be blended only once to produce FFV test fuel. In such cases, a sample of the subject gasoline
blendstock must be tested after the addition of ethanol to produce a finished standard E10
gasoline test fuel to demonstrate that the blend meets all of the requirements for standard
gasoline test fuel.
Substantial publicly available literature exists to demonstrate that the ethanol content of
fuel used in FFVs has a significant effect on vehicle emissions. The effect of ethanol content on
FFV emissions becomes more pronounced with increasing ethanol concentration. The current
ASTM specification for E85 provides that the ethanol content of E85 may vary from 51 to 83
volume percent depending on climactic conditions.35 Consistent with our long standing policy
regarding the emissions testing of FFVs, FFVs must comply with all applicable emissions
control requirements for each consumed fuel or blend of fuel. This policy ensures vehicles are
designed and calibrated for emissions performance across the full range of potential in-use fuel
formulations. To ensure that FFV certification testing adequately accounts for in-use emissions
performance, we are finalizing the ethanol content of FFV emissions test fuel at 81-83 volume
percent as proposed. Exhaust emissions testing conducted using a fuel containing 81-83 volume
percent ethanol will provide results that represent the effect of ethanol on FFV emissions
performance when this effect is most pronounced. The complimentary emissions certification
testing for FFVs on E10 will ensure that the effect on FFV emissions from the full range of
potential in-use ethanol concentrations is represented. Although certain local jurisdictions may
34 The specifications for commercial grade butane are contained in 40 CFR 80.82.
35 ASTM International D5798-13, "Standard Specification for Ethanol Fuel Blends for Flexible-Fuel Automotive
Spark-Ignition Engines".
4-335
-------
Chapter 4: Proposed Vehicle Emissions Program
limit the ethanol content of FFV fuel to below the maximum 83 volume prevent set by ASTM,
EPA must set test fuel specifications to ensure adequate emissions control performance
throughout the U.S., including areas that do not have such additional limitations. Given the need
to ensure that FFV emissions certification testing is representative of the full range of potential
in-use ethanol blends, it would be inappropriate to set the required ethanol concentration for FFV
emissions test fuel based on typical in-use levels as suggested by some of the commenters.
Similarly, we are setting the RVP of FFV emissions test fuel to assure emissions
performance over the range of in-use fuels. When ethanol and gasoline are blended to produce
high level ethanol blends, the RVP can be and often is very low. As a result, ASTM included a
minimum RVP for E51-83 of 5.5 psi. Given that low volatility fuels can make the control of
cold start emissions more challenging, we are finalizing the RVP of FFV emissions test to be
near the minimum RVP that will be encountered in-use. The 6.0 to 6.5 RVP specification will
help to ensure that FFVs are designed and calibrated to maintain their exhaust emissions
performance across the range of in-use fuels. A 9 RVP specification of FFV emissions test fuel,
as suggested by some commenters, would not provide assurance of exhaust emissions
performance with lower RVP fuels that are present in-use. We are also finalizing separate RVP
specifications for the test fuel used for FFV evaporative and refueling emissions testing as
discussed in Sections IV.C. and IV.F. of the Tier 3 final rule preamble.
The levels of other fuel parameters for FFV test fuel are determined by the levels of these
parameters present in the gasoline blendstock used as diluted by the addition of ethanol.36
Therefore, we believe that requiring that the levels of these other fuel parameters present in FFV
emissions test fuel be determined by the dilution of the levels present in gasoline emissions test
fuel appropriately reflects their potential effect on emissions performance. Given the
considerations discussed above in determining the FFV emissions test fuel specifications, we do
not believe that there would be a substantial benefit in waiting for the completion of the E51-83
fuel quality survey currently underway to finalize FFV test fuel specifications.
As discussed in Section V.H. of the Tier 3 final rule, although we sought comment on in-
use standards for E51-83 fuel for use in FFVs, we are not finalizing such standards at this time.
To the extent future implementation of in-use fuel standards for E51-83 might impact in-use
E51-83 quality, we may revisit the need to amend the FFV test fuel specifications finalized in the
Tier 3 FRM.
With respect to the comment on the development of vehicles that are designed to operate
on high octane fuels, mid-level ethanol blend fuels, EPA allows manufacturers to request
approval certify on alternative certification fuels. Responses to comments on this topic are
discussed in more detail in Chapter 4.5.1.5 of this Summary and Analysis of Comments
document.
36 Additional ethanol may either be added to the E10 emissions test fuel or to the gasoline blendstock used to
formulate a compliant E10 emissions test fuel to meet the 80-83 volume percent ethanol specification that we are
finalizing for FFV emissions test fuel.
4-336
-------
Tier 3 Summary and Analysis of Comments
4.5.3. Implementation Schedule
What Commenters Said:
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
EPA/California Mutual Reciprocity Should be Made Permanent - to Assure The Automaker
Option to Certify on Either the LEV III or the Tier 3 Test Fuel/Test Regime. EPA and California
each should extend permanent reciprocity (mutual acceptance) of either LEV III test fuel/testing
or Tier 3 test fuel/testing for purposes of both LEV III and Tier 3 certifications. EPA should not
require use of Federal test fuel/test procedures after MY 2019 (28) for Tier 3 certification, as
proposed. This is critical to facilitate OEM use of one fuel/one testing regime per vehicle model,
a longstanding goal for harmonization between LEV III and Tier 3 standards toward One
National Standard. The purpose of the arbitrary date cutoff in the proposal is not clear.
For in-use compliance purposes, the same fuel (LEV III or Tier 3) selected for certification
should also be used for in-use testing.
Broadening future Federal Test Fuel options for OEMs should not negate reciprocity with
California, or the option to use either LEV III test fuel/test regime or Tier 3 test fuel/tests for
both California and Federal certification.
(If no permanent reciprocity) EPA should accept carry-over certifications on LEV III
certification gasoline beyond MY 2019
ARB cannot amend the LEV III regulations until after EPA finalizes Tier 3. Thus, there will be a
significant period (at least a year or two) of uncertainty where each agency has different
certification gasoline requirements.
The proposed Tier 3 regulations allow manufacturers to use LEV III certification gasoline only
through the 2019 model year (MY). Beginning in the 2020 MY, all vehicles must be certified
using Tier 3 certification gasoline and carryover of emissions data is not allowed. (See §86.113-
07, Fuel Specifications). If there is no permanent reciprocity, despite our strong support of that,
then EPA should allow carryover of emissions data after 2019 MY. Manufacturers tend to
carryover emissions data for several years.
Without an extension, this could result in manufacturers retesting and recertifying vehicles for no
other reason than to change the certification gasoline. For example, a 2016 MY vehicle certified
to LEV III standards on LEV III gasoline might be required to recertify on Tier 3 gasoline, rather
than carry over the LEV III certification data through the 2020 or even 2021 MY until a redesign
requires a new certification. This would be an unnecessary waste of resources and we do not
believe this is EPA's intent.
Consequently, we recommend that EPA allow carryover of certification data using LEV III
gasoline through at least 2022. Assuming there is no reciprocity, new certifications beginning in
4-337
-------
Chapter 4: Proposed Vehicle Emissions Program
2020 could be required to certify on Tier 3 gasoline and Tier 3 evaporative emissions standards.
This will provide sufficient time for ARB to amend its regulations to accept Tier 3 gasoline and
test procedures. In the event ARB chose not to amend their regulations, EPA, ARB, and industry
could develop an alternative that would prevent unnecessary testing.
There should be permanent reciprocity between California LEV III and Tier 3 so that automakers
can test one time on one test fuel to certify for both LEV III and Tier 3, as originally envisioned
toward One National Program.
28 - 78 Fed. Reg. 29908 at 29869 (May 21, 2013).
Organization: California Air Resources Board (CARB)
Additionally, CARB recommends reciprocity with California's gasoline certification fuel beyond
2020 and throughout the lifetime of the Tier 3 program. CARB offers the following comments in
support of our recommendation in an effort to further harmonize certification fuel program
requirements of the two agencies.
As proposed, Tier 3 provides reciprocity for certification fuel between the LEV III and Tier 3
programs until model year 2020, at which time U.S. EPA will only recognize emission
certification on federal certification fuel. Traditionally, the two agencies have granted reciprocity
for certification fuel when demonstrating emission compliance, thus providing a degree of
compliance flexibility to vehicle manufacturers without impacting the emission benefits of the
California or federal programs. CARB believes this should continue to hold true whether Tier 3
certification fuel is E15 as currently proposed or E10 (see our previous comment though on the
impact of an RVP of 10 on reciprocity concerns for evaporative emission testing) and strongly
recommends continuing the practice of providing reciprocity for certification fuel in the Tier 3
program beyond 2020 and throughout the lifetime of the Tier 3 program.
Organization: Chrysler Group LLC
Chrysler supports EPA's proposal to allow as an option certification using ARB test procedures,
which include E10 fuel at 7.0 psi RVP, but believes that this option should be available
permanently (past Model Year 2019). Notwithstanding this option, Chrysler believes that the
primary certification test fuel, using federal test procedures, should be E10 fuel at 9.0 psi RVP.
Recommendation: Chrysler recommends that EPA adopt an E10, 9.0 psi RVP certification test
fuel requirement, which is representative of real-world in-use conditions. Chrysler further
recommends that EPA extend the option of using E10 fuel at 7.0 psi and ARB test procedures to
be permanent (past Model Year 2019), at least until EPA and California can agree on a
harmonized test fuel with a methodology for designating a certification fuel that is the
predominate fuel in the marketplace.
Organization: General Motors LLC (GM)
4-338
-------
Tier 3 Summary and Analysis of Comments
GM supports the harmonization of Federal Tier 3 and California LEV III vehicle emission
programs. While there are some differences in requirements, we continue to support permanent
mutual reciprocity that allows OEMs to use one test fuel and its related testing regime per test
vehicle, either LEV III or Tier 3, for purposes of both Federal and California certification.
Reciprocity with CARB's LEV III regulation should be applicable to all aspects of EPA's Tier 3
rulemaking: all test fuels, evaporative requirements, exhaust requirements, etc.
Tier 3 and LEV III - Full Reciprocity should be made Permanent: Full and permanent
reciprocity assures OEMs the option to certify to either Tier 3 or LEV III regulations (all
aspects). Reciprocity is an essential compliance flexibility that minimizes duplication of
emissions compliance efforts and minimizes development costs and workloads, without risking
increase in vehicle emissions.
We encourage and need EPA and California to extend permanent reciprocity (mutual
acceptance) of either LEV III test fuel/testing or Tier 3 test fuel/testing. There shouldn't be a
requirement to use Federal test fuel/test procedures after MY 2019 for Tier 3 certification, as
proposed. This is critical to facilitate OEM use of one fuel/one testing regime per vehicle model,
a longstanding goal for harmonization between LEV III and Tier 3 standards toward "one
national standard". The purpose of the arbitrary date cutoff in the proposal is not clear.
If permanent reciprocity is not possible, EPA should accept carry-over certifications on LEV III
certification gasoline beyond MY 2019:
CARB cannot amend the LEV III regulations until after EPA finalizes Tier 3. Thus, there will be
a significant period (at least a year or two) of uncertainty where each agency has different
certification gasoline requirements.
The proposed Tier 3 regulations allow manufacturers to use LEV III certification gasoline only
through MY 2019. Beginning in MY 2020, all vehicles must be certified using Tier 3
certification gasoline and carryover of emissions data is not allowed. (See 40 CFR 86.113-07,
Fuel Specifications). If there is no permanent reciprocity, despite our strong support, then EPA
should allow carryover of emissions data after MY 2019. Manufacturers tend to carryover
emissions data for several years. Without an extension, this could result in manufacturers
retesting and recertifying vehicles for no other reason than to change the certification gasoline.
For example, a MY 2016 vehicle certified to LEV III standards on LEV III gasoline might be
required to recertify on Tier 3 gasoline, rather than carry over the LEV III certification data
through MY 2020 or even 2021 until a redesign requires a new certification. This would be an
unnecessary waste of resources and we do not believe this is EPA's intent. Consequently, we
recommend that EPA allow carryover of certification data using LEV III gasoline through at
least MY 2022.
Organization: Mitsubishi Motors R&D America, Inc (MRDA)
In the current Tier 2 program, certification test fuel is Indolene (EO). The proposed Tier 3
program introduces an ethanol-based certification fuel, i.e. El5 with a 9 psi RVP (Reid Vapor
Pressure), while California's LEV III program includes a California E10 certification test fuel
4-339
-------
Chapter 4: Proposed Vehicle Emissions Program
with a 7 psi RVP. In order to avoid duplicative certification testing and to support the production
of 50-state vehicle fleets, we request that the Tier 3 Final Rule include permanent reciprocity of
certification test fuel with California's LEV III program. Specifically, we request that EPA
permanently allow the optional use of LEV III certification gasoline and LEV III evaporative
emissions test procedures as equivalent to federal certification fuel and evaporative emissions
test procedures. This would allow OEMs the option of performing either procedure with its
appropriate certification fuel to satisfy the regulatory requirement of BOTH agencies.
-Toward EPA's goal to harmonize with LEV III, we ask that EPA permanently allow the
optional use of LEV III certification test fuel (E10, 7 psi RVP) and LEV III evaporative
emissions test procedures, or Federal certification test fuel (E10, 9 psi RVP) and federal test
procedures.
Our Response:
We recognize that test fuels are an important element of a national program. Vehicle
manufacturers have emphasized in their comments the desire to reduce their test burdens by
producing one vehicle that is tested on a single test procedure and on a single test fuel and that
meets both California and federal requirements. Although we have been able to reasonably align
the Tier 3 program with the LEV III program in most key respects, we recognize that the Tier 3
and LEV III test fuels are different, and that some differences may still exist in emissions
performance between vehicles tested on the two fuels. The largest difference between the two
fuels is the Reid Vapor Pressure (RVP), and other differences in distillation properties and
aromatic levels also exist (largely related to differences in actual in-use fuel nationally and in
California). We are finalizing as proposed the requirement that manufacturers certify vehicles on
the new Tier 3 E10 test fuels37 beginning with the first model year that a vehicle model is
r-M-, ^R r-M-,
certified to the FTP NMOG+NOx Bin 70 or lower. (The heavy duty and evaporative and
evaporative provisions are detailed in preamble Section IV.F.) This requirement may apply as
early as MY 2017 for vehicles up to 6000 Ibs GVWR and MY 2018 for vehicles greater than
6000 Ibs GVWR.39 This requirement also applies to vehicles certified at Bin 70 and lower that
are brought into the Tier 3 program under the Early Tier 3 option described in preamble Section
IV.A.7.b, with the exception of the specific provision allowing the use of LEV III fuels discussed
below. Beginning in MY 2020, all gasoline-fueled models will need to certify on the Tier 3 test
fuels for all exhaust emission requirements, regardless of their certification bin.40 As discussed
in preamble Section IV.A.7.C, manufacturers must also meet the 150,000 mile useful life
requirements for NMOG+NOx standards for these same vehicles as they are certified to Bin 70
and lower.
During the transition period from Tier 2 fuel to the new Tier 3 and LEV III E10 fuels,
manufacturers have indicated that they face a substantial workload challenge of developing and
37 This includes fuels used for cold temperature and high altitude testing and durability requirements. See preamble
Section IV.F.
38 The lower Bins are Bin 0, Bin 20, Bin 30 and Bin 50.
39 Vehicles above 6000 Ib GVWR choosing the alternative phase-in schedules described in preamble section
IV.A.2.C generally would begin using the Tier 3 test fuels for MY2019.
40 Diesel-fueled and alternative-fueled vehicles will continue to be tested on the fuels used under the Tier 2 program
except for E85 fueled vehicles, for which we are finalizing new test fuel specifications (see preamble Section IV.F).
4-340
-------
Tier 3 Summary and Analysis of Comments
certifying each vehicle model to the two new fuels simultaneously. We recognize this
transitional challenge and are including an additional option. We are finalizing as proposed an
option that vehicles certified in MYs 2015 through 2019 to California LEV III standards using
California LEV III E10 certification test fuels and test procedures can be used for certifying to
EPA Tier 2 or Tier 3 exhaust emission standards, including PM. A manufacturer may submit
LEV III test data on vehicles tested using the new LEV III E10 fuels for Tier 2 or Tier 3
certifications. Consistent with existing Tier 2 policy, EPA may test vehicles certified to Tier 2
standards using LEV III test results on Tier 2 fuel for confirmatory or in-use exhaust testing. For
vehicles certified in MY 2017 through 2019 to Tier 3 standards using LEV III E10 fuels, EPA
will only use LEV III El0 fuels for confirmatory and in-use testing (except for high altitude or
cold CO and hydrocarbons testing, as described below). Vehicles certified to the provisions of
Early Tier 3 (preamble Section IV. A.7.b) will be treated the same as Tier 3 vehicles certified in
MY 2017. For example, for MY 2015 and 2016, EPA will consider Early Tier 3 vehicles to be
part of the Tier 3 program for purposes of fuel-related testing obligations. We will not accept test
results using LEV II fuels for Tier 3 vehicle certification, including Early Tier 3 certifications,
with the exception of the PZEV exhaust carry-over provision described below.
California does not have fuel specifications for high altitude testing or cold CO and
hydrocarbon testing. For this reason, we are finalizing that for vehicles that manufacturers
choose to certify using LEV III fuel and test procedures, manufacturers must use program-
specific federal test fuels to comply with these federal-only requirements (i.e. Tier 2 vehicles will
use Tier 2 fuel and Tier 3 vehicles will use Tier 3 fuel). Similarly, high altitude and cold CO and
hydrocarbon confirmatory and in-use testing for these vehicles will be performed on the federal
fuel that the manufacturer is required to use at certification as specified above regardless of
whether LEV III or federal fuel is used for other testing.
We proposed the requirement that after MY 2019, all Tier 3 certification, confirmatory
and in-use emission testing be required to use only the proposed Tier 3 El 5 test fuel because it
was believed to be a worst case fuel for emissions. Because we are finalizing Tier 3 E10 test
fuels which are very similar as explained above to LEV III E10 test fuels, and not considered a
worst case fuel, we are not finalizing the requirement for all testing to be performed on Tier 3
E10 test fuel. Instead, for certifications after MY 2019, EPA will continue to allow LEV III test
results to be submitted for certification to Tier 3 standards, consistent with protocol under the
Tier 2 program. However, if a manufacturer chooses to submit certification results for
compliance with Tier 3 standards using the LEV III test fuel, then for confirmatory and in-use
testing we will hold vehicles to the Tier 3 standards while using the Tier 3 fuel in addition to the
LEV III test fuel; we will not allow new or carry-over certifications using LEV II or Tier 2
certification test fuels after MY 2019. CARB has indicated that they will accept Tier 3 test data
(on federal certification test fuels) to obtain a California certificate as early as MY 2015. In this
manner manufacturers should be able to avoid compliance testing on more than one fuel, since
vehicles certified to Interim or Final Tier 3 status using federal certification test fuels could also
obtain LEV III certification.
4-341
-------
Chapter 4: Proposed Vehicle Emissions Program
Auto industry commenters noted that the LEV III program provides an allowance for
manufacturers to carry over PZEV-certified vehicle exhaust data41 from the LEV II program into
LEV III compliance in MY 2015 through MY 2019. Thus, CARS allows these PZEV vehicles
to use emission testing results using LEV II fuel (i.e. California Phase II test fuel) to meet the
LEV III obligations. The commenters suggested that EPA allow manufacturers to carry over
such PZEV 150,000 mile useful life exhaust emission data to meet the Tier 3 standards. We
agree that this approach is appropriate during the transition, and we are finalizing this provision
for MY 2015 through MY 2019, including allowing Early Tier 3 compliance at the Bin 30 level
as a combined NMOG + NOX standard. EPA will hold vehicles certified using this provision to
the Tier 3 emission requirements when they are tested on the LEV II fuel for confirmatory and
in-use. Compliance testing of these vehicles for all other Tier 3 obligations (i.e., high-altitude
testing and Cold CO and hydrocarbons testing) must be performed using Tier 3 fuel, and these
vehicles will be required to meet the Tier 3 standards for Bin 30.
4.5.4. Potential Implications on CAFE Standards, GHG Standards, and Fuel Economy
Labels
See 78 FR 29912-29913, May 21, 2013 for full details.
4.5.4.1. Need for Adjustment to GHG, CAFE, and Fuel Economy Labeling Program
Calculations
What Commenters Said:
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
Maintain Consistency with CAFE and Gas Guzzler Regulations: While Tier 3 is predominantly a
criteria emissions regulation, proposed changes to the energy content (percent ethanol) of the
certification test fuel and proposed requirements to test on four wheel drive dynamometers
(dynos) significantly impact vehicle fuel economy. The changes proposed by EPA are not
consistent with statutory law, the regulatory history and intent of the CAFE and GHG programs,
or the 2017-2025 greenhouse gas and fuel economy agreements signed by EPA, NHTSA,
California, and many automobile manufacturers governing the calculation and stringency of fuel
economy values for CAFE and gas guzzler tax purposes. The Energy Policy and Conservation
Act (EPCA) specifies that EPA determine the manufacturer's CAFE value for passenger
automobiles using the "same procedures for passenger automobiles the Administrator used for
model year 1975.. .or procedures that give comparable results." Section 503(d) of EPCA
prohibits EPA from changing the stringency of the CAFE standards through test procedure
changes. We therefore request that EPA comply as soon as possible with its legal obligation to
issue fuel economy test procedure adjustments.
41 California's PZEV exhaust standards are the same as their SULEV standards and the Tier 3 Bin 30, and are
certified to a 150,000 mile useful life.
4-342
-------
Tier 3 Summary and Analysis of Comments
EPA's proposed changes to dyno requirements and to the ethanol content of test fuel would
directly impact the measurement of fuel economy, which is used for CAFE reporting, for fuel
economy labels and the Gas Guzzler Tax and for compliance with the two GHG rules. Industry
recommends that EPA reconsider these proposals and instead adopt industry's suggested
modifications below in order to avoid introducing significant and costly changes to the fuel
economy procedures.
Organization: BMW of North America, LLC
Furthermore, BMW urges EPA to adjust the fuel economy test procedures for Corporate Average
Fuel Economy (CAFE) in order to account for the substitution of EO with E10. If not adjusted,
fuel economy testing with E10 would result in a more stringent CAFE standard.
Organization: Volvo Car Group
Emissions and GHG require the same test procedures and certification fuels. Thus, it is not only
emissions that are becoming more stringent but also the GHG/CAFE indirectly through enhanced
requirements regarding testing.
Organization: Mitsubishi Motors R&D America, Inc (MRDA)
EPA needs to modify the fuel economy equations to account for the introduction of an ethanol -
containing certification fuel; therefore, we request the following: EPA must establish new fuel
economy and CREE equations for ethanol certification fuels in the Tier 3 rulemaking. An R-
factor of 1 in the fuel economy and CREE equations is considered appropriate based on industry
analysis.
Organization: National Automobile Dealers Association (NADA)
Clearly, engineering and technology resources necessary to meet Tier 3 compliance will directly
compete with those necessary to meet GHG reduction mandates. Given EPA's recognition of the
need for a mid-term review process to assess critical assumptions associated with it's MY 2017-
25 GHG mandates, the agency should recognize similarly the importance of revisiting critical
assumptions associated with its Tier 3 regulations. In addition, the apparent benefits that
appropriate fuel designs (i.e., low-sulfur content) offer for GHG reduction strategies should in
the context of the mid-term review and otherwise.
NADA concurs with suggestions made in the joint comments of the Alliance of Automobile
Manufacturers and the Association of Global Automakers regarding the need to avoid
introducing unnecessary, significant, and costly changes to fuel economy procedures that could
negatively impact fuel economy rating, fuel economy labeling, and CAFE requirements.
Organization: Advanced Biofuels USA (ABFUSA)
4-343
-------
Chapter 4: Proposed Vehicle Emissions Program
In order to provide equality in EPA fuel economy calculations for higher octane, higher ethanol
fuels, the "R" fuel energy content factor should be set at "1." This change is necessary since EPA
has previously recognized that the engine technology used to set the "R" in the 1970s has
reached levels of efficiency not envisioned at the time and as a result the factor does not
represent the current reality.
Organization: American Coalition for Ethanol (ACE)
While discussion of the 'R-factor' is not technically part of this rulemaking, ACE agrees that the
current fuel economy 'equation' unfairly punishes ethanol blends based on old data, primarily
due to an out-dated 'R-factor.' EPA correctly notes that change in fuel economy is not directly
proportional to change in energy content of the test fuel, and mentions that manufacturers have
suggested that a higher R value may be warranted. ACE believes more current and accurate data
has been available for some time, and supports EPA's continued investigation into the issue, and
looks forward to that hurdle being removed in the near future.
Organization: Clean Fuels Development Coalition (CFDC)
There are other impediments to widespread commercialization of higher octane ethanol fuels and
one is the "R" factor noted in the NPRM and what appears to be a failure to recognize increased
efficiencies since the original "R" factor formulas were established.
Our Response:
EPA agrees with the commenters that action is needed to assess and potentially address
the impact of changing test fuel on compliance with GHG, CAFE, and fuel economy labeling
program requirements. We further agree that the adjustments should be such that the stringency
of these programs is not affected. The preamble to the final rule discusses our future plans to
obtain the data for future action in these areas. Note that at this time EPA is not adopting any
new testing requirements for 4WD dynamometers.
4.5.4.2 Test Procedure Adjustments
What Commenters Said:
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
Applicable Law Requires Test Procedure Adjustments (TPAs): The Tier 3 NPRM proposes
lower octane ethanol containing test fuel and dyno requirements which impact CO2, fuel
economy, and the measurement of fuel economy for labeling, CAFE, the Gas Guzzler Tax,
GHG, and medium-duty chassis cert compliance. EPA did not address the changes to measured
fuel economy that will be caused by the proposed changes to test fuel and dyno requirements in
the Tier 3 NPRM, nor the costs and benefits of the proposals. Instead, EPA's supporting analyses
simply alluded to a "future action" which would address these impacts. Deferring consideration
4-344
-------
Tier 3 Summary and Analysis of Comments
of fuel economy impacts from test procedure changes runs counter to EPA's own rules (50 Fed.
Reg. 27183, 1985) that require EPA to issue test procedure adjustments at the same time that it
promulgates final regulations incorporating the test fuel and dyno changes. In addition to
statutory and regulatory requirements, EPA must adhere to the commitments entered into
between the agencies and manufacturers for achieving the 2012-2016 and 2017-2025 GHG rules.
The Alliance and Global Automakers request that EPA comply as soon as possible with its legal
obligation to issue fuel economy test procedure adjustments.
Whenever EPA modifies the test procedures for measuring fuel economy, it is required by law to
give adjustments for measured fuel economy to compensate for the changes. 49 U.S.C. Chapter
329 sets forth the statutory provisions relating to CAFE. The test procedures to be used for
calculating CAFE are addressed in 49 U.S.C. §32904(c), which reads in relevant part as follows:
"The Administrator shall measure fuel economy for each model and calculate average fuel
economy for a manufacturer under testing and calculation procedures prescribed by the
Administrator. However ... the Administrator shall use the same procedures for passenger
automobiles the administrator used for model year 1975 (weighted 55 percent urban cycle and 45
percent highway cycle), or procedures that give comparable results." Will not
The requirement for "procedures that give comparable results" indicates that when EPA
promulgates changes to the test procedures that have an impact on the CAFE results for
manufacturers, the method of calculating CAFE should be modified so that the results are
comparable to those that would have been obtained in the absence of any change to the test
procedures. Such modifications are commonly referred to as TPAs. This section imposes a
"comparability" requirement on EPA: EPA must issue adjustment factors upon changing
emissions or fuel economy test procedures so that measured fuel economy is comparable to
results obtained using the 1975 test procedure. Attached as Appendix 7 is a discussion of the
statutory and other requirements for EPA to issue TPAs for the test fuel and dyno requirement
changes being proposed in the Tier 3 rule.
Organization: Mitsubishi Motors R&D America, Inc (MRDA)
Fuel Economy and CREE Equations: With the introduction of an ethanol-containing certification
fuel in the Tier 3 NPRM, there will be implications upon vehicles' fuel economy due to the
lower energy content of ethanol-containing fuels. Under the Energy Policy and Conservation Act
(EPCA), EPA must ensure that fuel economy test procedures for passenger cars produce results
comparable to the 1975 fuel economy test procedures. The 1975 test procedures are based on the
use of EC) certification test fuel. A change in the ethanol content in the certification test fuel will
result in changes to fuel economy calculations, and therefore an adjustment factor is necessary to
produce comparable results to the 1975 test procedures. The current R-factor of 0.6 was
established using an EO certification test fuel. Changing the certification test fuel to an ethanol-
containing fuel severely impacts the calculation for fuel economy, and therefore requires an
adjustment to the R-factor.
Fuel Economy/CREE Equations: EPA must establish new fuel economy and CREE equations,
with an R factor of 1 for ethanol-containing certification fuels in the Tier 3 rulemaking.
4-345
-------
Chapter 4: Proposed Vehicle Emissions Program
Organization: National Corn Growers Association (NCGA)
Fuel economy testing originated in 1975 using the FTP-75 procedure and a pure hydrocarbon
certification fuel with a specific gravity of 0.739 and a carbon weight fraction of 0.866. EPA is
required to provide average fuel economy adjustments for test procedure changes that would
cause a vehicle to achieve fuel economy results different from those that would be achieved
under the 1975 test procedure. Changes in test fuel energy content qualify as a test procedure
change that requires an adjustment.
EPA has proposed new certification fuels containing ethanol with slightly lower net heating
values on a per gallon basis. A simple way to adjust for energy content for vehicles not designed
and optimized to use a mid-level blend high octane fuel is to use a ratio of the volumetric energy
content of the new test fuel to that of the 1975 test fuel. EPA has proposed this approach in the
past (10, 11). However, 30 years ago when EPA originally corrected for fuel changes, vehicle
technology did not fully compensate for fuel energy changes. EPA modified the approach by
adding the "R-Factor" which may be considered as the efficiency with which the vehicle engine
adapts to fuel energy variations. A factor of 0.6 was derived from data gathered from vehicles
from the 1970s and early 1980s. An R-Factor less than 1 implies the vehicle is less efficient on
fuels with higher energy content than the 1975 test fuel and more efficient on fuels with lower
energy content than the 1975 test fuel.
EPA first applied the equation containing the R-Factor of 0.6 to a test fuel with oxygenates,
California Phase 2 gasoline, in 1995 (12). In applying the 0.6 R-Factor EPA acknowledged
"Data indicate that this [corrected] mpg value is partially corrected for the difference in fuel
properties between Phase 2 test fuel and the 1975 EPA test fuel."(13) The R-Factor of 0.6 was
derived from vehicles that are now over 30 years old and nearly 20 years ago EPA acknowledged
that the 0.6 value did not fully compensate for the fuel energy change and needed to be revised.
Nonetheless, the factor of 0.6 is still in use today.
Recent test data indicate a somewhat higher R-Factor than 0.6. The Auto-Oil test program on
1989 model year vehicles (14) determined an R-Factor value of approximately 0.93. More
recently, The U.S. Department of Energy published its own analysis indicating values ranging
from 0.92 to 1.03 (15). For virtually all vehicles evaluated, the 95% confidence intervals shown
include the 1.0 value that EPA initially proposed in 1976. The authors noted that: "Additionally,
there is uncertainty present in the ASTM results for heating value, carbon weight fraction, and
specific gravity that have not been included in the confidence intervals. Inclusion of these
uncertainties would increase the confidence intervals." Of course, in all these tests, the vehicles
tested were not designed for a higher octane mid-level blend.
We request that EPA review the R-Factor and determine whether the use of the value of 0.6
would be an impediment to automakers to petitioning EPA for the use of a MLEB as a
certification fuel and revise the R factor as necessary.
Commenter footnotes
Impact of Gasoline Characteristics on Fuel Economy and its Measurements, 76-10 JLB.
11 Corrections for Variations in Test Fuel Properties, EPA-AA-SDSB-84-3.
12 EPA "Dear Manufacturer" Letter CD-95-09.
13 Ibid.
4-346
-------
Tier 3 Summary and Analysis of Comments
14 A. Hochhauser et al., "Fuel Composition Effects on Automotive Fuel Economy-Auto/Oil Air Quality
Improvement Research Program," SAE paper 930138, SAE International, March 1993.
15 S. Sluder and B. H. West, (2013) Preliminary Examination of Ethanol Fuel Effects on EPA's R-factor for Vehicle
Fuel Economy; http://info.ornl.gov/sites/publications/Files/Pub42819.pdf, (2013).
Our Response:
EPA has reviewed the text of the main comments from the Alliance and Global
Automakers as well as appendix 7. EPA acknowledges that changes in testing and calculation
procedures prescribed by the Administrator should give comparable test results to those used for
the 1975 model year for passenger automobiles. In the NPRM we expressed our commitment to
the principle that test procedure adjustments related to the change in test fuel properties will not
affect the stringency of CAFE requirements for passenger automobiles and other vehicle classes
covered by the CAFE requirements. EPA understands the need to make these adjustments in a
timely manner and the manufacturers' perspective that EPA should implement the new
requirements for criteria pollutant emission testing, CAFE, and GHG at the same time.
However, in the NPRM we deferred action on adopting a new R factor and other potential
related adjustments because we did not have the requisite data on the new test fuel (E10) and the
current test fuel (EO) on Tier 3 technology vehicles. Without the requisite data we could not be
certain that the principles outlined in Appendix 7 would be followed. By requiring CAFE and
GHG testing on Indolene or EO, the interim requirements established in the FRM assure that the
stringency for the CAFE and GHG standards are not affected. This approach will also help to
ensure that the data needed to make the test procedure adjustments based EO and E10 emission
test data on Tier 3 technology vehicles is available in a timely manner.
4.5.4.3. R-Factor Value
What Commenters Said:
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
In addition, EPA proposes to change the definition of gasoline test fuels to include E10 and E15
test fuels, but EPA does not propose to change the fuel economy equation used for CAFE to
compensate for the change in volumetric energy content on Tier 3 compliant vehicles. EPA is
obligated to release a new fuel economy equation for ethanol containing test fuels within the Tier
3 Final Rule, and data supports that the R factor should be set to 1.0. Changing the fuel economy
equation would be consistent with statutory law and the regulatory history governing the
calculation of fuel economy values for CAFE and gas guzzler tax purposes. [EPA-HQ-OAR-
2011-0135-4461-A1, pp. 93-94]
EPA proposes to change the test fuel used by manufacturers to test vehicles for criteria emissions
and GHG and Fuel economy from the current "Indolene" test fuel containing no ethanol to a test
fuel with ethanol added. EPA proposes El5 for the ethanol level; elsewhere in these comments,
the Alliance and Global Automakers discuss specific recommendations regarding the correct test
4-347
-------
Chapter 4: Proposed Vehicle Emissions Program
fuel, level of ethanol, and Reid Vapor Pressure (RVP). The following discussion focuses on
implementing required TPAs for ethanol containing fuel.
Data Supports R Factor = 1.0: A factor used to represent the response of a vehicle's fuel
economy to changes in the fuel's ethanol content as measured by the fuel's energy content is the
oo
"R Factor." In a separate memorandum to the docket EPA concludes that an appropriate range
for the R-factor for ethanol containing fuels is 0.8-0.9. This conclusion was based on an
OQ
examination of the data in the EPAct study. The Alliance performed a similar analysis on this
dataset and identified an error in the measured properties for one of the EPAct fuels that biases
the R-factor results - correcting this error results in an R-factor of 0.96. Industry met with EPA
on May 20, 2013 to review this new information and to provide the results of a reanalysis that
included all the fuels included in the EPAct dataset except the above mentioned fuel (Attached as
Appendix 8 is a summary of that reanalysis).
Industry met with EPA again on June 13, 2013, and reviewed a new report from Oak Ridge
National Labs (ORNL) to determine R Factor.90 The report was based on an analysis of test data
gathered in the U.S. Department of Energy (DOE) mid-level blends test program.91 The ORNL
report concluded that:
"Analysis of city cycle Federal Test Procedure (FTP) fuel economy test results and fuel analyses
from the catalyst durability study indicates that the average R-factor for modern vehicles is very
close to unity at about 0.94±0.04 (or 0.96±0.04 if the one problematic vehicle omitted)."
In summary, available data strongly supports an R-Factor equal to 1.0. Therefore, industry
recommends that EPA adopt a fuel economy equation for gasoline fuels containing ethanol (i.e.
E10, E15) with an R-Factor of 1.0 in the Tier 3 Final Rule.
Commenter footnotes
OQ
EPA Technical Memorandum "Analysis of the Effects of Changing Fuel Properties on the EPA Fuel Economy
Equation and R-Factof by Aron Butler, et al.; also available in the docket (Tier 3 Docket #EPA-HQ-OAR-2011-
0135).
89
EPA Technical report "EPAct/V2/E-89: Assessing the Effect of Five Gasoline Properties on Exhaust Emissions
from Light-Duty Vehicles Certified to Tier 2 Standards, Final Report on Program Design and Data Collection",
Report Number EPA-420-R-13-004, March 2013; also available in the docket (Tier 3 Docket #EPA-HQ-OAR-
2011-0135).
90
Oak Ridge National Laboratory, US Department of Energy, report "Preliminary Examination of Ethanol Fuel
Effects on EPA's R-factor for Vehicle Fuel Economy," Report Number ORNL/TM-2012/198, June 2013; accessed
at: http://info.ornl.gov/sites/publications/Files/Pub42819.pdf
West, Brian H., Scott Sluder, Keith Knoll, John Orban, Jingyu Feng, Intermediate Ethanol Blends Catalyst
Durability Program, ORNL/TM-2011/234, February 2012, available at
http://info.ornl.gov/sites/publications/Files/Pub31271 .pdf
Organization: General Motors LLC (GM)
On page 29913 of the proposed rule, EPA discusses the potential impacts that the proposed
gasoline test fuel change will have on corporate average fuel economy (CAFE). Additionally,
EPA states that "it is almost certain that the same vehicle tested on a test fuel with 15 percent
ethanol content will yield a lower fuel economy value relative to the value if it were tested on the
4-348
-------
Tier 3 Summary and Analysis of Comments
current test fuel with zero ethanol content." Indeed, the Department of Energy concluded after
extensive testing of ethanol blended fuels on a fleet of late model U.S. vehicles: '"Fuel economy
was decreased ... consistent with the energy density of the test fuel."1
Pursuant to 49 U.S.C. Section 32094 (c) and the provisions of 40 CFR 600.510-12(f), EPA is
required to provide average fuel economy adjustments for test procedure changes that would
cause a vehicle to achieve fuel economy results that would differ from results that would be
achieved under the 1975 test procedure. However, EPA has proposed to defer any modifications
to the average fuel economy calculations to a possible 'future action' based on the fact that the
current gasoline fuel economy equation already contains an adjustment for the energy content of
the fuel. This would be true and acceptable if the total adjustment in this equation was solely
based on energy content. However, the current gasoline fuel economy equation includes a factor,
commonly referred to as the R-factor, which provides a partial energy adjustment to more
accurately represent the fuel economy that would be achieved if the vehicle were to be tested
under 1975 conditions. The R-factor is part of the current test procedure adjustment that accounts
for changes in test fuel energy content; therefore, changes to the R-factor must be made
concurrently with changes to the test fuel.
One issue with the current R-factor of 0.6 is that it was developed under conditions where the
energy content of the test fuel was tending to increase. It should not be applied when the energy
content of the fuel is decreasing. As a matter of fact, in EPA technical report 'Corrections for
Variations in Test Fuel Properties' (EPA-AA-SDSB-84-3), EPA states this concern as follows:
"A second area of questionable accuracy with the GM approach [referring to the inclusion of an
R-factor] occurs when the energy content per unit volume of the fuel decreases. In this case, the
efficiency term reduces or discounts the calculated correction... It is unlikely that vehicles could
tolerate a significant decrease in energy density of the fuel without experiencing an equivalent
decrease in fuel economy."
Therefore, continued use of an R-factor designed for use with higher energy content test fuel is
inappropriate when migrating to a lower energy content test fuel.
Commenter footnote
K. Knoll, B. H. West, et al, Effects of Intermediate Ethanol Blends on Legacy Vehicles and Small Non-Road
Engines, Report 1 - Updated.
Organization: Ford Motor Company (Ford)
Since discussions began in June 2012, the EPA and manufacturers have made considerable
progress in developing the FE equation for low-level ethanol blends. This equation differs from
the current "EO" equation in two respects: (1) an updated "R-factor" to better reflect the
efficiency of current vehicle technology when operated on the proposed fuel, and (2) a factor for
non-methane organic gases (NMOG) that accounts for the increase in oxygenated hydrocarbons
resulting from the addition of ethanol to gasoline. While the inclusion of the NMOG factor in the
equation has been a relatively straight-forward exercise, determining the most appropriate value
for the revised R-factor has been a more complex task. Ford's analysis of the EPAct2 data set
4-349
-------
Chapter 4: Proposed Vehicle Emissions Program
(Alliance/Global Comments, Appendix 8) used an established R-factor methodology 3 to show
that once an anomalous fuel is removed, the data set strongly supports an R-Factor very close to
"1". This result was corroborated using a second, independent data set in a paper published by
Oak Ridge National Laboratory4. Based on these findings, Ford agrees that the most appropriate
value for the revised R-factor is 1, as proposed in the Alliance/Global comments.
Recommendation: Ford recommends that EPA adopt the fuel economy equation for E10 and E15
test fuels as proposed in the Alliance/Global comments, with an R-factor equal to "1", at the
earliest possible date. Additionally, Ford recommends that EPA permit optional usage of the
current "EO" fuel for all fuel economy purposes through 19MY for vehicles required to use E10
or El5 for certification testing.
Organization: Mitsubishi Motors R&D America, Inc (MRDA)
Recent analyses of data have demonstrated the Net Heating Value of ethanol-containing fuels
result in an R-factor of 1. An R-factor of 1 is necessary to adjust the fuel economy values based
on the proposed ethanol-containing certification fuel in order to be equivalent to the baseline
1975 CAFE values.
An R-factor of 1 in the fuel economy and CREE equations is considered appropriate based on
industry analysis.
Organization: Renewable Fuels Association (RFA)
EPA should set the "R-factor" for fuel economy calculations at 1.0 to reflect the latest research
and to avoid unnecessarily complicating automaker compliance with CAFE standards.
In its discussion of proposed changes to the certification test fuel, EPA briefly references
potential implications for compliance with the CAFE program and GHG emissions standards.
EPA states it is "...committed to the principle of ensuring that the proposed change in test fuel
would not affect the stringency of either the CAFE or GHG emissions standards, and that the
labeling calculations would be updated to reflect the change in test fuel (emphasis added)." In
spite of this commitment, EPA fails to propose any actions to revise the "R-factor" used in the
existing fuel economy equation. Maintaining the existing R-factor of 0.6 would indeed "affect
the stringency" of CAFE standards and make it considerably more difficult for automakers to
meet fuel economy requirements if the certification test fuel is El 5 or other ethanol blend.
The R-factor was developed to account for a vehicle's relative ability to compensate for changes
in fuel energy content. The existing R-factor of 0.6 is based on data from vehicles tested more
than 30 years ago and was first used in fuel economy calculations in 1988. Control systems in
modern light duty vehicles have evolved substantially since establishment of the 0.6 factor, and
recent testing data demonstrate modern vehicles are able to entirely compensate for changes in
fuel energy content. A detailed analysis published in June 2013 by the Department of Energy
(DOE) evaluated fuel economy data collected during mid-level ethanol blend catalyst durability
testing to determine the proper value for the R-factor.1
4-350
-------
Tier 3 Summary and Analysis of Comments
According to the DOE work, "Calculating modern vehicle fuel economy with the existing CFR
equation with an R factor of 0.6 when ethanol blends are used produces significant errors
(emphasis added)." The DOE analysis found that the actual average R-factor for modern vehicles
is "very close to unity" at about 0.94±0.04 (or 0.96±0.04 with one problematic vehicle omitted).
DOE determined that fleet-wide average R-factor values for vehicles using E10 and E15 ranged
from 0.95 to 1.03. Accordingly, the DOE researchers concluded, "The current factor of 0.6
which is called out in CFR is clearly too low, and a proper factor for modern vehicles is closer to
unity [i.e., 1.0], as might be expected from improved air/fuel ratio control common for more
modern vehicles." Based on the DOE work, EPA should revise its R-factor to 1.0.
Commenter footnote
i
S. Sluder and B. H. West, (2013) Preliminary Examination of Ethanol Fuel Effects on EPA's R-factor for Vehicle
Fuel Economy; http://info.ornl.gov/sites/publications/Files/Pub42819.pdf
Organization: Mercedes-Benz USA, LLC on behalf of Daimler AG
Given that the volumetric energy content of an E25 Tier 3 fuel would be almost 9% lower than
an EO fuel, it is also necessary then to consider an R factor = 1. Per 'Corrections for Variations in
Test Fuel Properties, EPA-AA-SDSB-84-3', the R factor, 'considered as the efficiency with
which the vehicle engine adapts to fuel variations', should = 1 when an engine applies precise
closed-loop feedback control for A/F ratio. Since the engine can take thermodynamic advantage
of fuel properties but not fully adjust for volumetric energy content, as in the case of an E25 fuel,
variations in volumetric efficiency are more related to engine-specific hardware and other
vehicle attributes designed to optimize thermodynamic efficiency than to the specific test cycle
itself. An R-factor of 0.6, as is currently the case, would result in approximately 5% volumetric
fuel efficiency loss for an E25 fuel, which mathematically hinders any manufacturer seeking to
certify a vehicle on such a fuel. Thus an R factor = 1 is a necessary step for the acceptance of
mid-blend ethanol fuels.
Organization: International Council on Clean Transportation (ICCT)
The ICCT has similar concerns on any future use of engines using E30. In addition, we are
concerned that E30 could open the door to E30 credits against the CAFE and GHG standards,
similar to what has already occurred for FFVs. It would be much better for EPA to focus on
increasing the octane rating of all gasoline. The ICCT supports updating the R-factor in the
carbon balance equation for NHV changes. Specifically, the R-factor determined by ORNL for
Tier 2 vehicles without the data outlier should be used, or R=0.96.
The R-factor is an adjustment made by EPA to maintain compatibility with the test procedures
used in 1975 for calculating fuel economy, as required by the 1975 EPCA. The R-factor is used
to adjust the FE test results for changes in the net heating value (NHV) of the test fuel. EPA is
accomplishing this by incorporating the R-factor into the carbon balance equation, instead of
doing the adjustment after calculation of the fuel economy.
4-351
-------
Chapter 4: Proposed Vehicle Emissions Program
Specifically, the R-factor is the sensitivity of the fuel economy to changes in NHV. Ethanol
blends reduce the NHV of the test fuel. To the extent that NHV affects engine efficiency, this
change must be accounted for under the 1975 EPCA.
Current, an R-factor of 0.6 is in the regulations. This was based upon data submitted by GM and
others in 1985 on primarily carbureted vehicles. In the past there was not a lot of variation in fuel
properties of Indolene, so this R-factor had a minor effect and there was no need to revisit it.
However, switching to El0 as the test fuel will cause a much larger change in the NHV - about
3.5% for E10 (and over 5% for E15). Using an R-factor of 0.6 would give an artificial upward
adjustment of over 2% (0.4 x 5%) to El 5 vehicles.
ICCT commends EPA for having the foresight to readdress the R-factor. A contractor report for
EPA by ORNL, based upon test data on recent vehicles using EO, E10, E15, and E20, calculated
R-factors for the entire fleet, Tier 2 vehicles only, and with and without exclusion of an outlier
vehicle (12).
The existing R-factor is clearly outdated and needs to be replaced. Vehicles certified to Tier 2
are representative of current vehicles and are the closest representation of future Tier 3 vehicles,
so data from older non-Tier 2 vehicles should not be used. The Honda Accord data is clearly an
outlier and is likely the result of an error in testing. It is standard practice in statistics to exclude
such obvious outliers. Thus, the ICCT recommends that the R-factor found in the report for Tier
2 vehicles without the outlier, 0.96, be adopted in the final rule.
Commenter footnote
Sluder, S. and West, B., "Preliminary Examination of Ethanol Fuel Effects on EPA's R-factor for Vehicle Fuel
Economy", ORNL report for EPA, June 2013.
Our Response:
As indicated by the comments, there is disagreement between EPA and the industry with
regard to the appropriate value for the R-factor. The NPRM cited two memos discussing the
issue, but in EPA's view there was not adequate data at the time of the NPRM to settle on an R-
factor revised value for the proposal. Thus, no specific value was proposed and EPA indicated
that a revised R-factor would not be included in the final rule. EPA met with the commenters on
August 7, 2013 to further discuss the analysis they provided which supported their perspective
that an R= 1.0 was justified. EPA's view of the data was that a value of approximately 0.9 was
more appropriate42. EPA and GM met the following day on the same issue, and at EPA's
request, GM provided additional input to clarify a few technical perspectives.43'44 There was
agreement that further work would be helpful to understand the data relative to the effects of
42 Passavant, G. (August, 2013). EPA, "Alliance, and Global Automakers Meeting on Issues Related to
Tier 3 Cert Fuel and the CAFE "R" Factor". Memorandum to the docket.
43 Passavant, G. (August, 2013). "EPA and GM Meeting on Issues Related to Tier 3 Cert Fuel and
"R" Factor". Memorandum to the docket.
44 Fronetti, S. (August 2013)."GM Input on R-factor". E-mail from S Fronetti to G Passavant.
4-352
-------
Tier 3 Summary and Analysis of Comments
other fuel qualities on the R-factor and that the data set had several limitations related to
technology types and ethanol content of the test fuels.
The current fuel economy equations already have a means by which to adjust for
elements such as volumetric energy density, specific gravity and net heating value which are fuel
specific. The R-factor is more vehicle technology related so it best assessed on vehicles using
Tier 3 technology.45 The approach laid out in the FRM and prescribed in §86.600.113 will help
to ensure that the R-factor is based on EO and E10 emission results for Tier 3 technology
vehicles. As mentioned in the GM comments, the use of an R=0.6 is not ideal with E10
emission test results and is thus providing the manufacturers the option to use EO fuel in litmus
assessments if it does not think E10 test results with R=0.6 is appropriate.
We acknowledge the comments of Mercedes Benz USA (MB) regarding the relationship
of E25 and the R-factor. Fundamentally, the issue of the appropriate R factor value needs to be
assessed since the test fuel is changing. We note MBs view that an R=l .0 is appropriate for E25,
and will consider test data if and when E25 becomes a primary or optional test fuel.
We do not agree with the comments from the ICCT. They seem to misunderstand the role of the
R value term in the fuel economy equation and believe it is already covered by the change in net
heating value (NHV) for E10 versus EO in the fuel economy equation. R factor is the efficiency
with which the vehicle/engine adapts to changes in fuel qualities. It is determined empirically
based on vehicle test data. We do not agree with ICCT that a value of 0.96 is appropriate
because it is not based on the results for Tier 3/LEV III technology vehicles tested on Tier
3/LEV III fuels. The approach laid out in the FRM will allow EPA to gather the requisite data
and conduct the analysis needed to see if a change in the R factor value is needed and if so, what
value should be proposed.
4.5.4.4. Related Test Procedure Adjustment Calculations
What Commenters Said:
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
EPA should address Fuel Economy and CREE Equations for E10-25 Certification Fuels as
follows:
The Tier 3 proposal requires that 50-state E10 certifications use ethanol test fuel for fuel
economy label litmus tests and for fuel economy. This is not possible without an updated R
factor and fuel economy equation. Measuring the oxygenates for all E10 or E15 tests is labor
intensive, requiring ethanol and aldehyde measurement in the chemistry lab which is very
expensive and delays results for a very small and negligible change in fuel economy.
Oxygenated species contribution to fuel economy and CREE with E10 or El 5 is insignificant
(less than 0.005%) but will be included in the NMOG term in the denominator of the proposed
45 The R-factor may be considered as the efficiency which the vehicle/engine adapts to changes in fuel qualities.
4-353
-------
Chapter 4: Proposed Vehicle Emissions Program
fuel economy equation below. The NMOG term is calculated based on the NMHC to NMOG
equations proposed in §1066.665 in the Tier 3 Part 1066 test procedures without the requirement
for direct measurement of ethanol and aldehyde. Accordingly, the Alliance and Global
Automakers recommend the following equations (with an R factor of 1.0) be used for E10 and
El 5 fuel economy, CREE and ADCREE conversion factor calculations:
10-25 = (5174 x I04x )/[[ x( + 4)+(0.429 x )+(0.273 x 2)]x ]
10-25 = ( /0.273)( + 4)+(1.571 x )+ 2
2x =(5174xlQ4x )/(0.273 x )=
Where:
- 5174xl04 = density of H20 @ 60°F * SG of 1975 reference fuel * NHV of 1975 reference fuel
- CWF is the carbon weight fraction of the test fuel
- NHV is the net heating value of the test fuel [Btu/lbm]
- HC is the total hydrocarbons [g/mi] in the exhaust gas
- CO is the carbon monoxide [g/mi] in the exhaust gas
- CO2 is the carbon dioxide [g/mi] in the exhaust gas
Organization: General Motors LLC (GM)
An added benefit of setting the R-factor at 1 .0 is that it would simplify the fuel economy
equation and avoid future controversy if EPA were to ever propose another test fuel.
GM proposes that EPA include this change in the final Tier 3 rule by establishing a new fuel
economy equation for vehicles tested on Tier 3/LEV III Elx test fuels that includes an R-factor
of 1.0. This equation would be:
FE = (5174xl04 x CWF) / {[(CWF x (NMOG + CH4)) + (0.429 x CO) + (0.273 x CO2)] x NHV}
where,
- 5174xl04 = density of H2O @ 60° F x specific gravity of 1975 reference fuel x NHV of 1975
reference fuel - CWF is the carbon weight fraction of the test fuel
- NHV is the net heating value of the test fuel [Btu/lbm]
- NMOG is the non-methane organic gas [g/mi] in the exhaust gas as determined in accordance
with 40 CFR 1066 test procedures
is the methane [g/mi] in the exhaust gas- CO is the carbon monoxide [g/mi] in the exhaust
gas
is the carbon dioxide [g/mi] in the exhaust gas
4-354
-------
Tier 3 Summary and Analysis of Comments
It should be noted that although oxygenated hydrocarbons are an insignificant contributor to the
fuel economy value, their effect has been included by virtue of using NMOG in the equation.
Organization: General Motors LLC (GM)
One of GM's primary concerns with the use of any new certification test fuel, regardless of
chemical composition, is the test fuel's effect on CAFE fuel economy, which is intimately linked
to EPA's exhaust emissions regulations. The current fuel economy adjustment R-factor of 0.6
was first developed by EPA and GM in 1984 as a means to compensate for drift in the
composition and energy content of Federal emissions certification fuel relative to the original
1975 emissions certification fuel. This practice established the precedent that each time the
emissions certification fuel formulation changed, the fuel energy content would be partially
normalized (to the extent of the R-factor) back to that of the 1975 certification fuel to reduce
changes in fuel economy regulation stringency. Going forward, no vehicles meeting Tier 3
emissions will use the open loop control systems that were used to develop the R-factor back in
the 1970s and early 1980s. Modern closed loop control systems fully adjust for changes in the
heating value of the fuel. Indeed they must do so to maintain stoichiometry and minimize
emissions of HC, CO and NOx. An R-factor of 1.0 reflects the full control system adjustment to
the fuel. However, the Test Fuel section of the introductory material of the NPRM states, "EPA
will continue to investigate this issue (R-factor) and if necessary address it as part of a future
action."
Our Response:
EPA did not propose to change the fuel economy equations of 40 CFR 600.113 in the
Tier 3 NPRM. The equations proposed by the commenter differ from those now in the CFR in
that the R factor value is set at 1.0 (and this is not shown), the HC value is replaced by
(NMOG+CH/t), and an additive value of 5174 is missing from the denominator of the FE 10-25
equation because the R factor is 1. The term ADCREE is not used in the current CFR, but is
reflected in an EPA guidance letter to the manufacturers.46 The manufacturers suggest that the
use of NMOG + CH/jin the fuel economy equations provides sufficient accuracy relative to the
measurement of aldehyde values. EPA is not adopting any changes to the fuel economy
equations in the Tier 3 rule, but the commenters should raise these ideas in comments on the
follow-on action where EPA proposes the test procedure changes related to the fuel economy
equation. In that forum adjustment to the terms and values in the fuel economy equations and
any related guidance letters can be fully considered.
4.5.4.5. Timing for the GHG, CAFE, and Fuel Economy Labeling Program
Adjustments
What Commenters Said:
46 CD 12-03.(February, 2012)." Analytically Derived CO2 and Carbon-Related Exhaust Emissions
(CREE) for Light-Duty Vehicles".
4-355
-------
Chapter 4: Proposed Vehicle Emissions Program
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
EPA is Obligated to Release a New Fuel Economy Equation for Ethanol Containing Test Fuels
Within the Tier 3 Final Rule:
Under 49 U.S.C 32904(c) and 40 C.F.R. 600.510-12(f), EPA is required to provide average fuel
economy adjustments for test procedure changes that would cause a vehicle to achieve fuel
economy results that would differ from results that would be achieved under the 1975 test
procedure. The current regulations contain a fuel economy equation with an R-factor equal to 0.6
that applies to testing on "gasoline." Additionally the regulations currently define E10 and El5
fuels [§86.113-04(a); §86.113-07(a); and §1065.710(a)] as being gasoline. Hence when the
NPRM proposed to change the test fuel to either E10 or El 5 without changing the applicable
mpg equation, the resultant proposed regulations could be interpreted as requiring use of the
current equation with an R-factor of 0.6. This creates a change in test procedures relative to the
1975 baseline required for the CAFE program. For this reason, EPA must adopt a test procedure
adjustment to account for the change. EPA either needs to provide a separate test procedure
adjustment to account for the effects of the test fuel change or it needs to make the necessary
revisions to the mpg equation, including incorporating the appropriate R-factor, to appropriately
adjust for the effects of the fuel change.
This situation arises as a result of the proposed regulations and hence must be resolved in the
final regulations. EPA has indicated that it does not believe it can finalize the R-factor in the Tier
3 rulemaking. While the agency has verbally committed to do a follow-up rulemaking to handle
this issue, this will result in an unacceptable delay. Such a delay would result in a changed test
procedure but without all of the conditions and consequences being established. This would
make planning for compliance difficult. We believe that language that already exists in the
current regulations plus what EPA has proposed in the NPRM sets up the conditions that
effectively should enable EPA to resolve the R-factor issue in this rulemaking. In fact these
conditions arguably require that it must be done in this rulemaking. The proposed test fuel
change satisfies the five criteria in the Rules for issuing CAFE test procedure adjustment. First,
preliminary EPA/auto industry testing has confirmed that a directional (downward) change in the
measured fuel economy of an average vehicle can be anticipated from EPA's proposed test fuel
change. Second, this directional change in measured fuel economy can be quantified once
sufficient testing has been performed to develop appropriate and reliable data. Third, the
proposed FTP changes do not fall within the parameters of 40 C.F.R. § 600.510-86(f)(3), which
is intended to close "loopholes" that may have permitted manufacturers to gain measured
improvements in fuel economy without corresponding actual improvements in fuel economy.
Fourth, the proposed FTP revisions are not the type referenced in 40 C.F.R. §600.510-86(f)(4),
which relate to test procedure changes providing manufacturers with improved ability to receive
credit for real fuel economy improvements. Fifth, the proposed FTP changes are being initiated
and required by EPA, and not implemented solely by a manufacturer on its own initiative, in its
own laboratory. Thus, EPA is required to issue appropriate CAFE adjustments for changes to the
test fuel, with the appropriate R factor of 1.0.
Organization: General Motors LLC (GM)
4-356
-------
Tier 3 Summary and Analysis of Comments
In the NPRM, EPA indicates it would be evaluating whether R-factor changes should be
included in a future action. GM sees no reason for delaying the establishment of an R-factor for
the new EPA Elx test fuel as well as CARB LEV III E10 test fuel. We believe that there is
sufficient data available for EPA to make the R-factor determination now and that the data
supports the establishment of an R-factor value of 1.0, including a new report from Oak Ridge
National Labs3 and data analysis provided in the Alliance/Global comments. 4
GM is very concerned with this language leaving the R-factor of 0.6 in place and believes EPA
must address the R-factor as part of Tier 3 rule-making instead of waiting and issuing a "Dear
Manufacturer's Letter" sometime in the future. The extra degree of certainty and expediency by
doing so is needed by the automakers to assist in their compliance planning.
(B) The manufacturer must also use this E10 fuel for fuel economy measurements, with any
appropriate corrections related to ethanol content in the fuel.
EPA has long promised that it will revisit the correction factor (R-factor) regarding ethanol
content.5 The final value of the R-factor and its timing are unknown. In Section II of these
comments, GM recommends that EPA set an R-factor of 1.0. If EPA does not set the R-factor in
time for these early certifications, then GM requests EPA give manufacturers the option to use
today's EO fuel for fuel economy testing . Although this is a burden for manufacturers (dual
testing, EO and E10), the penalty of requiring the use of an inappropriate correction factor is
much more damaging.
Commenter footnote
Oak Ridge National Laboratory, US Department of Energy, report 'Preliminary Examination of Ethanol Fuel
Effects on EPA's R-factor for Vehicle Fuel Economy,' Report Number ORNL/TM-2012/198, June 2013; accessed
at: http://info.ornl.gov/sites/publications/Files/Pub42819.pdf
4
See Alliance/Global comments, Appendix 6: Reanalysis of R Factor Data.
Our Response:
EPA agrees that an assessment of the impact of the change in test fuel qualities on the
measurement of fuel economy is needed and that test procedure adjustments should be made if
appropriate. Referring back to the comments, the second point above ".. .this directional change
in measured fuel economy can be quantified once sufficient testing has been performed to
develop appropriate and reliable data. " In the case of the Tier 3 rule there are two variables to
consider: changes in fuel quality and changes in vehicle technology as a result of the new
emission standards. Thus, we believe EPA's approach as laid out in the FRM is appropriate. It
does not change the fuel economy test requirements now specified in 600.113-12 are not
changed. Test data based on Tier 3/LEVIII fuel in Tier 3/LEVIII technology vehicles will
provide the data needed to determine any appropriate test procedure changes. The data available
now such as that discussed in Appendix 8 of the Alliance and Global Automakers comments
may inform the analysis but are not sufficient to meet the needs of the assessment since it does
not focus fully on the fuels and vehicles of interest. EPA agrees with GM that fuel economy
testing should be permitted on EO fuel if the R factor is not revised in the Tier 3 rule. In fact it
the use of EO will be required for fuel economy testing.
4-357
-------
Chapter 4: Proposed Vehicle Emissions Program
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
While no substitute for adopting a new fuel economy equation in the Tier 3 Final Rule, EPA
should also optionally allow manufacturers to continue to perform fuel economy testing on EO.
As vehicles are certified on the new test fuel, it becomes important to set up processes that will
minimize manufacturer testing burdens and maintain a level playing field for fuel economy
labeling because not all competing products will phase-in to the new test fuel at the same time.
The fuel economy label for a product adopting new ethanol test fuel should still accurately
reflect its comparative performance relative to a competing product certified on non-ethanol test
fuel. Without adopting a new fuel economy equation appropriate for ethanol-containing fuels,
EPA creates a period of uncertainty where vehicle programs, which begin development work
several years before certification, must continue to develop and establish targets based on the
assumption that EO will be used for all fuel economy measurements. Allowing optional use of EO
test fuel for a predefined period of time would allow manufactures to better manage the task of
migrating to the new fuels and prevent disruption of vehicles program that are underway when
the new fuel economy equation is made available. We recommend that EPA allow the optional
use of EO test fuel for fuel economy purposes as an option through MY 2019 to help to mitigate
level playing effects due to different products phasing-in at different times and to manage
uncertainly regarding the timing of the new fuel economy equation.
Organization: Ford Motor Company (Ford)
Although the Tier 3 NPRM proposes a new El 5 certification fuel, it does not provide a new fuel
economy (FE) equation that can be appropriately used for such low-level ethanol blends-
including CA LEV III ElO-for fuel economy, CAFE or "litmus test" purposes (EPA does note
that this could be accomplished via a "future action"). As a result, several aspects of the rule
cannot be carried out by manufacturers as proposed, including the requirement that El 5 be used
for all fuel economy purposes when required for certification (§86.113-07 (a)(l)) and that CA
E10 be used for fuel economy testing for 50-state certification in 2015 and 2016 model years
(MY) (§ 86.113-04 (a)(2)(ii)(B)). Although these requirements could be changed in the final rule
to alleviate the immediate concerns regarding their non-executable nature, the absence of firm
timing for the adoption of a fuel economy equation for E10 and E15 (Elx) still creates a period
of uncertainty for manufacturers, who must continue to use the current "EO" fuel not only for
official testing but also as a basis for development testing and program target setting, both of
which take place several years ahead of final certification. To address this uncertainty, we
recommend that in addition to prompt adoption of the new FE equation for Elx fuels, EPA also
specify a pre-defined period of time where testing on "EO" would remain optional for FE
purposes, independent of the date the new Elx FE equation is eventually finalized. Such a
provision would also allow manufacturers to phase the new fuel into FE testing in such way to
avoid mid-stream program disruptions, better manage carryover issues, and mitigate any
competitive issues related to competing products moving to the new test fuel in different model
years.
Our Response:
4-358
-------
Tier 3 Summary and Analysis of Comments
Manufacturers have presented two points of view with regard to when the potential new
requirements (including a revised "R" factor and other possible test procedure
changes/adjustments related to CAFE, GHG, and fuel economy labeling) should take effect
assuming a future rule is proposed and finalized. Some have stated that use of the new provisions
should be available for use as soon as possible after the rule is completed. This would minimize
the need for any future duplicate emission testing and put manufacturers on course for fully
aligning with the new requirements quickly. Others have asked that there be lead time provided
before the application of the new requirements becomes mandatory. The manufacturers have
expressed concern that the use of the new requirements more quickly by one manufacturer versus
another could create a competitive imbalance. At the same time, manufacturers do not
necessarily want to be forced to certify all products to the new requirements by a cut-off date
(e.g., 2020 model year) without EPA consideration of phase-in or phase-out provisions and data
carryover.
EPA understands the manufacturers' various issues and concerns in this area. Based on
the information available at this time, EPA is expecting to allow the optional use of any future
adjustments for compliance calculation and labeling purposes as soon as the potential future rule
mentioned above becomes effective. Furthermore, we expect that the mandatory use of any such
new adjustments with all Tier 3 certifications would be required for the 2020 MY. These initial
timing considerations are subject to revision based on timing of the completion of the future
action and the data and record developed in that future rulemaking.
4.5.5. Consideration of Nonroad, Motorcycle, and Heavy-Duty Engine Emissions Test
Fuel
What Commenters Said:
Organization: American Petroleum Institute (API) and the Association of Fuel &
Petrochemical Manufacturers (AFPM); Marathon Petroleum Company LP
(MFC)
EPA believes it is important that the emissions test fuel for these other categories reflect real-
world fuel qualities but has elected to defer moving forward now pending additional analysis of
the impacts of changing the test fuel specifications for the wide range of engines, vehicles,
equipment and fuel system components that could be impacted. We would suggest that one
benefit of moving to an E10 test fuel, instead of E15, would be that EPA can then identify a
single certification fuel for all gasoline engines and vehicles.
Organization: Outdoor Power Equipment Institute (OPEI)
The Outdoor Power Equipment Institute (OPEI) is pleased to provide initial comments in
response to EP A's Notice of Proposed Rule Making (NPRM) "Control of Air Pollution from
Motor Vehicles: Tier 3 Motor Vehicle Emission and Fuel Standards". Specifically, the OPEI
4-359
-------
Chapter 4: Proposed Vehicle Emissions Program
comments address changes to gasoline certification test fuels, and the implications to non-road
engines.
OPEI is the major international trade association representing the manufacturers and their
suppliers of small off-road engines and consumer and commercial outdoor power equipment
(OPE). These products are commonly found in most American households and include products
such as lawnmowers, garden tractors, utility vehicles, trimmers, edgers, chain saws, snow
throwers, tillers, leaf blowers and other lawn and garden implements. While OPE consumes a
small percentage of the nation's fuel supply, their ownership by the American consumer is
ubiquitous. As a result, there exists an inventory of as many as 200 million OPE legacy products
currently in use.
In response to Section IV.D.5 of the subject NPRM (page 29914) "Consideration of Nonroad,
Motorcycle, and Heavy-Duty Engine Emissions Test Fuel", and specifically whether a different
test fuel (other than E15) would be more appropriate for the subject categories, OPEI is pleased
to provide these initial comments:
Increasing the use of mid-level ethanol blended fuels in retail fuel distribution as well as for
future certification test fuel presents enormous risks, burdens, and challenges to OPE
manufacturers and users. As stated in EPA's El5 partial waiver decision notice (FR Vol. 75, No.
213, page 68098) "The Agency has reasons for concern with the use of E15 in nonroad products,
particularly with respect to long-term exhaust and evaporative emissions durability and material
compatibility" as a result of "more basic engine designs". Similarly, in section IV.D.5 of the
subject NPRM (page 29914), EPA recognizes small non-road engine concerns stating "Because
of lower level of technology, emissions from these engines are potentially much more sensitive
to changes in fuel quality." These concerns, as expressed by EPA, are consistent with those of
the small spark-ignited engine and outdoor power equipment industries.
Currently, there are numerous technical challenges and constraints for legacy and current
production that were developed, manufactured and certified to EPA's current EO certification
fuel. These constraints and challenges would first need to be resolved through practical
regulatory and engineering solutions (that are a long way from being developed or
implemented)—before making any certification test fuel change beyond E10 for these products.
With few exceptions, small spark-ignited engine powered OPE is not designed or warranted to
run on gasoline fuel containing more than 10 percent ethanol by volume (E10). At its fuel
distribution core, today's carburetor technology still widely limits the reliable ethanol operating
range of most small spark-ignited engines. While carburetors can be calibrated to allow product
to run on fuels exceeding E10, the limited 10% operating range does not change. Additionally,
sensitive and critical fuel system components and materials can be damaged by E-15.
Consequently, when running carbureted engines on fuel outside their designed operating range,
excessive heat, emissions and safety issues may result. Accordingly it is critical that every retail
gas station continue to offer dedicated E10 or less fuel pumps to support the growing
contingency of legacy product independent of future certification fuels.
4-360
-------
Tier 3 Summary and Analysis of Comments
Small spark-ignited engines may ultimately be designed to run on mid-level ethanol fuel blends,
given adequate lead time for design changes and assurance that the retail fuel ethanol content is
within a limited and acceptable range of the certification test fuel specified by EPA for emissions
compliance. However, any consideration of certification test fuel changes should recognize the
significant investments and lead time required of manufacturers. An extended time period would
be required to complete design and recalibration work as well as necessary consumer education
programs for each new certification test fuel. Therefore small spark-ignited engine certification
test fuel changes would need to be nationally harmonized based on a limited number of re-design
iterations, representative of established and consumer accepted ("real world") fuels, not based on
incremental changes as required by the Renewable Fuels Standard (RFS).
Non-harmonization of certification test fuel for small spark-ignited engines will further
complicate manufacturing and consumer decision making. Similar products each designed to
operate on unique certification test fuels will add consumer confusion and increase the potential
of detrimental misfueling. In line with today's market fuel, recent California Air Resources
Board (CARB) rulemaking established E10 certification test fuel requirements for 2020 and later
for small spark-ignited engines. CARB's new regulation permits the use of E10 certification test
fuel as early as 2013 and suggests no certification test fuel changes thereafter. Similarly, EPA
worked with the industry on a small spark-ignited engine direct rule amendment which permits
E10 certification fuel through 2019. While this effort is appreciated, the new rule does not
provide the harmonization certainty needed beyond 2019. OPEI recommends a continued
optional alignment path with CARB E10 certification test fuel beyond 2019.
In summary, today's E-15 certification test fuel proposed rulemaking for motor vehicles is not
appropriate for small spark-ignited engines. In response to EPA's request for comments
regarding appropriate certification test fuels for small spark-ignited engines, OPEI recommends a
continued harmonization path with CARB E10 certification test fuel for 2020 and later model
years. Furthermore, aside from further CARB E10 certification test fuel harmonization, OPEI
recommends that longer-term certification test fuel rulemaking for small spark-ignited engines
be postponed until the future of the RFS and retail fuel marketplace is clear.
At this time OPEI's recommendations are limited to EPA's specific request for comment from
the nonroad industry in section IV.D.5 of the subject NPRM (page 29914). Consequently, OPEI
may have further comments, concerns or issues with the NPRM as it continues to review the
potential impact on its members.
Organization: Truck and Engine Manufacturers Association (EMA)
Exhaust emissions certification test fuel for other heavy-duty, motorcycle, and nonroad
categories:
EPA has requested comment regarding the implications of changing the exhaust emission
certification test fuel for gasoline-fueled nonroad engines (large spark-ignited (LSI), small
sparkignited (SSI), recreational vehicle, marine spark-ignition), motorcycle, and "other" heavy-
duty engines in a future rulemaking. (78 Fed. Reg. at 29919.)
4-361
-------
Chapter 4: Proposed Vehicle Emissions Program
Those product categories fall into a broad category of engines, vehicles, or equipment that EPA
previously determined would be adversely affected by the use of gasoline blends greater than
E10. (75 Fed. Reg. No. 213, 68094-68510, November 4, 2010; 76 Fed. Reg. No. 17, 4662-4683,
January 26, 2011.) Those products are substantially different from the 2001 and newer light-duty
vehicles that EPA determined would not be adversely affected by gasoline with an ethanol
content up to El 5. EMA agrees with EPA's previous determination regarding product categories
deemed to be adversely affected by El 5, but we remain concerned about the misfueling controls
and fuel availability issues that we identified in comments provided during the El 5 waiver
determination process. (Docket No. EPA-HQ-OAR-2010-0448; Comments of the Engine
Manufacturers Association, January 3, 2011). During the waiver determination process, EPA
asserted that both the potential for misfueling, and continued availability of E10 and lower
ethanol blends, were adequately controlled. However, in the Proposed Rule, EPA suggests that it
is contemplating future rulemakings whereby the products deemed to require E10 and lower
blends would be redesigned to utilize the proposed light-duty El5 certification test fuel and the
misfueling mitigation measures would be revised.
As we have discussed with EPA, and as we set forth in EMA's comments submitted during the
El5 waiver decision process, engine manufacturers cannot make product changes associated
with frequently-changing certification test fuel or marketplace gasoline-ethanol blends. The
light-duty Tier 3 proposed change to El 5 certification test fuel does not appear to be the end but,
rather, the beginning of changes driven by the requirements of the Renewable Fuel Standard
(RFS-2). The stated premise for the proposed change to El5 certification test fuel in the pending
rulemaking is EPA's intent to reflect marketplace fuel for the next 10 to 15 years. (78 Fed. Reg.
at 29909.) However, increasing ethanol content requirements required by the RFS-2 are not
being met by increased consumption of E85 as originally projected by EPA. In addition, in
separate action, EPA and the National Highway Traffic Safety Administration have promulgated
regulations that require significant increases in light-duty fuel efficiency.
As noted in the preamble, the ARB LEV III standards require use of an El0 certification test fuel
for on-highway vehicles. Similarly, ARB recently adopted regulations that require all nonroad
spark-ignition exhaust emission certification requirements to be met utilizing E10 certification
test fuel beginning in model year 2020. To facilitate the transition to E10 fuel, ARB's nonroad
spark-ignition regulations allow the use of E10 certification test fuel at the manufacturer's option
for the 2013 through 2019 model years. EPA is in the final stages of promulgating a regulatory
change for nonroad small spark-ignition engines to allow an option to utilize the ARB E10
certification test fuel through the 2019 model year, after which time the option will no longer be
available. (Heavy-Duty Engine and Vehicle, and Nonroad Technical Amendments; Direct Final
Rule, June 17, 2013; 78 Fed. Reg. No. 116, 36369, 36384-36385.) Those changes will provide
for harmonization of EPA and ARB exhaust emission test fuel requirements in the near term, but
will result in disharmony between EPA- and ARB-required certification test fuels beginning in
the 2020 model year. As noted above, engine manufacturers cannot make product changes
associated with such a frequently changing certification test fuel scheme.
EMA recommends that EPA adopt the same E10, 9 psi RVP certification test fuel for heavy-duty
on-highway, motorcycle, and nonroad categories that EMA recommends EPA adopt for Tier 3
on-highway vehicles. EPA should adopt stable, harmonized certification test fuel requirements.
4-362
-------
Tier 3 Summary and Analysis of Comments
Evaporative emission certification test fuel for nonroad categories:
Currently, EPA specifies E10 certification test fuel for all nonroad spark-ignition engine
evaporative compliance determinations. (40 CFR Part 1060.) The current E10 test fuel is not
prescribed beyond a splash blend of EO certification test fuel and 10 percent volume ethanol.
EPA requests comments on the impact on evaporative emissions if EPA were to require
certification test fuel to be 10 psi RVP E10. (78 Fed. Reg. at 29895.)
EMA recommends that EPA specify that the certification test fuel for evaporative emission
certification for all nonroad categories be E10 with an RVP of 9 psi. The use of an E10 test fuel
with 9 psi RVP will provide manufacturers the ability to design and implement emission control
systems that meet both ARB Small Off-Road Engine and EPA Nonroad Small Spark- Ignited
Phase 3 requirements, which balances the lower 7 psi RVP California certification test fuel
against the higher testing temperature requirements imposed by ARB. EPA should propose and
adopt a change to 40 CFR Part 1060 to replace the splash blended E10 with the regulatory
certification test fuel specification. EMA looks forward to working with EPA to adopt such
changes.
Organization: National Marine Manufacturers Association (NMMA)
The California Air Resources Board (CARB) recently approved a staff recommendation
requiring all nonroad spark-ignition exhaust emission certification requirements to be met
utilizing E10 certification test fuel beginning in model year 2020. NMMA testified at the CARB
Board hearing requesting that the board also allow for the use of an alternative California
certified advanced biofuel, if available.
Organization: Renewable Fuels Association (RFA)
We are disappointed that EPA did not require that the proposed new certification fuel also apply
to non-road and small engines. We believe it is necessary to ensure new non-road and small
engine equipment are properly engineered for future commercial fuels.
Our Response:
Several manufacturers of gasoline nonroad equipment and small engines have
commented in support of harmonization with CARB's adoption of an El 0 test fuel for their
products. However, in discussions during the development of the rule, a number of
manufacturers raised concerns to EPA about the level of ethanol in the new fuel, the cost of
recertifying emission families on the new fuel, the impact on nationwide product offerings, and
the cost impact of complying with the existing standards on the new test fuel. While engines in
some of these categories employ levels of technology similar to light-duty vehicles and trucks,
the vast majority have much simpler designs, with many of the engines being carbureted with no
electronic controls. As a result, emissions from these engines are potentially much more
sensitive to changes in fuel quality.
While EPA believes it is important that the test fuel for nonroad equipment and small
engines reflects real-world fuel qualities, we have elected to defer moving forward at this time
4-363
-------
Chapter 4: Proposed Vehicle Emissions Program
with a change to test fuel for those products. It is important to better understand the effects of
test fuel specifications on the wide range of engines, vehicles, equipment and fuel system
components covered. This includes impacts on the levels of the emissions standards as well as
other issues raised by the manufacturers. EPA plans to explore such a change in a separate,
future action.
4.5.6. Consideration of CNG and LPG Emissions Test Fuels
What Commenters Said:
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
In the NPRM Preamble, EPA requests comment on the appropriateness of changing 40 C.F.R.
§86.113 to reference 40 C.F.R. Part 1065 for both natural gas and LPG test fuels. EPA further
requests comment on amending these specifications to better reflect in-use fuel characteristics,
and in particular on the appropriateness of aligning the sulfur specifications with those that apply
for gasoline test fuel. Changing the sulfur specifications would depend on establishing that the
new specification is consistent with the range of properties expected from in-use fuels (31).
Automakers agree with EPA that Sec. 86.113 should reference 40 C.F.R. Part 1065 for both
natural gas and LPG test fuels. For ease of comparison, we have included the two tables, one for
CNG and one for LPG. From "left to right" each table shows the specified fuel parameters of
interest, summaries of new final California LEV III and proposed EPA emissions cert fuel
standards, and then a brief summary of U.S. marketplace fuel quality (32).
In general, we support the proposed Tier 3 CNG and LPG certification fuel standards...
Additionally, for CNG, the dissolved Oxygen content should be raised slightly, to 0.5 mole %, to
better represent the recent market survey results. Particularly for LPG in-use vehicle operation, it
can be seen that the surveyed marketplace LPG fuel does not follow the total sulfur, reactive
sulfur (e.g., hydrogen sulfide), water content, and hydrocarbon residues detailed in the proposed
Tier 3 certification fuel specification.
U.S. CNG and LPG industries are at the front end of what appears to be the start of a long term
growth trend, so it is an appropriate time to establish market and test fuel standards, including for
sulfur.
Recomm endati ons:
- Current alternative test fuel regulations should be revised, as noted above.
- EPA should implement our recommended specifications for E85, CNG and LPG detailed
above. [This comment is also found in section 4.5.2. of this comment summary regarding FFV
("E85") emissions test fuel.] [EPA-HQ-OAR-2011-0135-4461-A1, p. 64, The suggested detailed
CNG and LPG test fuel specifications are location on pages 62 and 63 of the comment
document]
4-364
-------
Tier 3 Summary and Analysis of Comments
31 -78 Fed. Reg. 29908 (May 21, 2013).
32 - U.S. CNG and LPG Market Quality Surveys, SGS Labs, Germany, SGS Group Geneva, Switzerland
Organization: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM); Marathon Petroleum Company LP (MFC)
EPA seeks comment on the inclusion of sulfur standards for the test fuel used in natural gas
engines and certifying liquefied petroleum gas (LPG) vehicles. EPA is also seeking comment on
the appropriateness of aligning the sulfur specifications with those that apply for gasoline test
fuel (LPG test fuel sulfur levels are at 80 ppm max for heavy-duty highway engines and for
nonroad engines). The changes would also need to be consistent with in-use fuels. Lacking any
data from the in-use distribution system, we don't see how EPA can make changes to either of
these test fuel standards at this time.
Organization: National Propane Gas Association (NPGA)
EPA also requested comment on LPG emissions test fuel. Specifically, the Agency seeks input
on the appropriateness of revising 40 CFR 86.113 to reference 40 CFR 1065.720, which includes
an 80 ppm maximum sulfur specification for heavy duty highway engines and non-road
engines.5 Given that this is already a requirement and that we are unaware of any issues
associated with this sulfur level for test fuel, NPGA would not object to this change.
Organization: NGV Am erica
The proposal requests comment on whether natural gas should have to meet an in-use standard as
well as a test fuel standard. EPA regulations currently include test fuel standards for natural gas
but do not include any controls on sulfur levels.
Based on the foregoing and as explained here, it would be premature to establish test-fuel
specifications or in-use specifications for natural gas sulfur levels.
Organization: Robert Bosch GmbH, Gasoline Systems, Germany
The best approach would be a standardization on a basis of the actual CEN standardization for
CNG fuel in Europe.
Organization: General Motors LLC (GM)
GM fully supports the Alliance/Global comments on both E85 and CNG/LPG certification fuels.
Our Response:
As discussed at proposal, there are currently no sulfur specifications for the test fuel used
for certifying natural gas (CNG) vehicles. There is also no sulfur specification in §86.113 for the
test fuel used for certifying liquefied petroleum gas (LPG) light duty vehicles. The LPG
certification test fuel for heavy-duty highway engines and for nonroad engines in 40 CFR
4-365
-------
Chapter 4: Proposed Vehicle Emissions Program
§1065.720 includes an 80 ppm maximum sulfur specification. We requested comment on the
appropriateness of amending §86.113 to reference 40 CFR part 1065 for all natural gas and LPG
test fuels. We further requested comment on amending these specifications to better reflect in-
use fuel characteristics, and in particular on the appropriateness of aligning the sulfur
specifications with those that apply for gasoline test fuel. We noted that revising the sulfur
specifications would depend on establishing that the new specification is consistent with the
range of properties expected from in-use fuels.
As discussed in Chapter 5.7.2. of this Summary and Analysis of Comments document and in
Section VJ. of the preamble to the Tier 3 FRM, additional time is needed for EPA to work with
industry to collect data on current CNG/LPG sulfur content, to determine whether additional
control of in-use CNG/LPG sulfur content is needed, and to evaluate the feasibility and costs
associated with potential additional sulfur controls. Therefore, we are deferring finalizing in-use
quality and certification test fuel specifications for CNG and LPG.
4.6 Small Business Provisions
4.6.1. Lead Time and Standards
What Commenters Said:
Alliance of Automobile Manufacturers (Alliance) and Association of Global Automakers
(Global Automakers):
EPA proposed that small volume manufacturers (SVMs), defined as manufacturers with average
annual USA sales of 5,000 units or less, would be allowed a delayed start date for the proposed
Tier 3 standards. SVMs would continue to meet the current tailpipe, evaporative and refueling
emissions standards through MY 2021. Then in MY 2022, EPA proposes to require SVMs to
meet the same FTP fleet average as other manufacturers, as well as use the new certification fuel
and meeting the other Tier 3 requirements (i.e. SFTP, evaporative emissions, etc.), ultimately
requiring compliance with the 30 milligrams per mile (mg/mi) fleet average in MY 2025. EPA
would, however, also provide hardship relief on a case-by-case basis for SVMs.
EPA's proposal generally follows the Agency's past practice of allowing SVMs more lead time
to comply (because of the need for additional flexibility) but requiring SVMs to ultimately meet
the same standards as other manufacturers. This requirement ensures that SVMs are held to the
same technological requirements as large manufacturers. However, there is a flaw in this
argument, which is created by the current regulatory framework of fleet average emissions
standards, as compared to historical standards. The use of fleet average emissions standards
emerged when EPA adopted the Tier 2 fleet averaging approach for oxides of nitrogen (NOx)
emissions. Prior to Tier 2, all vehicles produced in a particular model year had to meet the same
NOx emissions standards. Starting with Tier 2, manufacturers were allowed to design individual
models to meet one of several emissions bins, each of which had varying NOx emissions values;
however, the weighted average mix of vehicles produced in a particular model year had to meet a
declining NOx fleet average year by year.
4-366
-------
Tier 3 Summary and Analysis of Comments
So a large manufacturer could produce a high performance, luxury model and certify it to a
higher emission bin and offset this model with sales of other lower emitting models. On the other
hand, a SVM, which only has limited models (in some cases, only one model) primarily sold in
niche markets, does not have other lower emitting models to use for averaging. As EPA
recognized in the proposal, "Small manufacturers that typically do not have more than one or
two emission families generally cannot use averaging to the same extent because of their limited
product offerings" (78 Fed. Reg. 29915). Consequently, the use of fleet average standards
penalizes SVMs by holding them to higher level of technology than large manufacturers that
produce the same types of vehicles. Under the proposed Tier 3 standards, EPA is expanding the
use of fleet average standards to include both NOx and tailpipe hydrocarbons (NMOG+NOx),
thus putting SVMs in even further jeopardy.
Another aspect of the Tier 3 proposal is that manufacturers, including SVMs, are allowed to
comply using an approach based on credits banking and trading. Thus, an SVM could
accumulate and bank credits that could then be used, if needed, in future years for compliance
purposes. Additionally, SVMs could purchase another manufacturer's excess credits, if available,
for compliance purposes. While these types of credits options have been available under EPA
regulations, it is our understanding that for criteria emissions, there has been very little, if any,
trading of credits among manufacturers. Some believe that the credits trading market may
become more robust in the future; however, today this can only be based on speculation. The
fundamental problem remains that, in a competitive market such as the auto industry,
manufacturers do not have much incentive to make credits (which they may need in the future)
available to competitors nor do they wish to rely on the possible availability of credits from
competitors as a regulatory compliance strategy.
Furthermore, the demands on SVMs should be balanced against their environmental footprint.
SVMs' vehicles are typically driven less per year, as well as over their lifetime, compared to
other vehicles, resulting in lower overall emissions. Plus due to the limited sales of these
vehicles, there are far less of these vehicles in the fleet. For example, a composite fleet of SVM
vehicles (5,000) driven 5,000 miles per year at a fleet average standard of 51 mg/mi would result
in approximately 1.4 tons of NMOG+NOx in one year.
footnote: According to Polk data, less than 4,000 new units were registered in
combination for Aston Martin, Ferrari, Lotus, and McLaren in calendar year 2012.
Company data suggests that average miles driven is less than 5,000 per year, on average.
footnote: For example, McLaren Mercedes SLR showed that for a sample of over 1000
vehicle between one and three years old, the annual VMT was 1,360. Aston Martin
estimated that worldwide average VMT of their vehicles since 2004 is 4,500 miles per
year, and Lotus used a range a 4,500 to 7,000 miles per year based on 200 units in
warranty.
The remaining fleet of vehicles (14.8 million) certified to an average of 30 mg/mi and driven an
average of 13,446 miles per year would result in approximately 6,580 tons of NMOG+NOx. The
difference between the two grows when considering the lifetime vehicle miles traveled by
SVMs' vehicles compared to the rest of the light-duty vehicle fleet; SVMs' vehicles typically do
4-367
-------
Chapter 4: Proposed Vehicle Emissions Program
not exceed 60,000 miles over the entire life of the vehicle, even though they will be certified for
150,000 miles of use under the Tier 3 standards. The resulting impact of SVMs fleets on air
quality is insignificant in comparison to that of the rest of the fleet.
In light of these considerations, while we support the initial relief proposed for SVMs through
MY 2021 and the hardship relief, we believe that EPA should provide additional lead time and
flexibility in MY 2022 and beyond for SVMs.
EPA continually states throughout the proposal that their main goal is to create a harmonized
program with ARB. Yet, EPA's proposed standards for SVMs do not align with ARB's
requirements and would result in more stringent requirements. In its final Low Emission Vehicle
(LEV III) rule, ARB required that SVMs meet a fleet average of 125 mg/mi in MY 2022-2024
and 70 mg/mi in MY 2025 and beyond. We recommend that EPA align with ARB's
requirements for SVMs for MY 2022-2024 to meet EPA's stated goal of harmonization and
provide the necessary lead time flexibility that SVMs need to comply with the Tier 3 regulations.
Furthermore, we believe that additional relief in MY 2025 is appropriate. We understand that the
SVMs do not believe that they can meet the 30 mg/mi fleet average in MY 2025, but, in
combination with the additional lead time and the flexibility in MY 2022-2024 (125 mg/mi fleet
average), they believe that they can meet a fleet average lower than 70 mg/mi in MY 2025.
Therefore, we support a fleet average of 51 mg/mi in MY 2025 and beyond for SVMs; this
level equates to the industry fleet average in MY 2022.
Recommendation: We recommend that EPA revise the requirements for SVMs according to the
following: In MY 2022-2024, harmonize with ARB's LEV III SVM requirements of a fleet
average of 125 mg/mi. In MY 2025 and beyond, adopt a fleet average of 51 mg/mi for SVMs.
Aston Martin Lagonda Ltd:
EPA have proposed that SVMs would enter the Tier 3 proposal for NMOG+NOx at 51mg/mi in
2022MY, then follow the slope through the following model years to 30mg/mi at 2025MY.
The NMOG+NOx standard decreasing year on year does not fundamentally work for SVMs with
their limited test groups and model ranges. It would effectively incur additional testing or early
compliance with lower standards than is necessary, mainly due to product lifecycle timings. For
example it may become necessary to certify to the proposed 2022MY 51mg/mi NMOG+NOx
requirement as early as 2018MY based on current program plans. This could reasonably be seen
as unfair.
We would propose that, as a minimum, 125mg/mi applies up to 2021MY, followed by 51mg/mi
from 2022MY onwards. Preferably, the possibility of the temporary Small Business Provision of
125mg/mi applying all the way to 2024MY would allow for even greater certification timing
flexibility.
4-368
-------
Tier 3 Summary and Analysis of Comments
During the Tier 3 and LEV III formulation periods Aston Martin, Mclaren and Lotus jointly
urged EPA and CARB to adopt less stringent standards for SVMs based on our inherently
limited fleet averaging capabilities. CARB understood this thinking and adopted a 70mg/mi
NMOG+NOx 2025MY SVM standard. EPA has ignored the opportunity to harmonise with
CARB in this ruling despite the fact that EPA always stated that this was their goal.
EPA made the assumption that SVMs would have ready access to technologies by 2022MY to
enable them to meet a 30mg/mi NMOG+NOx fleet average by 2025MY. The demonstration of
capability by the large volume manufacturers does not automatically mean that those same
technological enablers are available to SVMs in the given timeframe. It is therefore unreasonable
to base a rule on an assumption of this nature.
Within the Small Business Provision wording, EPA have acknowledged the CARB flexibilities
for SVMs and have requested comment on a relaxed standard proposal extending from 2022MY
(the entry point onto the slope for SVMs) thru 2024MY that is based on a 125mg/mi
NMOG+NOx FTP standard. Aston Martin fully supports this flexibility.
To clarify, Aston Martin would find the 2022 to 2025MY slope fleet average almost impossible
to meet. As a manufacturer with one or two test groups the slope based fleet average starting in
2022MY would effectively push us to meet the 2025MY 30mg/mi standard as early as 2022MY
in order that we do not incur burdensome extra testing of vehicles already certified for 2022MY.
There are many theoretical scenarios with regard to meeting a certain standard at a particular MY
but typically as an SVM we would have to treat a fleet average as a cap standard thereby
ensuring that all vehicles sold within either of our two test groups do not adversely contribute to
that fleet average requirement. As we have noted on a number of occasions previously the fleet
average nature of any criteria emissions standard is fine if you have many model lines, and
therefore test groups, over which to average, but SVMs generally have a (very) limited model
range leading to the fleet average becoming a cap standard. Additionally SVMs do not have the
capability to manipulate sales in order to meet the fleet average requirements and are reliant
upon sales and marketing strategies to gain sales of any and every model in their fleet without
recourse to sales manipulation.
Alternatively EPA should harmonise with the CARB SVM standard, again with a midterm
review to determine capability to meet a later, more stringent, standard.
Aston Martin still feels strongly about meeting the 30mg/mi NMOG+NOx standard from
2025MY. We would propose that SVMs are asked to meet a final 51mg/mi standard from
2025MY with a midterm review of capability with the individual SVMs in the 2022MY
timeframe to establish the possibility of meeting a 30mg/mi NMOG+NOx standard.
Aston Martin would propose that SVMs are only held to a 51mg/mi NMOG+NOx standard that
is applicable from 2022MY onwards and additionally that there is a midterm review of capability
to meet a more stringent standard after 2025MY.
There is a general understanding that due to the small number of vehicles sold by SVMs and the
low level of annual mileage accumulation by those vehicles their contribution to fleet criteria
4-369
-------
Chapter 4: Proposed Vehicle Emissions Program
emissions levels is de minimis. This fact should allow for a more lenient standard to be
promulgated without it being viewed as contentious.
California Air Resources Board (CARB):
California has traditionally provided relief to Small Volume Manufacturers (SVM) (primarily
producers of high performance vehicles), recognizing that they are at a competitive disadvantage
with larger manufacturers in terms of both investment and engineering resources. In addition,
they must compete with full line manufacturers who are able to offset the emissions of their low
volume high performance vehicles with higher volume, lower emission vehicles. For these
reasons, LEV III provides slightly less stringent emission standards for SVMs.
Unlike LEV III, Tier 3 requires SVMs to defer compliance with the Tier 3 fleet average
standards until 2022. The Notice of Proposed Rulemaking (NPRM) requests comment on this
and an alternative requirement that matches LEV III requirements through 2024 and then
requires that SVMs meet Tier 3 fleet average requirements in 2025. Since harmonization of
SVM requirements will allow these manufacturers to certify the same vehicle to both programs,
CARB recommends that U.S. EPA consider finalizing the alternative requirement by matching
Tier 3 requirements with LEV III through 2024. Since Tier 3 allows SVMs to apply for hardship
relief under certain conditions, thereby addressing their concerns on competing with full line
manufacturers, CARB will review the Tier 3 SVM requirements when finalized and determine
whether alignment between LEV III and Tier 3 requirements for SVMs beginning in 2025 is
appropriate.
Light-Duty NMOG+NOx Certification Options:
The LEV III program sets forth a separate option for certifying light-duty vehicles to the SFTP
NMOG+NOx standards, primarily to accommodate small volume manufacturer. While the
proposed Tier 3 program does not include an equivalent option, alignment between the two
programs is not necessary because Tier 3 would accommodate small volume manufacturers
through a different mechanism. Specifically, Tier 3 would delay SFTP compliance for small
volume manufacturers until model year 2022. Accordingly, CARB does not intend to propose
any changes to the LEV 111 program for this issue.
Ferrari:
Ferrari has built its reputation in the market segment in which we compete on production of
vehicles with key attributes—excellent power, superior handling, hand craftsmanship, cutting
edge to classic body design, signature engine sound and overall superior quality. This is the
DNA of all Ferrari vehicles, and these factors remain key to Ferrari's continued success in the
ultra-competitive, low-volume market segment in which we operate. Ferrari must design and
build vehicles that prospective owners aspire to purchase. Without the passion for motoring that
is built into every Ferrari vehicle, Ferrari would be "just another vehicle manufacturer." The key
to Ferrari's future success is to continue to design and build vehicles with the DNA that has
made Ferrari a vehicle people aspire to own and drive.
4-370
-------
Tier 3 Summary and Analysis of Comments
In addition, as Ferrari has previously noted in other EPA rulemakings and likely due to the DNA
of our vehicles, these are not driven to the same extent as other manufacturers' vehicles. For
example, it is exceptionally rare for a Ferrari to be driven for even 100,000 miles; indeed, an
exceedingly small number of vehicles, if any, would ever reach the full useful life of 150,000
miles. Typically, the average Ferrari owner will drive the vehicle for 3000-4000 miles per year.
It is with both the Ferrari DNA and the very low mileage of our vehicles in mind that Ferrari
offers the technical and policy comments and suggestions below.
EPA has proposed that SVMs (manufacturers with average annual USA sales of 5,000 units or
less) be permitted to delay meeting the proposed Tier 3 emission standards. During the period up
to and including 2021 MY, SVMs would have to meet current emission standards. During the
period of MY 2022-2024 EPA has indicated that a more stringent set of standards would apply.
Under the proposal, in 2025 MY, SVMs would be required to meet the same standards as large
volume manufacturers - 30mg/mi of NMOG+NOx.
The following comments are made supposing that EPA will treat the operationally-independent
manufacturers SVMs in the same way as other SVMs. As noted above, if EPA does not finalize a
provision that would treat operationally-independent SVMs similarly to how such manufacturers
are addressed for purposes of the GHG standards, then Ferrari will submit additional comments
explaining the very different impact this approach would have on our ability to meet Tier 3
standards.
Taking into account the technologies today available on the market, Ferrari has determined that
the fleet standards of 51 mg/mi of NMOG+NOx in 2021 MY and of 30 mg/mi in 2025 MY are
not technically feasible without affecting the Ferrari DNA.
The current technology of catalyst systems and relative wash-coating, even in combination with
higher loading of precious metals (200-250 g/ft3) could not guarantee our vehicles' compliance
with the fleet standards required during the MY 2022-2024 period. The only possible solution
could be a closed coupled warm-up catalyst and a main catalyst both with a high cells density
(600-800 cells/inch2), but this will affect performance, sounds and, in general, the typical overall
Ferrari engine characteristics.
Moreover, the reduction of exhaust emissions would require the development of a complex OBD
system, able to detect any defect of the catalytic converter system by analyzing the increase of
exhaust emissions in a much lower range.
In conclusion, taking into account Ferrari's expected status as an operationally-independent
small-volume manufacturer, the technologies today available on the market and the importance
of the keeping alive our models' DNA, Ferrari kindly requests that EPA establish two separate
fleet standards for SVMs/operationally-independent manufacturer:
- 120 mg/mi of NMOG+NOx in 2022 MY;
- 70 mg/mi of NMOG+NOx in 2025 MY.
4-371
-------
Chapter 4: Proposed Vehicle Emissions Program
Ferrari, being an operationally-independent SVM, typically does not have more than two engine
families. Therefore, unlike much larger manufacturers, Ferrari cannot use averaging to reach a
lower emissions standard due to its limited product offerings. In other words, for an SVM like
Ferrari, the fleet standard corresponds to the standard for each engine family. Moreover, Ferrari
models are characterized by a particularly long life cycle (about 7 years for each model). This
obviously means less capability to renew our fleet over a short period of time. As a result, it is
absolutely unfeasible for Ferrari to assure compliance with fleet standards on a year-to-year basis
as the standards become more stringent each model year.
Therefore, Ferrari kindly requests that EPA establish "stair-step" standards for SVMs and
operationally-independent SVMs that decrease over a longer time period. Ferrari suggests the
following time frames: 2017-2020 MYs, 2021-2024 MYs, 2025-2030 MYs.
Ferrari is continuously working in order to reduce the emissions in its vehicles. For this reason,
Ferrari suggests to EPA to reconsider in MYs 2017-2018 the emission standards for the MYs
2022-2030, by examining at that point in time whether SVMs and Operationally-Independent
SVMs could in fact meet a fleet standard below 70 mg/mi. Whether SVMs can meet a standard
below 70mg/mi will largely depend on the availability of new technologies able to reduce
emissions without affecting Ferrari DNA. Ferrari continues to work diligently toward this goal.
But at this time, we do not see a way to reach these standards while retaining the characteristics
that are key to Ferrari's continued success in the highly-competitive market segment in which we
operate.
In the LEV III rulemaking, CARB recognized the particular conditions of SVMs (and
operationally-independent SVMs) and, consequently, defined SVM-specific fleet NMOG+NOx
standards. Ferrari has determined that the LEV III standards for SVMs are challenging but still
feasible. Therefore, we strongly support EPA's harmonization with CARB.
Lotus Cars Ltd.:
Vehicle Services Consulting, Inc. (VSCI), who advises Lotus on small volume manufacturer
matters relating to safety and emissions issues, has prepared and submitted comments regarding
this proposed rule. Lotus fully endorses these comments.
McLaren Automotive Limited:
For manufacturers with limited product lines where fleet averaging is not possible, the proposed
30mg/mi target from MY 2025 is not a fleet average limit but effectively a minimum
requirement, hence the flexibility of fleet averaging afforded to Large Volume Manufacturers is
not available to Small Volume Manufacturers (SVMs).
There is no confidence to date that the 30mg/mi minimum requirement can be met without
jeopardising high performance characteristics of McLaren products.
There is no apparent reason for EPA to deal with SVMs differently from CARB.
4-372
-------
Tier 3 Summary and Analysis of Comments
Background:
EPA identifies Small Volume Manufacturers (SVMs) as those with average annual nationwide
sales of 5,000 units or less on the three year average of actual nationwide sales for MYs 2012-
2014.
In the NPRM EPA allow SVMs the following flexibilities.
- Postpone compliance with the standards and other Tier 3 requirements, including use of the
proposed new certification test fuel, until model year 2022.
- Apply, under the hardship provisions, for additional time to meet the 100 percent phase-in
requirements for exhaust and evaporative emissions.
- Use of assigned deterioration factors.
- Delay using the new test fuel until model year 2022.
For model year 2022 and later, Small Volume Manufacturers would be subject to the same Tier 3
exhaust and evaporative requirements as other manufacturers, including moving to the declining
FTP fleet average NMOG+NOX curve and complying with the fully phased-in standard of 30
mg/mi, as well as certifying on El5 test fuel.
EPA also propose that "in light of the CARB Board-approved implementation schedule for small
manufacturers described above, we also request comment on an option that would not provide a
permanent relaxed standard for small manufacturers, but would provide a temporary relaxed
standard matching the California standard from model year 2022 through 2024. This option
would apply to the Tier 3 exhaust emission standards starting in 2022, except that a relaxed
NMOG+NOX standard of 125 mg/mi would apply in model years 2020-2024 for FTP testing.
For model years 2025 and later, the standard would be the same as for all other manufacturers, or
30 mg/mi. Under this option, small manufacturers would have to take some action to reduce
emissions in 2022 and could postpone meeting fully phased-in Tier 3 standards until 2025."
The above can be briefly summarised as follows.
- A business with annual nationwide sales of 5,000 units or less (calculated on the three year
average of actual nationwide sales for MYs 2012-2014) will be allowed to comply with the
current standards during the period to and including MY 2021.
- During the period MY 2022-2024 EPA may apply more stringent standards.
- In MY 2025 SVMs would be required to meet the same standards as Large Volume
Manufacturers.
With regard to the above, McLaren Automotive Ltd. appreciate EPA effort to introduce
flexibility for small volume manufacturers in their NPRM regarding Tier 3 standards however,
we note that a number of concerns have not been addressed by EPA, to date. We believe that
EPA current position is strongly at risk of jeopardising Small Volume Manufacturers business
and competitiveness in the US market.
Fleet average mechanism unfairness to SVMs:
The limited product ranges of SVMs' make it impossible for them to make use of the fleet
average emissions standards applicable to large manufacturers.
4-373
-------
Chapter 4: Proposed Vehicle Emissions Program
As a consequence of fleet averaging being not feasible for SVMs, McLaren consider that the
proposed Tier 3 NMOG+NOx fleet average standard places SVMs at a significant disadvantage
compared to Large Volume Manufacturers (LVMs).
Manufacturers like McLaren will not have possibilities to fleet average their emission results
hence translating the actual EPA requirement to a fixed standard.
As a consequence, whilst being given the possibility to comply with the proposed standard later
in time, EPA is denying SVMs the flexibility allowed to the niche vehicles of LVMs that
compete directly with our products. Such flexibility will allow LVMs to produce niche vehicles
that do not have to meet what is effectively a criteria emission limit for equivalent products
produced by SVMs.
The policy issue in connection with the fleet average mechanism is simply one of fairness, since
LVMs are able to produce niche vehicles using their fleet average but SVMs are not able to do
the same.
For 30 years EPA has taken the position not to promulgate SVM-specific standards (contrary to
CARB). The issue now is whether history should take precedence over fairness.
The fairness issue is touched by EPA themselves in The Draft Regulatory Impact Analysis: Tier
3 Motor Vehicle Emission and Fuel Standards, March 2013, par 1.2 (FTP NMOG+NOX
Feasibility), EPA clearly recognise that "the proposed new emission requirements include
stringent NMOG+NOX standards over the FTP that would require new vehicle hardware and
additional control of gasoline sulphur levels in order to achieve the 30 mg/mi fleet average level
in 2025. The type of new hardware that would be required would vary depending on the specific
application and emission challenges.
Smaller vehicles with corresponding smaller engines would generally need less new hardware
while larger vehicles and other vehicles with larger engines may need additional hardware and
improvements beyond what would be needed for the smaller vehicles. Additionally, the fleet
average nature of the standards would allow more challenged vehicles to be offset by vehicles
that could outperform the required fleet averages."
Clearly, small volume manufacturers like McLaren will have to face the challenge of engineering
additional hardware and improvements without taking any advantage of any "fleet average
nature of the standards".
In the NPRM, EPA reports their position on a series of comments raised. Citing from the NPRM,
"during the SBAR Panel process, one small entity representative (SER) recommended that EPA
adopt relaxed exhaust standards for small manufacturers.
The SER noted that the exhaust emission averaging program being proposed by EPA would
allow large manufacturers that have many engine families to certify their small, niche products at
emission levels numerically higher than the standards.
4-374
-------
Tier 3 Summary and Analysis of Comments
Small manufacturers that typically do not have more than one or two emission families generally
cannot use averaging to the same extent because of their limited product offerings. The SER was
concerned that the high-performance vehicles produced by large manufacturers which they
compete against would be able to certify at numerically higher levels at less cost than the SER
would incur."
Whilst we appreciate the consideration that EPA have shown in reporting these comments, it is
notable that the EPA did not provide any comment with respect to the SER concern that "EPA
would allow large manufacturers that have many engine families to certify their small, niche
products at emission levels numerically higher than the standards."
McLaren strongly believe that this is a key issue in the extent that it highlights how McLaren
(and SVMs in general) must compete with niche products from large OEMs that need not
comply with the same standard limit. This is an unfair situation created by the current Tier 3
proposal.
Technological feasibility assessment pending:
As a Small Volume Manufacturer with limited development resources, McLaren must develop a
high performance product that produces class-leading CO2 emissions, fuel economy and
driveability.
In general the relatively low sales volumes of SVMs mean that a return on investment must come
from astute platform engineering and an extended life cycle for the base powertrain.
In EPA comprehensive "Draft Regulatory Impact Analysis: Tier 3 Motor Vehicle Emission and
Fuel Standards", main technological aspects for FTP NMOG+NOX have been studied and EPA
anticipates that "to achieve the proposed NMOG+NOX FTP emissions standards, vehicle
manufacturers will focus on three areas to reduce emissions:
reducing the emissions produced by the engine before the catalyst reaches the light-off
temperature;
- reducing the time required for the catalyst to reach the light-off temperature; and,
improving the NOX efficiency of the catalyst during warmed-up operation."
Whilst we appreciate EPA analysis on available and future technologies that will be developed in
order to meet the new requirements and whilst we understand and share the effort in controlling
the country's pollution, we believe that EPA is incorrect in believing "that these standards are
technologically feasible and can be readily achieved with the additional lead time we are
proposing as the technology would have already been demonstrated by other manufacturers, in
some cases on the very same engines used by the small manufacturers."
Given the inapplicability of fleet averaging to SVMs, the EPA's assumption that SVMs will be
able to achieve the 30mg/mi target is incorrect because SVMs cannot rely upon the technology
demonstrated by other manufacturers. In fact, as stated in previous paragraphs, the Tier 3
proposal would allow LVMs to produce vehicles with similar characteristics and performance
but with higher NMOG+NOX emissions than SVMs.
4-375
-------
Chapter 4: Proposed Vehicle Emissions Program
Independently from LVMs, SVMs are exploring technologies that would enable the Tier 3
standards to be achieved without compromising the essence of our products and are constantly
working on assessing the potential of new technologies and innovations that could contribute
towards achieving future legislative requirements and provide product differentiation in the
market place.
SVMs like McLaren have already developed their own new engine using optimised
technological and engineering solutions (Variable Valve Timing (VVT), Secondary Air Injection
(SAI), electronically controlled twin thermostats, very high precious metal loading on catalysts,
close-coupled catalyst, etc.).
With reference to the aforementioned EPA Impact assessment, after extensive research, McLaren
would respectfully suggest that more time is needed to assess the feasibility of the 30mg/mi limit
which EPA is effectively setting as a minimum requirement for SVMs.
McLaren studies to date show that, although combustion system, emissions and after treatment
research and development projects are being suitably funded and pursued aggressively by
McLaren, a technology package that will enable a niche sport car engine to both:
deliver the 30mg/mi proposed Tier 3 NMOG+NOx emissions requirements; and,
meet the very high specific power and low weight requirements demanded by this market
segment has not yet been identified.
Conclusions:
McLaren believe that the Tier 3 rulemaking should provide an alternate SVM standard.
The proposed 30mg/mi target from MY 2025 is not a fleet average limit but a minimum
requirement, hence the flexibility of fleet averaging afforded to Large Volume
Manufacturers is not available to Small Volume Manufacturers
There is no confidence to date that the 30mg/mi minimum requirement can be met
without jeopardising high performance characteristics of McLaren products.
There is no apparent reason for EPA to deal with SVMs differently from CARB
McLaren would therefore suggest that a fair and reasonable Tier 3 SVM solution would be as
follows.
Appropriate phase-in and interim SVM NMOG+NOx fleet standards
- Model year 2014-2016: 160mg
-Model year 2017-2021: 125mg
- Model year 2022+: 51mg
At this time it is not demonstrably feasible to design a high-performance niche-market
powertrain to achieve a 30mg/mi NMOG+NOx standard and as such, SVMs must be permitted
the time to assess the outstanding technical issues.
McLaren recommends that the 51mg/mi standard, consistent with CARB policy, should be set in
Tier 3 for SVMs.
4-376
-------
Tier 3 Summary and Analysis of Comments
McLaren proposes that time be permitted to allow industry development to demonstrate
feasibility of meeting standards below 51mg/mi in high performance engines from MY2025 on.
It is proposed that a review of feasibility to reduce targets below 51mg should be held by 2018.
Vehicle Services Consulting, Inc (VSCI):
We focus on two key points - That the 'fleet average' system is unfair in how it affects SVMs
compared to the niche carlines of large volume manufacturers (LVMs); and
That to remedy (in part) this inherent unfairness, for SVMs the MY 2022 51mg fleet standard
should not continue downward to 30mg unless and until there is clear evidence that SVMs - with
their limited model offerings (and hence limited number of test groups) - can in fact attain the
30mg level.
EPA has proposed that SVMs — manufacturers with average annual USA sales of less than
5,000 units - be permitted to delay meeting the proposed Tier 3 emission standards. 2
During the period to and including MY 2021, SVMs would have to meet current Tier 2 emission
standards.
During the period MY 2022-2024 EPA has indicated that a set of more stringent interim
standards would apply.
In MY 2025, SVMs would be required to meet the same standards as large volume
manufacturers — 30mg of NMOG+NOx. In other words, in MY 2025, a company that sells 25
cars per year in the US would have to meet the same standards as a company like GM that sells
2.5 million cars per year in the US.
footnote: Indeed, EPA is not even proposing that businesses which meet the Small Business
Administration definition of 'small entity' would be subject to more lenient requirements. This is
manifestly unfair.
SVMs have urged EPA to adopt less stringent standards specifically applicable to small volume
manufacturers — the philosophy that the California Air Resources Board (CARB) has adopted.
The basis for this SVM request is a key fact which EPA itself has acknowledged - that the
emission averaging program used by EPA allows large manufacturers, which have many engine
families, to certify their small, niche products at emission levels numerically higher than the fleet
standard. SVMs, on the other hand, which typically do not have more than one or two engine
families, generally cannot use averaging to reach the same result due to their limited product
offerings. SVMs have therefore been concerned that LVM high-performance vehicles with
which SVMs compete will be certified at numerically higher emissions levels and at less cost
than the levels and costs that SVM would face. Indeed, CARB recognized this very issue in
adopting its LEVIII standards, and promulgated an SVM-specific fleet NMOG+NOx standard. 4
In view of the above, SVMs recommended to EPA that Tier 3 include less stringent, SVM-
specific NMOG+NOx standards in order to reach an equitable result for SVMs, while at the
same time harmonizing with California — a clearly stated goal of EPA.
4-377
-------
Chapter 4: Proposed Vehicle Emissions Program
No one disputes that the total number of SVM vehicles per year in the US has an insignificant —
de minmis — impact on US air quality.
The NPRM's Small Volume Manufacturer Language:
From the NPRM, it is clear that EPA has not agreed with the SVMs' suggested approach.
Although EPA is proposing a delay in Tier 3 requirements as they apply to SVMs, EPA is not
proposing an SVM-specific fleet standard when the Tier 3 rulemaking phase-in ends. We believe
that EPA's decision is insufficient and inequitable.
EPA believes that the large volume end-of-phase-in standards are technologically feasible for
SVMs, and that they can be readily achieved with the additional SVM lead time being
proposed.5 We strongly disagree.
We have not seen hard evidence to support EPA's conclusion. EPA alleges that by 2022 'the
technology would have already been demonstrated by other manufacturers, in some cases on the
very same engines used by the small manufacturers'. EPA is therefore assuming that the
necessary technology will be available to SVMs. But such assumptions are an insufficient
foundation on which to base a rule's applicability. Under the Administrative Procedure Act it
would be impermissibly arbitrary and unreasonable to base the SVM standards on an assumption
of this magnitude.
Indeed, what would happen if an SVM used a Tier 2 Bin 5 engine sourced from a large volume
manufacturer? Under Tier 3, the LVM could continue to use and certify this engine based on the
LVM's fleet average. The SVM, however, would be precluded from using the engine under Tier
3 since the Bin to which the engine was certified would be above the fleet standard. This result is
simply unjust.
EPA also notes in the Tier 3 preamble that 'the compliance costs for many of these [SVM]
vehicles, even if higher on an absolute basis, may still be lower on a relative basis given the
higher average cost of the vehicles'. But this remark begs the question, since there are niche
products of large manufacturing groups that indeed compete with SVM products and are sold at
similar prices to SVM vehicles.
In short:
We strongly believe that fairness requires that small manufacturers be allowed to meet less
stringent, SVM-specific, NMOG+NOx exhaust emission standards over the long term, as
reflected in the LEVIII rules from CARB. We support EPA's harmonizing with CARB.
footnote: In view of EPA's firm stance against harmonization, however, as discussed below, we
propose herein an SVM solution that diverges from CARB, notwithstanding our firm support
both for CARB's decision and for harmonization
But the heart of the problem remains that EPA seeks to compel SVMs to meet the fully phased-
in large volume Tier 3 NMOG+NOx fleet standard for model years post 2022 (and specifically
the MY 2025 standard of 30 mg/mi). This is neither fair nor feasible.
footnote: EPA has asked SVMs to provide evidence of infeasibility. But few people would argue
that it is easy - or even possible - to prove a negative.
4-378
-------
Tier 3 Summary and Analysis of Comments
EPA's justification for only proposing extra lead-time for SVMs — even in the face CARB
deciding otherwise — has been that extra lead-time has always been EPA's policy for SVMs and
has worked adequately under Tier 2. But this response is hollow — amounting to 'we have
always done things this way so we will continue to do them this way.' EPA's rationale overlooks
how, as standards become more and more stringent, new flexibilities must be considered.
footnote: Justice Oliver Wendell Homes wrote in 'The Common Law' that: 'The truth is that the
law is always approaching and never reaching consistency.'
Conclusions:
SVMs would therefore suggest that a fair and reasonable Tier 3 SVM solution would be as
follows:
1) Appropriate phase-in and interim SVM NMOG+NOx fleet standards
Model year 2014-2016 2017-2021 2022+
160mg 125mg 51mg
2) We note that, for MY 22 an SVM NMOG+NOx fleet standard of 51 mg would be the
same as the large volume fleet standard for that year.
3) But under our suggested approach, the SVM standard would remain at 5 Img as the
interim SVM standard, while the large volume standard would continue to tighten down
to 3Omg for MY 2025.
footnote: For small entities, we urge EPA to allow a 125mg standard through MY 2024 and then
to require the 5 Img standard in MY 2025.
4) The above proposal actually brings forward by four years a more stringent (ULEV) SVM
standard (compared to the NPRM and CARB).
5) The above SVM solution is in reality a leadtime issue. We are not saying that SVMs
should not be required to meet a 30mg standard. Rather, we are taking issue with WHEN
SVMs should be so required. We believe that it should be after MY 25 and thus that any
SVM standard lower than 5 Img should be set by a further rulemaking.
6) The SVM longer term issue can be revisited in CY 2017-18 to examine at that point in
time, whether SVMs could in fact meet a fleet standard below 51 mg in the post MY
2025 timeframe.
In closing, we raise the following query: if EPA promulgated the proposed SVM language as a
final rule, then would not SVMs find themselves in the awkward position of either having to
accept an EPA conclusion that 30mg is doable in MY 2025 or to challenge immediately the EPA
rule in order to avoid waiving their right to seek judicial review of the issue?
Our Response:
As described above, commenters provided a variety of information and a number of
conclusions drawn from this information in support of our assessment that Tier 3 implementation
flexibility for small entities is warranted. These commenters include trade associations that
4-379
-------
Chapter 4: Proposed Vehicle Emissions Program
represent both large and small manufacturers of motor vehicles. None of the commenters
opposed our granting such flexibility.
We received comments from SVMs (but none from non-SVM small businesses subject to
the Tier 3 vehicle standards) about our specific proposed small entity phase-in provisions, as
well as from the Alliance of Automobile Manufacturers and the Association of Global
Automakers, arguing that the proposed phase-in for small entities did not provide adequate lead
time relief for SVMs, and that the long-term Tier 3 standards for light-duty vehicles are not
technologically feasible for SVMs. They referred to the ability of large manufacturers to offset
high emissions from high-performance, luxury models by averaging with their low-emitting
models, while competing SVM products must be designed to actually achieve low emissions
while still meeting customers' performance expectations. The limited production of these SVM
models can also result in emission control technology suppliers placing a lower priority on SVM
orders than on those of larger, high-volume manufacturers.
Because of these factors, SVMs suggested that their companies meet a slightly more
stringent NMOG+NOx standard (125 mg/mi) than what we proposed for SVMs in the early
years of the program and a permanently relaxed standard of 51 mg/mi beginning in MY 2022.
Ferrari suggested a compliance schedule for SVMs similar to that of the California LEV III
program, with either a permanently relaxed standard (matching the California LEV III 70 mg/mi
long-term standard) or a delay until MY 2030 to meet the primary 30 mg/mi Tier 3 standard
(when they suggest that SVMs could potentially comply). VSCI suggested that EPA's long-term
standard-setting be deferred to a subsequent rulemaking, characterizing it as a leadtime issue—
not a matter of "if' the SVMs should be required to meet a 30 mg/mi standard but of "when".
CARB comments supported Tier 3 adoption of its LEV III provisions for SVMs, including the
long-term 70 mg/mi standard beginning in MY 2025.
After considering the comments, we agree with SVMs that their unique logistical and
technological challenges, especially in the later years of the primary NMOG+NOx FTP
standards phase-in schedule, warrant a significant period of relaxed standards for these
manufacturers. However, we have found no fundamental reason why, given sufficient lead time,
all manufacturers, regardless of company size and vehicle characteristics, will not be able to
meet the Tier 3 standards. Thus, we are finalizing an optional program for SVMs, available to
non-SVM small businesses as well, under which they can choose an alternative 3-stage FTP
NMOG+NOx fleet average standard phase-in schedule: an initial standard of 125 mg/mi for
MYs 2017 through 2021, a more stringent standard of 51 mg/mi for MYs 2022 through 2027,
and the final Tier 3 standard of 30 mg/mi thereafter.
Commenters' argued against a 30 mg/mi long-term standard for SVMs based on their
fundamental inability to use emissions averaging to build high performance cars that have
emissions above the standards over the long run, in contrast to competing large manufacturers
with high-performance products, who sell many lower-performance vehicles with which to
average. Commenters also point out that the lack of opportunity for SVMs to average emissions
is paralleled by the lack of a robust credit trading market in this sector. Manufacturers with
credits have thus far not been greatly interested in selling them.
4-380
-------
Tier 3 Summary and Analysis of Comments
However, none of the commenters addressed the most recent experience with this issue in
the Tier 2 program, which phased in several years ago based on a bin and fleet average concept
that, as pointed out in comments from the Alliance and Global Automakers, is very similar to the
Tier 3 phase-in concept central to the SVMs' expressed Tier 3 concerns. In a 2010 review of the
impacts of the Tier 2 rule on small businesses, performed under Section 610 of the Regulatory
Flexibility Act, we concluded that the Tier 2 rule needed no revisions at that time to minimize
impacts on small entities (75 FR 79853, December 20, 2010). No comments from SVMs were
submitted in the public comment opportunity for that review. We have not received any other
information or comment that SVMs have become non-competitive in the high-performance car
market over these years as the result of Tier 2 regulations and the long-standing ability of large
competitors to gain an advantage through fleet-wide averaging under the Tier 2 program.
Though we acknowledge additional factors such as differences in the overall economic climate,
the recent setting of greenhouse gas emissions standards, and the relative stringency of Tier 2
and Tier 3 standards, this experience under the Tier 2 program bolsters our confidence that, with
the additional implementation flexibility provided in the final Tier 3 rule, SVMs will be able to
meet the long-term 30 mg/mi standard by MY 2028 without becoming non-competitive in their
market segment.
Ferrari also raised technical concerns with application of the 30 mg/mi standard, arguing
that the use of a close-coupled warm-up catalyst and main catalyst with 600-800 cells/inch2 cell
density to meet the standard would affect engine performance and acoustics. McLaren also
provided technical comments, pointing to progress with variable valve timing, secondary air
injection, electronically controlled twin thermostats, very high precious metal loading on
catalysts, and close-coupled catalysts, but suggesting that more time is needed to assess the
feasibility of the 30 mg/mi limit while maintaining the very high specific power and low weight
requirements demanded by this market segment, and to develop associated OBD systems. We
agree that the successful application of these technologies to high-performance vehicles will
require some time, and have concluded that the additional extension to the phase-in period we
are adopting for SVM compliance with the final standard in MY 2028 (under the 3-stage
alternative), coupled with the extension of credit life and availability of hardship relief,
adequately address the SVMs' concerns.
What Commenters Said:
VNG.CO (VNG):
Extended Lead Time: The NPRM proposes extending to SVMs only a special lead time of five
additional years (until 2022) before they must meet the full Tier 3 exhaust and evaporative
standards, based on the rationale that small manufacturers have limited resources available for
developing new designs to comply with new emission standards.
OEMs will also face significant engineering requirements unique to the production of NGVs
under the proposed rules, and may have similarly limited resources to devote to these
requirements if the market for NGVs is still relatively small in 2017. One of the chief impacts of
the Tier 3 program will be to move OEM gasoline vehicles from the use of port fuel injection
4-381
-------
Chapter 4: Proposed Vehicle Emissions Program
(PFI) to gasoline direct injection (GDI) technology, which can achieve significantly lower
emission levels as well as higher fuel economy. Howver, as noted in the NPC Future
Transportation Fuels study, additional investments in research and development and engineering
will be necessary for OEMs to utilize this technology for gaseous fuels because of several new
challenges unique to converting vehicles to operate on CNG, including:
- Thermal management of GDI fuel delivery system components such as the high-pressure pump
and injectors needs to be addressed as a result of no cooling offered to the gasoline injector when
operating on natural gas.
- Engine oil pumps, nozzles, and coolers may require design changes in order to manage heat
due to the high combustion chamber temperatures in natural gas mode.
- Adapting base vehicle GDI controls to CNG PFI, and accommodating PFI components in base
engine GDI architecture.
- Turbocharger designs may need to change in order to achieve a faster response time to
minimize performance differences between gas.
While it will clearly be necessary for these OEM investments to be made in order to ensure a
long-term future for NGVs, it is not clear that market growth in the near term will be sufficient to
justify these investments in time for the 2017 model year. Thus, EPA should also grant OEM
production of gaseous-fuel small-volume test groups an option for an extended phase-in period
for the Tier 3 standards identical to that proposed for SVMs.
Our Response:
We disagree with VNG's assessment that small-volume test groups of large
manufacturers should have until 2022 to comply with Tier 3. The technical challenges outlined
by VNG have to do with converting gasoline vehicles to run reliably and durably on natural gas.
Although these conversion challenges may increase for the new generation of turbocharged GDI
vehicles, we have no evidence or vehicle manufacturer comments indicating that meeting Tier 3
standards is significantly more difficult for natural gas vehicles than for gasoline vehicles. We
are providing some reduction in regulatory burden for small volume test groups in the form of
assigned deterioration factors, but not because of feasibility concerns. Rather, we believe that
assigned deterioration factors provide a sufficient alternative to the extensive process of
developing a unique factor for each low-volume vehicle model. We find no justification to delay
compliance with Tier 3 standards for larger manufacturers' low-volume models as requested by
VNG. We did not receive comments from vehicle manufacturers on this issue raised by VNG.
4.6.2. Useful Life and Assigned Deterioration Factors
What Commenters Said:
McLaren Automotive Limited
Useful life requirements are unrealistic for SVMs:
EPA is proposing extension of the regulatory useful life to 150,000 miles, from 120,000 miles.
4-382
-------
Tier 3 Summary and Analysis of Comments
Data from MP4-12c customers has shown average annual Vehicle Miles Travelled of 4100
miles. Consequently, it is extremely unlikely that McLaren vehicles will reach 150,000 miles in
their lifetime.
McLaren believe that lower VMT per annum and over a lifetime, is typical for an SVM produced
vehicle, as these are niche vehicles bought by enthusiasts and collectors who typically own a
large number of vehicles.
The combination of low sales volumes and low annual mileage means that the actual pollution
impact of McLaren vehicles as a proportion of the USA fleet is extremely small.
Consequently, a 150,000 mile useful life for SVM vehicles is unrealistic and such a durability
requirement would place an excessive liability on SVMs.
Our Response:
The Tier 3 useful life requirement is 150,000 miles or 15 years, whichever occurs first
(alternatively 120,000 miles/10 years for vehicles under 6,000 Ibs GVWR certified to a
numerically lower NMOG+NOx standard level). For vehicles driven 4100 average annual miles,
the Tier 3 useful life would elapse after 15 years, not the 36 years it would take to accumulate
150,000 miles. We expect that a 15 year vehicle life is not uncommon for such vehicles, and that
emissions control systems designed to perform well for this period are feasible in the Tier 3
timeframe. Key elements of emissions control design that are more prone to mileage-based
degradation than to time-based degradation, such as irreversible catalyst poisoning, would be
more challenging for higher annual mileage vehicles than for vehicles like those built by
McLaren that, at 4100 average annual miles, would surpass their 15 year useful life before
accumulating 62,000 miles. We believe any remaining concerns SVMs may have about
establishing durability for EPA certification will be mitigated by our allowing the use of EPA-
assigned deterioration factors and, where useful, pre-certification discussion between the
manufacturer and EPA regarding useful life durability demonstration requirements. McLaren
did not provide any specific technical information to support its statement about excessive
liability.
What Commenters Said:
Aston Martin Lagonda Ltd:
We would welcome the opportunity to use EPA assigned DFs [deterioration factors] as an
alternative to manufacturer DFs but must reiterate the need for those DFs to be kept up to date
with LVM capabilities in order for them to be useful to SVMs. EPA have indicated that they
recognize this, can it be confirmed that EPA assigned DFs will be regularly updated?
Our Response:
4-383
-------
Chapter 4: Proposed Vehicle Emissions Program
In response to Aston Martin Lagonda's request that the Agency confirm that it will keep
these factors up to date as durability data accumulates, we can state that we are committed to
periodically updating and publishing these assigned deterioration factors.
4.6.3. Reduced Testing Burden
What Commenters Said:
VNG.CO (VNG):
Reduced OBD Testing Requirements - Today: Current EPA rules allow alternative fuel
conversion SVMs to meet much less stringent OBD requirements than OEMs certifying small
volume test groups, with as few as four tests of major diagnostic monitors (including fuel trim
lean & rich, catalyst deterioration, engine misfire, and oxygen sensor) instead of the many more
typically required for OEM certification. As illustrated above, the NPC Future Transportation
Fuels report notes that "certification, OBD, aftertreatment and calibration" are a significant
source of per-vehicle incremental costs for current OEM NGV production. This is particularly
the case for pickup trucks, where it is the single largest cost component at about $4,500 per
vehicle, or nearly 40% of the total incremental cost.
Harmonizing OEM small vehicle test group OBD testing procedures with the reduced OBD
testing requirements for alternative fuel converters is consistent with the 'level playing field'
provided in most other respects between these categories of vehicles, and represents a major
opportunity to reduce NGV incremental costs for consumers without sacrificing the efficacy of
the certification process.
footnote: Existing EPA regulations (40 CFR 86.1806-01) already afford flexibility in OBD
testing requirements and allow EPA to reduce them if requested to do so by the applicant,
particularly in the case of alternative fuel vehicles. EPA should encourage both SVM converters
and OEMs certifying NGVs in small-volume test groups to request and utilize these reduced
OBD requirements.
Our Response:
We proposed to apply CARB's OBD requirements to Tier 3 vehicles, with an exception
for small alternative fuel vehicle converters to instead meet our existing OBD requirements (40
CFR 86.1806-05). VNG objected that the proposed exception disadvantages larger
manufacturers and should be made equally available to all vehicle manufacturers' small volume
test groups. We expect that larger manufacturers that produce alternative fuel vehicles will be
familiar with CARB's OBD requirements and well-positioned to implement these requirements
in Tier 3. We note that larger OEMs themselves did not request an extension of this provision to
themselves. We are finalizing the exception to the Tier 3 OBD requirements as proposed.
However, as discussed in Chapter 4.4, we are willing to consider reduced OBD requirements for
small volume test groups of OEMs (perhaps on a transitional or interim basis) in the future,
although we cannot at this time commit to applying the changes under consideration in
California.
4-384
-------
Tier 3 Summary and Analysis of Comments
What Commenters Said:
Aston Martin Lagonda Ltd:
We accept the rationale behind the small business waiver for PM testing, but of course the
statement of compliance for PM standards also requires some knowledge of the capability of the
EDV (and indeed all engines/permutations contained in that test group) for each test group which
in turn drives the need for testing!
Our Response:
We agree that a manufacturer's statement of compliance with the PM standard could
depend on testing to provide confidence in its validity. It is not our intent to prescribe or limit
actions a manufacturer considers necessary to gain this confidence. Nevertheless, we believe the
waiver from certification testing will reduce burden for small businesses.
4.6.4. Hardship Relief
What Commenters Said:
Alliance of Automobile Manufacturers (Alliance) and Association of Global Automakers
(Global Automakers):
In light of these considerations [detailed above in Chapter 4.6.1], while we support the initial
relief proposed for SVMs through MY 2021 and the hardship relief, we believe that EPA should
provide additional lead time and flexibility in MY 2022 and beyond for SVMs.
Aston Martin Lagonda Ltd:
Aston Martin appreciates the Hardship provision being proposed and the flexibilities contained
in the wording. If we are talking about more time to comply as a hardship provision I would
point EPA towards the temporary relaxed NMOG+NOx standard noted in the NPRM and
commented on in the first paragraph above, and the fact that this provision alone would give
SVMs added flexibility to accommodate the need to meet the 30mg/mi standard by 2025MY.
Additionally (and also as noted above) I believe that SVMs would appreciate a midterm review
of capability that would allow them to give EPA adequate notice of any hardship requirement.
Ferrari:
Ferrari faces significant technical challenges in meeting the 30mg/mile standard. Ferrari
produces only two engine families, meaning that the ability to average emissions and offset an
engine family with higher emissions is extremely limited, particularly when compared to large
volume manufacturers. In addition, Ferrari must continue to produce vehicles that still retain the
Ferrari DNA and meet customer expectations for performance.
4-385
-------
Chapter 4: Proposed Vehicle Emissions Program
To reflect the unique challenges faced by a small volume manufacturer producing high-
performance vehicles under the proposed Tier 3 NMOG+NOx standards, Ferrari proposes
revised regulatory language that would provide compliance flexibility to small volume
manufacturers facing technical hardship. As currently written, the economic hardship provisions
do not adequately address the technical issues we face. Therefore, Ferrari suggests that EPA
establish a similar but separate provision allowing a small volume manufacturer to request an
extended compliance deadline for the NMOG+NOx fleet average standard beginning in MY2028
if the manufacturer demonstrates technical—rather than economic—hardship.
Lotus Cars Ltd.:
Vehicle Services Consulting, Inc. (VSCI), who advises Lotus on small volume manufacturer
matters relating to safety and emissions issues, has prepared and submitted comments regarding
this proposed rule. Lotus fully endorses these comments.
Vehicle Services Consulting, Inc (VSCI):
We commend EPA for proposing an SVM hardship provision which will allow SVMs to apply
for additional lead-time on a case-by-case basis.
footnote: EPA has proposed a hardship provision for SVMs — see reference in footnote 5 above.
Under the hardship provision, small businesses would be allowed to apply for additional time to
meet the 100 percent phase-in requirements for exhaust and evaporative emissions. All hardship
requests would be subject to EPA review and approval. Petitions for hardship relief would need to
be made in writing and submitted before the earliest date of potential noncompliance. The request
would need to identify how much extra time is requested. It must also include evidence that the
noncompliance would occur despite the manufacturer's best efforts to comply, and must contain
evidence that severe economic hardship would be faced by the company if the relief were not
granted. Thus, the hardship provision could provide the opportunity for small businesses to obtain
more time to comply with the new Tier 3 standards. While the existing hardship provisions in part
1068 limit the extra time that can be requested to 1 year, EPA is proposing that such limit not be
included in the Tier 3 rule. SVMs support this proposal. However, we point out that while a
hardship provision is a significant step in the direction of equity and fairness, the hardship
provision by itself does not afford SVMs a sufficient basis on which to reasonably develop
business and engineering plans beyond 2021.
Our Response:
Commenters generally supported the proposed provisions for hardship relief, within the
context of a revised approach to small volume manufacturer lead time, discussed in Chapter
4.6.1. In response to Ferrari's additional request to expand the proposed hardship relief
provision, we believe the hardship provision already addresses Ferrari's concern over non-
economic technical hardship. The economic case we are requiring, involving a showing of
"severe economic hardship" will not be difficult for an SVM having problems certifying to Tier
3 standards to make, considering the limited product offerings of the SVMs. Inability to certify
and therefore sell vehicles would be almost certain to create a severe economic hardship.
4-386
-------
Tier 3 Summary and Analysis of Comments
4.6.5. Applicability of Flexibilities
What Commenters Said:
Advanced Biofuels USA (ABFUSA):
While nothing in these regulations should throw up unneeded barriers to true low-volume
manufacturers, EPA should proceed cautiously on the determination of what is a "low-volume"
manufacturer. There are three circumstances that need to be considered.
1. While a manufacturer listed as a US company may fall in the <5,000 or 5,001-15,000
US vehicles sold category it's worldwide sales total may well exceed that. Aston Martin
most probably falls in that category.
2. While a manufacturer listed as a US company may fall in the <5,000 or 5,001-15,000
US vehicles sold category it is actually a part of a larger motor vehicle company that sells
many more vehicles that 5,000 or 15,000 worldwide. Ferrari USA, which is part of Fiat,
Porsche USA, which is part of Volkswagen/Audi, and Jaguar USA, which is part of Tata
all fall in that category.
3. While a manufacturer listed as a US company may fall in the <5,000 or 5,001-15,000
US vehicles sold category it is actually a part of a larger motor vehicle company that has
a business plan to greatly increase US sales above the initial "beachhead" number of a
few thousand in a short timeframe.
In each of these cases, especially number 3, EPA would run the risk of approving a once low-
volume vehicle that turned into a US market 250,000/year seller two or three years later without
adequate long-term performance data that could predict consumer in-use issues including
potential recalls.
California Air Resources Board (CARB):
Independent operational manufacturers are a subset of small volume manufacturers. These
manufacturers would qualify for small volume manufacturer status, except they are owned by a
larger manufacturer and their sales are currently aggregated with the larger manufacturer when
determining emission compliance. Both the California and federal greenhouse gas programs
allow these manufacturers to meets small volume manufacturer emission requirements if they
can demonstrate that they operate completely independently from the owner manufacturer
(design and manufacture of powertrains, receive no fiscal support from owner manufacturer,
etc.). The LEV 111 program extends SVM status to these manufacturers for criteria emission
requirements and CARB recommends that this manufacturer category be included in the Tier 3
program. [EPA-HQ-OAR-2011-0135-4261-A1, attachment pp. 1-2]
Ferrari:
An Operationally-Independent Manufacturer for GHG Purposes Should Have the Same Status
for Tier 3 Emission Standards. EPA has proposed to allow small volume manufacturers that
have demonstrated operational independence for purposes of the Agency's greenhouse gas
standards to also have the option of complying with the SVM emission standards established
4-387
-------
Chapter 4: Proposed Vehicle Emissions Program
under the Tier 3 program (78 Fed. Reg. at 29,917). Ferrari supports this proposal and
recommends that EPA issue a final rule that considers manufacturers who have met the criteria
for operational independence under 40 C.F.R. § 86.1838-01(d) for purposes of the GHG program
as also operationally-independent and similarly eligible for SVM provisions under the Tier 3
program.
Ferrari supports this approach for several reasons. First, there is no reason why a manufacturer
that has been found by EPA to be an operationally independent SVM for purposes of the GHG
program should not be treated the same way under the Tier 3 exhaust emission standards
program. Further, this option would be consistent with EPA's past practice of allowing SVMs to
follow an alternate path to certification using assigned deterioration factors and other, more
streamlined approaches. It also would ensure that manufacturers are subject to the same
certification procedures for both the GHG and Tier 3 emission standards. Otherwise, a
manufacturer could potentially be treated as an SVM for GHG purposes but not for the Tier 3
program; this could result in unnecessarily duplicative or even conflicting certification test
procedures.
In addition, the comprehensive examination of a manufacturer's relationship (or lack thereof) to
other related manufacturers that is required before a determination of operational independence
can be made highlights the comparatively low sales volume of SVMs and the need for the more
flexible SVM compliance options. Also, the extremely detailed application process for a
determination of operational independence ensures that manufacturers cannot abuse the system
and the SVM compliance option.
Finally, adopting this option would be consistent with the approach taken by the California Air
Resources Board (CARB) in its 2012 LEV III final rule. In that rule, at 13 C.C.R. § 1900(22),
CARB revised the definition of SVM to include operationally-independent manufacturers such
as Ferrari. CARB's definition applies both to the GHG program and to emission standards for
conventional exhaust pollutants. Harmonizing with California would allow manufacturers to
more efficiently certify vehicles to both EPA and CARB emission standards.
To accomplish this goal, Ferrari recommends that EPA revise the language in 40 C.F.R. §
86.1838-01 (a) to clarify that a manufacturer that has fulfilled the requirements for being
classified as an operationally-independent SVM shall be treated as an SVM for the purposes of
all EPA emission standards. Ferrari suggests that EPA revise § 86.1838-01(a) as follows, with
our recommended revision in underlined text:
§ 86.1838-01 Small volume manufacturer certification procedures.
(a) The small-volume manufacturers certification procedures described in paragraphs (b)
and (c) of this section are optional. Small-volume manufacturers, including small-volume
manufacturers that have satisfied the criteria for operational independence under
paragraph (d) of this section, may use these optional procedures to demonstrate
compliance with all of the general standards and specific emission requirements
contained in this subpart.
Finally, these comments are drafted based on the assumption that EPA will treat operationally-
independent SVMs as eligible for the Tier 3 SVM provisions, similar to the way in which EPA
4-388
-------
Tier 3 Summary and Analysis of Comments
has treated such SVMs under the GHG program. If EPA does not issue a final rule adopting this
approach, then Ferrari will submit additional comments because this would have a vastly
different impact on our operations and vehicles.
Our Response:
We requested comment on extending eligibility for the Tier 3 SVM provisions to small
manufacturers that are owned by large manufacturers but are able to demonstrate that they are
operationally independent. We established such a provision in the light-duty greenhouse gas
(GHG) program, and CARB did so in LEV III. Comments from CARB and Ferrari supported
this extension. No commenters opposed it; however, Advanced Biofuels USA recommended
caution to avoid advantaging SVMs capable of leveraging parent company resources to
drastically increase U.S. market share within 2-3 years. Given the establishment of this
provision in our GHG program, and the value of this extension for harmonization with LEV III,
we are adopting this change into Tier 3 using the same eligibility criteria as in our GHG
program, set forth in 40 CFR 86.1838-01(d). We believe these criteria are sufficiently strict and
objective to address the concerns expressed by Advanced Biofuels USA.
What Commenters Said:
National Propane Gas Association (NPGA):
Among the changes proposed by EPA is the incorporation of CARB's OBD requirements into
the federal EPA regulations, thereby requiring compliance with CARB's provisions. EPA is
proposing that small business alternative fuel converters have the option to continue to comply
with EPA's existing OBD requirements when the Tier 3 standards become effective. EPA goes
on to state that the OBD requirements would then have to be met by small entities by the 2022
Model Year. Likewise, as it pertains to exhaust and evaporative emission standards, EPA
proposes to allow small manufacturers to postpone compliance with the standards and other Tier
3 requirements until MY 2022, but then require them to meet all the Tier 3 standards for MY
2022 and later.
NPGA believes that EPA should increase the threshold for small volume manufacturers beyond
the 15,000 units per year level and that the flexibilities allowed small businesses as noted in the
NPRM should be allowed to remain in place beyond MY 2022. Small businesses and small
volume manufactures have limited resources, lower production volume over which to spread
their compliance costs and limited product lines, which hinder their ability to take advantage of
the phase-in and averaging, banking and trading provisions. These changes would facilitate the
increased availability of alternative-fueled vehicles.
VNG.CO (VNG):
Small-Volume Manufacturer Threshold: Under current Tier 2 rules, compliance flexibilities are
available to SVMs producing up to 15,000 vehicles per year or OEMs producing up to 15,000
units of a given small volume test group. The NPRM proposes to decrease this threshold to 5,000
4-389
-------
Chapter 4: Proposed Vehicle Emissions Program
vehicles for SVMs (but not small volume test groups), based on the rationale that this definition
was used in the recent 2017-2025 greenhouse gas regulations and that "the 5,000 unit cut-off for
small volume manufacturers would include all of the vehicle manufacturers, ICIs, and alternative
fuel converters that currently meet the applicable SBA definition as well as a few additional
companies that have similar concerns to small businesses."
While EPA is certainly correct that a 5,000 vehicle cutoff would pose no risk to conversion
companies currently, the proposed rules will cover the years 2017 and onwards, at which point
there could be significant growth in the NGV market. Under such a scenario, this changed
threshold could potentially limit the ability (or willingness) of SVMs to scale up production by
forcing them to choose between increased sales and increased regulatory burdens at much lower
production level. Finally, 15,000 vehicles is still just 0.1% of an annual US light-duty vehicle
market of 15 million vehicles. For all these reasons, EPA should continue to define SVMs as
producing up to 15,000 vehicles per year.
Our Response:
We do not believe that the SVM relief provisions are so significant as to cause the self-
limiting of sales suggested by VNG, except perhaps for a company very near the threshold.
Even if this were to happen, moving the threshold to 15,000 would not prevent the same dynamic
from potentially occurring at that sales level. Furthermore, our use of a three-year average of
sales for determining SVM eligibility protects the SVMs from being disadvantaged for having an
especially good year not reflective of its long-term sales trend. See the 2017 and later light-duty
GHG final rule for a discussion of our basis for adopting the 5,000 vehicle threshold (77 FR
62793, October 15, 2012). NPGA suggested that the small business sales threshold be increased
to a level over 15,000 but did not provide any detailed reasoning or indicate what that level
should be. For these reasons, we are not altering the SVM sales threshold from what we
proposed. The issue of extending small business flexibilities beyond MY 2022 is discussed in
Chapter 4.6.1.
What Commenters Said:
VNG.CO (VNG):
Level Playing Field Between Small-Volume Manufacturers and Small-Volume Test Groups:
While existing regulations ostensibly maintain a "level playing field" between converters (small
volume manufacturers) and OEMs producing limited volumes of a vehicle model (small volume
test groups), there is an important difference in current administrative requirements for on-board
diagnostic (OBD) systems testing that disadvantages OEM small-volume production. Moreover,
the current Tier 3 proposals risks entrenching this inequality for the long term and introducing
others for both OEMs and converters.
Because of the current lack of CNG refueling services and a more general uncertainty over the
market appeal of NGVs, automakers have been cautious about producing NGVs in high
4-390
-------
Tier 3 Summary and Analysis of Comments
volumes. They therefore utilize the same types of low-volume manufacturing processes used by
alternative fuel conversion companies - and face similar per-vehicle cost challenges.
In its recent, comprehensive Future Transportation Fuels study, the National Petroleum Council
of the U.S. Department of Energy developed a detailed breakdown of the sources of incremental
costs for NGVs under both the current low-volume production model as well as under a medium-
term (2020) high-volume original equipment manufacturer (OEM) production scenario. As
shown below in the examples of a small car and a pickup truck, at low volumes (2010) these
costs are dominated by fixed costs for design as well as certification and on-board diagnostic
(OBD) system costs that must be spread over a small volume of vehicles. These incremental
costs can fall to zero or near-zero with the move to high-volume production (i.e., the 2020
scenario), under which these fixed costs would be spread across a similar number of vehicles as
high-volume gasoline-fueled cars and trucks.
Sources of NGV Incremental Costs
Cost Component
CNG Fuel Storage (Type I)
Compound Mass
Engine Hardening
Certification, OBD, Aftertreatment and Calibration
Engine & Vehicle Fuel Delivery System & Safety Integration
Total Incremental Cost
Pickup Trucks
2010
$2,144
$945
$310
$4,481
$3,788
$11,668
2020
$2,144
$899
$0
$0
$419
$3,462
Small Cars
2010
$1,590
$655
$155
$1,535
$3,670
$7,605
2020
$1,365
$623
$0
$0
$345
$2,333
Source: National Petroleum Council, Future Transportation Fuels: Natural Gas Analysis pp. 53-58
While OEMs are much larger entities than conversion companies and could potentially produce
NGVs at much lower incremental costs with high-volume, assembly line production, automakers
are highly competitive and are unlikely to risk committing resources required to move to high-
volume production until natural gas refueling availability improves and the mass-market appeal
of NGVs is better established. This results in a "vicious cycle," with automakers and converters
producing low volumes of high-priced NGVs which limits their market appeal and which in turn
limits the development of natural gas fueling services infrastructure, which in turn reinforces the
low-volume/high-cost production model.
This vicious cycle could be changed into a "virtuous cycle" by ensuring a truly level playing
field for both OEMs and converters that would allow for fully-harmonized, reduced regulatory
barriers to entry for both at production volumes of up to 15,000 vehicles per SVM or per OEM
small-volume test group. Encouraging gradually greater quantities of small-volume NGVs to be
introduced by both converters and OEMs would stimulate accelerated development of CNG
fueling services infrastructure, which in turn would increase the market appeal of NGVs, which
would ultimately lead to the production of low-cost NGVs on a high-volume basis that would
then meet full large-volume certification requirements. This dynamic has already been successful
in building the market for NGVs in Europe, where several OEMs have developed the market
through the production of a wider range of models on a small-volume conversion basis.
4-391
-------
Chapter 4: Proposed Vehicle Emissions Program
40 CFR §86.1838-01 ostensibly provides such a path by allowing OEMs to use small-volume
manufacturer (SVM) certification procedures - including the use of assigned deterioration
factors and reduced in-use testing requirements - for test groups of vehicles which total less than
15,000 units, called small-volume test groups. However, as discussed below, in the area of OBD
system demonstration, EPA allows for significantly reduced requirements for only SVM
converters. Moreover, the current NPRM risks entrenching this arbitrary distinction between
OBD requirements for SVM converters and OEM small-volume test groups while creating other
inequalities between these two classifications. This runs contrary to the EPA's own goal of
maintaining "a level playing field for OEMs and conversion manufacturers," as expressed in its
revised aftermarket certification rules in 2011:
EPA also believes that a certification demonstration requirement for new vehicle and
engine conversions is prudent to maintain a level playing field for OEMs and conversion
manufacturers. The certification requirement for new vehicle and engine conversions
reduces any incentive that might otherwise exist for OEMs to circumvent requirements
by certifying a traditional configuration and then converting it, rather than certifying the
alternative fuel configuration in the first place.
This level playing field should work both ways, ensuring that both types of entities can produce
small volumes of alternative fuel vehicles with reduced and harmonized regulatory obligations.
Some disharmonies included in the NPRM risk undermining incentives for OEMs to produce
NGVs in small volumes; others have the potential to risk discouraging conversion companies
from scaling up their own production. In both cases, it is critical for EPA to avoid placing
unnecessary regulatory burdens on the production of NGVs at this early stage of the market, or it
risks making the present "vicious cycle" a permanent feature of the market, with NGV
production limited to small-volume, high-cost vehicles with limited customer appeal.
A true level playing field can be accomplished by making changes to the proposed Tier 3 rules in
the following three areas:
1. OBD testing requirements
2. Extended lead time provisions
3. Small-volume production thresholds
The Tier 3 rules are projected to have enormously positive impacts for local air pollution issues.
However, if not implemented carefully, they could also inadvertently impede the development of
vehicles fueled by natural gas - a critical clean fuel for the highly efficient internal combustion
engines of the future as well as a bridge to zero-emission hydrogen FCEVs. Conversely, if these
rules are implemented thoughtfully, they could provide an opportunity for EPA to reinforce the
support for NGVs it showed in the 2017-2025 GHG rulemaking, as well as the support for this
technology that President Obama has repeatedly asserted.
The proposals outlined in these comments would significantly reduce the regulatory barriers to
entry for OEMs producing limited quantities of NGVs by more fully harmonizing rules for
small-volume test groups with the rules for small-volume conversion manufacturers -
particularly in the critical area of OBD system testing requirements. Additionally, by avoiding
unnecessary sulfur requirements for CNG fuel, EPA will assure gas producers, utilities, and
retailers that the development of the NGV fueling market can proceed without unexpected
4-392
-------
Tier 3 Summary and Analysis of Comments
regulatory roadblocks. In combination, these efforts can help transform the current "vicious
cycle" of low-volume, high-cost NGVs and sparse CNG fueling into a "virtuous cycle" of high-
volume, low-cost NGVs with plentiful CNG availability.
Our Response:
The flexibilities we are adopting for small companies are prompted by their limited
access to engineering resources and credit opportunities, as discussed in Section IV.G of the
preamble to the final rule. These issues are far less present for larger OEMs, even for their small
volume test groups. Many new technologies must undergo start-up challenges similar to those
described by VNG for NGVs, and manufacturers will often discount the retail price of vehicles
using these technologies so as to help overcome these challenges, or will partner with others to
spread development and implementation costs. Given that manufacturers have significant
latitude in defining test groups, extension of small business flexibilities to large manufacturers'
small volume test groups could provide large unintended incentives to artificially create small
volume test groups and seriously hamper the Tier 3 program, and could introduce new
competitive issues, including for the small companies these provisions are designed to help. We
believe that the small manufacturer flexibilities should remain firmly focused on the specific
issues that arise for small businesses under the Tier 3 program. We note that larger OEMs
themselves did not argue for extension of these flexibilities to their small volume test groups.
4.6.6. Credits
What Commenters Said:
Ozone Transport Commission (OTC):
EPA's proposal would prevent the carryover of Tier 2 credits for use in meeting the Tier 3
standards. However, while large volume manufacturers may be well-positioned to achieve
compliance with the Tier 3 standards using only the credit flexibility mechanisms included in the
proposed rule, small volume manufacturers may have more difficulty achieving the reductions
necessary for early credit generation. These manufacturers have limited model lines with which
to comply with fleet average requirements, and lesser availability of investment and engineering
resources to meet more stringent standards. Therefore, they may be disadvantaged under the
current credit structure. Therefore, OTC recommends that EPA allow carryover of Tier 2 credits
for small volume manufacturers with a mechanism similar to California's carryover provisions
between the LEV II and LEV III programs, under which credits are initially available at full
value, but then discounted over time. OTC believes that this option best accounts for the
challenges small volume manufacturers may face in complying with the program, and levels the
playing field for firms that may be unable to utilize the currently proposed flexibility
mechanisms.
Our Response:
4-393
-------
Chapter 4: Proposed Vehicle Emissions Program
Under our Tier 2 program, credits may be earned for NOx but not for NMOG+NOx, so
allowing manufacturers to port excess Tier 2 credits into Tier 3 to demonstrate compliance with
NMOG+NOx standards would introduce significant complexities. Instead we decided to allow
the generation of Tier 3 NMOG+NOx credits early, and this flexibility is available to large and
small manufacturers alike. However, as pointed out by a number of manufacturers, the ability of
SVMs to take advantage of averaging, banking, and trading flexibilities of any sort is limited by
their limited product offerings, and for this reason we are adopting special flexibilities for SVMs
that do not depend on credits. Additional discussion of comments requesting conversion of pre-
Tier 3 credits to Tier 3 NMOG+NOx credits can be found in Chapter 4.2.5.
Ferrari:
Ferrari suggests a debit/credit trading mechanism in order to be able to guarantee compliance
with the emissions standards proposed by EPA for 2022 and subsequent MYs. A debit/credit
trading scheme would enable Ferrari to internally generate credits through over-compliance, or
obtain credits from other manufacturers for their over-compliance with the fleet average
emission standards. This approach would not result in any harm to the environment as the same
overall net emissions level would be maintained. In fact, due to the very low mileage that is
typical of most Ferrari vehicles, a credit trading program could result in a net environmental
benefit because the credits would cover full useful life emissions while the mileage of most
Ferraris does not even approach 150,000 miles.
Ferrari proposes that EPA allow NMOG+NOx credits earned by small volume manufacturers to
expire after ten years instead of five. As explained in our November meeting with you, meeting
the 30mg/mile standard in 2028 would require the introduction of vehicles with much lower
emissions as early as 2022, which may be extraordinarily difficult to do. The extended credit
lifespan would provide Ferrari with the lead time to ensure compliance with the proposed
standards without sacrificing the performance or character that is expected of each Ferrari
vehicle. Ferrari agrees that any credits earned during an extended compliance period may not be
sold, traded, or otherwise transferred to another manufacturer.
Our Response:
As discussed in Section IV.A.T.m of the preamble to the final rule, in response to
comments we received on the proposal, we are extending the life of NMOG+NOx credits earned
in MY 2017-2024 (MY 2018-2024 for larger light-duty vehicles), to as long as 8 years,
compared to the 5 years we proposed. For SVMs choosing to take advantage of the 3-stage
compliance option, these credits could be earned beginning in MY 2022 when the second-stage
standard takes effect. We believe that this credit life extension substantially addresses Ferrari's
concerns. The low-emissions vehicles they would introduce in MY2022 (as well as those sold in
MY2023-2025) can earn credits that may be used anytime through MY2030. This provides 3
model years, MY2028-MY2030, in which these banked MY 2022-2025 credits can be used to
demonstrate compliance with the 30 mg/mi standard under the SVM 3-stage compliance option.
4-394
-------
Tier 3 Summary and Analysis of Comments
4.7 Compliance Provisions
4.7.1. Exhaust Emission Test Procedures
What Commenters Said:
Alliance of Automobile Manufacturers (Alliance) and Association of Global Automakers
(Global)
NMOG+NOx contribution factor for Plug-In Hybrid Electric Vehicles (PHEVs): PHEVs only
use electricity for a portion of their driving time and generate no exhaust pollution (NMOG,
NOx, PM, etc.) during this portion. Thus, these vehicles produce less real-world pollution than a
vehicle certified to the same standard that does not have the off-vehicle charge capability. For
example, a PHEV that operates half the time on electricity only, will emit half the criteria
exhaust emissions. Although the benefit is clearly operator dependent, it could be roughly
approximated by the battery-only range of the vehicle (i.e., a longer battery-only range will result
in more battery-only miles).
Recognizing this characteristic of PHEVs, the LEV III regulations include the NMOG+NOx
Correction Factor for Off-Vehicle Charge Capable Vehicles in §1961.2(b)(l)(B)l.c.2. Under
LEV III, the NMOG+NOx emissions used to calculate the fleet average emissions are reduced
by up to the next lowest category based on the battery-only range (i.e., longer range = greater
reduction). For example, this could allow a PHEV certified to Bin 125 to be treated in the fleet
average as a Bin 70. Regardless of range, the lowest value that a Bin 125 could use in the fleet
average is 70 mg/mile; it could not, for example, be treated as a Bin 50. Thus, the correction
factor in ARB's LEV III regulations is conservative but is an appropriate recognition of the
technology's benefits. EPA did not include a similar correction factor at all in its proposed
regulations and should adopt this allowance in the final rule.
Manufacturers and ARE are gathering data on PHEV usage. This data is expected to influence
the NMOG+NOx contribution factor. For example, it's possible that a Bin 70 PHEV with a long
range should use 35 or 25 mg/mile for the fleet average calculations.
Manufacturers will introduce significant volumes of PHEVs in the coming years. A change to
incorporate the NMOG+NOx contribution factor for PHEVs is essential to harmonize the
programs and prevent manufacturers from being forced to manage separate federal and
California fleets.
Recommendation: We recommend that EPA harmonize with the LEV III allowance for PHEVs.
Once more data is available, we would also like to work with both agencies to determine if the
cap at the next lowest category is appropriate or should be eliminated.
BMW of North America, LLC
Plug-in hybrid vehicles offer a promising technology on the path towards ambitious GHG targets
and further reduction of oil dependency in the transport sector.
4-395
-------
Chapter 4: Proposed Vehicle Emissions Program
Depending on the trip length and zero emission driving capability of the vehicle, a portion of
vehicle miles travelled (VMT) with PHEVs is based on mere use of electricity hence reducing
the overall exhaust emissions of the PHEV substantially. CARB acknowledges and rewards the
electric drive share of PHEVs through allowing for an HEV contribution factor which based on
Zero-Emission VMT reduces the NMOG+NOx value for PHEVs.
BMW proposes harmonizing Tier 3 with LEV III through adoption of the PHEV allowance in
Tier 3 final regulation for all 50-state certified PHEVs.
Our Response:
The CARB ZEV mandate requires manufacturers to produce and sell zero emission
vehicles. The LEV III program provides manufacturers with emission adjustments for advanced
vehicles like PHEVs. The adjustments are based on expected emissions reductions from the
electric operation portion of driving in the California market. Similar information was not
provided to EPA and since Tier 3 does not have a ZEV requirement, we did not propose and are
not adopting PHEV adjustment factors.
What Commenters Said:
Alliance of Automobile Manufacturers (Alliance) and Association of Global Automakers
(Global):
PHEVs were introduced just a few years ago, and both the agencies (EPA and ARE) and
automakers are still developing the vehicles and the test procedures for those vehicles. In 2009,
before any PHEVs had been introduced, ARB adopted PHEV test procedures for California. In
June, 2010, the Society of Automotive Engineers (SAE) J1711 Committee (which includes EPA,
ARB, automakers, and automotive suppliers) approved a recommended practices document that
contains detailed test procedures for PHEVs. The test procedures contained in SAE J1711
represent the latest and best collaborative effort to test these new vehicles. Most recently, EPA
adopted PHEV test procedures in July, 2011. The EPA requirements reference many of the test
procedures contained in SAE J1711.
Since EPA adopted PHEV test procedures in 2011, manufacturers and the agencies have gained
significant experience developing, validating, testing, and certifying these vehicles, with a
growing number of PHEVs are on the road from more manufacturers with more attribute sets. As
this occurred, industry worked to develop a set of recommended changes to the PHEV test
procedures that we believe maintain the stringency of the standards and the benefits of these
vehicles, while streamlining the testing.
We have met with EPA and ARB staff several times over the last few months and sincerely
appreciate the efforts of both agencies to consider and discuss our recommendations. Appendix 2
[See the attachment in Docket number EPA-HQ-OAR-2011-0135-4461-A3] contains a
comprehensive list of the changes we have recommended to the EPA (and ARB) test procedures.
We highlight two important issues below, although these are also addressed in Appendix 2.
4-396
-------
Tier 3 Summary and Analysis of Comments
CS Worst Case, Manufacturers Attest That CD Emissions Will Be Lower when the Utility Factor
(UF) is Used: PFIEVs have two primary operating modes - charge depleting (CD) and charge
sustaining (CS). In CD mode, the battery provides all, or at least some, depending on the
PHEV's design, of the power to drive the vehicle. In CS mode, the gasoline engine provides all
of the power to drive the vehicle either through a direct connection to the drive train, or by acting
as a generator to deliver electricity to the electric motor.
Manufacturers must certify that PHEVs will meet the emission standards in CS mode, which is
real-world worst-case since it represents a driver who never charges the vehicle and thus never
uses the battery for propulsion. To the extent that a vehicle uses the battery for propulsion, the
emissions during that period would be zero and thus would lower the overall emissions from the
vehicle. In addition to CS mode, manufacturers must also certify that their vehicles will meet the
emissions standards in the "worst-case" CD mode.
The method used to test "worst-case" CD mode is to drive the vehicle from a full charge over
consecutive Urban Dynamometer Driving Schedules (UDDS)13 until the engine starts and then
run one additional UDDS. Emissions are measured during the UDDS when the engine starts and
over the next UDDS. This can lead to a situation where a vehicle has two cold starts (one at the
very end of the first UDDS and another during the second UDDS) and appears to not meet the
emission standard regardless of the zero emission miles traveled prior to the first cold-start. This
does not represent any real-world condition. In fact, PHEVs will have far lower emissions than
non-PHEVs certified to the exact same standard.
For example, GM provides data for the Chevrolet Volt on the total number of miles driven
compared to the total number of all-electric miles (see http://www.chevrolet.com/volt-electric-
car.html). As of May 27, 2013, Volts have driven a total of 310 million miles, of which 192
million miles were all-electric. Thus, more than 60 percent of the miles driven on the Volt
produce zero emissions. If the Volt is certified to a Bin 30, this would translate into real world
emissions of about 12 mg/mile (= 30 * 0.4). We recognize that this will vary depending on the
vehicle design, all electric range, etc. However, it is indisputable that a PHEV will emit less than
a conventional vehicle certified to the exact same emission standards. Yet manufacturers spend
two to three times as much effort to certify a PHEV compared to a conventional vehicle. This is
especially significant given that PHEV test groups are relatively low volume compared to
conventional vehicle test groups and manufacturers face greater challenges in trying to make
PHEVs profitable.
The driving cycles used for decades - Federal Test Procedure (FTP), UDDS, US06, etc. - were
developed by monitoring driving behavior and choosing what would be an average driving cycle.
No one drives a vehicle exactly according to an FTP cycle. However, on average, the FTP cycle
is a good approximation. Likewise, the SAE J1711 and Part 600 procedures include a UF that
approximates the amount of driving on the battery and the amount of driving with the engine.
Again, perhaps no single vehicle mirrors the exact UF split in electric versus gasoline propulsion,
but on average the UF is a good approximation of the average. In fact, EPA uses the UF for
GHG emissions, but does not allow its use for criteria emissions.
4-397
-------
Chapter 4: Proposed Vehicle Emissions Program
Because the UF effectively mirrors total real-world emissions (SAE J2841) from a PHEV, EPA
has expressed concern that manufacturers could use this to increase the air-to-fuel ratio, improve
fuel economy and reduce GHG emissions at the expense of increased NOx emissions when
compared to a CS test. This is a legitimate concern that EPA could address with the current
defeat device regulations. If EPA needs further assurance, it could specify that the vehicle
emissions system must perform in a similar manner at the same auxiliary power unit (APU)
speed and load point on both the CS and the CD mode.
Recommendation: The following three requirements should address EPA's concerns with the
criteria emissions. These assure that a vehicle that is never charged meets the emission standards,
and that vehicle emission controls operate in a similar manner whether in CD or CS operations:
1. Vehicle must meet the emission standards in CS operating mode.
2. Vehicle must meet the emission standards in CD operating mode when the UF is applied. The
manufacturer may attest to meeting this requirement.
3. The vehicle emission system must operate in a similar manner when operating at the same
APU speed and load point in either CS or CD mode.
Net Energy Change (NEC) Tolerance and Use of Appendix C: CS tests are intended to determine
emissions when the gasoline engine is providing all of the propulsion power. However, PFIEV
systems are complex, and the gasoline engine is designed to operate at maximum efficiency. This
means that all of the power of the gasoline engine may go to providing propulsion at some times,
while at other times it may be used to charge the propulsion battery; at other times, the
propulsion battery may discharge further to provide peak or transient power. Consequently,
while the vehicle is operating in CS mode, the battery state of charge may slightly change. To
verify that the gasoline engine is providing all of the power over the CS test, SAE J1711, ARB,
and EPA procedures require that the vehicle operate within a NEC tolerance of 1 percent of the
total fuel energy.
As propulsion batteries become larger and the control systems more sophisticated, maintaining
NEC within 1 percent becomes more difficult. The SAE J1711 committee recognized this
difficulty and developed a method to correct an NEC of greater than 1 percent but less than 5
percent back to an equivalent of NEC = 0. This NEC correction of up to 5 percent is contained in
Appendix C of SAE J1711.
The CS test is difficult, time consuming, and resource intensive, and if the vehicle misses the
NEC tolerance, a complete retest over that cycle is required. Allowing use of Appendix C to
correct NECs of less than 5 percent would significantly reduce the amount of retests conducted.
Moreover, it would allow manufacturers to develop vehicles with better performance.
We understand that both EPA and ARB are concerned that if a vehicle is certified over the CS
test with a -5 percent NEC, then emissions would be under-reported, since 5 percent of the
vehicle's propulsion came from the zero emission battery. However, SAE J1711, Appendix C,
corrects the measured fuel consumption back to NEC=0. Moreover, manufacturers would be
required to meet the emission standards provided at any NEC between ± 5 percent. Thus, the
vehicle must meet the emission standard if the NEC is +5 percent as well.
4-398
-------
Tier 3 Summary and Analysis of Comments
We agree that manufacturers should declare the use of Appendix C at an early point in the
certification process. Automakers would agree to provide this information in the pre-certification
meetings.
Recommendation: We recommend that EPA allow the use of SAE J1711, Appendix C,
correction for NEC tolerance greater than one percent, provided the manufacturer declares its
intention to use this during the pre-certification meeting. Moreover, manufacturers support a
requirement that the vehicle must meet the emission standards within the full range of the NEC
±5 percent (with fuel consumption corrected per SAE J1711, Appendix C).
Our Response:
With regard to the recommendations on how PHEV's should be assessed in charge
depleting and charge sustaining modes; we have addressed these recommendations in a memo to
the docket which summarizes our test procedure changes. In summary, we have not adopted
these recommendations as part of Tier 3 as we did not have adequate information on which to
base our decision. We will continue to work with manufacturers to address concerns with
current PHEV test procedures.
With respect to the NEC tolerance, EPA has adopted the recommendation made by the
Alliance and Global Automakers, however, manufacturers wanting to exercise the option of
using NEC +/-5% will be required to apply for permission to use this procedure and receive prior
approval from EPA.
What Commenters Said:
Organization: BMW of North America, LLC
In-Use Testing of PHEV's:
The in-use requirements for Carbon Related Exhaust Emissions (CREE) of PHEV's were
changed in 2012. Prior to this modification, the CREE was calculated separately for depleting
and sustaining operation modes. The new language requires combining depleting and sustaining
modes by the utility factor. However, the utility factor is determined by the depleting range;
hence the new language makes the depleting range the key influence on in-use performance.
Over useful life the depleting range is reduced as a result of aging of the traction battery.
To date very little experience exists on battery deterioration in actual customer use.
BMW finds the new in-use requirement premature and recommends that the feasibility of this
approach be evaluated on the variety of PHEV concepts manufacturers are introducing into the
market to better assess the effects of long-term customer use. Further, BMW recommends re-
introducing the PHEV in-use requirements from May 2010 wherein depleting and sustaining
operations are treated separately in CREE calculations.
4-399
-------
Chapter 4: Proposed Vehicle Emissions Program
Once long-term behavior is better understood through sufficient field data and impact analysis, a
combined CREE calculation (depleting and sustaining modes) based on the depleting range
could be re-introduced, provided available battery technologies support the approach.
Our Response:
BMW commented that in the 2012 light-duty Greenhouse Gas (GHG) regulations the in-
use requirements for measuring and calculating Carbon Related Exhaust Emissions (CREE) for
plug-in hybrid vehicles (PHEV) was calculated separately for charge depleting and charge
sustaining operation modes, but that in the 2017 GHG rule, EPA changed the requirements so
that in-use CREE is calculated by combining the depleting and sustaining modes using a utility
factor. Their concern is that the utility factor is determined by the depleting mode driving range
thus making the depleting range the key influence on in-use performance. They argued that there
is very little experience with battery deterioration in actual use and that it is premature to use this
approach. They suggested that more information on battery deterioration was necessary and that
in the future it may be more appropriate to determine in-use performance for PHEVs by using a
utility factor and combining depleting and sustaining modes.
BMW's comments are related to the GHG regulations and are not directly related to the
Tier 3 rule. While their comments do discuss in-use emissions and how they would be calculated
for IUVP, they are not related to anything that was proposed or discussed for the Tier 3 rule.
However, EPA will address some of their concerns. BMW is fundamentally wrong when they
state that EPA originally specified separate in-use CREE standards for charge-depleting and
charge-sustaining modes. The 2012-2016 GHG final rule specified, for dual fuel vehicles
(including PHEVs), that there should be an in-use standard for each fuel, not for each mode. (See
75 FR 25687, May 7, 2010.) PHEVs were included with other dual fuel vehicles, where the
CREE (or CAFE) value is based on test results on each fuel (e.g., gasoline and E85). Obviously,
each in-use test for most such vehicles should be done on one fuel or the other, especially for
dual fuel vehicles where the fuels can be mixed in the same tank (e.g., gas-E85 FFVs). Such an
approach might have worked for the Volt (a series PHEV), which at the time was the only PHEV
with which EPA had any experience. However, by the time of the 2017-2025 GHG rule (see 77
FR 63158, October 15, 2012), we realized that PHEVs did not generally fit the mold of other
dual fuel vehicles. With blended PHEVs, for example, it is simply impossible to test either on
gasoline or on electricity, or to specify standards separately for each fuel. It was in the 2017
GHG rule that we changed the language for PHEVs, requiring that PHEVs be tested as they are
for certification, using a utility factor weighted composite of charge-depleting and charge-
sustaining operation.
Contrary to BMW's assertions, we never considered separate charge-depleting and
charge-sustaining standards. PHEVs are a unique technology. While the in-use performance of
FFVs can be typically assessed by testing on either or both fuels, this is not the case for PHEVs,
where the two fuels can be inextricably intertwined, and where the battery performance over
time can have an impact on in-use GHG emissions. The GHG emission standards are full useful
life standards, and the regulations require manufacturers to account for battery deterioration in
their certification values (which are likewise required to represent full useful life values). While
we agree with BMW that battery performance over time can affect in-use GHG emissions, it is
4-400
-------
Tier 3 Summary and Analysis of Comments
precisely for this reason that in-use testing must assess the composite of charge-depleting and
charge-sustaining operation. If manufacturers appropriately account for battery deterioration in
their certification values, then they should have no concerns. BMW is in effect asking that the
fundamental basis of in-use testing - to produce a fair comparison between certification and in-
use emission levels - be set aside for PHEVs. In a fleet-averaging program in particular, EPA
believes it is important to assess in-use performance against certification levels.
What Commenters Said:
Alliance of Automobile Manufacturers (Alliance) and Association of Global Automakers
(Global)
EPA proposes to confirmatory test vehicles on four wheel (4WD) dynos as soon as 90 days after
the Final Rule, even if they were originally certified by the manufacturer on a two-wheel (2WD)
dynos. Such a change in dyno requirements would be a change to the yardstick used to determine
fuel economy, and would significantly impact the measurement of fuel economy. While EPA
would be obligated to release a Test Procedure Adjustment (TPA) for CAFE, EPA would need to
also change GHG rule targets and label procedures. Since the yardstick at the EPA lab would be
changed, manufacturers would be effectively forced to install 4WD dynos at their facilities to
ensure correlation with the new EPA yardstick, with a massive capital cost for new equipment
and new facilities. But studies and test data have not yet laid to rest basic questions concerning
the higher variation in results on 4WD dynos or their effects on fuel economy or even their
representativeness of on road behavior. 4WD dynos, as they are proposed to be used in the
NPRM, may turn out to not be a better representation of on-road behavior, but just another,
different representation, with new and different measurement offsets and correlation issues.
Given that basic questions have yet to be answered, and that these questions were not addressed
at all in the NPRM or in the RIA, we disagree with EPA adopting 4WD confirmatory testing, as
this change in test procedure would require appropriate application of the dyno factor and CAFE
TPA, GHG standards adjustments, and would be coupled to significant lead time, dramatic cost
impacts, loss of emissions testing capacity and increased variability.
EPA's Proposal for 4WD Dyno Confirmatory Tests is Premature and Should not be Adopted:
Not discussed in the preamble part of the Tier 3 NPRM, but included only in the draft regulations
section (78 Fed. Reg. 30163, § 1066.410 (g), Dynamometer Test Procedure) is a significant
change to current test procedures. EPA proposes to confirmatory test vehicles on 4WD dynos
even if they were originally certified by the manufacturer on a 2WD dyno. Such a change would
impact the measured fuel economy of a vehicle and effectively forces manufacturers to install
4WD dynos in order to minimize the risk of not correlating with the results of the EPA lab. This
EPA lab test procedure change measurably impacts fuel economy and thus is a change to test
procedures which requires a TPA. [EPA-HQ-OAR-2011-0135-4461-A1, p. 100]
First, the Alliance and Global Automakers recommend that EPA not adopt the proposed change
and that EPA instead continue to follow established policy of confirmatory testing vehicles as
OEMs do on the same dyno type as used by the manufacturer. EPA should continue the current
4-401
-------
Chapter 4: Proposed Vehicle Emissions Program
policy of only using 4WD dynos for non-confirmatory testing and for data collection purposes
for the following reasons:
It is premature to require confirmatory testing on 4WD dynos because there has been insufficient
analysis and research performed on 4WD dynos to fully understand their operation and minimize
test result variability to acceptable levels. For the switch from the Clayton twin-roll dyno to the
48" single-roll electric dyno in the 1990's, both EPA and industry cooperated in a large study to
determine the necessary changes to calibration and setup procedures to ensure good correlation
between EPA and industry's dynos (92). Based on a preliminary EPA and Alliance/Global
Automakers study, attached as Appendix 9, 4WD dyno correlation is not predictable or well
understood for even simple, "passive" 2WD vehicles. For this reason, meaningful conclusions
cannot be drawn from 4WD dyno test results at this time. This may be due to the great variations
in designs (four motors, motor in the middle, location of load cells, location of speed acquisition,
parasitic loss adjustments, roll speed synchronization, flexible versus rigid vehicle tie downs,
etc.), or it may be due to setup and testing procedures not being defined and standardized. The
Alliance and Global Automakers strongly recommend that much more testing and analysis must
be performed before 4WD dynos can be adopted for confirmatory testing.
By instituting a test procedure change to confirmatory test vehicles on a 4WD dyno even if the
manufacturer certified on a 2WD dyno, EPA is effectively requiring manufacturers to install
4WD dynos in order to minimize the risk of not correlating with the EPA lab. This would impose
an extraordinary burden on industry in terms of cost and practicality for manufacturers to install
4WD dynos for fuel economy and certification testing. Since the requirement could be
implemented by EPA 90 days after publication of the Tier 3 Final Rule, over a very short time, a
large number of 4WD dynos would need to be purchased and installed because of the great
volume and frequency of certification and fuel economy tests performed by manufacturers. In
many instances this would not be practicable, since many facilities in current use cannot be
modified to accept the increased space requirements for 4WD dyno testing with their
requirements for large underground pits. New buildings and test cells would be required in many
cases to accommodate a 4WD dyno at a test facility. 4WD dynos would be required to be
installed in environmental chambers to be able to perform 20°F, SC03 and AC 17 tests. Many
existing environmental chamber test cells will not be able to accommodate the installation of a
larger pit and extra roll for a 4WD dyno. Finally, the cost for each 4WD dyno is more than
double the cost of two 2WD dynos, due to the necessary and more complex hardware/software to
synchronize two rolls and to move one roll to accommodate varying wheelbase vehicles.
4WD dynos have a particular burden associated with them in terms of the greater floor area
required for their installation. The footprint area requirement for a 4WD dyno is partly a result of
the two rolls, with one of the rolls being a movable roll, but is more a result of the unique vehicle
tie-down arrangements necessary on a 4WD dyno. Differently than on a 2WD dyno, a vehicle is
required to be restrained in all directions to maintain its contact at the top of the two rolls. Such a
vehicle restraint system, be it a rigid type or a flexible type, requires sufficient space around all
the sides of the vehicles to be able to resist the substantial forces during a test. On average, 4WD
dyno test cells require approximately 3 to 4 times more square feet of floor space than do 2WD
dynos. Appendix 10 provides examples of predicted floor space requirements for manufacturers
to implement the 4WD dyno confirmatory proposal. As can be seen graphically in the first
4-402
-------
Tier 3 Summary and Analysis of Comments
example, the floor space that is currently able to accommodate five 2WD dynos would only be
able to accommodate three 4WD dynos. A new facility would be required to be built to
accommodate the other two 4WD dynos which are needed to maintain the same test production
capacity.
For some manufacturers, expansion of an existing facility may be possible, and for some
manufacturers entirely new facilities would be required. However, some manufacturers are so
real estate constricted that to add 4WD dynos at an existing location would require them to build
vertically rather than horizontally, and the cost of expansion for those manufacturers would be
significantly higher.
In order to better quantify the potential burdens associated with this proposal, manufacturers
developed an overall industry cost estimate based on a survey of cost estimates from individual
manufacturers for the purchase and installation of 4WD dynos. The cost estimate includes
estimated costs for 4WD dynos (including individual selections of 4WD dyno technologies, such
as location and number of power absorbers, and power ratings) and estimated costs to prepare the
test sites to accept the 4WD dyno. These costs include digging larger pits in existing test cells,
reconfiguration of existing equipment to accommodate 4WD dynos, and, where necessary,
construction of entirely new facilities to maintain production capacity during and after
construction. Based on this industry cost estimate, attached as Appendix 11, the one-time cost
burden to implement the EPA proposal would be in a range from $1.031 billion to $1.246 billion.
The cost estimate is conservative and an actual cost would likely be much greater, since the
number of dynos to be replaced is not actually proportional to vehicle sales.. The ongoing cost
burden to operate and maintain 4WD dynos with their reduced throughput and higher
maintenance needs would be significantly increased as well, but is not quantified in the cost
estimate.
These high costs make it very clear that before a switch to 4WD dynos can be contemplated, a
rigorous analysis of costs and benefits needs to be performed, including evaluations of
potentially less costly alternatives. Since such an analysis was not included or even addressed by
the NPRM or the RIA, it is unclear how procedurally such a costly requirement could even be
proposed.
The Alliance and Global Automakers also have concerns with a reduced test throughput from a
4WD dyno confirmatory requirement. The vehicle setup burden would be increased, not reduced.
Previously, EPA has not released any guidance on standardizing methods for vehicle tie down on
4WD dynos. There are increased concerns with safety because of the more complex vehicle tie
down arrangements required. Wheel chocks cannot be used on 4WD dynos and most vehicles are
not equipped with trailer hitches and front tow hooks, requiring ad-hoc welding/securing of tie-
down fixtures to vehicle frame - often impractical for customer vehicles and inherently unsafe.
Specific and detailed procedures are necessary for how to tie down a vehicle, align the vehicle to
the dyno rolls to minimize skewing, and how to account for any potential interaction between the
4WD dyno and 4WD/AWD vehicle torque distribution control systems, which in some cases
may result in on-dyno AWD system activation that is not representative of its "on-road"
behavior. Such guidance is needed before EPA and Industry can successfully implement a 4WD
dyno confirmatory testing requirement.
4-403
-------
Chapter 4: Proposed Vehicle Emissions Program
Organization: Ford Motor Company (Ford)
Four-Wheel Drive (4WD) Dynamometer Testing: In the NPRM, EPA proposes to change its
policy regarding 4WD dynamometer testing. There are two components of the proposal:
(1) EPA proposes to perform confirmation testing of four-wheel drive vehicles using four-wheel
drive dynamometers, even if a manufacturer tests such vehicles using a two-wheel drive
dynamometer.
(2) EPA has also proposed that starting in 22MY, manufacturers may continue to test four-wheel
drive vehicles using 2WD dynos, but only if there is no decrease in emissions or energy
consumption relative to on-road operation.
Our concerns with these proposals are provided below:
Added Complexity - The core argument in support of 4WD dynamometer testing is that it
should, in theory, better represent a vehicle's "on-road" performance. This is true in the sense
that all four wheels are rotated when testing on a 4WD dyno, however, the benefits of including
this additional vehicle component must be weighed against the added complexity that is
necessary to accomplish this seemingly simple task.
This added complexity includes:
(1) a second 48" dynamometer roll electrically coupled to the primary roll that requires the
dynamometer control system to continually distribute and rebalance the target road load between
the two axles in response to both changing roll speed and changing vehicle torque distribution,
while at the same time maintaining a high degree of speed synchronicity between the rolls;
(2) fundamentally more complex and error prone vehicle positioning and securing process due to
the inability to secure the tires of a "non-drive" axle;
(3) varied and complex restraint systems (chains, cables, and rigid fixtures);
(4) non-existent or inconsistent attachment points for these restraints on the vehicle and/or
vehicle frame;
(5) four tires rotating on rounded surfaces (vs. a "flat" road) resulting in less contact area and
higher rolling resistance, which requires the dyno to "assist" the vehicle at low speeds;
(6) dyno simulation of "negative" inertia for vehicles with test weights that are less than the
combined base inertia of two dyno rolls; This complexity adds variability, and also likely
contributes to correlation differences between 4WD and 2WD testing that should not be
attributed to a vehicle tested in 4WD dyno mode. With sufficient study the inherently greater
complexity of 4WD dyno testing can be better managed than it is today, but it is not certain that
4WD testing will be able to achieve the same level of consistency as the existing 2WD
procedures.
Test Procedure Capability - Manufacturers have had very little experience using 4WD
dynamometers, and as a result, best practices needed to ensure repeatable 4WD testing results
that correlate with existing 2WD procedures are non-existent or not well defined. These practices
cannot simply be carried over from the 2WD testing paradigm since they must address issues
unique to 4WD dyno testing (see sources of added complexity above). EPA and Industry have
been working to develop such procedures, but this work is incomplete and should be continued.
4-404
-------
Tier 3 Summary and Analysis of Comments
Correlation to Current 2WD Procedure - EPA has expressed an interest in understanding
whether the existing 2WD dyno method satisfactorily accounts for vehicle systems that must be
"disabled" or modified in order to permit operation on a 2WD dynamometer (for example, AWD
systems). Although 4WD dyno testing may provide a means to answer this question once reliable
test procedures are in place, the effect of "enabling" such a system can only be isolated if 4WD
dyno testing in general produces results equivalent to a 2WD dyno for vehicles that do not
possess such systems. Otherwise, any general testing bias between the dyno modes could be
falsely attributed to the vehicle system being studied.
Additionally, since the parasitic drag of a test vehicle's "second" axle is already included in the
existing 2WD testing process via the on-road coastdown test, future 4WD dyno testing studies
should be focused on vehicle technologies that provide some "active" attribute on this axle that
cannot be characterized by coastdown parasitics.
To better understand these issues, EPA and Industry have been conducting a joint, 2WD-to-4WD
correlation program. Ford commends EPA staff for its continued interest and participation in this
program, which we believe will ultimately lead to more capable and robust 4WD dynamometer
test procedures.
Cost/Facility Impact - 4WD dyno testing is a complex and expensive testing option. Beyond the
substantial cost of the dynamometer equipment itself, the installation process requires extensive
site construction and often building modifications, and the additional time needed to align and
secure test vehicles also reduces the number of tests that can be performed per shift. These costs
and facility impacts must be carefully weighed against any demonstrated enhancement of
measurement capability associated with 4WD dyno testing.
Lead Time - Whenever any new test method is introduced that has the potential to influence test
results, manufacturers need adequate lead time to prepare for, and transition to, the new method.
This time is needed to study the impact, develop test procedures, and if necessary, install new
equipment. It also prevents disruption of programs currently in development using the existing
procedures.
Recommendation:
(1) EPA should continue its current policy of conducting confirmation testing in the same
dynamometer test mode used by the manufacturer. Meanwhile, EPA should continue to work
with Ford and Industry to:
• Develop and refine best practices for 4WD dynamometer test methods to ensure meaningful
and repeatable results.
• Complete the ongoing effort between EPA and Industry to establish a robust correlation factor
between 2WD and 4WD dyno mode to ensure that any future 4WD dyno test results are
appropriately adjusted to account for overall (i.e., non-vehicle specific) differences between the
two dyno test methods.
• Once best practices and repeatable correlation factor are in place, conduct testing to
demonstrate which specific vehicle technologies, if any, would be significantly better
represented by using 4WD dyno test methods.
4-405
-------
Chapter 4: Proposed Vehicle Emissions Program
• For the technologies that have been clearly demonstrated to benefit from 4WD dyno testing,
explore more cost effective options for including this effect in the fuel economy label that would
not require manufacturers to install and test on 4WD dynamometers.
(2) Additionally, EPA should continue to define the 2WD dynamometer as the default test mode
for all vehicles, and refrain from adding any language that would require manufacturers to test
4WD vehicles on 4WD dynos.
In light of (1) and (2) above, we recommend that the language of proposed § 1066.410 (g) be
either deleted in its entirety, or revised as follows:
(g) For vehicles which provide four-wheel drive or all-wheel drive operation, manufacturers may
either utilize the vehicle's normal (default) mode of operation or test the vehicle on a single roll
by deactivating the second set of drive wheels. Any confirmatory testing will be conducted in the
testing mode used by the manufacturer.
(3) If EPA proceeds with using 4WD dynamometers for confirmation testing in the future:
• EPA should provide manufacturers with adequate lead time to prepare.
• All fuel economy and CO2 results (certification, FE label, CAFE, GHG) generated using a
4WD dynamometer should be adjusted using the dyno correlation factor currently under
development by EPA and Industry.
• EPA should apply appropriate test procedure adjustments for CAFE and GHG/CO2 purposes to
account for any impact the 4WD dyno test methods have on the stringency of these standards.
Organization: General Motors LLC (GM)
The proposed Tier 3 regulations would effectively require testing of multiple drive axle vehicles
on 4WD dynamometers (dynos) (ref Section 1066.410(g)). Additionally the agency has stated
that it reserves the right to include other vehicle configurations like hybrid vehicles in this
'4WD' population, and may, or will confirm manufacturers emissions and fuel economy data on
4WD dynos, even if the manufacturer did its testing on a 2WD dyno. This is different than past
practices by the agency which were adopted to minimize sources of variability between
manufacturers' testing and the agency's testing. For OEMs to eliminate any risk at confirmatory
testing, 4WD dynos will have to be used for a significant percentage of the fleet.
4WD Dynamometer Test Procedure Adjustment: GM supports the Alliance/Global comments
with respect to adverse fuel economy impacts associated with 4WD testing, which would require
EPA to issue a test procedure adjustment (TPA). However, data indicate that test to test
variability along with cycle to cycle variability may prohibit the establishment of a robust TPA
value at this time. The source of this variability is not understood, and likely is a function of
several equipment and setup parameters. 4WD dyno testing is more complex than 2WD testing,
and to state it simply, more things can go wrong. This needs to be thoroughly researched before
4WD testing is used for confirmatory testing. To this end, GM believes that EPA should
continue with today's practice of testing the same way the vehicle is certified by the
manufacturer.
4WD Dynamometer Facility Impacts: Implementation of 4WD dyno testing is a major facility
impact, conservatively estimated to cost over $1 billion industry-wide, as detailed in the
Alliance/Global comments. In addition, installation of the new dynos will adversely affect test
throughput at a time when manufacturers need all of their test sites available to develop, validate
4-406
-------
Tier 3 Summary and Analysis of Comments
and certify vehicles meeting the GHG/FE and Tier 3/LEV III requirements. The following
outlines the facility impacts.
4WD Facility Impacts to Conventional 75o F Test Sites: Implementation of 4WD dynos will
have dramatic and adverse effects on emissions test facilities and OEM's certification and
development testing. Additionally, there is insufficient time to convert existing test sites to 4WD
capability for the following reasons:
4WD dynos are more than twice the size of current 2WD dynos. 4WD dynos require very large
pits compared with current, much smaller 2WD pits to house and service the dyno. These 4WD
large pits require four times the surface area of today's conventional 2WD pit, are twice as deep,
require two egresses for safety, and require a huge inertial mass to mount the dyno to.
4WD dynos will require new building structures and utilities. Many OEM's facilities were built
in the 1970's and were sized for much smaller twin roll water brake dynos. The much larger pit
size required for 4WD dynos will easily exceed the available area on many of today's test sites,
and will clearly require much larger test sites. To accomplish this, existing sites will need to be
expanded into neighboring cells, thereby significantly reducing the number of test sites available
for Tier 3 certification and development testing. Given the increased stringency of the Tier 3
regulations along with new GHG/FE standards, reducing the number of emissions test sites will
definitely have an adverse impact on OEM's ability to meet these new requirements.
Additionally, besides encroaching on adjacent test sites, these larger pits will also dramatically
protrude out into routine vehicle movement lanes with safety egress stairs/hatches. When these
hatches are open for dyno service or monthly calibration, vehicle traffic to adjacent test sites or
evaporative sheds will be blocked, thereby adversely affecting emissions and fuel economy
testing. Evaporative sheds need to be in close proximity to the chassis dynos due to the 40 CFR
86.130-96 requirements that no more than 7 minutes time elapse from the end of chassis testing
to the start of evaporative testing.
4WD Facility Impacts to Climatic Test Chambers: Cold 20o F testing, hot 95o F SC03 testing,
and the new A/C 17 testing all require specialized test chambers, and will also have to be
updated with 4WD dynos. This is very problematic because these facilities are limited in
number, and modifying them for 4WD testing will have a significant and adverse impact on
current climatic emissions testing. The SC03 test sites are even more limited in number and
4WD implementation is further complicated by the CFR required large road speed fan (wind
tunnel), which is likely to cause additional structural issues and longer implementation time.
4WD Construction, Implementation and Testing Down Time: We believe from start of
construction to test site commissioning will take 1.5 years per test site, and as mentioned above,
will need to be done in pairs of sites or more likely multiple test sites at any one time. The test
sites removed from service for this construction are active certification and development test
facilities, some of which are designed for specific vehicles like diesels. This will have an adverse
impact on emissions testing capability. Additionally, because of this long downtime,
manufacturers will only be able to convert a few sites at any given time. For a full line
manufacturer, with dozens of test sites, this 4WD conversion will take a decade or more to
4-407
-------
Chapter 4: Proposed Vehicle Emissions Program
complete. In fact, it is likely that 4WD dynos will force green field annexes or require the
construction of new buildings. Manufacturers simply cannot afford to have test sites unavailable
during this time when so much testing is needed for the development and certification of vehicles
to meet the GHG/FE and Tier 3/LEV III standards.
4WD Pits - Non Dynamometer Impacts: Installing these very large machines will undoubtedly
have major effects on virtually all of the other equipment located in the test sites such as exhaust
sampling systems (CVS/Tunnel/bulk stream and raw exhaust tubing), modal sample conditioning
systems plus close coupled modal analyzers, overhead XY support systems to hang vehicle
cooling fans, drivers aids, remote mixing tees, etc. It is likely that these will be replaced with the
site construction or require modification. With such large scale changes, additional changes to
computer programs to control these new devices will be needed and modifications (or likely
replacement) of emissions benches will be needed as well.
4WD Technology Not Well Understood: Many technical challenges still need to be resolved
with the use of 4WD dynos. The dyno controls, set up parameters, calibrations, diagnostics and
vehicle restraints all need to be studied. When 48-inch single roll dynamometers were adopted as
part of the SFTP rulemaking, an extensive industry/government research program was launched
to look into best practices and resolve any implementation and usage issues. Before 4WD dynos
can be used for certification testing, it is highly recommended that a similar program be initiated
to understand 4WD dynos.
4WD Dyno Adverse Impact on Long Term Testing Throughput: In addition to the impact on test
throughput during the installation of 4WD dynos, it also takes much longer time to properly set
up a vehicle on a 4WD dyno, which increases testing burden and slows down emissions testing
on an ongoing basis. Please see the Alliance/Global comments for more detail.
Recommendation: Given the magnitude of the impact on test facilities and test throughput, along
with the adverse yet not fully understood impacts on fuel economy and GHG emissions that
would require accurate test procedure adjustments, GM does not believe that it is appropriate
make changes that would require 4WD dyno testing in the Tier 3 rulemaking, particularly since
Tier 3 is an accelerated rulemaking focused on criteria emissions. Instead, GM recommends that
the current practice of EPA performing confirmatory testing the same way the manufacturer did,
whether 2WD or 4WDwd, continue, and that EPA work with industry to study 4WD dynos
further as detailed in the Alliance/Global comments.
Organization: Mitsubishi Motors R&D America, Inc (MRDA)
As part of the Tier 3 NPRM, EPA has included language indicating they intend to begin
Confirmatory Testing on 4WD dynos. If EPA conducts Confirmatory Testing on 4WD dynos,
manufacturers will need to perform their certification and development tests on 4WD dynos to
ensure proper correlation to EPA tests. Manufacturers will incur a massive cost burden to replace
their 2WD dynos with 4WD dynos. Additional costs will be associated with the operation of the
4WD dynos, the reduced testing throughput due to the time needed to upgrade facilities for 4WD
dyno installation, as well as the longer vehicle set up time associated with 4WD dynos. Neither
4-408
-------
Tier 3 Summary and Analysis of Comments
EPA nor Industry has completed analysis to verify that there is any air quality benefit to the
overall cost of installing and using 4WD dynos as opposed to the 2WD dynos currently in use.
The language in the Tier 3 NPRM states that manufacturers may continue to test on a 2WD
dyno, 'but only if this mode of operation does not decrease emissions or energy consumption
relative to normal in-use operation'. (Page 30163 Federal Register / Vol. 78, No. 98 / Tuesday,
May 21, 2013). This implies that manufacturers could be required to test on 4WD dynos.
Mitsubishi Motors strongly recommends that an extensive test program be conducted to
determine if there are both dyno and vehicle losses associated with testing on 4WD dynos. If
4WD dyno testing is deemed necessary, Mitsubishi Motors believes that 2WD dyno testing
would still be appropriate if the resulting data is accurately adjusted with a TPA.
NPRM Comments: Based on the above information, we recommend that:
-EPA should continue to allow certification and fuel economy label testing to be conducted on
2WD dynos.
-EPA should continue confirmatory testing the same way manufacturers perform their
certification and fuel economy label testing.
-A TPA should be established to correlate the 2WD and 4WD dyno data.
-The language in 1066-410, stating a manufacturer may test on a 2WD dynamometer, 'but only
if this mode of operation does not decrease emissions or energy consumption relative to normal
in-use operation' should be eliminated.
4WD Dyno Testing: EPA should continue to allow certification and fuel economy label testing
to be conducted on 2WD dynos. EPA should continue confirmatory testing the same way
manufacturers perform their certification and fuel economy label testing. Any inclusion of 4WD
dyno testing into the Tier 3 program must be accompanied by a TPA to correlate the 2WD and
4WD dyno data.
Organization: Volkswagen Group of America, Inc.
VW understands the need for a level playing field and a real world testing scheme. However,
without the proper factors, testing vehicles on a 4WD dynamometer will not relate to the 1975
testing methodology as required. VW requests that much more work be done before beginning
any confirmatory testing on vehicles with a 4WD dyno. The requirement for FPA to test in the
same manner as Manufacturers should be retained in Tier 3 final rules. Industry and EPA are
testing vehicles to determine the proper methodology moving forward and VW supports this
cooperative spirit. While we do hove 4WD dynos in our labs, the capacity may not be large
enough to support testing a potential future W/ fleet. Pathways need to be available to continue
to test vehicles on 2WD dynos.
New EPA testing protocols involving 4WD dynos and the apparent split from the agreement to
perform testing in the manner of OEM test methods may lead VW to install more 4WD dynos. It
is the unpredictable nature of these confirmatory tests that may lead to abandoning of 2WD
dynos. As an estimate, it may cost $1.3 Million and approximately 1 year to convert a 2WD dyno
cell. Typically only 1 cell at a time is completed to maintain throughput of testing. Should
climatic elements or solar cells need to be installed, this price would increase. Thus, Industry
4-409
-------
Chapter 4: Proposed Vehicle Emissions Program
would need more than 90 days post Tier 3 signature for this conversion. We echo the Industry's
request to continue working with EPA on these issues. We also echo Industry's comments that
this program would potentially provide for factors for future technology vehicles, and not apply
retrospectively to prior vehicles. We would continue to encourage EPA to engage with dyno
manufacturers to develop proper dyno inertia error simulation procedures.
Organization: Volvo Car Group
In addition, in order to enable future technology development it is critical that there be
harmonization with CARB with respect to test criteria and design requirements.
If manufacturers are required to utilize several different test procedures and to test against
several different fuels, the lack of harmonization will generate unreasonable and unnecessary
burdens that will impede the overall goal of facilitating low emission technology.
During the workshops with EPA during 2012 and 2013, the Alliance has consistently and
repeatedly shown that there is a FE decrease when the same 2WD-vehicle is run on 4WD-
compared to 2WD dynamometer.
VCG supports the concerns of the Alliance regarding these phenomena. At the EPA industry
meeting on May 20, 2013, EPA expressed very clearly that a crucial aspect and goal for EPA is
to achieve real world fuel consumption reductions. Part of this strategy is that EPA now has the
ability to test on 4WD due to major investments.
Through substantial investments, VCG has developed similar testing capability, but VCG would
like to highlight several issues that have been identified with regard to this testing capability.
- 4WD testing is more complex then 2WD.
- The correlation between 2WD and real road performance has been thoroughly investigated over
the past 20-30 years, yet more remains to be done. There are aspects of 4WD and actual road
performance which now need to be investigated, understood and correlated. (Car fixation, tie
down etc.)
- Possible interaction between vehicle and dyno not seen on road.
Due to these issues, it would be premature to say that 4WD testing automatically represents real
world fuel economy. Therefore, the investigation of a correlation factor to address the 4WD dyno
aspects vs 2WD is essential.
Dyno to road correlation may not be more accurate on 4WD dynos than on 2WD; it could just be
different. The exact impact on FE for the all issues is not clearly known and this shows that
further method development is needed for 4WD testing. VCG believes the EPA proposal is
premature and that further analysis of 4WD dyno testing is necessary so all remaining issues are
resolved. VCG would like to work with EPA to support this future work.
4-410
-------
Tier 3 Summary and Analysis of Comments
Our Response:
We proposed that EPA may test vehicles with the capability of all-wheel drive operation
with dynamometers operating in either two-wheel drive or four-wheel drive mode, regardless of
the type of dynamometer that the manufacturer used for certifying the vehicle. However, the
final regulations specify that we will conduct our testing using the same drive mode as the
manufacturer. Vehicle manufacturers commented that differences in test results between a
vehicle tested on a two-wheel drive and a four-wheel drive dynamometer might be due to
differences in dynamometer characteristics more than in vehicle operation. Results of a
government-industry study that tested vehicles on both two-wheel and four-wheel drive
dynamometers indicated fuel economy differences in the range of ±4%, although the study was
inconclusive with respect to the cause of the differences. Based on the results of this study, we
will continue to test vehicles during confirmatory tests using the manufacturer's dynamometer
configuration for that vehicle, and that test will be the official certification result. We are,
however, finalizing revisions to 40 CFR 1066.410(g) to clarify that we may also test the
manufacturer's vehicle in a different dynamometer configuration than what was used for
certification testing for information-gathering purposes. If we decide to perform this testing, we
will depend on the manufacturer to cooperate in reconfiguring the test vehicle for our testing.
We will continue to investigate the effects of four-wheel drive dynamometers on emission results
and will not rule out possible future test procedure changes that might require certification of, or
allow EPA to perform confirmatory testing on, any vehicle on a four-wheel drive dynamometer.
4.7.2. Reduced Test Burden
What Commenters Said:
Organization: Ford Motor Company (Ford)
Where alignment cannot be achieved, it is important that mechanisms be provided to mitigate
any unnecessary complexity and testing burden. For example, we agree with the proposed
reciprocity provisions that will be crucial in helping manufacturers to manage the complexity in
multiple certification testing and test procedures.
Our Response:
EPA has worked closely with CARB and vehicle manufacturers to limit test complexity and
burden in the Tier 3 Final Rule.
In addition, we are updating the regulatory provisions that allow manufacturers to omit testing
for certification, in-use testing, and selective enforcement audits in certain circumstances.
Sections IV.A.3, IV.B.6, and IV.G.3 of the Preamble describe how this applies for demonstrating
that vehicles meet the Tier 3 PM standards. We are also allowing manufacturers to omit PM
measurements for fuel economy and GHG emissions testing that goes beyond the testing needed
for certifying vehicles to the Tier 3 standards. Requiring such measurement would add a
significant burden with very limited additional assurance that vehicles adequately control PM.
4-411
-------
Chapter 4: Proposed Vehicle Emissions Program
We are also allowing manufacturers to ask us to omit PM and formaldehyde measurement for
selective enforcement audits. If there is a concern that any type of vehicle would not meet the
Tier 3 PM or formaldehyde standards, we will not approve a manufacturer's request to omit
measurement of these emissions during a selective enforcement audit.
The existing regulations have allowed for waived formaldehyde testing for gasoline- and diesel-
fueled vehicles. The Tier 3 NMOG+NOX emission standards are stringent enough that it is
unlikely that vehicles will comply with the NMOG+NOX standards while exceeding the
formaldehyde standards. We are therefore continuing this waiver practice, such that
manufacturers of Tier 3 vehicles do not need to submit formaldehyde data for certification.
4.7.3. Miscellaneous Provisions
What Commenters Said:
Organization: General Motors LLC (GM)
GM is concerned with EPA's definition of non-conformity of in-use vehicles under Tier 3 when
vehicles are tested in-use. The non-conformity definition is a critical component of the
stringency of the regulation. GM supports the Tier 3 regulation based on the premise that Tier 3
and LEV III are of equal stringency as well as harmonized. The industry widely recognizes that
emissions control systems must be engineered with a degree of compliance margin to assure in-
use compliance. EPA conformity language differs significantly from the conformity language of
CARB. CARB regulations for LEV III recognize the average emissions test results of each
constituent as a method to evaluate conformity of a test group. Average emissions are a sound
statistical criterion upon which GM and other manufacturers can base compliance margins.
EPA's criteria are potentially subjective and statistically undefined, particularly given the
relatively small sample sizes used for in-use test programs. Therefore, GM is concerned about
both the stringency of Tier 3 and the ambiguity of the compliance definition, and would like to
work with EPA to ensure an approach that is statistically sound and harmonized with LEV III.
Our Response:
With the exception of addressing in-use fuel sulfur implications to emissions discussed in
Chapter 4.1.5.12 of this Summary and Analysis of Comments document, the IUVP requirements
and determinations of conformity are outside of the Tier 3 rule. EPA is committed to working
with industry to address Tier 3 related in-use concerns that may arise in the future.
4.8 Leak Standard Comments
4.8.1. Introduction of New Leak Standard
What Commenters Said:
4-412
-------
Tier 3 Summary and Analysis of Comments
Organization: Michigan Department of Environmental Quality (MDEQ)
The MDEQ, Air Quality Division supports the more stringent evaporative, leak and refueling
emissions standards, as well as the new test procedures described in the proposed rule. Control
of these significant sources of volatile organic compounds, particulate matter, and toxic air
contaminants have been demonstrated to be cost effective and attainable. California appears to
have already implemented a program with many similar requirements, showing that the limits
can be met.
Organization: National Association of Clean Air Agencies (NACAA)
Evaporative Emission Standards — EPA proposes new evaporative emission standards to reduce
total evaporative emissions from all gasoline-powered highway vehicles to near-zero levels. The
program would require new evaporative emission control technology on new vehicles as well as
system design improvements to achieve improved in-use system performance and extend useful
life. The proposed approach also introduces a new canister leak emission standard and would
apply California's onboard diagnostic requirements nationwide. NACAA endorses these
proposed requirements.
Organization: California Air Resources Board (CARB)
CARB supports the proposed evaporative leak test and standard, and believes that the leak test
program will be effective in reducing in-use evaporative emissions. Accordingly, CARB intends
to propose alignment with this proposal once the Tier 3 program is finalized.
Our Response:
EPA acknowledges the support of State and local governments for the new leak standard
as well as California's intent to propose the new standard for adoption into their requirements,
thereby maintaining a harmonized vehicle program for 50 States.
4.8.2 Nature and Scope of the Requirement
What Commenters Said:
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
EPA asked for comment on a third option that would require that a manufacturer meet the Tier 3
evaporative emissions standard on 20 percent of its fleet less than 6,000 pounds GVWR and also
meet the leak check standard on those vehicles. We support this option and appreciate that EPA
understands our concerns.
4-413
-------
Chapter 4: Proposed Vehicle Emissions Program
We agree that manufacturers could meet the leak check standard on 20 percent of their MY 2017
fleet; however, since these are independent requirements, there is no reason that manufacturers
need to overlap the vehicles meeting Tier 3 evaporative emission standards with vehicles
meeting the leak check standard.
Our Response:
EPA concurs and has incorporated this MY 2017 flexibility into the final Tier 3 rule.
4.8.3. Leak Test Procedure
What Commenters Said:
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
The connection point for most IUVP vehicle leak testing will be located at or near the purge
valve. This will test the complete fuel system. For leak testing conducted in this manner only one
test should be required. No additional testing is required to test the fuel cap. Also, for dual tank
vehicles that have the vapor space of both tanks connected, repeat testing should not be required.
Our Response:
EPA asked the Alliance for data to support this comment and did not receive a
response47. Therefore we are relying on the laboratory data that was gathered by EPA48 which
was referenced in the NPRM Appendix to the RIA. This EPA report shows that location of test
connection points does matter for leak detection depending on where the leak is located. The leak
standard requires two or more test points depending on the fuel evaporative system
configuration.
What Commenters Said:
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
Assuming implementation of the OBD regulation changes proposed in Section IV.C.S.d, EPA is
proposing an optional approach to a portion of the leak emission test procedure discussed in
Section IV.C.S.c. This optional testing approach would be included in the proposed IUVP/IUCP
testing program for the leak emission standard, but would not be used for certification testing for
1 Passavant, G. (June 2013) EPA and Auto Industry Meeting Related to Tier 3 Evap and OBD NPRM.
Memorandum to the docket
48 Smith, P. and Passavant, G., "Recommended Test Procedure and Supporting Testing Data for the Evaporative
Emissions Leak Test", December 2013.
4-414
-------
Tier 3 Summary and Analysis of Comments
the leak emission standard. It would be considered an approach which could be used by the
manufacturers to assess compliance with the leak emission standard. EPA could also use this
procedure for conducting assessments and asks for comment on using this procedure for
compliance purposes with a 0.02 inch cumulative equivalent diameter orifice standard.
We support this optional approach that would allow manufacturers to use the OBD evaporative
system leak detection hardware in lieu of running the stand alone in-use leak test. This option
allows manufacturers to rely on the OBD system, which is designed for this purpose, as the
primary compliance method, thereby reducing and simplifying testing requirements, as well as
reducing test time and costs.
Our Response:
EPA concurs and this option is part of the final Tier 3 rule.
4.8.4. Certification and Compliance
What Commenters Said:
Organization: National Association of Clean Air Agencies (NACAA)
Also related to evaporative emissions, with respect to the In-Use Verification Program (IUVP)
requirements for the leak emission standard, EPA notes in the proposal that fuel and evaporative
control system leaks are influenced to a significant degree by age as well as design and other
factors. The agency, therefore, seeks comment on whether to extend leak emission IUVP testing
to vehicles beyond the four-year age point, perhaps to six or eight years. Because of the
importance of in-use confirmatory testing for older vehicles, NACAA believes EPA should
extend testing beyond the four-year point.
Organization: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
EPA asks comment on the viability of extending leak emission IUVP testing beyond the nominal
four year point.
We do not support this change at this time. As EPA recognizes in the preamble, "in the past,
manufacturers have expressed concern about the implications of testing older vehicles and about
finding vehicles still within their warranty and recall liability periods" [78 FR 29905]. These
issues continue to be problematic. In fact, obtaining vehicles at the required high mileage point at
four years under the current IUVP test can be difficult. Any changes to vehicle age should be
considered as a broader and separate IUVP regulatory action that would include reductions in
test burden along with consideration of procurement of older vehicles still within warranty.
4-415
-------
Chapter 4: Proposed Vehicle Emissions Program
Industry is opposed to evap leak testing of customer vehicles at 6 to 8 years of age. Requiring
evap testing beyond the current 4 year IUVP point would require a second evap only leak test
fleet of vehicles. This would be costly and represent a new burden to industry.
Organization: California Air Resources Board (CARB)
The NPRM proposes additional measures to identify evaporative leaks on in-use vehicles based
on studies performed by the Coordinating Research Council and Colorado Department of Public
Health and Environment. Specifically, the NPRM proposes adding a test procedure to the IUVP
focused on testing a broader sample of, fuel/evaporative systems leaks in in-use vehicles.
Moreover, the NPRM proposes the addition of an IUVP evaporative leak check requirement
beginning with 2018 model year vehicles. This requirement would be included in the lUVP's
low and high mileage tests for any volatile fuel powered vehicle (except diesel and natural gas)
with a requirement that there be at least one representative vehicle for each
evaporative/refueling/leak family evaluated at each year point. EPA is proposing this approach in
lieu of creating a new set of requirements which would require a separate set of vehicles to be
procured for testing. In addition, the NPRM is considering extending the age point for leak
emission testing for IUVP beyond the current four year test point (e.g., testing vehicles six to
eight years old) to better assess aged vehicles for evaporative leaks.
U.S. EPA is also proposing to conduct an evaporative leak test on vehicles that are 6 to 8 years
old. Although this is doable for IUVP, the challenge will occur if enough vehicles fail the leak
check for a given test group and the manufacturer must then conduct an In-Use Confirmatory
Program. Under an In-Use Confirmatory Program, the manufacturer must procure ten properly
maintained and used vehicles which creates a challenge because many of these vehicles will be
out of warranty, sold to a second owner and/or operated beyond 75 percent of the certified useful
life and may not be properly maintained. Accordingly, CARB recommends that U.S. EPA
include language to require testing on vehicles no more than 6 years after the certified model
year and within 75 percent of the useful life mileage. For example, a manufacturer would be
required to conduct evaporative leak testing on 2015 MY vehicles within 75 percent of the useful
life mileage and before the end of the 2021 calendar year.
Our Response:
EPA has taken the conflicting comments into consideration and will not be including an
extension of the IUVP age in this rule. Given the benefits which can be achieved through
extending of the IUVP testing age to 6-8 years or more, our intention is to look at this more
closely in the future. At that time EPA will be able to consider the issues raised above. EPA
notes that with the useful life for emissions certification levels extended to 15 years and 150,000
miles, 75% is now over 110,000 miles and 11 years which is far beyond the 6-8 years identified
in the solicitation for comment, we would expect manufacturers to be able to find vehicles in
these age ranges suitable for IUVP testing.
4-416
-------
Tier 3 Summary and Analysis of Comments
5 Tier 3 Fuel Sulfur Program
What We Proposed:
The comments in this chapter correspond to Section V of the preamble to the proposed
rule and address the economic impacts of the program. A summary of the comments received
and our response to those comments are located below.
5.1. Proposed Tier 3 Gasoline Sulfur Standards
What Commenters Said:
Commenter: Chevron Products Company
If EPA does proceed with reducing gasoline sulfur content, we generally agree with the
mechanism of the program outlined in the proposal. We agree with maintaining the structure of
the current Tier 2 gasoline sulfur program and extending that structure to the new Tier 3
program. The use of an annual average standard per refinery, accompanied by separate
maximum per gallon cap standards at the refinery gate and downstream of the refinery, is the
appropriate way to implement the new program. This structure has worked well under Tier 2
and provides consistency for regulated parties as they transition to Tier 3.
Commenter: Manufacturers of Emission Controls Association (MECA)
MECA believes that a 10 ppm average gasoline sulfur standard with a gasoline sulfur cap of no
higher than 50 ppm provides the refining industry with an adequate, cost-effective compliance
pathway.
Our Response:
The final Tier 3 fuel program consists of a 10 ppm refiner annual average sulfur standard
(which allows a refiner to average over all of its refineries), with an 80 ppm refinery per-gallon
cap and a 95 ppm downstream cap. Please see Chapter 5.1.1 for more discussion on the 10 ppm
annual average sulfur standard, and Chapter 5.1.2 for more discussion on the per-gallon sulfur
caps.
5.1.1. Annual Average Sulfur Standard
5.1.1.1. General (incl. General Support & Opposition for 10 ppm Sulfur Standard)
What Commenters Said:
Support
5-1
-------
Chapter 5: Tier 3 Fuel Sulfur Program
Commenter: American Coalition for Ethanol (ACE)
We support EPA's proposal to ensure that gasoline and ethanol-blended gasoline contain no
more than 10 parts per million (ppm) sulfur on an annual average basis by January 1, 2017.
Commenter: American Thoracic Society
We are particularly supportive of the proposal to reduce the fuel sulfur content from 30 parts per
million to 10 parts per million.
The policy will also have the double benefit of allowing auto manufacturers to use improved
technology in catalytic converters and other pollution control technology to make future vehicles
emit even less pollution.
Commenter: Chrysler Group LLC
In the April 2013 Tier 3 hearings in Philadelphia and Chicago, Chrysler testified in support of
the Environmental Protection Agency's (EPA's) lowering the sulfur content of market fuel
gasoline.
Recommendation: Chrysler recommends that EPA adopt the 10 ppm sulfur market fuel standard.
Commenter: Clean Air Council
More specifically, the proposed rules would reduce the refinery annual average sulfur limit in
gasoline from 30 parts per million to 10 parts per million, which is a 66 percent reduction.
The Council applauds EPA's proposed reduction of the annual average sulfur limits from 38 to
10 parts per million.
Commenter: Ford Motor Company (Ford)
The proposed Tier 3 emissions standards program are cited by the EPA as a "systems approach
to reducing vehicle-related exhaust and evaporative emissions by addressing the vehicle and fuel
as a system." We agree with the EPA's identification of the need to reduce sulfur levels in
market fuel to 10 parts per million (ppm).
Commenter: Michigan Department of Environmental Quality (MDEQ)
The MDEQ, Air Quality Division is very supportive of the proposal to reduce sulfur
concentrations in gasoline to 10 parts per million (ppm) on an annual average basis by January 1,
2017.
Commenter: Truck and Engine Manufacturers Association (EMA)
5-2
-------
Tier 3 Summary and Analysis of Comments
In addition, EMA and its members have also been long-time proponents of high quality fuels,
and, as such, we are advocates for the removal of sulfur from both gasoline and diesel fuel.
Commenter: Union of Concerned Scientists (UCS)
The oil industry, which stands alone in its opposition to the Tier 3 program, would have you
believe that these standards are unnecessarily burdensome - that the challenge is too great, and
the costs are too high. This is simply not true. We have the technical capacity to meet these
standards. Scientists and engineers in the U.S. and around the world are already achieving the
goals of these standards, and doing so in a cost-effective way. And that is why the standards are
supported by health groups, labor organizations, the environmental and faith communities, the
auto industry, and state and local officials, among others. Don't let the oil industry's misleading
analysis stand in the way of improving public health for everyone.
Our Response:
We appreciate the comments and, as previously stated, are finalizing an annual average
sulfur standard of 10 ppm. For more information on the level of the standard, please see Chapter
5.1.1.2, below.
5.1.1.2. Level of the Standard
What Commenters Said:
Commenter: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
As EPA has noted, sulfur is a catalyst poison and precludes the full functioning of catalytic
exhaust after-treatment systems. Indeed, EPA assumed the existence of 10 ppm average sulfur
gasoline when it determined the feasibility of the Tier 3 emission standards EPA's new study of
the effects of sulfur on vehicle emissions found that there are strong benefits when reducing
sulfur from 28 ppm to 5 ppm, confirming earlier studies. The agency noted in the Regulatory
Impact Analysis that lower levels of sulfur can lead to "significant reductions" of emissions in
the in-use fleet.
The specific catalyst damage caused by sulfur and the concurrent reduction in the functioning
and efficiency of emission control equipment has been assessed in a number of studies, most
recently in the EPA study noted above. EPA noted that the relative effectiveness of NMOG and
NOx control are constrained by the sulfur level of the fuel, and found that there is clear damage
to the catalysts from sulfur exposure, and further found that catalysts never fully desulfurize,
even after a clean out procedure, as long as there is sulfur in the fuel.
Prior sulfur studies have also assessed the effects of sulfur on vehicle/fuel systems. The Ball
study found that 3 ppm sulfur fuel would reduce NOx emissions by 40% compared to the 33
ppm fuel, and/or allow savings because lower levels of precious metals would be required in the
5-3
-------
Chapter 5: Tier 3 Fuel Sulfur Program
catalyst. In addition, the Association for Emissions Control by Catalyst (AECC) study found
"The promising NOx adsorber technology that diesel and lean burn engines need requires sulfur
levels significantly below 10 ppm. This will avoid compromising the lower fuel consumption
and CO2 emissions by requiring frequent regeneration to remove the sulfur that is clogging the
NOx adsorption capacity".
The 10 ppm refinery average and a reduced retail cap are necessary to optimize and maintain the
function of engine and exhaust after-treatment technologies. During the combustion process,
sulfur, in both elemental and in compound forms, is present in the exhaust stream and is readily
deposited on the surface of exhaust after-treatment devices, including oxidation and reduction
catalysts as well as NOx adsorption components. It is on these surfaces precious metals are
placed by design to provide the conversion sites for the chemical reaction which converts
pollutants to harmless gases. Surface contamination with sulfur inhibits the conversion efficiency
of exhaust after treatment components. In fact, even low levels of sulfur impede the function of
these devices especially if they were previously subject to higher sulfur fuels.
While it is possible to dislodge the embedded sulfur through various mechanisms which raise the
exhaust gas temperature sufficiently to "burn off such deposits, in practice, it is far more
difficult to accomplish this removal. This is especially true given the nature of emerging, higher-
efficiency combustion technologies required under the new CAFE/GHG standards, which extract
more energy from the fuel. Reduced sulfur is necessary not only to help catalysts function better,
but also to prolong their life and to achieve a 25% increase in catalytic converter durability
(120K-150K miles useful vehicle life).
Commenter: National Association of Clean Air Agencies (NACAA)
Tier 3 Vehicle Technologies and Gasoline Are Already Available: Not only is Tier 3
tremendously effective from an air quality perspective — low in cost, high in cost effectiveness
and good for the economy — it is feasible today. EPA's proposed Tier 3 vehicle tailpipe
standards are modeled on California's LEV III program. The potential technologies for this
program are consistent with, and almost entirely the same as, those on today's California's
vehicles, including precious metal catalyst loading, optimized close-coupled catalysts, secondary
air injection pumps and evaporative control systems.
Further, California's gasoline already achieves 10-ppm sulfur on average. Finally, gasoline in
other nations, including those in the European Union and Japan, is subject to a 10-ppm cap.
China has also adopted requirements for 10-ppm sulfur, to take effect in 2017.
Commenter: Natural Resources Defense Council (NRDC)
To achieve the maximum cost-effective pollution reductions, the agency must consider the
interaction between gasoline sulfur levels and vehicle pollution control equipment because sulfur
buildup can limit exhaust catalysts effectiveness and shorten its useful life.
Commenter: Union of Concerned Scientists (UCS)
5-4
-------
Tier 3 Summary and Analysis of Comments
This will be particularly important over time as new vehicle greenhouse gas standards begin to
reduce exhaust temperatures, which can degrade catalytic converter performance. Reducing
gasoline sulfur can help offset this problem. UCS worked hard to secure the greenhouse gas
standards, and we are again committed to achieving the lower sulfur standards needed to protect
public health, and ensure the efficiency of the greenhouse gas standards as well.
Commenter: Flint Hills Resources, LP (FHR)
As a member of the American Fuel and Petrochemical Manufacturers (AFPM) trade association,
FFIR for the most part supports the comments submitted by AFPM, including opposition to a
more stringent average gasoline sulfur standard, due to inadequate technical justification and
negligible environmental benefits. Additionally, FFIR is submitting the attached comments to
emphasize, or perhaps clarify, our position related to certain detailed aspects of the proposal.
Commenter: Irving Oil Terminals Inc.
Assuming that EPA plans to adopt a reduced sulfur standard for motor gasoline, that standard
should be no more stringent than an annual average of 10 ppm sulfur. EPA has already
determined that such a standard is sufficient to enable vehicles to reach the proposed Tier 3
standards. Moreover, a standard below 10 ppm would be too difficult and costly to meet. For
example, EPA estimated that the cost of sulfur control compliance with a standard of 5 ppm
sulfur would be at least 50 percent higher than compliance with the proposed 10 ppm standard —
an unnecessary substantial financial burden. In addition, industry studies have estimated that a
standard below 10 ppm would result in substantial refinery shutdowns and supply disruptions.
Commenter: Countrymark; Small Business Refiners (SBRs)
As we examined the proposed rule, we found no technical support for the recommended limit of
10 ppm. The rule suggests that 10 ppm is a good break point where sulfur removal is cost
effective, but no data was provided to support this position. In addition, no commercial data
could be found by searching literature to support this position. CountryMark requests why the
limit is not 15 ppm to match diesel sulfur content, if the limit has to be changed.
We believe that the present sulfur levels are even lower than the above data because 1) it
includes small refiner gasoline which had a higher standard until 2011 and was estimated by
DOE to be approximately 11.9% of the gasoline supply in 2010 and 2) the above data represents
summer gasoline which would overstate the annual sulfur level as low sulfur butane is added to
winter gasoline in levels of 3-4 volume percent.
Pennsylvania Department of Environmental Protection (DEP )
The EPA should fully explain the justification for choosing a fuel sulfur limit of 10 ppm.
The EPA should better explain why 10 ppm was chosen. The EPA explained why a lower level
was not chosen, but did not fully explain why a higher level was not chosen.
5-5
-------
Chapter 5: Tier 3 Fuel Sulfur Program
Commenter: Sierra Club
EPA has provided robust evidence that regulating fuel quality, and specifically, reducing the
sulfur content of fuels from 30 ppm to 10 ppm pursuant to its Section 211 authority, will obtain
significant pollution reductions that could not otherwise be achieved by setting direct pollution
control standards for tailpipe exhaust and evaporative emissions alone. For example, in the
context of NOx reductions under the proposed rule, EPA's studies conclude that "the impact of
gasoline sulfur poisoning on exhaust catalyst performance and the relative stringency of the Tier
3 standards, particularly for larger vehicles and trucks, when considered together make a
compelling argument for the virtual elimination of sulfur from gasoline ... a gasoline sulfur
standard of 10 ppm ... represents the highest level of gasoline fuel sulfur that will allow
compliance with a national fleet average of 30 mg/mi NMOG + NOx." As such, EPA has
appropriately proposed direct pollutant emission limitations and fuel quality standards to set a
standard that is "achievable" and that meaningfully addresses the problems posed by motor
vehicle pollution. (42)
Moreover, EPA's studies also demonstrate gasoline sulfur content controls are a necessary
component to ensuring NOx reduction feasibility under the proposed rule. EPA's studies
conclude that "1) reversible sulfur loading is occurring in the existing Tier 2 vehicle fleet and has
a measurable effect on emissions of NOx, hydrocarbons, and other pollutants; 2) the
effectiveness of high speed/load procedures in restoring catalyst efficiency is a function of fuel
sulfur content; and 3) reducing fuel sulfur levels from 28 ppm to 5 ppm is likely to achieve
significant reductions of emissions of NOx, hydrocarbons, and other pollutants of interest in the
in-use fleet." EPA also provides that: [T]he impact of gasoline sulfur on NOx emission control of
catalysts in the fully-warmed-up condition is particularly of concern for larger vehicles.
Manufacturers face the most significant challenges in reducing cold-start NMOG emissions for
these vehicles. Because of the need to reach near-zero NOx levels, any significant degradation in
NOx emissions control over the useful life of the vehicle would likely prevent some, if not most,
largest vehicles from reaching a combined NMOG + NOx low enough to comply with the 30
mg/mi fleet-average standard ... [a]ny degradation in catalyst performance due to gasoline sulfur
would reduce or eliminate the margin necessary to ensure in-use compliance with the proposed
Tier 3 emissions standards.
Based on studies, EPA concludes that "A gasoline sulfur standard of 10 ppm also represents the
highest level of gasoline fuel sulfur that will allow compliance with a national fleet average of 30
mg/mi NMOG + NOx."
EPA adequately demonstrates that the 10 ppm gasoline sulfur content standard is feasible. EPA
makes the case that currently available technologies, some of which are employed in U.S.
refineries are achieving significant reductions in gasoline sulfur/
42 It is important to note that EPA is proposing an average sulfur standard of 10 ppm. However, the
actual sulfur content of fuel can vary substantially, based on variations in refinery operations and
contamination from gasoline being transported through pipelines to its final destination. Automakers must
then design vehicles to function with higher-sulfur gasoline. In order to give certainty to automakers
designing vehicles and to ensure emissions reductions, setting caps on per-gallon sulfur content is critical.
Currently, sulfur is capped at 80 ppm per gallon at the refinery gate and 95 ppm per-gallon downstream.
5-6
-------
Tier 3 Summary and Analysis of Comments
In this proposal, EPA is proposing two options: keep the per-gallon caps the same, or lower them to 50
ppm at the refinery gate and 65 ppm downstream in 2020, to allow small refinery provisions to take
effect. We urge EPA, as it lowers the average sulfur standard from 30 ppm to 10 ppm, to also lower the
per-gallon sulfur caps to at least 50 ppm at the refinery gate and 65 ppm downstream. Lowering the per-
gallon caps will allow automakers to employ additional advanced vehicle technologies that can increase
vehicle efficiency and lower emissions, while still allowing compliance flexibility for refiners.
Commenter: Emissions Control Technology Association (ECTA)
We agree with EPA's conclusion that the average sulfur content of gasoline must be reduced
from 30ppm to lOppm to ensure the effectiveness of catalysts necessary to reduce emissions.
EPA concludes that: "...the impact of gasoline sulfur poisoning on exhaust catalyst performance
provides a compelling argument, particularly for larger vehicles and trucks, that these vehicle
standards would be achievable only with a reduction of gasoline sulfur content from the current
30-ppm average down to a 10-ppm average'.
Section III of the FR Proposed Rule is an excellent review of the state-of-the-art understanding
of sulfur impacts on gasoline vehicle emissions. Further, achieving the Tier 3 levels of emissions
in all vehicle classes will require this level of sulfur. The description in Section IV of immediate
reductions in fleet criteria pollutants when pump fuel sulfur is reduced to <10 ppm is new,
relevant, and well conducted. We cannot add to these thorough analyses, but strongly endorse
them.
Let me get technical. The fundamental catalyst problem we are talking about here is that sulfur
is extremely surface active. It's unique to these types of chemicals. It absorbs readily on active
catalyst sites, blocking the amazing action we need and want to clean up toxins. Surface active
species are essentially — are especially critical in low concentration ranges where most of the
valuable active sites are sulfur poisoned in the sub 20 ppm sulfur range, depending on catalyst
design. Dropping from 30 to 10 ppm sulfur can have a huge impact on catalyst performance and
durability because of this fundamental physical chemistry behavior.
Commenter: Robert Bosch GmbH, Gasoline Systems, Germany
The limits for sulfur compounds needs to be fixed at least at the lowest level for sulfur in the
respective liquid fuel (diesel/gasoline) due to the same requirements for the exhaust gas
treatment systems, no matter what fuel is used.
Our Response:
As explained in Sections II and V.M of the preamble to the final rule, we are lowering the
existing gasoline sulfur standards under section 211(c)(l), because emission products of gasoline
with current levels of sulfur cause or contribute to air pollution which may reasonably be anticipated
to endanger public health or welfare, and because emission products of gasoline with current levels
of sulfur will impair to a significant degree the emissions control device or systems on the vehicles
subject to the final Tier 3 standards. As discussed in Section II.B of the preamble to the final rule,
control of gasoline sulfur to 10 ppm will lead to significant reductions in emissions of these
5-7
-------
Chapter 5: Tier 3 Fuel Sulfur Program
pollutants, with the benefits to public health and welfare significantly outweighing the costs. As
also discussed in Chapter 4 of this Summary and Analysis of Comments document (and further
in Section IV.A.6 of the preamble to the final rule and Chapter 5 of the RIA), we believe that a
standard of 10 ppm is appropriate, and when combined with the advances in emissions control
technologies, will be sufficient to meet the Tier 3 emissions standards. We also believe that the
10 ppm annual average standard will bring significant immediate benefits by reducing emissions
of the existing fleet. This sulfur standard is also feasible and is the level which appropriately
balances costs with the emission reductions that it provides and enables.
With regard to the comment on why a higher sulfur level was not chosen, we note that the
feasibility of the 30 mg/mi NMOG+NOx fleet average depends on exhaust catalyst systems that
require gasoline with average sulfur levels of 10 ppm or less. Further, annual average sulfur
levels greater than 10 ppm would significantly impair the emission control technology that we
expect will be used to meet the Tier 3 standards and to ensure in-use compliance over a vehicle's
useful life. This is particularly a concern for some larger vehicles that will need to reduce NOx
to near-zero levels, due to greater difficulty in reducing cold-start NMOG, in order to meet a
combined NMOG+NOx standard. As discussed in Section IV.A.6 of the preamble, increasing
gasoline sulfur to 20 or 30 ppm would make it impossible for vehicle manufacturers to meet the
Tier 3 standards. Achieving Tier 3 standards would require offsetting the resultant higher
emissions but EPA is not aware of existing technology or developing technology that could
address these higher emissions when taking into consideration the entire vehicle fleet. A higher
average sulfur standard would also forego the very large immediate reductions from the existing
fleet.
We also disagree with the comment that we did not provide data to support our position
that a sulfur standard below 10 ppm is not cost-effective. As discussed in Section V.B of the
preamble to the Tier 3 final rule, we also do not believe a sulfur standard lower than 10 ppm is
necessary to enable vehicles to meet the Tier 3 standards. While reducing sulfur below 10 ppm
would further reduce vehicle emissions and allow the Tier 3 vehicle standards to be achieved
more easily, we believe that a 10 ppm average standard is sufficient to allow vehicles to meet the
Tier 3 standards given the significant challenges associated with reducing sulfur below 10 ppm.
Our analysis shows that a 10 ppm annual average is sufficient to enable vehicles to reach the Tier
3 standards. Consequently, while reducing sulfur levels further would continue to yield
reductions from the in-use fleet, they would not be necessary to enable the new Tier 3 vehicle
standards to be met. Further, as explained in Section V.L of the preamble, while sulfur levels
would continue to reduce emissions from the existing fleet, reducing sulfur further below 10 ppm
becomes increasingly difficult and costly. FCC naphtha is very rich in high-octane olefms. As
the severity of desulfurization increases, more olefms are saturated, further sacrificing the octane
value of this stream and further increasing hydrogen consumption. Making up for this lost
octane represents a significant portion of the sulfur control costs. Furthermore, as
desulfurization severity increases, there is an increase in the amount of sulfur removed (in the
form of hydrogen sulfide) which recombines with the olefms in the FCC naphtha, thus offsetting
the principal desulfurization reactions. There are means to deal with the recombination
reactions, but they result in even greater capital investments. In addition, while FCC gasoline
contributes the majority of sulfur to the finished gasoline, as the sulfur level drops below 10
ppm, the sulfur level of the various other gasoline streams within the refinery also become
5-8
-------
Tier 3 Summary and Analysis of Comments
important. Any necessary treatment of these additional streams increases both capital and
operating costs.
U.S. refineries are currently in different positions technically and financially, and vary
greatly in size. It can be less costly per gallon for some to get down to 10 ppm, so a 10 ppm
average standard plus the flexibility afforded by the averaging, banking, and trading (ABT)
program helps those refineries with very high costs. However, if the gasoline sulfur standard
were lower, this would essentially end the ability of refiners to average sulfur reductions across
their refineries, because there would not be enough opportunity to generate credits at levels much
below 10 ppm. Further, as discussed in detail in Chapter 5 of the RIA, our analysis shows that
sulfur control costs for refineries to meet a standard below 10 ppm could be on the order of two
times more costly per ppm-gallon of gasoline sulfur reduced. A standard below 10 ppm could
also be cost-prohibitive for more challenged refineries. Lastly, such a standard would also
introduce additional costs to address the contribution to gasoline sulfur from gasoline additives,
transmix, ethanol denaturants, and contamination in the distribution system.
Therefore, we believe that the 10 ppm annual average standard will help reduce current
levels of sulfur that contribute to ambient levels of air pollution that endanger public health and
welfare. It will also help prevent significant impairment of the emission control systems
expected to be used in Tier 3 technology, significantly improve the efficiency of emissions
control systems currently in use, and continue prevention of the substantial adverse effects of
sulfur on the performance of vehicle emissions control systems.
5.1.1.3. Vehicle Need/In-Use Benefits/GHG & CAFE
What Commenters Said:
Commenter: Advanced Engine Systems Institute (AESI)
As MECA has pointed out, compliance with the proposed standards is both technically feasible
and cost-effective. However, as automakers deploy new and increasingly more efficient engines
that produce less waste heat, the catalysts and other emissions control devices that now depend
on using that excess heat to burn off damaging deposits will be significantly more vulnerable to
sulfur poisoning and degradation, unless the amount of sulfur in fuel is simultaneously reduced
to 10 parts per million or lower, as has been proposed.
Commenter: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
EPA acknowledges that reducing the sulfur content of gasoline to a 10-ppm annual refinery
average is a vital component of the proposed rule. We agree. In fact, our ability to effectively
meet the Tier 3 requirements depends on EPA reducing the sulfur in market fuel. Sulfur inhibits
the catalytic converter's ability to reduce vehicle emissions, so lower sulfur at the pump means
fewer exhaust emissions in the air. And because lower sulfur reduces emissions from all
5-9
-------
Chapter 5: Tier 3 Fuel Sulfur Program
vehicles, the proposed sulfur reductions would achieve Day One benefits, immediately reducing
emissions from every gasoline-powered vehicle on our roads, no matter how old.
Support for Reductions in Tier 3 Market Fuel Gasoline Sulfur Standards: There are a number of
issues related to market gasoline which are appropriately within the scope of the Tier 3 final rule,
or a supplemental Tier 3 rulemaking. Such issues are integral to product design, planning, and
implementation for all OEMs because vehicles and fuels are a system and clean, efficient
operation of vehicles is limited by the fuels on which they depend. EPA itself echoed this
concept in the NPRM Preamble noting that its proposal promotes a "systems approach to
addressing the impacts of motor vehicles and fuels on air quality and public health' (33).
We agree with changing the first of three prongs of the Tier 2 market sulfur standards, the annual
refinery average, from 30 ppm to 10 ppm. As EPA notes, there is ample evidence that sulfur is a
catalyst poison, and there is a "compelling case" that significant reductions of sulfur are needed
to meet increasingly stringent criteria pollutant requirements in the Tier 3 standards. The Agency
noted that "because any amount of sulfur in fuel can [damage catalysts] the lower sulfur the
better." Sulfur reduction also supports increased vehicle efficiency to meet CAFE/GHG
requirements.
To be clear, automakers support EPA reducing the annual refinery average for gasoline sulfur
from 30 to 10 ppm.
OEMs require further reductions of the content of sulfur in gasoline, to continue to innovate and
meet important environmental goals included in Tier 3 and CAFE/GHG requirements. Based on
recent projections, the internal combustion engine (ICE) will continue to be the predominant
technology through the next several phases of vehicle emission standards, at least until 2025.
Thus, it is increasingly important to determine the best way for the internal combustion engine to
meet more stringent emission and fuel efficiency requirements.
Automakers support EPA in its efforts to reduce the sulfur content of fuel, and the 10 ppm
annual average in the proposed rule is a step in the right direction.
We support the 10 parts per million average and a low cap for sulfur and gasoline.
In addition to needing ultra-low sulfur fuel to achieve the proposed emission standards, ultra-low
sulfur fuel is also critical to auto manufacturers' effort to meet the combined 54.5 miles per
gallon fuel economy and greenhouse gas emission standards by 2025. And, most importantly,
the ultra-low sulfur fuel will provide immediate and substantial public health benefits from the
existing fleet of nearly 250 billion vehicles on the road today.
According to the National Association of Clean Air Agencies, the impact on the fleet is
equivalent to taking over 30 million vehicles off the road. And today vehicles are lasting longer,
and the average age of all vehicles on the road is 11 years. While the proposed standards will be
another great step in reducing the fleet's emissions, it will take time for fleet turnover to fully
realize the benefits. In the meantime, the ability to immediately and directly reduce emissions
from the vehicles on the road today by reducing sulfur and gasoline cannot be matched.
5-10
-------
Tier 3 Summary and Analysis of Comments
EPA acknowledges that reducing the sulfur content of gasoline to 10 part per million average is a
hugely important component of the proposed rule. We agree. We know that sulfur inhibits the
catalytic converter's ability to reduce emissions. Lower sulfur fuels at the pump mean fewer
exhaust emissions in the air. It's that simple.
Further, sulfur reductions in the market fuels produce day one benefits. So while it will take a
couple of decades for the current fleet to be replaced by Tier 3 compliant vehicles, reducing
sulfur at the pump immediately will reduce emissions from every gasoline powered vehicle on
our road, no matter how old.
Gasoline sulfur reduction is also a critical element needed to enable benefits of future alternative
vehicle technologies, for example, Direct Injection Lean Burn engines which have greater energy
efficiency, as well as HCCI (homogenous charge compression ignition), pHCCI (premixed
HCCI), RCCI (reactivity controlled compression ignition), and lean cruise/idle technology.
Without this change, fuels and vehicles CANNOT operate as an integrated system and
synergistically achieve emissions and energy efficiency goals. It is appropriate for regulators to
put some burden on fuel suppliers to do their part to invest and reduce sulfur to optimize the
emissions reductions of the system. The projected cost of the Tier 3 system is far higher for
automakers than for the refinery industry (See April 24, 2013 Alliance Tier 3 Testimony, and
EPA RIA at ES-8 ff.)
But these standards cannot be met with vehicle changes alone, and the Tier 3 proposal recognizes
another critical component in reducing vehicle emissions, and that is fuel quality, or, more
specifically, the sulfur content in gasoline. We believe the proposed fuel standards for a
nationwide supply of clean, ultra-low sulfur gasoline is critical for auto makers to meet the new
and highly stringent vehicle emission standards.
33 - 78 Fed. Reg. 29908 at 29816 (May 21, 2013).
34 - As noted in Alliance testimony for this rulemaking, according to EPA, automakers will spend about
$15 billion in just ten years to meet the requirements of this proposed rule, and effectively remove tailpipe
emissions from the environmental equation. As EPA further states, combined with the fuel economy and
greenhouse gas program, this rule would bring auto investment in the vehicle emission reduction
programs through 2025 to more than $216 billion -about 100 times the investment that the oil industry is
being asked to make under the NPRM.
47 - 78 Fed. Reg. 29908, 29858 (May 21, 2013).
Commenter: American Honda Motor Co., Inc.
As the agency is aware, through the use of sophisticated emissions control technology, the
automotive industry has been able to achieve extraordinarily low exhaust levels throughout a
vehicle's life. The anticipated Tier 3 standards - and the LEV III standards of California - will
push those emissions to near-zero levels. But doing so requires the use of vehicle technology and
fuels that complement one another. Sulfur in gasoline has long been known to adversely affect
catalyst performance and, as a result, achieving the Tier 3 and LEV III stringencies is highly
dependent on the vehicles' use of low-sulfur fuel.
5-11
-------
Chapter 5: Tier 3 Fuel Sulfur Program
Lowering sulfur will enable new vehicles to better meet the cleaner proposed emissions
standards for the duration of their useful life, improving air quality as those vehicles work their
way into the vehicle population. It's also worth mentioning that lowering the sulfur content of
gasoline has another, less gradual benefit. As the agency notes in the proposed rule, cleaner fuel
"facilitates immediate emissions reduction from all vehicles on the road at the time the sulfur
controls are implemented." Failing to bring cleaner fuel to market in a timely manner would miss
a fundamental opportunity presented by this systemic approach to regulating both vehicles and
fuels.
Setting fuel sulfur levels comparable to those currently set in California would ensure that the
Tier 3, LEV III, fuel economy, and vehicle greenhouse gas policies are working not at odds with
one another, but rather in a complementary manner, opening doors to future designs that are both
cleaner and more efficient.
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM)
Further reducing sulfur is not necessary for meeting new stringent vehicle emission standards,
and auto makers are unlikely to introduce the vehicle emission technologies that are enabled by
the lower sulfur fuel. And I'm talking about lean burn GDI.
Auto makers insist they need 10 ppm gasoline to enable lean burn GDI engines, and yet in
Europe and Japan where 10 ppm fuel is available, lean burn GDI engine penetration peaked at
about two percent and then tailed off. Research shows it maxing out at about three percent of the
U.S. fleet. Diesel sulfur was kept at 15 ppm to enable an emissions control technology that never
took off in the marketplace. Don't make the same mistake with gasoline. Auto makers can meet
the emission standards without further reductions to gasoline sulfur.
Commenter: Appalachian Mountain Club (AMC)
The adjustment in sulfur fuel content is needed as sulfur fouls the efficiency of vehicle catalytic
converters, a key system component that reduces smog forming emissions and toxic air
pollution.
As stated above, AMC is very supportive of the provision that will reduce the sulfur content of
gasoline from 30 parts per million (ppm) to 10 ppm . This will result in immediate emission
reductions when implemented in 2017.
Commenter: BMW of North America, LLC
To assist automakers in their efforts to comply with these stringent Tier 3 emission standards as
well as with the upcoming MY 2017-2025 GHG/CAFE requirements, BMW strongly supports
EPA's proposal that Federal gasoline, including any ethanol-gasoline blend, contain no more
than 10 parts per million (ppm) of sulfur on an annual average basis by January 1,
2017. Uniformly low gasoline sulfur levels throughout the U S marketplace will enable
5-12
-------
Tier 3 Summary and Analysis of Comments
automakers to effectively optimize vehicle emission control technology to meet these rigorous
standards in the required timeframe.
Commenter: Chrysler Group LLC
Sulfur is the single most important gasoline parameter to enable manufacturers to broadly
introduce gasoline direct injection ("GDI") technologies to meet EPA's air quality and
greenhouse gas (GHG) requirements. The cause and effects are well known: Sulfur inhibits
exhaust after-treatment conversion efficiency, so lower sulfur means lower emissions.
In fact, EPA considered these very issues in the heavy-duty diesel context, so the discussion has
already taken place. Specifically, in the 2007 Heavy-Duty Highway Rule, EPA recognized the
necessity of adopting low sulfur diesel fuel to support diesel direct injection technology with
advanced exhaust aftertreatment systems, the very same technology (but as applied to gasoline
engines) that EPA seeks to foster in the current proposed rule, which, again, necessitates the
adoption of low sulfur gasoline.
Additionally, EPA's creation of new fuel economy and GHG standards anticipated the
nationwide availability of low sulfur gasoline. Consistent with that basis, Chrysler supported the
concept of One National Program to improve fuel economy and reduce GHG emissions with the
understanding that low sulfur gasoline would be called for in this next round of establishing
tailpipe emissions standards, and which would ensure availability to enable advanced lean-burn
GDI technologies.
Advanced lean-burn GDI systems are an enabling technology to meet EPA and the National
Highway Traffic Safety Administration's (NHTSA's) 2025 model year (MY) GHG and fuel
economy requirements. As such, low sulfur gasoline, at 10-ppm or lower, is a prerequisite for
lean-burn GDI systems achieving EPA/NHTSA's GHG and fuel economy benefits. Conversely,
without low sulfur gasoline, lean-burn GDI systems incur a significant fuel economy (and
corresponding GHG) penalty due to more frequent sulfur regeneration events.
The auto industry has already invested hundreds of millions of dollars to design, develop, tool
and in some cases manufacture and produce lean-burn GDI systems. Without absolute assurance
of low sulfur gasoline in the marketplace, the performance and GHG benefits of this promising
technology will be significantly compromised.
As in previous rulemakings, Tier 3 continues to treat vehicles and fuels as a system, recognizing
that fuels are critically important to achieve the full potential of advanced emission control
technologies, especially during these historic time when regulations are driving near zero
emissions and doubling fuel economy. Sulfur is the single most important gasoline parameter to
enable vehicle technologies to meet these new requirements. The cause and effects are well
known. Sulfur inhibits exhaust after treatment conversion efficiency, so lower sulfur means
lower emissions.
Commenter: Clean Air Council
5-13
-------
Chapter 5: Tier 3 Fuel Sulfur Program
The Alliance of Automobile Manufacturers, which includes BMW Group, Ford Motor Company,
and others, has publicly stated that it supports 'the lowest possible sulfur content at retail for
uniformity at the pump for planning future technologies, and also to get the immediate benefit of
existing vehicles as well as future ones, reducing sulfur poisoning of the catalyst and, therefore,
minimizing exhaust emissions of the vehicle.'
Commenter: Delaware Department of Natural Resources (DNRC)
Sulfur in fuel inhibits the performance of catalysts that control vehicle exhaust, resulting in
increased vehicle emissions. Reducing sulfur in fuel will reduce mobile source emissions of
nitrogen oxides, carbon monoxide, fine particulate matter, and volatile organic compounds and
continue the effective, proven approach of treating vehicles and fuels as an integrated system.
Importantly, these reductions will begin immediately with the existing fleet so long as the
vehicles are equipped with a functioning catalytic converter. Lower sulfur gasoline will also
facilitate the deployment of advanced technologies that will improve fuel economy, reduce
gasoline consumption and save consumers money. The effectiveness of EPA's proposed Tier 3
standards will continue to grow over the years as newer vehicles replace older ones.
Commenter: Emissions Control Technology Association (ECTA)
The EPA neglects other benefits from low-sulfur gasoline. To meet the 2025 GHG regulations,
automakers will leverage all their engine tools. Lean-burn direct-injection gasoline (lean GDI) is
one not mentioned by the EPA. In the 2020-25 timeframe, CO2 reductions will cost about $50-
75 per percent (Corning Review, SAE 2013-01-0538). Lean-burn GDI (gasoline, direct injection)
engines will deliver >10% CO2 reductions (Daimler SAE 2013-01-1299; Mercedes Kemmler
Vienna Motorsymposium 2013) versus stoichiometric GDI engines, and it is our estimate that
incremental costs will be less than $500. Most of the engine costs are in calibration, and the
incremental exhaust costs reside in replacing a TWC with two LNT (lean NOx traps). This
technology is not being considered for the United States due to high-sulfur gasoline (Mercedes,
Kemmler, Vienna Motorsymposium, 2013).
Commenter: Environmental Defense Fund (EDF)
Cleaner fuel will also enable the technologies needed to meet the Tier 3 exhaust standards and
enable the vehicles to meet the standards in-use for the duration of their useful life. Finally,
lower sulfur gasoline facilitates the development of lower cost technologies to improve fuel
economy.
The preamble to the proposal and the RIA clearly outline the technical need for additional sulfur
requirements. As EPA states in the preamble, "[R]obust data from many sources shows that
gasoline sulfur at current levels (i.e., around 30 ppm on average) continues to degrade vehicle
catalytic converter performance during normal operation. The most significant problem is for
NOx. Today's proposed NMOG+NOx vehicle emission standards, an 80 percent reduction from
current Tier 2 standards, would not be possible without the gasoline sulfur controls we are
proposing today." This determination has been echoed by the Manufacturers of Emission
5-14
-------
Tier 3 Summary and Analysis of Comments
Controls Association in their recent report describing the adverse impact of fuel sulfur on
emission control technology (97).
In addition to protecting catalysts on existing vehicles and Tier 3 vehicles, automakers and
emissions control manufacturers support cleaner fuel because it will allow for additional cost-
effective technologies to meet the later-year fuel economy and GHG standards. For example, the
lean burn gasoline direct inject (GDI) engine, which offers a 5-15 percent fuel economy
advantage over the stoichiometric GDI, is currently being used in Europe where 10 ppm fuel is
available. The advanced NOx catalysts needed to maintain low NOx emissions in a lean GDI
engine are especially sensitive to sulfur. The lean burn GDI could be deployed by automakers to
meet the 2025 fuel economy and GHG emissions standards in the US if the 10 ppm fuel standard
is finalized. Indeed, many manufacturers in the U.S. are developing the lean GDI technology in
anticipation of Tier 3 fuel standards (98).
97 - MECA, The Impact of Gasoline Fuel Sulfur on Catalytic Emission Control Systems (June 2013).
98 - Id. ;.MECA, LEV III and Tier 3 Exhaust Emission Control Technologies for Light-duty Gasoline
Vehicles (Apr. 2013).
Commenter: Ford Motor Company (Ford)
Sulfur, which occurs naturally in crude oil, is known for its notorious ability to poison catalysts
and dramatically reduce the functionality of advanced emissions systems. Laboratory research
has also shown sulfur to increase light-off temperature and decrease conversion efficiency, in
addition to slowing the rich to lean air/fuel (A/F) transition. Sulfur in gasoline is also known to
cause corrosion of metallic components of the engine and fuels system. A reduction in market
gasoline sulfur levels can immediately reduce the emissions from all vehicles on the road.
Improved fuel economy and reduced CO2 emissions are made possible in vehicles through
engine operation at lean air-fuel ratios. While lean A/F ratios result in decreased hydrocarbon
(HC) and carbon monoxide (CO) emissions, they increase oxides of nitrogen (NOx) emissions as
three-way catalysts (TWC) have their highest efficiency for NOx reduction during stoichiometric
or fuel rich conditions. Catalysts undergo sulfur poisoning when the sulfates formed during
combustion (from the sulfur in the fuel) compete with NOx for the catalytic active sites on the
TWC; this in turn reduces the lean NOx trapping functionality of the catalyst. Laboratory tests
have shown a greater than 40% reduction in NOx emissions when decreasing sulfur from 30 ppm
to 5 ppm.(5)
In order to meet the increased CAFE requirements of tomorrow, manufacturers are developing
lean-burn engines that will reduce fuel consumption, but this requires sophisticated sulfur-
sensitive NOx control technologies. Thus, the lowering of market sulfur levels in fuels to 10 ppm
maximum has been proposed in countries with advanced fuel economy and emissions
requirements that utilize such technologies.
In addition to worldwide efforts in mobile source emissions reductions, ambient air quality
continues to be a global issue of interest. Air quality improvement efforts are often made
possible through maintaining ozone and smog levels to acceptable ranges. Smog consists of fine
5-15
-------
Chapter 5: Tier 3 Fuel Sulfur Program
particulate matter such as sulfate and nitrate aerosols. As stated above, fuel sulfur has a
significant impact on NOx emissions that contribute to the formation of such aerosols, which
furthers the importance for lowering fuel sulfur levels. Sulfur in the fuel has also been linked to
the formation of sulfates in the exhaust stream. The lowering of sulfur levels in gasoline (and
also diesel fuel) has been linked to lower smog-forming particulate matter emissions and
improved air quality.
Recommendation: As communicated at the EPA and NHTSA public hearing on April 29, 2013
on the Tier 3 NPRM, Ford strongly supports a reduction of sulfur levels to 10 ppm maximum for
market fuels in order to meet the strict emissions standards and maintain the integrity of the
engine and emissions system throughout the vehicle's useful life and fully supports the detailed
Alliance comments on sulfur.
5 SAE 2011-01-0300, D. Ball, et al., Effects of Fuel Sulfur on FTP NOx Emissions from a PZEV 4 Cylinder
Applications.
Commenter: General Motors LLC (GM)
There are a number of issues related to the market gasoline which are appropriately within the
scope of the Tier 3 final rule, or a supplemental Tier 3 rulemaking. Such issues are integral to
product design, planning and implementation for all OEMs because vehicles and fuels are a
system and clean, efficient operation of vehicles is limited by the fuels on which they depend.
EPA itself echoed this concept in the NPRM Preamble noting that its proposal promotes a
"systems approach to addressing the impacts of motor vehicles and fuels on air quality and
public health."
While OEMs discussed the interconnection of fuels and vehicle issues with EPA during the
negotiations leading to the MY 2017-2025 GHG/FE requirements, EPA declined to include fuel
issues in that rulemaking. However, at that time EPA gave the industry assurance that these
issues would be addressed promptly within the Tier 3 rulemaking in order to facilitate the needed
product changes to support the GHG/FE standards, as well as Tier 3 requirements. OEMs are
spending billions of dollars on compliance with the GHG/FE rules and require the cleanest fuels
possible to meet these obligations. (13) We urge EPA to support the progress toward cleaner
vehicles by ensuring that the proper fuels are widely available.
GM supports EPA in its efforts to reduce the sulfur content of fuel, and the 10 ppm annual
average in the proposed rule is a step in the right direction.
Another key aspect of the proposal is the cleaner fuels that are included. Auto makers will need
these cleaner fuels to meet the stringent Tier 3 emission standards EPA and California envision.
We must have clean fuels because the vehicles and fuel function as a single system in
determining the emission performance of the vehicle.
As EPA moves ahead to finalize the Tier 3 requirements, GM believes the provision for low
sulfur fuel levels and other clean fuel properties is essential, as is outlined in the Alliance's
comments. And while clean fuels are needed to meet the Tier 3 standards on new vehicles, they
also provide the added benefit of reducing emissions immediately across the entire on road fleet.
5-16
-------
Tier 3 Summary and Analysis of Comments
13 According to EPA, automakers will spend about $15 billion in just ten years to meet the requirements
of this proposed rule, and effectively remove tailpipe emissions from the environmental equation. As EPA
further states, combined with the fuel economy and greenhouse gas program, this rule would bring auto
investment in the vehicle emission reduction programs through 2025 to more than $216 billion -about
100 times the investment that the oil industry is being asked to make under the NPRM.
Commenter: Private Citizen
Vehicles and fuels are a system. Setting a strict gasoline sulfur standard of 10 ppm will cleanup
every car on the road and enable the next generation of cleaner vehicle technologies. The result
would be like taking 33 million cars off the road, and these benefits can be achieved for a penny
or less per gallon of gasoline.
Commenter: Hyundai Motor Group
The Hyundai Motor Group supports EPA's proposal to lower the annual average sulfur retail
gasoline levels to 10 ppm. We believe a systems approach is necessary to reduce emissions to
the stringent levels called for in the proposal. There is a large body of evidence that shows that
with reduced sulfur levels, we will see an immediate benefit from all vehicles in the existing
fleet, not just on vehicles with the latest technology.
Sulfur inhibits the ability of after-treatment technology to reduce emission levels of NOx, VOCs,
PM, carbon monoxide, and air toxics, by depositing on the precious metal catalysts.
As we work to meet all of the low emission levels proposed in the Tier 3 rule, particularly the
ppm standards. We need the catalyst to be working as efficiently as possible to meet both
certification and in use requirements. Also as sulfur degrades the catalyst, it reduces useful life.
Auto makers are required to meet useful life standards of 150,000 miles for emission-related
equipment as part of this regulation, so it is important that the technology works as intended.
Additionally, as auto makers introduced lean burn technologies that we see as necessary to meet
stringent future GHG regulations, low sulfur levels must be implemented to minimize
detrimental impact to NOx after treatment.
Commenter: Johnson Matthey
One of the important aspects of the Tier 3 proposal is to reduce the sulfur level in the
gasoline fuel. Sulfur compounds cannot be transformed into something harmless in the way that
a NOx molecule is, turning to NOx catalysts. Therefore, all sulfur emissions from the engine are
bad for the environment, and the only way they can be prevented from entering the environment
is by removal from the fuel itself.
The catalyst technology Tier 3 proposal is across the whole vehicle fleet. That's gasoline and
diesel vehicles. It exists today, and it's already in the marketplace. New generations of catalysts
are continually being developed, and they have further enhancements and performances.
Implementation of these catalyst technologies to meet Tier 3 proposals is what our industry
5-17
-------
Chapter 5: Tier 3 Fuel Sulfur Program
thrives on. We have a long track record of achieving these tighter emissions limits successfully,
and this proposal would be no different.
Sulfur also poisons the efficiency of the catalyst for criteria emission control. To ensure that the
full range of light duty vehicles comply with Tier 3 proposals, both at time of certification and
during their use in the real world, the proposed lower gasoline sulfur levels, 10 ppm average, are
needed. The EPA has recently completed an excellent and extensive evaluation of sulfur effects
on the current Tier 2 vehicle fleet that clearly shows significant emission benefits from reducing
gasoline sulfur levels on those existing vehicles.
And as we move to the future, vehicle manufacturers are working hard to meet greenhouse gas
requirements. Exhaust temperatures are expected to get cooler, exacerbating these sulfur
poisoning effects on the catalyst. This remaining clean fuel is increasingly important as we
move forward.
Having low sulfur fuel may also enable new technologies to be introduced, such as NOx
absorber catalysts, which are leading catalyst technology for highly fuel efficient clean burning
gasoline engines, and need these ultra-low sulfur fuel levels to function at peak effectiveness,
and allow these engines to meet their fuel economy potential. Sulfur in diesel fuel is already
limited to a maximum of 10 ppm for these reasons, and sulfur and gasoline should also be
limited to the proposed 10 ppm average.
Commenter: ECTA
To be sure, we have cars today that are meeting the Tier 3 average fleet requirements with 30
ppm sulfur fuel. However, all these cars are small and compact. It is a very different story in
trying to get a large vehicle to Tier 3 average emission levels. Not only are the catalysts
compromised, but the sensors are affected, causing complexities in on board diagnostics, which
are so critical to proper fleet emissions monitoring. We demand 150,000 durability in our cars.
High sulfur fuel makes this very difficult.
We also have catalyst formulations that are not as sensitive to sulfur as others. This is partly the
result of unique formulations derived from — through catalyst R&D. These formulations are
usually intellectual property of the developers. They do not have 100 percent of the market
share.
Further, for many reasons, the catalyst might be designed differently to meet the the market or
technical requirements, or be placed in different conditions. We all drive different vehicles with
different engines and different designs. The field is very complex. This is why we need the low
sulfur fuel because the technologies and engines are becoming so complex.
I just attended the SAE International Annual Congress in Detroit. Fantastic. Developments are
coming very quickly, but — very important — most researchers are using clean fuels in their
development of engines and catalysts. We will see commercialized gasoline engines by 2020
that consume 20 to 30 percent less fuel. These new engines are complex, and many are lean, like
diesel. Lean NOx mitigation for gasoline requires less than 10 ppm sulfur fuel. It doesn't make
5-18
-------
Tier 3 Summary and Analysis of Comments
sense to tie one arm behind the back of the auto engineers by requiring them to use relatively
dirty fuel.
Commenter: Manufacturers of Emission Controls Association (MECA)
A critically important element to ensuring that future gasoline vehicles will be able to comply
with EPA's proposed Tier 3 emission limits is EPA's proposed reduction of gasoline fuel sulfur
levels to a 10 ppm national average starting in 2017.
MECA has consistently supported the introduction of the lowest possible fuel sulfur limits used
with mobile sources to enable the use of best available exhaust emission controls and minimize
the impacts of fuel sulfur on catalyst performance and durability.
Commenter: Mercedes-Benz USA, LLC on behalf of Daimler AG
The impact of a 10 ppm retail average sulfur limit in the Tier 3 proposal is immediate and
enduring for both new vehicles as well as legacy fleets.
Mercedes-Benz strongly supports the 10 ppm annual average for gasoline in the Tier 3 proposal
and requests that this element be incorporated in the final EPA Tier 3 rulemaking. This
advancement in fuel quality will immediately benefit the existing vehicles on the road by
improving the function of exhaust aftertreatment components and by decreasing the formation of
smog-producing pollutants.
In 2006, Mercedes-Benz first capitalized on the value of low sulfur gasoline and the reduction to
fuel consumption it enables when it introduced in the EU the 350CGI CLS luxury sedan, which
is equipped with stratified lean burn combustion. Depending on the driving mode, engine speed,
and load, gains in the fuel efficiency can range between five and 15 percent as compared to
conventional stoichiometric combustion. Incorporation of this technology in the EU is enabled
by a 10 ppm refinery limit in comparison to the 10 ppm annual average as proposed in the Tier 3
rulemaking.
But sulfur reduction is not a lean burn discussion. Even existing power train architectures,
including downsized turbocharged gasoline direct injected engines would also benefit from
reduced sulfur and fuel, so no longer would a portion of the fuel savings realized need to be
expended by periodically burning off sulfur to maintain optimum effectiveness in exhaust after
treatment systems.
Additionally, prior testimony highlights the fact that power train technologies which increase
efficiency result in core exhaust gas temperatures, which promote sulfur deposition and inhibit
burn off. Thus, the effect of low sulfur on the current fleet is immediate and enduring. No other
single step can provide retroactive and future benefits in criteria emissions and greenhouse gas
reduction.
Commenter: Motor & Equipment Manufacturers Association (MEMA)
5-19
-------
Chapter 5: Tier 3 Fuel Sulfur Program
The steady phase-in of lower sulfur fuel will provide greater opportunities for deploying new
technologies, such as lean-burn engines, and for improving emissions and fuel economy
performance.
Commenter: National Automobile Dealers Association (NADA)
Vehicle emissions control is a direct function of the interaction of motor vehicle components and
systems and motor vehicle fuels. The time-honored tug-of-war between 'auto' and 'oil' is a
function of the basic laws of chemistry and physics. To the extent fuels can be controlled to
make them cleaner or easier to clean, less need be done by vehicle components and systems. The
reverse is also true.
Commenter: New York State Department of Environmental Conservation
Maximizing the benefit to the environment from this investment in catalytic exhaust
aftertreatment requires appropriate reductions in fuel sulfur content. The Department urges EPA
to implement the annual average gasoline sulfur standard as proposed, with most gasoline being
subject to a 10 ppm sulfur standard in 2017.
However, the proposed gasoline sulfur standard is a significant component of the emission
control system. The reduction is gasoline sulfur provides substantial benefits, particularly in the
early years of the program, due to emission reductions from the existing vehicle fleet. Lower
gasoline sulfur content will also minimize the in-use emissions of Tier 3 and LEV III vehicles
throughout their lives.
The engine exhaust after treatment fuel needs to be treated as a system, as EPA has recognized
and done with the Tier 3 and now the Tier 3 standards and heavy duty standards as well. And we
fully support the proposal here to regulate the system again. Reduction in the sulfur content of
gasoline facilitates improvements in engine and exhaust after treatment signs that can further
reduce emissions of ozone precursors, as well as reductions in greenhouse gas emissions
improvements in fuel economy as other technology is enabled.
Sulfur is well-known poison catalysts that are vital for meeting existing and proposed standards.
There's been concern expressed about this for many years now, and Tier 3 made a lot of progress
in reducing standards that - sulfur standards, and now it's time to go the next step and bring it
down even further.
Commenter: Pennsylvania Department of Environmental Protection (DEP)
EPA's proposal would establish more stringent vehicle emissions standards and reduce the sulfur
content of gasoline beginning in 2017. According to EPA, the proposed gasoline sulfur standard
of 10 parts per million (ppm) average, down from the current 30 ppm average, would make
vehicle emission control systems more effective for both existing and new vehicles, and would
enable more stringent vehicle emissions standards.
5-20
-------
Tier 3 Summary and Analysis of Comments
DEP realizes the potential for significant emission reductions from the proposal as soon as the
Tier 3 gasoline standards take effect. A 10 ppm average sulfur level in gasoline would allow
Tier 3 vehicles' three-way catalysts to perform at an optimum level and potentially allow the use
of more cost-effective emission controls on vehicles. In addition, a 10 ppm average sulfur level
in gasoline will immediately lower emissions from vehicles that comply with Tier 2 standards.
These reductions would occur because the lower sulfur levels in Tier 3 gasoline will reduce
sulfur 'blinding' of the catalysts that has occurred from the use of current gasolines. Catalyst
'blinding' inhibits the performance of the catalysts and consequently Tier 3 gasoline will provide
for almost immediate additional NOx emissions reductions from the existing fleet.
Commenter: Union of Concerned Scientists (UCS)
One of the unique benefits of Tier 3 is that, by lowering sulfur in the gasoline that goes into
vehicles already on the road, the program will have an immediate positive impact. Reducing the
sulfur content of gasoline helps existing catalytic converters work more efficiently, reducing the
emissions from the legacy fleet even while we wait for the vehicle fleet to turn over. By setting
standards for the fuel and the vehicle together (what is referred to as the "system based
approach"), the Tier 3 standards are able to achieve the greatest emission reductions at the lowest
cost to society.
Commenter: Utah Air Quality Board
The Board recognizes that the proposed Tier 3 Program addresses the vehicle and its fuel as an
integrated system. The proposed low-sulfur gasoline component (Tier 3 Gasoline) — which
would reduce sulfur content to an average of 10 ppm - is required to enable the advanced
emissions reduction technologies that will be installed on new vehicles by automobile
manufacturers beginning in 2017. In addition to enabling these advanced pollution cutting
technologies, the Board recognizes that the Tier 3 Gasoline with 10 ppm sulfur content will also
yield immediate and significant emissions reductions in our existing fleet of vehicles beginning
on the day the fuel becomes available at the pump. We support the penetration of Tier 3 Gasoline
into the market in Utah as soon as possible.
Commenter: Sierra Club Southeastern Pennsylvania Group
According to the Delaware Valley Regional Planning Agency, the area MPO, conformity with
the State implementation plan is only being achieved through the continuing improvements in
vehicle-related pollution controls. Now we can assist the manufacturers of the nation's vehicles
and advance clean air efforts by adopting these standards and clean up fuel at the refinery.
Commenter: State of Utah
The proposed Tier 3 Program addresses the vehicle and its fuel as an integrated system. The
proposed Tier 3 gasoline—with a reduced average sulfur content of 10 ppm—would enable the
advanced catalytic converters to function more effectively. Therefore, Tier 3 gasoline should be
made available at the pump as soon as possible.
5-21
-------
Chapter 5: Tier 3 Fuel Sulfur Program
Commenter: United Automobile Workers
A national standard for ultra-low sulfur gasoline is an especially cost-effective way to achieve
substantial emission-reductions from vehicles. Low-sulfur gasoline allows a vehicle's catalytic
converter to operate at higher efficiency, in turn leading to lower tailpipe emissions and
decreasing the formation of smog-producing pollutants from cars and trucks. Looking forward, a
reduction in sulfur will also enable the introduction of new fuel-saving technologies in the next
generation of vehicles. Since manufacturers know that all vehicles will be using the same
standardized low-sulfur gasoline across the country, they can include technologies in new
vehicles that will capitalize on this fuel development.
Our Response:
As explained in Sections II and V.M of the preamble to the final Tier 3 rule, we are
lowering the existing gasoline sulfur standards under section 21 l(c)(l), because emission products of
gasoline with current levels of sulfur cause or contribute to air pollution which may reasonably be
anticipated to endanger public health or welfare, and because emission products of gasoline with
current levels of sulfur will impair to a significant degree the emissions control device or systems on
the vehicles subject to the final Tier 3 standards. As discussed in Section II.B of the preamble to the
final rule, control of gasoline sulfur to 10 ppm will lead to significant reductions in emissions of
these pollutants, with the benefits to public health and welfare significantly outweighing the
costs. As also discussed in Chapter 4 of this Summary and Analysis of Comments document
(and further in Section IV.A.6 of the preamble to the final rule and Chapters 4 and 5 of the RIA),
a reduction in gasoline sulfur from the Tier 2 30 ppm average to the Tier 3 level of a 10 ppm
average is necessary to enable the new vehicle fleet to meet the Tier 3 standards, as well as
achieve immediate emission reductions from the existing fleet. Lower sulfur gasoline will
enable the needed reductions in vehicle emissions and the 10 ppm average sulfur standard is a
feasible standard for refiners to meet taking into consideration costs and other factors. Further,
the 10 ppm average sulfur standard will also help prevent the significant impairment of the
emission control systems expected to be used in Tier 3 technology, significantly improve the
efficiency of emissions control systems currently in use, and continue prevention of the
substantial adverse effects of sulfur levels on the performance of vehicle emissions control
systems.
As discussed in Chapter 5.1.1.2 of this Summary and Analysis of Comments document
(and in Sections III. A and III.B of the preamble to the final rule), a 10 ppm annual average
standard will help reduce current levels of sulfur that contribute to ambient levels of air pollution
that endanger public health and welfare. Lowering gasoline sulfur will not only provide
emissions benefits when combined with Tier 3 vehicle technologies, but it will also help to
reduce emissions of pollutants from vehicles currently on the road today. As also discussed in
Section IV. A of the preamble to the final rule, we have tested a wide range of vehicles to better
understand the impact that even lower gasoline sulfur could have on emissions. Our test data
showed significant NOx and VOC reductions when vehicles were tested on low sulfur gasoline.
Further, Section III.B of the preamble, shows that lowering average gasoline sulfur from 30 to 10
ppm will result in approximately 260,000 less tons of NOx and 50,000 less tons of VOC almost
immediately when the Tier 3 gasoline sulfur standards take effect.
5-22
-------
Tier 3 Summary and Analysis of Comments
Comments on GHG, CAFE, and CC>2 emissions are beyond of the scope of the Tier 3
rulemaking.
5.1.1.4. Impacts on Cost, Supply, and RFS Program
What Commenters Said:
Commenter: American Fuel & Petrochemical Manufacturers (AFPM)
EPA proposes that refiners further lower the amount of sulfur from gasoline despite the fact that
our industry already removed 90 percent since 2004. The remaining incremental reduction of
trace amounts of sulfur will cost refiners almost as much as the Tier 2 reduction 10 years ago,
which removed 15 times more sulfur than the proposed Tier 3 regulation would require.
Commenter: Medical Advocates for Healthy Air
Other arguments mention the need for heavy investment in the newer technologies to achieve the
standards of Tier 3. Yet oil refineries supplying gasoline to California are already reducing the
sulfur content to 10 ppm. The technology already exists, is working, and can be installed at the
national level with minimal increase—if any—to the cost of gasoline. As mentioned above, the
new standards will increase employment and contribute positively to the nation's GDP.
Commenter: Small Business Refiners (SBR)
EPA Enforcement continues to pursue its National Petroleum Refinery Initiative that focuses on
NSR/PSD, NSPS, LDAR, and benzene. The refining industry has spent over $6 billion to install
additional control technologies, $80 million in civil penalties, and $75 million in supplemental
environmental projects. Through investments, the refining industry has reduced sulfur emissions
in gasoline from 300 ppm to less than 30 ppm and sulfur emissions in diesel from 500 ppm to
less than 15 ppm. Tier 3 Fuels reduction in gasoline sulfur will be significantly more expensive
per pound of sulfur removal, than previous regulations. Tier 2 sulfur reduction cost[s]
approximately $100,000 per ppm of sulfur, where Tier 3 fuels is expected to cost $900,000 to
$1,000,000 per ppm of sulfur reduction.
Commenter: Sutherland Asbill & Brennan LLP (Sutherland)
Instead, the Proposed Rule may hasten the impending blend wall under the Renewable Fuel
Standard ('RFS') by exacerbating the increase in exports to foreign markets and decrease in
supply to the U.S. markets brought on by the RFS and lack of consumer demand.
Our Response:
5-23
-------
Chapter 5: Tier 3 Fuel Sulfur Program
As discussed in Chapter 7.2 of this Summary and Analysis of Comments document (and
in Section VII.B of the preamble to the final rule), our cost analysis shows that refiners will
make about $2.025 billion in capital investments to achieve sulfur reductions to meet the Tier 3
gasoline standards. These capital investments are expected to be made over the 6 years that the
Tier 3 program is expected to be phased in, which would spread out the capital costs to average
about $330 million per year. When fully phased-in, we estimate that the national average
refinery cost of compliance with the Tier 3 sulfur control program, with a nationwide credit
trading program, will be 0.65 cents per gallon averaged over all gasoline.
We disagree with the comment that the costs associated with complying with Tier 3 may
indirectly "hasten the "impending blend wall" under the RFS program. As highlighted in the
2014 RFS Annual Rule proposal ("2014 Standards for the Renewable Fuel Standard Program;
Proposed Rule", 78 FR 71732, November 29, 2013), the Agency believes that the E10 blendwall
was reached in 2013. Therefore, the Tier 3 program can do nothing to hasten its arrival. Even if
this were not the case, there would be virtually no impact. The commenter highlights the
potential for the costs of complying with Tier 3 to increase exports as a reason why Tier 3 may
hasten the arrival of the blend wall. However, even if Tier 3 were to encourage some US refiners
to export more of their gasoline, the supply would then be made up with reductions in exports by
other refiners or increases in imports from abroad. This volume would also be subject to the
same RFS standards, resulting in no impact on the RFS program as a whole. To the extent that
Tier 3 does increase overall compliance costs for gasoline, and prices to consumers rise along
with it, this could have an impact on consumer demand for gasoline. Given the very small
projected increase in cost for Tier 3 (0.65 cents/gallon on average) the impact on consumer
demand would be very small. However, even then it would have no meaningful impact on the
RFS program as a whole, since the RFS standards are percentage standards that are adjusted on
an annual basis to reflect whatever happens to gasoline and diesel fuel demand for whatever
reason. The impact of Tier 3 would be imperceptible in comparison to other changes in the
marketplace (e.g., changes in crude oil prices).
While we recognize that there are other regulations that are imposing compliance costs
on the refining industry, including other EPA regulations, these costs do not preclude
implementation of the Tier 3 standards, and such comments are otherwise beyond the scope of
this rulemaking.
5.1.1.5. Harmonization with California & Other Countries
What Commenters Said:
Commenter: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
Ultra-low sulfur gasoline has been available in California for years, and the ability to comply
with emissions standards in California is premised on the use of such fuel, recognizing the need
to treat the vehicle and the fuel as a system.
5-24
-------
Tier 3 Summary and Analysis of Comments
One final point on sulfur. You've already heard it this morning, and that is that we have fallen
behind other countries. The proposed rule would help Americans to get gasoline with sulfur
limits approaching 10 parts per million cap, which is like the retail sulfur - the retail levels of
sulfur already available in Europe and Asia.
Commenter: American Honda Motor Co., Inc.
California has recognized this in its LEV III standards, requiring an average of 15 parts per
million, and a cap of 20 parts per million. It is critical that EPA set a national gasoline sulfur
average and cap levels equivalent to those specified in California regulations.
Commenter: Boulder County Board of Commissioners and Boulder County Board of Health
These are cost-effective, proven technologies that are almost entirely the same as those already
in place on California vehicles. Further, California's gasoline already achieves 10-ppm sulfur on
average, as do other nations, including the European Union, Japan, and - by 2017 - China.
Commenter: California Air Resources Board (CARB)
CARB staff strongly supports the proposed Tier 3 gasoline sulfur standards and applauds the
efforts of U.S. EPA staff for bringing our fuel programs in alignment.
U.S. EPA is proposing to lower the sulfur standard in gasoline and ethanol-gasoline blends to a
10 parts per million (ppm) annual average. California's gasoline already meets the proposed 10
ppm sulfur standard. For many years California has led the nation in gasoline quality, including a
low-sulfur standard. On December 31, 2003, California lowered the sulfur cap on gasoline from
80 ppm to 60 ppm. Then on December 31, 2005, the sulfur cap was lowered from 60 ppm to 30
ppm. And finally, on December 31, 2011, the sulfur cap was lowered to 20 ppm. The 20 ppm
sulfur cap brought the in-use fuel's sulfur average in California to about 9 ppm (2).
(2) http://www.arb.ca.qov/reqact/2012/leviiiqhq2012/levisorpdf
Commenter: Chrysler Group LLC
California recognizes the environmental benefits of low sulfur gasoline and requires 10 ppm
sulfur in its LEV III regulation as an enabler to meet the Air Resources Board's super ultra-low
emission vehicle (SULEV) fleet average requirements. Requiring low sulfur gasoline in these
Tier 3 rules will further harmonize Tier 3 and California LEV III to realize the emission benefits
and to maximize new vehicle development and manufacturing efficiency.
We will continue to work with EPA to bring California LEV III and Tier 3 as close as possible,
with the goal of a completely harmonized California and Federal, one national criteria pollutant
program. Low sulfur gasoline is a vitally important and critical element in achieving this goal.
Commenter: City of Philadelphia Department of Public Health Management Services (AMS)
5-25
-------
Chapter 5: Tier 3 Fuel Sulfur Program
So in summary, I would like to have the rule be adopted as quickly as possible because the
standards are consistent with California and other European countries already have.
Commenter: Delaware Department of Natural Resources (DNRC)
The Tier 3 standards support our efforts by reducing the average gasoline sulfur concentration
from 30 ppm to 10 ppm, which is consistent with the international trend to reduce sulfur in
gasoline.
Commenter: Emissions Control Technology Association (ECTA)
From a general global perspective, Japan, Europe, and even China are moving to 10 ppm sulfur
gasoline. By not doing this, the United States will be a significant major outlier and cause vehicle
options and costs to be higher than in other major markets.
Commenter: Environmental Defense Fund (EDF)
Reducing the sulfur content in gasoline from 30 parts per million (ppm) to 10 ppm will align the
federal standard with those of California, the European Union, and Japan. Timely
implementation of a national, unified sulfur standard will drive investment and development of
emissions control technology, while allowing manufacturers to efficiently align technology
upgrades with fuel efficiency and GHG emissions standards.
Finalizing the tighter fuel standard would bring the U.S. in line with the cleaner fuel already
required in California, Europe, and Japan. Canada is also proposing 10 ppm sulfur fuel.
Commenter: Environmental Law and Policy Center (ELPC)
In reducing the sulfur content from 30 parts per million to 10 parts per million, that is a critical
step to improving air quality. And we know that that standard is achievable because it is being
met in California, Europe, and Japan.
Commenter: Ford Motor Company (Ford)
As mentioned earlier today, other markets with advanced emissions requirements, such as those
found in Asia and Europe, are particularly concerned with high quality market fuels containing
ultra-low sulfur fuels of 10 ppm to ensure catalysts maintain their high efficiencies throughout
the expected useful life of the vehicle. Improved fuel economy and reduced CO2 emissions are
made possible in vehicles through engine operations at lean air fuel rations, as Frank just
mentioned.
Additionally, the proposed rule enables Americans to enjoy immediate air quality benefits
through operating on gasoline with sulfur limits approaching the 10 ppm cap similar to those
already in place around the world.
5-26
-------
Tier 3 Summary and Analysis of Comments
Commenter: Hyundai Motor Group
While we understand some refineries may face challenges in meeting the level proposed in this
rule, we want to point out that by setting a 10 ppm average annual standard, EPA will be
bringing the U.S. gasoline specifications in line with California, Japan, Korea, and the European
Union.
In particular, South Korea has an average 10 ppm standard in place - has had an average 10 ppm
standard in place since 2009. Also California has a cap of 20 ppm, and the average sulfur level
in the fuel in the State is 10 ppm. Lastly, Japan and the EU have set a cap of 10 ppm, which is
much more stringent than an average annual level proposed by EPA. All of these jurisdictions
have recognized the importance of ensuring low sulfur levels in fuel. It is a reasonable and
necessary standard for the U.S. as well.
Commenter: International Council on Clean Transportation (ICCT)
The ICCT also commends the EPA for proposing to reduce gasoline sulfur to 10 ppm. Japan and
South Korea have required 10 ppm sulfur in gasoline since 2007 and Europe since 2009. Even
Chile has required 15 ppm sulfur in gasoline since 2010. Given the leadership shown by the US
in most environmental areas, it is important for the US to catch up on gasoline sulfur.
Commenter: ECTA
It's essential the U.S. refinery industry enter the 21st century by providing an average of 10 ppm
fuel. Europe and Japan have had it for years. China is doing so now, and will have it throughout
the country in 2017. And India and Brazil are considering moving in this direction.
It would be shameful if the U.S. still had 30 ppm sulfur and these countries have clean fuel.
They all realize less than 10 ppm sulfur fuel is critical to clean vehicles. It all ties together.
Commenter: Manufacturers of Emission Controls Association (MECA)
Adoption of a 10 ppm average gasoline sulfur limit by EPA will bring the U.S. in line with other
major vehicle markets in Europe, Japan, South Korea, and China where 10 ppm gasoline and
diesel fuel sulfur caps are already in place or will be in place by 2018.
Commenter: Union of Concerned Scientists (comment campaign)
I support Tier 3 standards that will keep the United States on pace with global trends to reduce
the sulfur concentration of gasoline from 30 parts per million (ppm) to 10 ppm.
Commenter: Michigan Department of Environmental Quality (MDEQ)
Once again, given California's regulations, we know this concentration is achievable.
5-27
-------
Chapter 5: Tier 3 Fuel Sulfur Program
Commenter: National Association of Clean Air Agencies (NACAA)
Further, California's gasoline already achieves 10-ppm sulfur on average. Finally, gasoline in
other nations, including those in the European Union and Japan, is subject to a 10-ppm cap.
China has also adopted requirements for 10-ppm sulfur, to take effect in 2017.
Commenter: New York State Department of Environmental Conservation
The gasoline sulfur standards are clearly feasible.
Many jurisdictions (European Union, Japan, South Korea, Beijing, China and California) are
meeting comparable sulfur standards now. The processing technology necessary to produce
gasoline meeting a 10 ppm sulfur standard is well established and commercially proven.
Hydrodesulfurization has been used in oil refining since at least the early 1950s. Processes
tailored to a wide variety of refinery streams, including those that contribute significant amounts
of sulfur to the gasoline pool, have been developed and are in commercial use.
We also fully support adoption of the proposed standards regulating the sulfur content of
gasoline. When we adopted the California motor vehicle emissions standards in 1990, the Clean
Air Act pretty much hinders New York and other States from adopting the California fuel
standards. I won't say it's impossible, but it is a high hurdle.
The proposed gasoline sulfur limited of 10 parts per million is widely accepted, developed
around the developed world. It's currently the standard in the EU, Japan, South Korea, and in
California, and China is even moving forward with the 10-part per million sulfur standard.
Commenter: Philadelphia Physicians for Social Responsibility
Now 13 years after the Tier 2 final rules on sulfur were published, it's time to update these
regulations. We know that we have the technical ability to do this because we already require
these standards in California, Japan, South Korea, and most of Western Europe. Because you
will hear from many others on the Tier 3 rules, I will limit my few minutes to the harmful effects
of sulfur.
Commenter: Sierra Club
Refiners in other countries, such as Japan and 30 European nations, already are meeting a 10
ppm gasoline sulfur cap standard. In addition, California refiners are already producing gasoline
that averages 10 ppm sulfur content. Indeed, even without the proposed averaging, banking and
trading (ABT) improvements, EPA demonstrates that a 10 ppm gasoline sulfur standard is
feasible.
Commenter: Southern California Association of Governments (SCAG)
5-28
-------
Tier 3 Summary and Analysis of Comments
There appears to be some differences between the proposed Tier 3 standards and California's
LEV III Regulations and California gasoline sulfur content standards. While acknowledging the
complexities and nuances of these programs, harmonization of the proposed Tier 3 with the
California program is important from a business and manufacturer continuity perspective. We
believe that such harmonization may yield savings to California residents through the
efficiencies of scale in the production of vehicles that meet one national standard.
Commenter: Sutherland Asbill & Brennan LLP (Sutherland)
Annual Average Sulfur Standard. The Proposed Rule would lower the annual average sulfur
standard for gasoline refined, blended or imported from 30 ppm sulfur to 10 ppm sulfur on
January 1, 2017. EPA states that Europe, Japan and other countries have moved to 10 ppm
sulfur, so it is clear that U.S. refineries can also adapt. The make-up of refineries in the United
States and Europe differs dramatically, with certain European markets having very little or no
refining capacity and are thereby dependent on other markets for their refined products. These
markets may have been able to purchase certain grades from refineries in other countries without
the refineries having to undergo any dramatic changes to meet these markets' needs.
Furthermore, European markets use diesel fuel rather than gasoline for their transportation needs
at a much higher rate than the United States. Simply because other countries have lowered their
sulfur standard in gasoline does not necessarily mean that the U.S. refinery complex will be able
to supply lower sulfur gasoline in the quantities required by the U.S. markets without
considerable expense and cost to U.S. gasoline prices and the U.S. economy.
Commenter: Union of Concerned Scientists (UCS)
Tier 3 is also consistent with the global trend to reduce sulfur in gasoline. Refiners are already
producing ultra-low sulfur fuel in California and much of the developed world - demonstrating
that the technology to meet these standards has already been deployed. In 2012, the Hart Energy
International Fuel Quality Center ranked the top 100 countries based on sulfur limits in gasoline
(12). The United States, at the current level of 30 ppm, is ranked 47th, with higher sulfur than
Japan, South Korea, Turkey, Taiwan, Thailand, Chile and all of Europe.
12 - International Fuel Quality Center. 2013. International Gasoline Rankings - Top 100 Sulfur. Online at
http://www.ifqc.org/NM_Top5.aspx accessed on July 1, 2013.
Commenter: United Automobile Workers
We applaud the EPA for taking the right approach, one that recognizes the connection between a
vehicle's emissions and its fuel, and treats them as a connected system. This approach will also
create a nationwide harmonization of our fuel standards and will bring us in line with low-sulfur
fuel requirements already available in California, Europe, and Japan. Our nation's automakers
will then be able to use a broader range of technologies to meet the significant environmental
challenges facing our industry.
Commenter: United Steelworkers Union (USW)
5-29
-------
Chapter 5: Tier 3 Fuel Sulfur Program
Once the proposed regulations to lower overall sulfur emissions in gasoline to lOppm are
finalized, the European Union and China will either implement in tandem or have in place
standards of ultra-low sulfur gasoline equivalent to the EPA Tier 3 Motor Vehicle Emission and
Fuel Standards. By aligning emission standards with such large economic regions, U.S. refiners
will be able to compete both domestically and globally in providing superior refined products to
consumers.
Commenter: State of Utah
Experience in other countries and in California demonstrates that the lower-sulfur Tier 3 gasoline
will yield immediate and significant emissions reductions in our existing vehicle fleet. Therefore,
Tier 3 gasoline should be made available at the pump as soon as possible.
Our Response:
The commenters are correct that gasoline desulfurization technologies are well known,
readily available, and have been further demonstrated by current fuel programs in California,
Asia, and Europe. Under California's Phase 3 Reformulated Gasoline program (CaRFG3) and
the Predictive Model that California refiners use to demonstrate compliance, California gasoline
currently contains approximately 10 ppm sulfur on average. Europe has a 10 ppm sulfur cap that
has been adopted by the 30 Member States that comprise the European Union (EU) and the
European Free Trade Association (EFTA) as well as Albania and Bosnia-Herzegovina. Japan
also has a 10 ppm gasoline sulfur cap, and a 10 ppm sulfur limit for gasoline was recently
introduced in Beijing, China. As these are all 10 ppm sulfur cap standards, they are considerably
more stringent than the 10 ppm annual average standard being finalized today because each
batch of gasoline produced at every refinery must meet the 10 ppm cap. Therefore, every
refinery must be designed to meet this cap regardless of changes in crude oil supply, operation
conditions, or product mix.
While we acknowledge that the fact that other countries have a similar sulfur standard is
not a reason for the U.S. to adopt it, this is not the reason we are lowering the sulfur content. As
discussed in Chapters 2 and 3 of the Summary and Analysis of Comments document, we are
controlling sulfur because emission products of gasoline with current levels of sulfur cause or
contribute to air pollution which may reasonably be anticipated to endanger public health or welfare,
and because emission products of gasoline with current levels of sulfur will impair to a significant
degree the emissions control device or systems on the vehicles subject to today's final Tier 3
standards.
In response to the comment that the gasoline desulfurization technologies are not
appropriate for the U.S. gasoline market, we note that many of these technologies were
demonstrated under the current Tier 2 program. However, we agree that many oil refineries
outside of the United States operate differently from their U.S. counterparts. U.S. refiners have
invested more heavily in fluidized catalytic cracker (FCC) units than the rest of the world to
maximize gasoline production. Because the FCC unit is responsible for nearly all the sulfur that
ends up in gasoline, U.S. refineries can face a bigger challenge in achieving 10 ppm gasoline
sulfur levels, however that is changing. The 2013 Annual Energy Outlook produced by the
5-30
-------
Tier 3 Summary and Analysis of Comments
Energy Information Administration supports the view that the U.S. demand for diesel fuel is
increasing while the demand for gasoline is decreasing, starting the process that will make the
U.S. more like Europe. Thus, U.S. refiners seem to be beginning to move away from relying on
the FCC unit as the most important refinery unit. Further, as discussed in more detail in Section
V.L of the preamble to the final rule, the challenge of complying with more stringent gasoline
sulfur control will decrease over time. The review of gasoline sulfur control in California and
elsewhere, and the future trend for gasoline demand, show that achieving 10 ppm is feasible.
The Tier 3 requirements are less demanding than those of Europe or Japan, for a number of
reasons. First, the Tier 310 ppm standard is a refiner annual average standard, rather than sulfur
cap. Second, the Tier 3 program has an 80 ppm refinery gate cap, which allows individual
gasoline batches to vary in sulfur level (as long as they meet the 10 ppm annual average). Lastly,
the Tier 3 ABT program allows refiners to generate credits for overcompliance with the standard
and to bank or sell those credits to other refiners with more challenged refineries that need to
purchase credits to assist with compliance.
5.1.1.6. In-Use Gasoline
What Commenters Said:
Commenter: National Automobile Dealers Association (NADA)
Importantly, dealerships and their customers deserve to know that state-of-the-art emissions-
related components and systems will not be undermined by substandard or 'bad' fuels. Thus, it is
incumbent upon EPA to set appropriate limits on motor fuel characteristics in order to enable
effective vehicle emissions control and to protect in-use vehicle performance against substandard
or inappropriate real-world fuels.
Our Response:
With regard to concerns about in-use sulfur levels, we believe that vehicles will see sulfur
levels closer to the 10 ppm average rather than the 80 ppm cap due to the fact that the 10 ppm
average will drive reductions in gasoline sulfur levels. As discussed further in Chapter 5.1.2
below, we believe most refineries will still average less than 20 or 30 ppm in their physical
gasoline production. Further, our cost analysis (in Section VII.B of the preamble to the final
rule) projects that nearly 40% of the gasoline pool will be at 5 ppm, about 45% at 10 ppm, and
the remaining 15% at levels higher than 10 ppm. Thus, the majority of in-use sulfur levels are
expected to be at or below 10 ppm sulfur. And as discussed in Section V.C.I.b of the preamble,
we are committing to monitor and further evaluate in-use sulfur levels and their impact on
vehicle emissions. If it is warranted, we will reassess the sulfur cap level and the need for
potential future regulatory action.
Further, while we are taking action to require lower sulfur content in in-use gasoline
(which emissions test fuel should reflect) under authority given in Clean Air Act sections
21 l(c)(l) and (2), any action related to other in-use fuel standards is outside the scope of this
rulemaking."
5-31
-------
Chapter 5: Tier 3 Fuel Sulfur Program
5.1.2. Per-Gallon and Downstream Sulfur Caps
We received comments on both of the proposed per-gallon cap options of 80/95 ppm and 50/65 ppm,
as well as comments on finalizing lower caps of 20/25 ppm and a 20 ppm overall cap. Comments
supporting lower caps noted potential environmental benefits. Comments in support of maintaining
the current Tier 2 caps cited concerns on cost, flexibility for turnarounds/unplanned shut downs (due
to refinery fire, natural disaster, etc.), and potential impacts on gasoline supply and pricing.
What Commenters Said:
Commenter: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
We also support reducing the refinery gate and retail pump per gallon caps to 20 and 25 ppm
respectively for an interim time period, rather than leaving them at Tier 2 levels (80 and 95 ppm
respectively), or partially reducing them to 50 and 65 ppm, respectively.
In the final rule, EPA should also reduce the per-gallon refinery gate and retail pump cap to 20
and 25 ppm, respectively.
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM)
API and AFPM support EPA's proposed option for maintaining the current per gallon sulfur
caps: 80-ppm at the refinery gate and 95-ppm downstream. As EPA notes, the annual average
sulfur standard is a factor that limits the amount of sulfur in gasoline; per gallon caps are
important to manage planned and unplanned refinery unit downtime.
Commenter: ExxonMobil
If the Agency decides to proceed with this rulemaking, we recommend that EPA retains the
sulfur caps at the current levels of 80 ppm at the refinery gate and 95 ppm downstream.
Commenter: Kinder Morgan Transmix Company, LLC
Yes, we support EPA's proposed downstream cap implementation date of 2020. We concur that
based upon a 2017 refinery gate cap of 50-ppm, transmix processors would have sufficient time
to evaluate the impact of the regulation on our operations prior to 2020.
Commenter: General Motors LLC (GM)
In the final rule, EPA should also reduce the per-gallon refinery gate and retail pump cap to 20
and 25 ppm, respectively.
5-32
-------
Tier 3 Summary and Analysis of Comments
Commenter: Manufacturers of Emission Controls Association (MECA)
MECA is supportive of a 20 ppm gasoline sulfur cap for Tier 3 that would be aligned with
California's current gasoline sulfur cap.
Commenter: Mercedes-Benz USA, LLC on behalf of Daimler AG
Mercedes-Benz further requests that a 20 ppm refinery gate and 25 ppm retail cap be enacted to
ensure that a dependable, uniform fuel quality is available nationwide.
Commenter: Natural Resources Defense Council (NRDC)
EPA should evaluate even lower the caps to add more certainty to pollution reductions from low
sulfur gasoline. EPA requests comment on caps of 20 ppm at the refinery gate and 25 ppm
downstream. EPA's analysis, as discussed in the proposed rule, indicates that the 20 ppm and 25
ppm levels will render the flexibilities of the ABT program unusable because all refineries will
have to produce gasoline at 10 ppm, thus increasing the compliance cost of the Tier 3 program.
There may be, however, cap levels below 50 ppm at the gate and 65 ppm downstream that would
maintain sufficient compliance flexibility through the ABT program but also enhance
downstream fuel quality certainty. EPA should provide an evaluation of additional cap levels and
finalize caps tighter than 50 ppm and 65 ppm if feasible and cost-effective.
Commenter: New York State Department of Environmental Conservation
We also believe that the per-gallon sulfur caps for gasoline can, and should, be reduced from the
current levels of 80 ppm sulfur at the refinery gate, and 95 ppm sulfur at downstream locations.
The per-gallon refinery gate and downstream sulfur standards should be reduced from 80 and 95
ppm respectively to at most 50 and 65 ppm.
In addition to an annual average of 30 ppm sulfur, the current Tier 2 standards set a per-batch
cap of 80 ppm, and a downstream cap of 95 ppm. While proposing to reduce the annual average
standard to 10 ppm sulfur, EPA proposes to leave the 'per-gallon' sulfur limits unchanged, while
requesting comment on more stringent per-gallon limits.
The EPA should allow the current 80 ppm fuel sulfur level refinery gate cap and the 95 ppm
downstream per gallon cap.
It is evident that many refiners need the 80 ppm fuel sulfur cap to be able to purchase credits
while they transition to the 10 ppm average. It is also evident that the full emission reduction
potential for Tier 3 vehicles claimed in the proposed rule cannot be realized if the caps are kept
indefinitely at the 80 ppm level. EPA should consult with the refining industry to develop a
phase-in schedule to lower the caps to the 20 ppm to 30 ppm level by the year 2025, when Tier 3
vehicles will comprise most of the in-use fleet.
5-33
-------
Chapter 5: Tier 3 Fuel Sulfur Program
Our Response:
The final Tier 3 program retains the 80 ppm per-gallon refinery gate and 95 ppm per-
gallon downstream caps that are currently in place under the Tier 2 program. As discussed in
more detail in Chapters 5.1.2.1 through 5.1.2.3 below, and in Section V.C of the preamble to the
final rule, we believe this is the most prudent approach for lowering in-use sulfur while
maintaining refinery flexibility, considering cost and other factors (such as refinery turnarounds,
unplanned upsets, etc.). The per-gallon caps provide an upper limit on the sulfur concentration
that vehicles must be designed to tolerate, and they also limit downstream sulfur contamination
and enable the enforcement of the gasoline sulfur standard in-use.
With regard to concerns about in-use sulfur levels, we believe that vehicles will see sulfur
levels closer to the 10 ppm average rather than the 80 ppm cap due to the fact that the 10 ppm
average will drive reductions in gasoline sulfur levels. As discussed further in Chapter 5.1.2
below, we believe most refineries will still average less than 20 or 30 ppm in their physical
gasoline production. Further, our cost analysis (in Section VII.B of the preamble to the final
rule) projects that nearly 40% of the gasoline pool will be at 5 ppm, about 45% at 10 ppm, and
the remaining 15% at levels higher than 10 ppm. Thus, the majority of in-use sulfur levels are
expected to be at or below 10 ppm sulfur. And as discussed in Section V.C.l.b of the preamble,
we are committing to monitor and further evaluate in-use sulfur levels and their impact on
vehicle emissions. If it is warranted, we will reassess the sulfur cap level and the need for
potential future regulatory action.
5.1.2.1. Sulfur Cap Impacts on Vehicle Emissions and Enabling New Vehicle
Technology (GHG, GDI, etc.)
We received comments on both of the proposed per-gallon cap options of 80/95 ppm and
50/65 ppm, as well as comments on finalizing lower caps of 20/25 ppm and a 20 ppm overall cap.
Comments supporting lower caps noted greater certainty that vehicles would see lower and more
uniform gasoline sulfur levels, and enabling new vehicle technologies that require very low sulfur
levels. Comments against lower caps disagreed with the effect of sulfur levels on vehicle
technologies and asserted that sulfur effects on such technologies were largely reversible.
What Commenters Said:
Commenter: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
Furthermore, automakers support EPA reducing the retail per gallon cap to at least 25 ppm,
which is close to the 20 ppm cap in California and should be feasible to achieve nationwide in
the foreseeable future.
The auto industry supports EPA's proposed reduction of the annual refinery average sulfur limit
for gasoline in the proposed Tier 3 rule, but we also support further reduction in the other two
5-34
-------
Tier 3 Summary and Analysis of Comments
prongs of the federal market gasoline sulfur standard, the per gallon cap at the refinery gate, and
the per gallon cap downstream at the retail pump.
However, while we commend EPA for proposing reduction of annual average sulfur content in
gasoline, the proposed rule does not go far enough in reducing the absolute cap on downstream
levels of this harmful component. As proposed, the rule does little to assure that there is uniform
gasoline sulfur content at the gas pump. Automakers need a clear, harmonized standard, with
consistently low sulfur levels at the pump to optimize vehicle technology to meet the emissions
reductions and GHG/fuel economy goals set by the Agency, and thus request that EPA take
further action on sulfur to facilitate compliance with such requirements.
In the proposed rule, EPA has focused on reducing the annual refinery average for sulfur from 30
ppm to 10 ppm (38). While reducing this annual average is important, without a concurrent
reduction in the other two prongs of the standard the rule does not achieve a predictable capped
amount of sulfur in each gallon of gasoline.
EPA is considering leaving in place the Tier 2 refinery gate per-gallon cap at 80 ppm and the
per-gallon downstream cap at 95 ppm, and indeed has written the proposed C.F.R. regulatory
language based on this approach (39). However, the Agency is also taking comment on reducing
the refinery gate per-gallon cap to 50 ppm [or 20 ppm], and the downstream per-gallon cap to 65
ppm [or 25 ppm] (40). As EPA's own data show, the lower the sulfur the better, and we support
the lowest possible levels for these latter two standards.
Predictable, low, downstream levels of sulfur are necessary for vehicles to run on the most
advanced engine/emissions control systems, and OEMs need assurance of such levels to
facilitate long-term product planning.
The highest sulfur level recorded among gasoline in the contiguous U.S. in the July 2012
Alliance fuels survey was 43 ppm, (41) and in the January 2013 survey was 72 ppm. Only 1 of
the 394 retail samples in January 2013 exceeded 65 ppm, suggesting that EPA would be able to
transition the refinery gate and retail pump cap limits without imposing significant costs. A retail
sulfur cap of 25 ppm should be workable within the time period covered in the rule. As discussed
below, this change would also come closer to harmonization with California's 20 ppm sulfur
retail cap (42).
It will not be possible for emissions control equipment to work at the required levels found
within Tier 3 while operating with sulfur as high as 95 ppm. Therefore a predictable national
sulfur standard at the point of sale to the consumer, the retail pump, is crucial for ensuring that
vehicles will have the fuel they need to meet the required fuel economy increases and emissions
reductions.
EPA's own new study data show clear benefits from reducing gasoline sulfur from 28 ppm to 5
ppm (43). The remaining issues are technical and economic feasibility for establishing a more
stringent sulfur cap at retail. EPA's own data shows catalyst sulfur poisoning reversibility is
limited, and reduced sulfur in the fuel provides better emissions reductions (44). EPA must take
action to reduce the per-gallon retail cap of sulfur to 25 ppm to support the OEMs' emissions
5-35
-------
Chapter 5: Tier 3 Fuel Sulfur Program
reduction requirements, and also to plan for adjustments to achieve a 10 ppm retail sulfur cap as
soon as possible, as noted above.
38 - 78 Fed. Reg.29908 at 29921 (May 21, 2013).
39 - 78 Fed. Reg. 29908 at 30023-30024 (May 21, 2013).
40 - 78 Fed. Reg. 29908 at 29920 (May 21, 2013).
41 - Annual Alliance of Automobile Manufacturers North American Fuels Survey, July 2012.
42 - California Reformulated Gasoline Regulations, 13 C.C.R. §2262.
43 - See EPA study, "The Effects of Gasoline Sulfur Level on Emissions from Tier 2 Vehicles in the In-
Use Fleet," April 2013.
44 - Id.
Commenter: American Fuel & Petrochemical Manufacturers (AFPM)
The AFPM recommends that the current refinery gate cap for gasoline remain at 80 ppm sulfur.
This would not interfere with an engine technology. Even the Agency has proposed that 80 ppm
is prudent for 2017 to 2019.
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM)
API engaged Turner, Mason & Company (TM&C) to evaluate the economic, supply, and overall
gasoline pool quality implications for imposing more stringent per-gallon sulfur caps on U.S.
gasoline in addition to the assumed reduction in the annual average sulfur limit to 10 ppm. (33)
The complete TM&C report titled "Economic and Supply Impacts of a Reduced Cap on Gasoline
Sulfur" has been submitted as Attachment No. 12.
The automakers have expressed concern about potential impacts on emissions performance if
individual vehicles are exposed to gasoline above 10 ppm S. We believe that this concern is
unfounded. The per-gallon limits on the concentration of sulfur in gasoline should not be
changed from the current levels of 80 ppm at the refinery gate and 95 ppm downstream.
Research recently completed by SGS Environmental Testing Corp. indicates that the increase in
exhaust emissions from late model vehicles exposed to as much as 80 ppm S in gasoline is fully
reversible within a short period of time (i.e., -70 miles of driving) following a return to operation
on gasoline containing 10 ppm S (34). The study has been submitted as Attachment No. 11 for
EPA's review.
The API study focused on six passenger cars, of which five were certified to California SULEV
II/PZEV emissions standards, and one vehicle which complied with the federal Tier 2/Bin 5
exhaust emissions standard. The test vehicles represented a range of emission control and engine
technologies and were equipped with catalytic converters that had been aged to the equivalent of
120,000 to 150,000 miles of driving. The reversibility test sequence included: (a) four baseline
emissions tests run on 10 ppm S fuel, (b) three tests using 80 ppm S gasoline following 300
miles of operation on this high sulfur fuel, and (c) three tests after the vehicles were switched
back to 10 ppm S fuel. The base fuel was a California LEVIII certification gasoline containing
10% ethanol by volume. The 80 ppm S fuel was produced by doping the base fuel with a
representative mixture of sulfur compounds.
5-36
-------
Tier 3 Summary and Analysis of Comments
Figure 13 [of EPA-HQ-OAR-2011-0135-4276-A2] evaluates the reversibility of the sulfur
effects by comparing emissions before and after the exposure to 80 ppm fuel. The mean
emissions after the exposure to the high sulfur fuel are subtracted from the mean emissions for
the initial baseline on 10 ppm sulfur fuel, 95% confidence intervals are calculated for the
difference, and the results are plotted for each of the six individual test vehicles and for the fleet
as a whole. If the confidence interval for the emissions difference does not include zero, then
emissions on 10 ppm sulfur fuel after exposure to 80 ppm fuel are statistically different from the
initial 10 ppm baseline. If the entire confidence interval is less than zero then emissions are
statistically higher after exposure to 80 ppm sulfur, indicating that sulfur effects are not fully
reversible. If the confidence interval includes zero, then mean emissions before and after
exposure to the 80 ppm fuel are not statistically different and the hypothesis that the sulfur
effects on emissions were irreversible would be rejected.
Based on the statistical analysis of the results as summarized in Figure 13, the study concluded
that, for each vehicle tested and for the test fleet as a whole, the change in NMOG, NOx, CO,
Soot and PM emissions resulting from exposure to 80 ppm S fuel was quickly reversed upon
returning to operation on 10 ppm S gasoline. There was greater than 95% confidence that the
differences in the mean emissions values measured before and after the high sulfur fuel exposure
were not statistically different.
33 - Auers, J.R. et al.Turner, Mason & Company Consulting Engineers, Economic and Supply Impacts of
a Reduced Cap on Gasoline Sulfur Content, February 2013.
34 - SGS Environmental Testing Corp., Reversibility of Gasoline Sulfur Effects on Exhaust Emissions
From Late Model Vehicles, prepared for American Petroleum Institute, June 2013
Commenter: Appalachian Mountain Club (AMC)
Related to this provision, AMC supports EPA's proposed option to lower the per-gallon sulfur
caps to 50 ppm at the refinery gate and 65 ppm downstream in 2020. This provision allows
flexibility for some refiners but will also provide opportunities for advancements in the
automobile technologies that result in lower emissions.
Commenter: BMW of North America, LLC
We also recommend that EPA promulgate a final rule that reduces the refinery gate and
downstream sulfur caps to 20 and 25 ppm, respectively, to ensure low levels of sulfur at the retail
pump.
Commenter: Chrysler Group LLC
In addition to EPA's 10 ppm sulfur national average gasoline, Chrysler urges EPA to adopt 20
ppm refinery sulfur and 25 ppm retail sulfur caps. Having a predictable national standard at the
retail pump is crucial for vehicle design and development to achieve intended GHG and fuel
economy benefits from lean-burn GDI systems. Again, without low sulfur (both national refinery
average and at retail), lean-burn GDI systems will be subject to more frequent regeneration
events leading to significant fuel economy (and corresponding GHG) penalties.
5-37
-------
Chapter 5: Tier 3 Fuel Sulfur Program
Recommendation: Chrysler recommends that EPA adopt a 20 ppm refinery sulfur cap coupled
with a 25 ppm downstream (retail) sulfur cap.
Commenter: Emissions Control Technology Association (ECTA)
Given EPA's two options regarding per-gallon sulfur caps, we agree with EPA's
recommendation for lowering the caps from 80 ppm at the refinery gate (and 95 ppm
downstream) to 50 ppm at the refinery gate (and 65 ppm downstream) (3). EPA's other option of
maintaining the current caps is not acceptable for two reasons: First, it will have a negative
impact on the performance and durability of emissions and catalyst systems, as explained above.
Secondly, reducing the caps in EPA's proposal from 80 ppm at the refinery gate (and 95 ppm
downstream) to 50 ppm at the refinery gate (and 65 ppm downstream) would not materially
increase the cost of compliance to refineries. Under the EPA's cost model that assumes intra-
company trading only, just eight refineries consume credits rather than reduce their sulfur levels
down to 10 ppm. To the extent that the current level of sulfur for each of these refineries is less
than 50 ppm—a highly likely outcome given the nationwide average of 21 ppm according to
MOVES 2010 data—then requiring these refineries to reduce the sulfur content to 50 ppm is a
non-binding constraint that would impose no incremental costs.
Because EPA's cost model did not permit refineries to choose sulfur levels between their current
levels and 10 ppm, one can infer that for these eight refineries that choose to consume credits and
maintain their current sulfur levels, the associated compliance cost would be less than the cost of
achieving 10 ppm. It is possible, however, that for those refineries, the cost of reducing sulfur
slightly, to say 20 or 25 ppm, and consuming fewer credits would be an even better option. If so,
then compelling these refineries to achieve at most 50 ppm would not impose any additional
compliance costs.
Fortunately, it is possible to estimate an upper bound on the incremental compliance costs
associated with imposing a cap of 50 ppm. Assuming conservatively that all eight of these
refineries currently generate sulfur in excess of 50 ppm, one can estimate the upper bound by
comparing EPA's no-ABT scenario with its limited ABT scenario: $2.527 billion (or 0.97 cents
per gallon) versus $2.203 billion (0.89 cents per gallon), a difference of just $324 million. In
other words, forcing these eight refineries to achieve 10 ppm generates an added cost of $324
million relative to a system with limited credit trading. Because EPA's proposed caps would
require these eight refineries to achieve a more modest 50 ppm, the expected incremental
compliance cost would range from $0 (assuming either they are already below 50 ppm, or they
are above 50 ppm but prefer reducing sulfur slightly and consuming fewer credits) to $324
million (assuming they are above 50 ppm and prefer to stay above 50 ppm and consume more
credits).
Indeed, the same comparison provides the upper bound of incremental compliance costs for any
cap up to 10 ppm, including EPA's alternative 20 ppm cap. Accordingly, so long as the social
benefits of imposing a cap of either 20 ppm or 50 ppm exceed $324 million, then the EPA should
impose a more stringent cap than 80 ppm. EPA notes that a higher cap means that vehicles still
have to be designed to operate on high-sulfur fuel, which would undermine emission
performance and likely increase the cost of compliance for automobile manufacturers.4 Because
5-38
-------
Tier 3 Summary and Analysis of Comments
the benefits of greater emission compliance increase with the cap stringency—a more stringent
cap generates less variability in sulfur levels—it is conceivable that a 20 ppm cap (which would
allow little variability in sulfur levels) would generate greater net surplus than a 50 ppm cap
(which would allow substantial variability).
3 - Id. at 29927.
4 - Id. at 29927.
Commenter: Marathon Petroleum Company LP (MFC)
The automakers have expressed concern about potential impacts on emissions performance if
individual vehicles are exposed to gasoline above 10 ppm S. We believe that this concern is
unfounded. The per-gallon limits on the concentration of sulfur in gasoline should not be
changed from the current levels of 80 ppm at the refinery gate and 95 ppm downstream.
Research recently completed by API indicates that the increase in exhaust emissions from late
model vehicles exposed to as much as 80 ppm S in gasoline is fully reversible within a short
period of time (i.e., -70 miles of driving) following a return to operation on gasoline containing
10 ppm S. (21)
The API study focused on six passenger cars, of which five were certified to California SULEV
II/PZEV emissions standards, and one vehicle which complied with the federal Tier 2/Bin 5
exhaust emissions standard. The test vehicles represented a range of emission control and engine
technologies and were equipped with catalytic convenors that had been aged to the equivalent of
120,000 to 150,000 miles of driving. The reversibility test sequence included: (a) four baseline
emissions tests run on 10 ppm S fuel, (b) three tests using 80 ppm S gasoline following 300
miles of operation on this high sulfur fuel, and (c) three tests after the vehicles were switched
back to 10 ppm S fuel. The base fuel was a California LEVIII certification gasoline containing
10% ethanol by volume. The 80 ppm S fuel was produced by doping the base fuel with a
representative mixture of sulfur compounds.
Figure 11 (below) evaluates the reversibility of the sulfur effects by comparing emissions before
and after the exposure to 80 ppm fuel. The mean emissions after the exposure to the high sulfur
fuel are subtracted from the mean emissions for the initial baseline on 10 ppm sulfur fuel, 95%
confidence intervals are calculated for the difference, and the results are plotted for each of the
six individual test vehicles and for the fleet as a whole. If the confidence interval for the
emissions difference does not include zero, then emissions on 10 ppm sulfur fuel after exposure
to 80 ppm fuel are statistically different from the initial 10 ppm baseline. If the entire confidence
interval is less than zero then emissions are statistically higher after exposure to 80 ppm sulfur,
indicating that sulfur effects are not fully reversible. If the confidence interval includes zero, then
mean emissions before and after exposure to the 80 ppm fuel are not statistically different and
the hypothesis that the sulfur effects on emissions were irreversible would be rejected.
Based on the statistical analysis of the results as summarized in Figure 13, the study concluded
that, for each vehicle tested and for the test fleet as a whole, the change in NMOG, NOx, CO,
Soot and PN emissions resulting from exposure to 80 ppm S fuel was quickly reversed upon
returning to operation on 10 ppm S gasoline. There was greater than 95% confidence that the
differences in the mean emissions values measured before and after the high sulfur fuel exposure
5-39
-------
Chapter 5: Tier 3 Fuel Sulfur Program
were not statistically different.
21. SGS Environmental Testing Corp., Reversibility of Gasoline Sulfur Effects on Exhaust Emissions
From Late Model Vehicles, prepared for American Petroleum Institute, June 2013
Commenter: Mercedes-Benz USA, LLC on behalf of Daimler AG
The impact of a 10 ppm retail average limit in the Tier 3 proposal... and can be strengthened by
reducing the current refinery gate and retail sulfur caps to 20 and 25 ppm, respectively.
Commenter: Michigan Department of Environmental Quality (MDEQ)
Because poisoning of the catalyst can occur when larger amounts of sulfur are in the fuel and
these poisonings may, in some instances, be irreversible, the MDEQ, Air Quality Division
advocates the modification of the current refinery gate and downstream caps of 80 ppm and 95
ppm, respectively. The MDEQ, Air Quality Division would prefer the USEPA chose to lower the
refinery gate and downstream caps to 50 ppm and 65 ppm, respectively. The downstream sources
of sulfur should be able to take steps to decrease sulfur contamination, and have no incentive if
the caps remain where they currently are.
Commenter: New York State Department of Environmental Conservation
While catalyst sulfur poisoning is to a degree reversible, restoration of catalyst activity is not
instantaneous. High sulfur gasoline batches may have a negative impact on in-use emissions
beyond the fraction of the gasoline pool they represent. Given that the total amount of sulfur in
the gasoline pool will be reduced, the per-gallon maximum levels should also be reduced. These
levels should be set no higher than 50 ppm sulfur at the refinery and 65 ppm sulfur downstream.
Lower levels should be considered.
We concur with EPA that this change should take effect in 2020.
Commenter: Sierra Club, Clean Air Watch, Respiratory Health Association
It is important to note that EPA is proposing an average sulfur standard of 10 ppm. However, the
actual sulfur content of fuel can vary substantially, based on variations in refinery operations and
contamination from gasoline being transported through pipelines to its final destination.
Automakers must then design vehicles to function with higher-sulfur gasoline. In order to give
certainty to automakers designing vehicles and to ensure emissions reductions, setting caps on
per-gallon sulfur content is critical.
We urge EPA, as it lowers the average sulfur standard from 30 ppm to 10 ppm, to also lower the
per-gallon sulfur caps to at least 50 ppm at the refinery gate and 65 ppm downstream.
Lowering the per-gallon caps will allow automakers to employ additional advanced vehicle
technologies that can increase vehicle efficiency and lower emissions, while still allowing
compliance flexibility for refiners.
5-40
-------
Tier 3 Summary and Analysis of Comments
Commenter: Union of Concerned Scientists (UCS)
Lowering the per-gallon caps from 80 ppm at the refinery gate (and 95 ppm downstream) to 50
ppm at the refinery gate (and 65 ppm downstream) would not materially increase the cost of
compliance to refineries. Under the EPA's cost model, no refineries would be producing product
over 50 ppm per-gallon due to the 10 ppm average target, and only one refinery would likely be
over 40 ppm (6). Therefore, requiring these refineries to meet a 50 ppm per-gallon cap would
impose no incremental costs under normal operations, and would further ensure public health
protections by minimizing the impacts of the sulfur on the performance and durability of the
catalytic converter.
Commenter: Volkswagen Group of America, Inc.
As automakers, we must constantly balance market fuel availability and vehicle durability when
deciding on which concepts to bring to the US market. Having caps on things such as sulfur set
low enough would allow a range of vehicle technologies to be offered in a wide variety of
products. A 10 ppm cap is appropriate, but we support 20 ppm & 25 ppm caps at the refinery
gate and downstream, respectively.
Tier 3 vehicle emission standards are arguably the most stringent on earth; OEMs must have the
most state-of-the-art fuels to support the state-of-the-art vehicles. Therefore, VW supports
reductions in market fuel sulfur measured at the pump to a maximum 10 ppm. Other areas of the
world with less-strict regulations already have very low sulfur market gasoline.
Our Response:
As explained in Section V.C.b of the preamble to the final rule, we are retaining the
current Tier 2 80 ppm refinery gate cap. The refinery gate cap provides flexibility for batch-to-
batch variability that naturally occurs at a refinery due to the varying types of crude that
refineries process, variations in unit operations, and variations in product mix. It further provides
for flexibility during unit turnarounds, and unplanned upsets (e.g., refinery fires, natural
disasters, etc.), to avoid a complete refinery shutdown. A lower cap could create situations where
refiners would need to store more off-spec gasoline for future processing. However, if a refinery
does not have adequate tankage for storing this product, and/or if its processing units are not
large enough to "catch up" in refining off-spec product, it could result in significant impacts to
fuel pricing or supply. For a refiner that produces multiple products, any potential supply impacts
could also impact other fuel markets (e.g., diesel, jet fuel, etc.). Additionally, the refinery gate
cap is a "hard" limit—a refinery's actual production has to be well below this limit to account for
in-use testing tolerances, safety margins, and any additives that a refiner may need to add prior to
the fuel leaving the refinery. An 80 ppm refinery gate cap will provide refiners needed
flexibility, and more certainty that they will be able to continue producing and distributing at
least some gasoline during turnarounds/upsets to avoid a total shutdown. It will further provide
more certainty for transmix processors, additive manufacturers and other downstream parties.
As also described below in Section VII of the preamble to the final rule, we believe that
most refineries will be able to meet the 10 ppm average sulfur standard largely through revamps
5-41
-------
Chapter 5: Tier 3 Fuel Sulfur Program
and operational changes at their facilities, rather than installing grassroots units. Lowering to a
cap of 50 ppm would directionally increase the costs of the Tier 3 program. They provided a
detailed study quantifying the additional costs associated with successively more stringent per-
gallon caps. While we do not agree with the study's overall cost analysis, we do agree that with a
refinery gate cap of 50 ppm, a number of refiners would incur higher capital costs due to the
decreased ability to handle off-spec product with a lower refinery gate cap. As refiners must
ensure that they can continue to produce saleable product and meet demand in the event of an
upset or an off-spec batch of fuel, the need for installation of additional tankage and/or increased
refinery processing capability would be greater with a 50 ppm refinery gate cap. While at the
time of the proposal we believed that a cap of 50 ppm would have little cost impact, our more
recent analysis shows that a 50 ppm cap would increase the cost of the Tier 3 gasoline sulfur
standards by approximately 10 percent (see RIA Chapter 5.2.2.4). At the same time, the more
stringent cap with its associated increase in cost would be unlikely to provide significant
additional emission benefits nationwide. As discussed previously in Sections III and IV of the
preamble, the emissions benefits associated with the Tier 3 program are mainly driven by the
reduction in the average sulfur content of gasoline from 30 to 10 ppm, since vehicle emissions
are proportional to the sulfur content of the fuel. Changes in the cap would not affect this. In the
context of the final ABT provisions, a higher cap does allow for increases in emissions on a
temporal basis as one batch of fuel is allowed to have higher sulfur levels. However, this is then
offset by reductions in emissions from batches of fuel that are then required to be below the 10
ppm average standard. Similarly, the final ABT provisions allow for the possibility that the fuel
from different refineries will cause varying emission reductions as one refinery's higher average
sulfur levels would lead to less emission reductions in-use. However, this is then offset by
greater reductions in emissions due to the fuel produced by refineries with sulfur levels below
the average standard. In sum, we continue to believe that vehicles will see sulfur levels closer to
the 10 ppm average rather than the 80 ppm cap due to the fact that the 10 ppm average will drive
reductions in gasoline sulfur levels.
Thus, we believe it is prudent at this time to retain an 80 ppm refinery gate cap.
However, we are committing to monitor and further evaluate in-use sulfur levels and their impact
on vehicle emissions. If it is warranted, we will reassess the sulfur cap level and the need for
potential future regulatory action. Such ongoing evaluation will include analyses of in-use fuel
surveys, batch data that refineries are required to submit, and the sulfur credit market. It will
also include an evaluation of any issues or concerns that might arise during implementation of
the program. Finally, we will also carry out an ongoing evaluation of data submitted by the
vehicle manufacturers on the performance of their Tier 3 vehicles in-use.
As also discussed in Section IV.A.6 of the preamble to today's final rule, we believe the
80 ppm sulfur cap is sufficient to allay concerns over irreversibility or sulfur poisoning.
5.1.2.2. Impacts of Lower Caps on Fuel Supply/Cost
We received comments on both of the proposed per-gallon cap options of 80/95 ppm and
50/65 ppm, as well as comments on finalizing lower caps of 20/25 ppm and a 20 ppm overall
cap. Comments in support of maintaining the current Tier 2 caps cited concerns on cost, flexibility
5-42
-------
Tier 3 Summary and Analysis of Comments
for turnarounds/unplanned shut downs (due to refinery fire, natural disaster, etc.), and potential
impacts on gasoline supply and pricing.
What Commenters Said:
Commenter: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
The highest sulfur level recorded among gasoline in the contiguous U.S. in the July 2012
Alliance fuels survey was 43 ppm, (41) and in the January 2013 survey was 72 ppm. Only 1 of
the 394 retail samples in January 2013 exceeded 65 ppm, suggesting that EPA would be able to
transition the refinery gate and retail pump cap limits without imposing significant costs. A retail
sulfur cap of 25 ppm should be workable within the time period covered in the rule. As discussed
below, this change would also come closer to harmonization with California's 20 ppm sulfur
retail cap (42).
41 - Annual Alliance of Automobile Manufacturers North American Fuels Survey, July 2012.
42 - California Reformulated Gasoline Regulations, 13 C.C.R. §2262.
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM)
API and AFPM support EPA's proposed option for maintaining the current per gallon sulfur
caps: 80-ppm at the refinery gate and 95-ppm downstream. As EPA notes, the annual average
sulfur standard is a factor that limits the amount of sulfur in gasoline; per gallon caps are
important to manage planned and unplanned refinery unit downtime.
Beyond the refinery gate, the introduction of sulfur into gasoline occurs during pipeline shipment
through multiproduct pipelines and back-to-back shipments with higher sulfur content jet fuel. A
recent study by Turner Mason shows that reductions below the current sulfur cap standards, will
result in even higher capital costs, reduced compliance flexibility, and potential loss of gasoline
supplies.
API and AFPM support EPA's first proposed option for maintaining the current per gallon sulfur
caps: 80-ppm refinery gate sulfur cap and 95-ppm downstream sulfur cap. As EPA notes, the
annual average sulfur standard is a factor that limits the amount of sulfur in gasoline, and also
that per gallon caps have an important role. Beyond the refinery gate, the introduction of sulfur
into gasoline occurs during pipeline shipment through multiproduct pipelines and back-to-back
shipments with higher sulfur content jet fuel.
However, it is important to understand potential impacts of tighter per gallon sulfur caps on
compliance cost and supply impacts. A tighter per gallon sulfur cap, either with the current
annual average sulfur limit or a tighter standard as proposed for Tier 3, results in less flexibility
and could lead to supply reductions. This is because that in addition to downstream sulfur
introduction into gasoline, increased sulfur can occur inside the refinery gate due to planned or
unplanned upsets, unit turnarounds or unit shutdowns.
5-43
-------
Chapter 5: Tier 3 Fuel Sulfur Program
Another advantage of maintaining the current per gallon sulfur cap standard is that it allows
current transmix operations to continue. As noted by the EPA in the Tier 3 proposed rule,
transmix that occurs from pipeline shipping accounts for only a small amount of gasoline
consumption and most transmix batches of gasoline are approximately only 10 ppm above the
current Tier 2, 30-ppm refinery sulfur average. This indicates that sulfur content in transmix is
relatively small, even in the context of current refinery annual average sulfur standard.
API engaged Turner, Mason & Company (TM&C) to evaluate the economic, supply, and overall
gasoline pool quality implications for imposing more stringent per-gallon sulfur caps on U.S.
gasoline in addition to the assumed reduction in the annual average sulfur limit to 10 ppm. (33)
The complete TM&C report titled "Economic and Supply Impacts of a Reduced Cap on Gasoline
Sulfur" has been submitted as Attachment No. 12 for EPA's review.
TM&C's conclusions are summarized below:
- The cost to manufacture gasoline will increase as the sulfur cap is reduced from the current 80
ppm standard; capital costs range from approximately $2 billion to over $6 billion and annual
operating costs are estimated at $900 million for a 20 ppm cap. These costs are in addition to
those required to meet a 10 ppm annual average limit.
- Overall potential loss of gasoline supply will increase tenfold as the sulfur cap is reduced from
the current 80 ppm standard, resulting in 130 MBPD of supply loss at a 20 ppm cap.
- Regions served by just a few refineries could experience shortages of 25% - 50% during
outages of gasoline sulfur reduction units at a 20 ppm cap, while outages would be minimized at
sulfur caps exceeding 50 ppm.
Maintaining the current sulfur cap is a reasonable approach, but won't lower costs very much.
33 - Auers, J.R. et al.Turner, Mason & Company Consulting Engineers, Economic and Supply Impacts of
a Reduced Cap on Gasoline Sulfur Content, February 2013.
34 - SGS Environmental Testing Corp., Reversibility of Gasoline Sulfur Effects on Exhaust Emissions
From Late Model Vehicles, prepared for American Petroleum Institute, lune 2013
Commenter: BP Products North America Inc.
Maintain Current Sulfur Cap -BP recommends that EPA maintain the current 80 ppm refinery
gate per gallon cap and 95 ppm downstream per gallon cap as proposed. Reducing the sulfur cap
is unnecessary and adds to potential supply issues during turnarounds and upsets.
Commenter: Chevron Products Company
If EPA proceeds with the gasoline sulfur reduction, we do support several of the proposed
elements including: per gallon sulfur cap.
We agree that the most appropriate per-gallon cap would be Cap Option 1, to maintain the
existing 80 ppm refinery gate cap and the existing 95 ppm downstream cap. Under a 10 ppm
annual average, the sulfur cap will be largely self-limiting, as refiners will not be able to produce
individual batches near the sulfur cap and still meet the annual average. An arbitrarily low sulfur
5-44
-------
Tier 3 Summary and Analysis of Comments
cap like 20 or 25 ppm will result in a significant increase in capital requirements in the refinery
and will likely decrease refinery utilization and impact the supply of gasoline to the market.
Regarding Cap Option 2, while we do not feel that it would be necessary to reduce the refinery
gate sulfur cap to 50 ppm with a 65 ppm downstream cap in 2020, we do not strongly oppose this
option. A 50 ppm refinery gate cap still provides some flexibility to manage variability in the
sulfur content of the gasoline pool without imposing too stringent of a cap. Because the
maximum per-gallon sulfur is self-limiting due to the 10 ppm average, a 50 ppm cap would be
reasonable. The EPA should not consider any refinery gate sulfur caps below 50 ppm regardless
of timing. Reducing the sulfur caps to 50 ppm will add significant cost to the refinery and
downstream systems. However, going below this level is prohibitively expensive and not
warranted by the Tier 3 vehicle technology requirements (2). A Turner Mason & Co. study
commissioned by API indicates a significant increase in implementation cost for a sulfur cap
below 50 ppm.
2 - Turner Mason & Co., 'Economic and Supply Impacts of a Reduced Cap on Gasoline Sulfur', 2012
Commenter: CHS, Inc.
CHS is providing the following comments regarding the Tier 3 proposed gasoline refiner gate
per-gallon sulfur cap and downstream per-gallon sulfur cap. Although CHS prefers Cap Option
1, we can support the lower sulfur levels in Cap Option 2. CHS does not support lowering the
refinery gate per-gallon cap to 20 ppm sulfur and lowering the downstream per-gallon cap to 25
ppm sulfur. These lower limits do not allow enough room for refinery unit disruptions where
gasoline could still be produced at the higher sulfur per-gallon levels. The unintended
consequences of the proposed lower levels will affect the consumer's ability to buy gasoline and
cause regional gasoline price spikes that will put unnecessary economic burden on the consumer.
Commenter: Countrymark; Small Business Refiners (SBR)
EPA has proposed to keep the 80 ppm sulfur refinery cap in the draft regulatory text of the
proposed rule but asked for comment on reducing this cap to 50 ppm in 2020, and potentially
even further to 20 ppm in 2020. The SBRs are opposed to any reduction of the present cap and
believe EPA should keep this successful flexibility provision. This flexibility will particularly
benefit inland refiners that have limited ability to sell or ship blend stocks, should they have an
unplanned outage of the FCC post-treater.
EPA acknowledges in the proposed rule preamble that the unique US fuel distribution system
needs some level of cap room to allow for:
- Refinery flexibility to produce gasoline during temporary upsets and turnarounds
- Contamination challenges posed by pipelines and terminals that must also handle jet fuel with a
3,000 ppm sulfur specification
- Transmix processors that need ability to inject processed material back into the system.
With the existing 80 ppm cap, the Tier 2 sulfur rule has successfully reduced the average
gasoline sulfur to below 30 ppm as shown in the graph provided by EPA, included in the draft
Regulatory Impact Analysis (RIA):
5-45
-------
Chapter 5: Tier 3 Fuel Sulfur Program
Clearly EPA's own analysis shows very little incentive for lowering the 80 ppm cap. Keeping
this cap will allow refiners maximum flexibility in responding to unplanned events and outages.
This is especially true for inland refiners that lack the ability to sell/ship blend stocks should they
have an unplanned outage of the FCC post-treater. Lowering the cap could put the areas served
by inland refiners at increased risk for supply disruptions with no corresponding relief in average
gasoline sulfur level.
Commenter: Independent Fuel Terminal Operators Association (IFTOA)
Per Gallon Caps - Retain 80/95-ppm Standards: EPA has requested comments on two options
for the per-gallon sulfur caps. Under the first option, the Agency would retain the current 80
ppm refinery/import gate sulfur standard and the 95 ppm downstream (terminal) sulfur cap.
Under the second option, the refinery/import gate cap would be reduced to 50 ppm and the
downstream (terminal) cap would be set at 65 ppm.
As EPA is aware, to ensure the continuous flow of product, refineries and importers need
maximum flexibility. EPA recognizes that the U.S. has the longest and most complex gasoline
distribution system in the world, making it harder to control sulfur contamination than in other
countries. The 80/95 ppm sulfur caps provide such flexibility, allowing facilities to address
unanticipated problems such as a malfunctioning unit or a terminal hit by a superstorm, such as
Hurricane Sandy, or an anticipated shutting down of a unit due to a scheduled turnaround.
Lowering the per-gallon caps to 50/65 ppm sulfur reduces that flexibility and makes it harder to
meet supply obligations - often at times when supply is desperately needed.
Finally, IFTOA agrees with EPA that a per-gallon standard of 20/25 ppm would be impossible
for the industry to meet without almost all refinery facilities making substantial modifications to
their units. Significant changes to the nation's gasoline pipeline and terminal distribution system
would also be required. Such modifications would certainly compound the problems, burdens
and costs discussed above. Implementation of the annual average 10 ppm sulfur standard would
be delayed for many years.
Retention of the current 80/95 ppm per-gallon standards to provide maximum flexibility to the
industry. It is unlikely that industry would produce/import, ship or distribute gasoline meeting
those standards; however, it is important to have the ability to do so when unexpected events
occur and strain the distribution system.
Retain the current per-gallon sulfur caps of 80/95 ppm to provide the maximum degree of
flexibility to address unanticipated events.
Commenter: Irving Oil Terminals Inc.
EPA is proposing either to maintain the current 80 ppm refinery gate per gallon sulfur limitation
and the 95 ppm downstream per gallon cap or to lower them to 50 ppm and 65 ppm, respectively.
Irving Oil recommends that EPA retain the current per gallon caps of 80/95 ppm sulfur. These
limitations provide the most flexibility, allowing refiners to address operational problems such as
unexpected unit shutdowns and importers to address severe weather-related disruptions to the
5-46
-------
Tier 3 Summary and Analysis of Comments
downstream supply chain. For example, in a 'gasoline-short' supply scenario, an importer could
increase supply options beyond its typical resources. Considering the alternative options
proposed, the 50/65 per gallon limit is achievable but will reduce supply flexibility; and the
20/25 per gallon cap could result in a significant step-change in the level of capital investment
required by refiners. For this reason, Irving Oil strongly opposes the proposed 20/25 per gallon
cap. Retention of the 80/95 per gallon limits would provide as much compliance flexibility as is
possible while achieving the new Tier 3 goals of an annual average 10 ppm sulfur content.
Commenter: Magellan Midstream Partners, L.P.
Downstream Sulfur Cap - Under the Tier 3 gasoline sulfur program, EPA is proposing to (1)
maintain the current 80-ppm refinery gate and 95-ppm downstream per-gallon caps or (2) reduce
the cap to 50 and 65 ppm, respectively or (3) reduce the per-gallon caps to as low as 20 and 25
ppm.
Magellan Comment — We support EPA's proposal to maintain the current per gallon sulfur
caps, which are SOppm at the refinery and 95ppm downstream, for the following reasons:
The potential for downstream contamination is as prevalent today as it was at the time the Tier 2
rules were promulgated to regulate sulfur. In the proposed rule, EPA provides an extension to
small refiners for compliance with the reduced sulfur standards. Allowing small refiners to
continue production of gasoline with elevated sulfur levels provides further justification for
maintaining the existing downstream cap. We believe there will be a significant volume of
gasoline with elevated sulfur levels regardless of the cap for other refiners.
Lowering the downstream cap increases the chances of product outages at storage and
distribution terminals. If there is a sulfur excursion at a refinery, the flexibility associated with a
higher downstream cap will potentially reduce supply disruptions that could result from having
to reprocess gasoline.
Sulfur levels in jet fuel remain high. Major American airports rely almost entirely on pipelines,
and have dedicated pipelines to deliver jet fuel directly to the airport.
Maintaining the current per gallon cap allows current transmix operations to continue in certain
markets.
Commenter: Marathon Petroleum Company LP (MFC)
As to other specifics of the proposal, MFC agrees with EPA on the proposed cap of 80 ppm
sulfur and notes that any cap below this level has the potential to both limit gasoline supply
and/or increase the refinery investments required.
MFC supports EPA's first proposed option for maintaining the current per gallon sulfur caps: SO-
ppm refinery gate sulfur cap and 95-ppm downstream sulfur cap. As EPA notes, the annual
average sulfur standard is a factor that limits the amount of sulfur in gasoline, and also that per
gallon caps have an important role. Beyond the refinery gate, the introduction of sulfur into
5-47
-------
Chapter 5: Tier 3 Fuel Sulfur Program
gasoline occurs during pipeline shipment through multiproduct pipelines and back-to-back
shipments with higher sulfur content jet fuel.
However, it is important to understand potential impacts of tighter per gallon sulfur caps on
compliance cost and supply impacts. A tighter per gallon sulfur cap, either with the current
annual average sulfur limit or a tighter standard as proposed for Tier 3, results in less flexibility
and could lead to supply reductions. This is because that in addition to downstream sulfur
introduction into gasoline, increased sulfur can occur inside the refinery gate due to planned or
unplanned upsets, unit turnarounds or unit shutdowns.
Another advantage of maintaining the current per gallon sulfur cap standard is that it allows
current transmix operations to continue. As noted by the EPA in the Tier 3 proposed rule,
transmix that occurs from pipeline shipping accounts for only a small amount of gasoline
consumption and most transmix batches of gasoline are approximately only 10 ppm above the
current Tier 2, 30-ppm refinery sulfur average. This indicates that sulfur content in transmix is
relatively small, even in the context of current refinery annual average sulfur standard.
API engaged Turner, Mason & Company (TM&C) to evaluate the economic, supply, and overall
gasoline pool quality implications for imposing more stringent per-gallon sulfur caps on U.S.
gasoline in addition to the assumed reduction in the annual average sulfur limit to 10 ppm. The
complete TM&C report titled "Economic and Supply Impacts of a Reduced Cap on Gasoline
Sulfur" has been submitted to the docket for EPA's review.
TM&C's conclusions are summarized below:
-The cost to manufacture gasoline will increase as the sulfur cap is reduced from the current 80
ppm standard; capital costs range from approximately $2 billion to over $6 billion and annual
operating costs are estimated at $900 million for a 20 ppm cap. These costs are in addition to
those required to meet a 10 ppm annual average limit.
-Overall potential loss of gasoline supply will increase tenfold as the sulfur cap is reduced from
the current 80 ppm standard, resulting in 130 MBPD of supply loss at a 20 ppm cap.
Commenter: Monroe Energy, LLC
Lastly, EPA should not lower the SOppm per gallon cap. Lowering the per gallon cap is not
necessary to accommodate any engine technology, but would force some refiners, including
Monroe, to install additional controls that otherwise could be avoided via reliance on the credit
market for compliance.
Commenter: National Association of Convenience Stores (NACS) and Society of Independent
Gasoline Marketers of America (SIGMA)
NACS and SIGMA support the proposed option for maintaining the current per gallon sulfur
caps of 80 parts-per-million ("ppm") at the refinery gate and 95-ppm downstream.
5-48
-------
Tier 3 Summary and Analysis of Comments
NACS and SIGMA support EPA's proposed option for maintaining the current per gallon sulfur
caps of 80-ppm at the refinery gate and 95-ppm downstream. As EPA notes, pipeline transport
inevitably introduces the potential for sulfur contamination of the gasoline being shipped through
multiproduct pipelines. In addition, various gasoline additives also contain varying levels of
sulfur which contribute to the overall sulfur content of gasoline. If the final rule requires further
reductions below the current sulfur cap standards, it could result in higher capital costs for
refineries, which inevitably are passed down to retailers and consumers through higher prices.
The final rule should take every step possible to mitigate the adverse effect it may have on fuel
prices. Therefore, NACS and SIGMA support maintaining the current per gallon sulfur caps.
Commenter: PBF Energy Inc.
PBF supports the EPA option one proposal to keep per gallon sulfur cap at 80 ppm at the refinery
gate and 95 ppm downstream. PBF facilities have been able to comply with this standard during
any unplanned equipment outages.
PBF is opposed to the second option on the gasoline sulfur cap of lowering the per gallon cap to
50 ppm sulfur at the refinery gate or 65 ppm sulfur downstream or any lower per gallon sulfur
standard. PBF will make significant investments to comply with the proposed 10 ppm annual
sulfur standard and is not currently planning to invest further to ensure reliability at that level. In
the event of an unplanned outage on the new equipment installed, the refineries will likely fall
back to current gasoline product sulfur distribution levels. The EPA is aware of the sulfur content
distribution of PBF gasoline products through our required batch reporting. In the event of an
unplanned outage in the Northeast system there is a risk that the event could trigger a significant
supply disruption at the 50 ppm sulfur cap level and could possibly result in product price
increases to consumers as a result of the potential supply disruption.
Commenter: Phillips 66 Company
We support maintaining the 80 ppm refinery gate cap.
Phillips 66 supports maintaining the existing 80 ppm refinery gate cap as proposed. Providing an
80 ppm cap is beneficial in several ways. A lower cap with an effective date prior to the small
refiner/refinery compliance date would be problematic. The result would be a different sulfur cap
for small refiners/refineries causing complications in the distribution system. Also, reductions in
the sulfur cap could be constraining in certain operating situations. The result would be reduction
in supply until the operating problem was resolved.
On behalf of API, Turner, Mason & Company recently completed an evaluation entitled
"Economic and Supply Impacts of a Reduced Cap on Gasoline Sulfur". Phillips 66 participated
in this study by providing TM&C with data from our refineries. The study concludes that
additional capital costs would be required to meet a more stringent cap (in addition to capital
requirements to meet the 10 ppm sulfur average) and that the potential for loss of supply
increases with a reduction in the sulfur cap. API has provided a copy of this study with their
comments for inclusion in the docket.
5-49
-------
Chapter 5: Tier 3 Fuel Sulfur Program
Commenter: Shell Oil Products for Shell and Motiva
EPA proposes to leave the refinery per-gallon gasoline sulfur specification at 80 ppm, but is also
considering lowering the per-gallon specification to 50 ppm effective in 2020. EPA should not
change the refinery per-gallon sulfur specification from the existing 80 ppm.
A tighter per gallon sulfur cap will result in less flexibility and could lead to supply reductions.
Maintaining the per-gallon cap would provide important flexibility in the event of planned or
unplanned upsets, unit turnarounds or unit shutdowns, which can result in elevated sulfur levels.
If the per-gallon cap is too low, it could result in lost volumes of gasoline during such events.
Another advantage of maintaining the current per gallon sulfur cap standard is that it allows
current transmix operations to continue. Transmix that occurs from pipeline shipping accounts
for only a small amount of gasoline consumption and most transmix batches of gasoline are
approximately only 10 ppm above the current 30-ppm refinery sulfur average. This indicates that
sulfur content in transmix is relatively small, even in the context of current refinery annual
average sulfur standard.
Commenter: Sunoco Logistics Partners L.P.
In particular, we support maintaining the current 95 ppm downstream per gallon gasoline sulfur
cap to enable the current practices of pipeline interface handling. Pipeline and terminals will
continue to be constrained by interfaces of gasoline with jet fuel or home heating oil. Interfaces
with 3,000 ppm sulfur product are unavoidable, thus increasing the sulfur of the gasoline. Even
small volume interfaces with 3,000 ppm sulfur jet fuel can significantly increase the gasoline
sulfur. The northeast states are legislating the reduction of sulfur in home heating oil, which will
help alleviate the heating oil sulfur contribution, although no action has been taken or is being
contemplated to reduce the sulfur in jet fuel. Therefore we would not be supportive of lowering
the downstream sulfur cap as low as 25 ppm. Downstream facilities with the inability to change
their operations will not be able to handle interfaces of gasoline with jet fuel.
Commenter: Sutherland Asbill & Brennan LLP (Sutherland)
Per Gallon Sulfur Standards. EPA also proposes lowering the refinery gate per-gallon cap to 50
ppm sulfur and the downstream per-gallon cap to 65 ppm sulfur. Such reductions and any further
reductions may limit the availability and use of components used to blend gasoline. These
reductions could also limit imports of gasoline. It is important that EPA preserve the current
refinery gate and downstream caps to provide refiners, and perhaps especially gasoline
component blenders, the necessary flexibility to meet gasoline demand.
Commenter: Union of Concerned Scientists (UCS)
For these reasons, it is imperative that strict sulfur content limits be set on gasoline. This
standard will protect human health and enable the next generation of vehicle technologies be
implemented. We strongly support the EPA's proposal to limit the annual average sulfur content
5-50
-------
Tier 3 Summary and Analysis of Comments
of gasoline to 10 ppm and also urge EPA to lower the refinery gate and downstream per-gallon
sulfur caps to 50 and 65 ppm, respectively from the current levels of 80 ppm at the refinery gate
and 95 ppm downstream.
6 - Environmental Protection Agency (EPA). 2013. Draft Regulatory Impact Analysis Tier 3 Motor
Vehicle Emission and Fuel Standards. Online at http://www.epa.gov/otaq/documents/tier3/420dl3002.pdf
p. 4-22, accessed on June 28, 2013.
Our Response:
As explained in Section V.C.b of the preamble to the final rule, we are retaining the
current 80 ppm refinery gate and 95 ppm downstream sulfur caps in effect under the Tier 2
program. The refinery gate cap provides flexibility for batch-to-batch variability that naturally
occurs at a refinery due to the varying types of crude that refineries process, variations in unit
operations, and variations in product mix. It further provides for flexibility during unit
turnarounds and unplanned upsets (e.g., refinery fires, natural disasters, etc.), as noted by some
commenters, to avoid a complete refinery shutdown or supply issues. A lower cap could create
situations where refiners would need to store more off-spec gasoline for future processing.
However, if a refinery does not have adequate tankage for storing this product, and/or if its
processing units are not large enough to "catch up" in refining off-spec product, it could result in
significant impacts to fuel pricing or supply. For a refiner that produces multiple products, any
potential supply impacts could also impact other fuel markets (e.g., diesel, jet fuel, etc.).
Additionally, the refinery gate cap is a "hard" limit—a refinery's actual production has to be well
below this limit to account for in-use testing tolerances, safety margins, and any additives that a
refiner may need to add prior to the fuel leaving the refinery. Thus, we believe that retaining the
80 ppm refinery gate cap will provide refiners helpful flexibility, and more certainty that they
will be able to continue producing and distributing at least some gasoline during
turnarounds/upsets to avoid a total shutdown. Further, it will further provide more certainty for
transmix processors, additive manufacturers and other downstream parties.
As discussed more in Section VII.B of the preamble to the final rule, lowering to a cap of
50 ppm would direct!onally increase the costs of the Tier 3 program. The Turner Mason & Co.
study cited in API and AFPM's comment quantified the additional costs associated with
successively more stringent per-gallon caps. Although we do not agree with the study's overall
cost analysis, we do agree that with a refinery gate cap of 50 ppm, a number of refiners would
incur higher capital costs due to the decreased ability to handle off-spec product with a lower
refinery gate cap. As refiners must ensure that they can continue to produce saleable product and
meet demand in the event of an upset or an off-spec batch of fuel, the need for installation of
additional tankage and/or increased refinery processing capability would be greater with a 50
ppm refinery gate cap. While at the time of the proposal we believed that a cap of 50 ppm would
have little cost impact, our more recent analysis shows that a 50 ppm cap could increase the cost
of the Tier 3 gasoline sulfur standards (see RIA Chapter 5.2.2.4). At the same time, the more
stringent cap (with its associated increase in cost) would be unlikely to provide significant
additional emission benefits nationwide.
5.1.2.3. Air Quality and Local Impacts of Sulfur Cap Levels
5-51
-------
Chapter 5: Tier 3 Fuel Sulfur Program
We also received comments arguing that lower caps of 20/25 ppm and a 20 ppm overall cap
would address "hot spots" that tend to have higher than average sulfur levels.
What Commenters Said:
Commenter: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
The Alliance fuel surveys also demonstrate that the absence of a low downstream sulfur cap
facilitates the creation of "hot spots" with higher than average sulfur content. However, these hot
spots appear to be geographically transient over time, as confirmed by the differences just
between July 2012 and January 2013 sampling, suggesting that there are not predictable
locations that consistently struggle to meet sulfur limits. Nonetheless higher sulfur locations do
occur regularly. This underscores the uncertainties for OEMs building vehicles for a nationwide
market.
Commenter: American Lung Association
Lower the sulfur caps. The American Lung Association supports lowering the per-gallon sulfur
caps to 20 ppm at the refinery gate and 25 ppm downstream from the current 80 ppm per gallon
refinery gate and 95 ppm per-gallon downstream caps in 2020. Although EPA is setting an
average sulfur gasoline standard of 10 ppm, there will be refinery-by-refinery differences in the
sulfur content of fuel due to operational differences. Additionally, fuel can be contaminated
during transportation through pipelines to its final destination. As a result, the fuel quality
available may differ by location exposing some populations to higher tailpipe emissions than
others. EPA should set the refinery gate cap and downstream cap to 20 ppm per gallon and 25
ppm per gallon respectively to limit the exposure of vehicles in-use to sulfur levels that
significantly degrade pollution control performance and to ensure all communities enjoy the
benefits of cleaner air.
Commenter: Consumers Union
EPA's modeling indicates that regardless of the downstream sulfur cap selected between the two
options being considered (95 ppm or 65 ppm), nearly all gasoline that ends up at the pump would
be close to the 10 ppm average required of refineries. Automakers will provide insight into any
burdens that a higher downstream cap may place on them to design vehicles that can tolerate the
higher upper limit. Although sulfur-induced corrosion of catalytic converters is largely reversible
if lower sulfur gasoline is used with regularity (justifying the averaging approach), if there are
retail hot spots or regional high-sulfur clusters that consistently sell gasoline towards the upper
limit, car owners in these areas may find their catalytic converters fail at a higher rate, and
emissions may be significantly higher in such locales.
Commenter: CountryMark Cooperative, Small Business Refiners (SBR)
5-52
-------
Tier 3 Summary and Analysis of Comments
EPA's own analysis in the RIA of the proposed nation-wide credit trading program predicts that
only 17% of the gasoline volume, mostly from smaller than average refineries, would exceed the
10 ppm standard and the vast majority would be below 30 ppm as shown in EPA's graph.
EPA also stated in this analysis: Under this scenario, most refineries with average sulfur levels
higher than 10 ppm supply areas that are also supplied by refineries with sulfur levels at 10 ppm
or below. We expect dilution of higher sulfur with lower sulfur in such areas to result in average
sulfur levels no higher than about 20 ppm. Of those refineries projected to have average sulfur
levels higher than 10 ppm under this scenario and which are the primary or only suppliers of
gasoline to a particular area, none of the affected areas have historically been in nonattainment
for ozone.
We agree with the above and believe this is a worst case scenario as it assumes perfect trading of
sulfur credits, which is not realistic. Some of these higher sulfur refineries will not be able to rely
on a supply of credits and therefore will take steps to reduce gasoline sulfur below the levels
discussed above.
Commenter: Environmental Defense Fund (EDF)
EDF requests that EPA finalize a refinery gate sulfur cap of 20 ppm and a downstream sulfur cap
of 25 ppm - standards that the Agency is officially requesting comment on. More lenient
standards, as proposed by EPA, will not provide as robust health protections and "vehicles still
have to be designed to operate on high sulfur fuel, and there will be less attention paid to limiting
contamination downstream of the refinery gate."
EDF believes that the 20 ppm and 25 ppm standards will better protect human health. EPA has
also stated that the tighter per gallon caps will benefit automakers by "limit[ing] the temporary
exposure of vehicles in-use to sulfur levels that would significantly degrade their emission
performance." EPA also notes that the more protective caps would "serve to provide added
assurance that all parts of the country would receive the full emission benefits of gasoline sulfur
control."
EDF believes that the proposal gives refiners and affected downstream parties ample additional
time and flexibilities to meet the more protective per-gallon caps and we urge the Agency to
finalize the most protective refinery gate and downstream caps.
In the event that EPA rejects the most protective 20 ppm refinery gate and 25 ppm downstream
caps in favor of one of the two proposed alternatives, then EDF believes that EPA must opt for a
50 ppm refinery gate cap and a 65 ppm downstream cap. In issuing the proposed Tier 3 rule, the
Administrator found pursuant to Clean Air Act section 21 l(c)(l) that the emission products of
higher-sulfur gasoline both: (1) cause or contribute to air pollution which may reasonably be
anticipated to endanger public health or welfare; and (2) will impair to a significant degree the
performance of any emissions control device or system which is either in general use or which
the Administrator finds has been developed to a point where in a reasonable time it will be in
general use were the fuel control to be adopted. In light of these findings and in order to "limit
the temporary exposure of vehicles in-use to sulfur levels that would significantly degrade their
5-53
-------
Chapter 5: Tier 3 Fuel Sulfur Program
emission performance" and assure that "all parts of the country would receive the full emission
benefits of gasoline sulfur control," it would be arbitrary and capricious for EPA to set refinery
gate and downstream caps that do not strengthen current standards.
Commenter: Environmental Justice Leadership Forum on Climate Change
I caution that the credit trading provision designed to offer flexibility during the transition to
more efficient vehicles and cleaner fuels seems counterintuitive to uniform national standards,
thereby uniform benefits. I propose that the success of the Tier 2 standards are well documented
and are all the transition that we ever needed.
We casually speak of trading credits, but if we look at the transaction closely, someone is
standing with a basket full of credits, and someone is standing with a basket full of disbenefits or
pollution.
Because all communities deserve their share of greenhouse gas reductions, fuel savings, energy
security, mitigation of disproportionate adverse health impacts, economic growth, and job
creation, I ask that you consider the benefits that - or the impact that the trading provision has
had on the Tier 2 standards as we move forward in Tier 3.
Commenter: Marathon Petroleum Company LP (MFC)
-Regions served by just a few refineries could experience shortages of 25% - 50% during outages
of gasoline sulfur reduction units at a 20 ppm cap, while outages would be minimized at sulfur
caps exceeding 50 ppm.
Commenter: Consumers Union (comment campaign)
EPA's modeling indicates that regardless of the downstream sulfur cap selected between the two
options being considered (95 ppm or 65 ppm), nearly all gasoline that ends up at the pump would
be close to the 10 ppm average required of refineries. Automakers will provide insight into any
burdens that a higher downstream cap may place on them to design vehicles that can tolerate the
higher upper limit. Although sulfur-induced corrosion of catalytic converters is largely reversible
if lower sulfur gasoline is used with regularity (justifying the averaging approach), if there are
retail hot spots or regional high-sulfur clusters that consistently sell gasoline towards the upper
limit, car owners in these areas may find their catalytic converters fail at a higher rate, and
emissions may be significantly higher in such locales.
Commenter: Natural Resources Defense Council (NRDC)
NRDC strongly supports the proposed requirement that refineries meet an annual average sulfur
limit of 10 parts per million but EPA should tighten refinery gate and downstream sulfur caps.
EPA Should Tighten Refinery Gate and Downstream Sulfur Caps. To help ensure that the
maximize the cost-effective pollution reductions are achieved by the lower sulfur gasoline
5-54
-------
Tier 3 Summary and Analysis of Comments
coming from the refineries, EPA should lower the refinery gate cap to no more than 50 ppm by
January 1, 2020 and the downstream cap to no more 65 ppm per gallon by March 1, 2020.
Tightening the caps from today's levels will help increase the certainty that gasoline's low sulfur
qualities will be maintained throughout the production and delivery processes. Tighter caps will
reduce the probability of large sulfur spikes due to production equipment outages and
maintenance and the probability of severe sulfur contamination during transport from the
refinery gate to the retail outlets. By enhancing the quality assurance of delivered gasoline,
automakers can better optimize the design of vehicle engines and exhaust systems within a
smaller range of fuel specifications and the probability that all regions of the country will see the
maximum pollution reductions from lower sulfur gasoline will be enhanced.
Commenter: Northeast States for Coordinated Air Use Management (NESCAUM)
The current Tier 2 refinery gate and downstream caps of 80 and 95 ppm should be lowered to 50
and 65 ppm, respectively. Given the stringency of the 10 ppm average standard, these lower caps
will provide sufficient flexibility for refiners, pipelines, terminals, transmix processors, and
gasoline additive manufacturers as achieved under the Tier 2 program, while ensuring maximum
reduction in downstream sulfur levels.
Long-term cap relief should not be made available in conjunction with lowering the sulfur caps
below the current 80/95 ppm level. The other shorter term hardship relief options should be
sufficient for all circumstances. Having noted that, it is important that refineries be given
sufficient time so as not to be faced with the need to undergo two turnarounds (shutdowns and
startups for major maintenance or equipment installation) in a relatively short time period. To the
extent reasonable, the equipment installation for low sulfur fuel equipment should be
accommodated during a normally scheduled maintenance turnaround. This avoids the excess
emissions that occur during multiple shutdowns and startups of a refinery.
Commenter: Union of Concerned Scientists (UCS)
EPA should lower the refinery gate and downstream per-gallon sulfur caps to 50 and 65 parts per
million (ppm), respectively from the current levels of 80 ppm at the refinery gate and 95 ppm
downstream.
Reducing the sulfur content of gasoline is critical to achieving the public health and air quality
benefits promised by the Tier 3 program. Combustion of sulfur in gasoline produces exhaust
containing sulfur dioxide (SO2), which can cause respiratory ailments, particularly in sensitive
populations like children, the elderly and people with asthma. Additionally, SO2 is a precursor to
particulate matter formation; the detrimental effects of paniculate matter on human health are
discussed in the next section. Unlike other compounds, most notably NOx, sulfur compounds in
vehicle exhaust cannot be converted into a less harmful molecule using emission control
technologies like catalytic converters. Therefore, all tailpipe sulfur emissions are harmful and the
only way to reduce sulfur emissions from the vehicle is to reduce its content in the fuel supply
(5).
5-55
-------
Chapter 5: Tier 3 Fuel Sulfur Program
Sulfur in fuel also leads to sulfur oxides in the exhaust that poison the catalytic converter,
reducing its effectiveness. Sulfur is adsorbed into the catalytic converter, greatly reducing the
efficiency of the converter for cleaning up exhaust emissions such as NOx, carbon monoxide,
and hydrocarbons. Therefore, reducing the sulfur content of gasoline will provide immediate
public health benefits by increasing the performance of existing catalytic converters, reducing
harmful emissions from the legacy fleet before complete fleet turnover has occurred. Moreover,
as the nation's vehicle fleet becomes more efficient, lower sulfur fuel will enable technology
advances that allow greater efficiency gains, as well as ensure that existing technology performs
as anticipated. Additionally, some technologies that help deliver efficiency improvements in the
vehicle fleet can also lead to cooler exhaust temperatures, which inhibit the regenerative abilities
of the catalytic converter, reducing its effectiveness. Lowering the sulfur content of fuel will
ensure that we are able to effectively capitalize on the full suite of technologies available to
improve both vehicle emissions and efficiency.
5 - Testimony of Dr. Phil Blakeman of Johnson Matthey. Public hearing on Tier 3 Standards.
Philadelphia, PA. April 24, 2013.
Our Response:
As discussed in Sections V.C.b and VII of the preamble to the final rule, our analysis for
the Tier 3 final rule shows that a lower refinery gate cap would be unlikely to provide significant
additional emission benefits. The emissions benefits associated with the Tier 3 program are
mainly driven by the reduction in the average sulfur content of gasoline from 30 to 10 ppm, since
vehicle emissions are proportional to the sulfur content of the fuel—changes in the cap would
not affect this. In the context of the final ABT provisions, a higher cap does allow for increases
in emissions on a temporal basis as one batch of fuel is allowed to have higher sulfur levels.
However, this is then offset by reductions in emissions from batches of fuel that are then
required to be below the 10 ppm average sulfur standard. Similarly, the ABT provisions allow
for the possibility that the fuel from different refineries will cause varying emission reductions as
one refinery's higher average sulfur levels would lead to less emission reductions in-use.
However, this is then offset by greater reductions in emissions due to the fuel produced by
refineries with sulfur levels below the 10 ppm average standard.
We anticipate that in most cases refineries will make operational changes and/or
investments in order to reduce their credit burden and reduce their compliance costs. Further, the
10 ppm average standard by definition limits the amount of gasoline that can remain at higher
sulfur levels (regardless of the cap). Thus, we anticipate that most refineries, including those
using credits, will still average less than 20 or 30 ppm. Nevertheless, the final ABT program
does allow for the possibility (regardless of the cap) that if higher sulfur fuel is concentrated in
any certain geographical area, it would not receive the full emission reductions from the Tier 3
program. We have considered the potential for areas to consistently receive fuel that might be
predominantly higher than the 10 ppm average. However, since refineries generating credits and
those using credits are interspersed across the country—and because most areas receive a
considerable portion of their fuel by pipeline, barge, rail, or truck from refineries in other areas—
we expect the variation in average sulfur levels across the country to be too limited to warrant
lowering the per-gallon cap to 50 ppm. Though, given the stringency of the 10 ppm average
standard, we predict that in-use sulfur levels will generally be well below 50 ppm.
5-56
-------
Tier 3 Summary and Analysis of Comments
Thus, we believe it is best to retain the current Tier 2 80 ppm refinery gate cap, however
we are committing to monitor and further evaluate in-use sulfur levels. As discussed above in
Chapter 5.1.2.1, we will reassess the refinery gate sulfur cap level and the need for potential
future regulatory action if it is warranted.
5.1.2.4. Downstream Cap Level Delta
What Commenters Said:
Commenter: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
EPA's RIA points out refiners (particularly smaller ones) may experience intermittent spikes in
sulfur in gasoline secondary to cleaning tanks and other equipment or during turnarounds (46).
However, options to dilute or otherwise reduce resultant higher sulfur content in this gasoline
and related costs are not discussed.
EPA also underscores that gasoline sulfur content can increase slightly from sulfur in additives,
from contamination of the pipeline itself, and during pipeline transport transmix when the
gasoline comes into contact with adjacent fuels such as jet fuel, that have very high sulfur
content. Similar types of fuel supplier concerns raised in the Ultra-Low Sulfur Diesel
rulemaking did not turn out to affect the ability to meet 15 ppm sulfur diesel requirements,
despite use of the same domestic pipeline distribution systems.
46-RIA at 10-7.
Commenter: Independent Fuel Terminal Operators Association (IFTOA)
Recommendations: Accordingly, the Association recommends - Retention of the 15 ppm
differential between the refinery/import gate and the downstream sulfur cap to address sulfur
contamination as gasoline moves through the lengthy distribution system; and
Commenter: Kinder Morgan Transmix Company, LLC
We support EPA's proposal to maintain a 15-ppm differential between the proposed refinery
gate sulfur cap of 50-ppm and the proposed 65-ppm downstream sulfur cap. Our support of the
15-ppm differential is based on gasoline sulfur data collected from our transmix processing
facilities over a 3-year period of time. A reduction or elimination of the 15-ppm differential
between the refinery gate cap and the downstream cap would have a significant impact on our
operations which could result in facility shut-downs or limited operations.
We also support EPA's proposed downstream cap implementation date of 2020. We concur that
based upon a 2017 refinery gate cap of 50-ppm, transmix processors would have sufficient time
to evaluate the impact of the regulation on our operations prior to 2020. Transmix typically
5-57
-------
Chapter 5: Tier 3 Fuel Sulfur Program
consist of varying quantities of refined petroleum products such as gasoline, diesel fuel, aviation
kerosene and home heating fuel. Reducing gasoline sulfur levels as presented in the proposed
regulation; combined with the existing 15-ppm sulfur limit on ultra low sulfur diesel; may result
in increasing sulfur levels in aviation kerosene or home heating fuel. Aviation kerosene and
home heating fuel (where not restricted by state regulations) may contain sulfur levels up to
3,000-ppm. The 2020 downstream implementation provides sufficient time to evaluate the
impact of reducing gasoline sulfur levels on transmix and take corrective action if necessary.
In summary: Yes, we support EPA's proposal to maintain a 15-ppm differential between the
proposed refinery gate sulfur cap of 50-ppm and the proposed 65-ppm downstream sulfur cap.
Commenter: New York State Department of Environmental Conservation
New York's fuel sampling program for 2011 and 212 shows no gasoline samples exceeding
either of the existing per-gallon standards (over 8500 samples). The highest measured sulfur
content was 79 ppm. There was no discernible difference between terminal samples and retail
samples, suggesting that a 15 ppm downstream tolerance is larger than necessary.
Our Response:
With regard to the downstream sulfur cap, we believe that maintaining a 15 ppm
differential between the refinery gate sulfur cap and the downstream sulfur cap will provide
pipeline operators, transmix processors, and gasoline additive users the same flexibility as was
provided under the Tier 2 program. As was the case under the Tier 2 program, allowing a 15
ppm differential is needed to ensure adequate flexibility in accommodating gasoline produced
from transmix, instances of contamination during distribution, and for the use of necessary
(sulfur-containing) additives. While we agree that sulfur levels of 95 ppm will not be the norm,
in those rare circumstances when the sulfur contribution from all these sources are concurrently
at their maximum levels, a very limited number of batches of gasoline at the 95 ppm downstream
sulfur cap may be present in the distribution system. However, we do not expect that this will
have a substantial impact on the average sulfur content of in-use gasoline.
Additionally, transmix processors must produce gasoline sufficiently below the 95 ppm
downstream sulfur cap to accommodate any downstream sulfur increases from the use of
gasoline additives and contamination from further distribution. The sulfur content of the
gasoline produced by transmix processors is determined by the sulfur content of the transmix
they receive, which is primarily a function of the sulfur content of gasoline and jet fuel
components in the transmix. Transmix processors do not handle sufficient volumes to support
the installation of desulfurization units, which would also increase the cost of the rulemaking
with little to no additional emissions benefit.
Regarding comments that the 15 ppm delta is not needed because there were not
contamination problems when the sulfur level in diesel fuel was reduced to 15 ppm, we note that
the diesel program is quite different than the final gasoline sulfur program. During
implementation of the diesel program, pipeline and downstream parties set "inlet specs" for the
sulfur level of fuel that they would accept (some pipelines had inlet specs as low as 7 ppm).
5-58
-------
Tier 3 Summary and Analysis of Comments
Thus, the diesel fuel that refiners needed to produce had to be much lower than the 15 ppm cap
to ensure that the fuel would still meet a 15 ppm sulfur standard after traveling through the
distribution system. Also, the diesel sulfur program was a phased-in program that also allowed
for two month transition periods for when each level of the fuel production and distribution
system was required to comply (e.g., refiners were required to comply by June 1, terminal
operators by August 1, etc.); the Tier 3 gasoline program does not have a transition period.
Lastly, contamination of 15 ppm diesel fuel was reduced by specific batch sequencing in
pipeline; however, gasoline remains susceptible to contamination by higher sulfur jet fuel and
heating oil.
5.1.2.5. Other (Alternate Phase-Down, SO2, etc.)
What Commenters Said:
Commenter: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
We also support reducing the refinery gate and retail pump per gallon caps to 20 and 25 ppm
respectively for an interim time period, rather than leaving them at Tier 2 levels (80 and 95 ppm
respectively), or partially reducing them to 50 and 65 ppm, respectively. In addition, we also
advocate that as part of the Tier 3 rulemaking process, EPA should commit to a supplemental
rulemaking to develop a pathway and timeline for reducing the retail cap for market gasoline
sulfur to 10 ppm, consistent with other developed countries. This is important so that refineries
can plan now for longer-term horizon of additional sulfur reduction, rather than seeing such
changes as "moving the goal posts again." It is also crucial to the product planning of Alliance
and Global member companies that there is assurance that over time EPA is moving toward
progressively lower sulfur standards at the retail pump toward a cap no greater than 10 ppm per
gallon (45).
Tier 3 sulfur reductions are also important because, if reduced to the lowest levels suggested by
EPA (25 ppm retail cap), the sulfur level will come closer to harmonizing the 20 ppm retail cap
in California and will create a predictable low sulfur standard nationwide. The results in the
January 2013 Alliance North American Fuel Survey show that the average sulfur content of the
samples outside of California was 25 ppm,65 or approximately three times higher than the
California samples average of 7 ppm, so it should be feasible to transition to a lower national
retail sulfur cap.
In addition, it is important that the EPA include a pathway to continue to ratchet down the sulfur
content of the per-gallon cap at the retail pump over time to reach a per-gallon retail pump level
of 10 ppm, in order to optimize emission reductions.
45 - We note that EPA's own data shows that 5 ppm is even better for reducing emissions, as evidenced
in its new study cited in the Preamble and RIA.
65 - The (octane) grade weighted (pool average for US (excluding California) would be closer to 27 ppm.
Survey can be purchased from the Alliance website www.autoalliance .org.
5-59
-------
Chapter 5: Tier 3 Fuel Sulfur Program
Commenter: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
We urge EPA to reduce the refinery gate and downstream caps to 20 and 25 ppm, respectively,
perhaps with a slightly longer phase in period than the refinery average, and at the same time to
develop a pathway toward a retail cap of 10 ppm per gallon. Effective standards should be put in
place for Tier 3, and the Agency can then provide ad hoc relief, with a notice and comment
process, for refiners and downstream distributors that can show evidence or reasons they actually
need it.
Because the country as a whole is already trending toward lower sulfur levels, EPA can and
should finalize the means to transition to a retail cap of 25 ppm, and then plan a pathway to reach
an ultimate goal of 10 ppm retail cap over the next 12-15 years. If it turns out that this goal
cannot be accomplished, then EPA can modify the limit, or extend the deadline, but the policy to
ultimately minimize retail gasoline sulfur to a 10 ppm cap per gallon (and reduce sulfur content
of other pipeline products mingled or trans-mixed with gasoline accordingly) should be
developed now, for implementation within the framework of Tier 3 authority.
Commenter: Consumers Union
EPA should consider issuing a prospective reporting or data collection requirement that could
help it identify whether such hot spots emerge and whether they warrant a lower downstream
sulfur cap in the future.
Commenter: General Motors LLC (GM)
In addition, is important that the EPA include a pathway to continue to ratchet down the sulfur
content of the per-gallon cap at the retail pump over time to reach a per-gallon retail pump level
of 10 ppm, in order to optimize emission reductions.
The Tier 3 rulemaking is the correct place to set the near-term standards for sulfur and also
create a roadmap of how the agency will further reduce sulfur in the coming years.
Commenter: Consumers Union (comment campaign)
EPA should consider issuing a prospective reporting or data collection requirement that could
help it identify whether such hot spots emerge and whether they warrant a lower downstream
sulfur cap in the future.
Commenter: Ozone Transport Commission (OTC)
In addition to proposing an average gasoline sulfur concentration of 10 ppm, EPA proposed
maintaining downstream and refinery caps of 95 ppm and 80 ppm, respectively. This was done
to allow refineries to have the flexibility to economically meet the sulfur requirements through
5-60
-------
Tier 3 Summary and Analysis of Comments
an allowance, banking, and trading (ABT) program. EPA also suggested the possibility of lower
downstream and refinery caps and requested comment on these lower caps.
OTC suggests that EPA consider evaluating a gradual reduction in the sulfur caps from 2017-
2025. The caps would begin at 95 and 80 ppm respectively for the downstream and refinery
levels, then be reduced to 65 and 50 ppm by 2020, and finally, in 2025, be reduced to a level that
harmonizes the sulfur cap nationally. This approach could still allow refineries to have the
economic flexibility that they need to successfully meet the caps at a low cost, while, in the long
term, limiting anomalies in the fuel market that could result in particular areas having fuels that
do not reduce NOx to a level necessary to reduce ambient ozone concentrations to attainment
status concentrations.
Commenter: Pennsylvania Department of Environmental Protection (DEP)
The EPA should allow the current 80 ppm fuel sulfur level refinery gate cap and the 95 ppm
downstream per gallon cap. The agency should consult with the refinery industry, especially
small refiners, and develop a phase-in schedule that will lower refinery gate and downstream
caps to the 20 ppm to 30 ppm range by 2025.
Commenter: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
EPA's RIA points out refiners (particularly smaller ones) may experience intermittent spikes in
sulfur in gasoline secondary to cleaning tanks and other equipment or during turnarounds (46).
However, options to dilute or otherwise reduce resultant higher sulfur content in this gasoline
and related costs are not discussed.
EPA also underscores that gasoline sulfur content can increase slightly from sulfur in additives,
from contamination of the pipeline itself, and during pipeline transport transmix when the
gasoline comes into contact with adjacent fuels such as jet fuel, that have very high sulfur
content. If, as the Tier 3 RIA indicates, the increase in the per gallon cap at retail from the per
gallon refinery gate cap is due to the sulfur content of other current products moving through the
pipeline, rather than older sulfur deposits within the pipeline itself, then one potential solution for
gasoline sulfur retail cap limits is to reduce the sulfur content of these other products, which
could have environmental benefits as well. If jet fuel can be as high as 2-3000 ppm sulfur, and
low sulfur fuel oil can be as high as 500 ppm, then EPA should lead efforts to reduce that content
and find substitute means for the lubricity loss from reduced sulfur.
46-RIA at 10-7.
Our Response:
As noted above in Chapter 5.1.2.1, we are committing to monitor and further evaluate in-
use sulfur levels and their impact on vehicle emissions and, if it is warranted, we will reassess
the sulfur cap level and the need for potential future regulatory action. This evaluation will
include analyses of in-use fuel surveys, refinery batch data, and the sulfur credit market. We will
5-61
-------
Chapter 5: Tier 3 Fuel Sulfur Program
also evaluate any issues or concerns that might arise during the implementing the program.
Finally, we will also carry out an ongoing evaluation of data submitted by the vehicle
manufacturers on the performance of their Tier 3 vehicles in-use.
5.1.3. Additives
What Commenters Said:
Commenter: Automotive Aftermarket Industry Association (AAIA)
The Automotive Aftermarket Industry Association (AAIA) appreciates this opportunity to
provide comments on the proposed Tier 3 Emissions and Fuel Standards for motor vehicles. In
particular, we would like to address the provision relating to sulfur content of automotive fuel
additives.
AAIA supports the provision within the Tier 3 standards that proposes 'gasoline additives that
are used downstream of the refinery at less than 1 volume percent must limit the sulfur
contribution to the finished gasoline from the use of their additive to less than 3 ppm when the
additive is used at the maximum recommended treatment rate.' At currently recommended use
levels, aftermarket fuel additives contribute less than the proposed 3 ppm. Furthermore, additives
are not applied every time a motor vehicle stops for a fill-up. Therefore, the sulfur contribution
continues to decrease over time.
Commenter: Consumer Specialty Products Association (CSPA) & Automotive Specialty
Products Alliance (ASPA)
Our organizations support the proposed rule as it pertains to the sulfur content of downstream
additives used at less than 1.0 percent by volume of the resultant additive(s)/fuel mixture
(Section A (3): Per-Gallon Sulfur Caps). When used at the maximum recommended treatment
rate, aftermarket gasoline additives contribute less than the proposed limit of 3 parts per million
sulfur to the finished gasoline. Considering most of these products are recommended for use
every 3,000-10,000 miles, the overall net increase in sulfur levels is further diluted.
Manufacturers of aftermarket gasoline additives have taken many proactive steps to protect the
environment, including adopting sustainable packaging initiatives - moving, where possible, to
more concentrated packages, which require using less solvent, less packaging and therefore
creating less freight. Additionally, aftermarket additives play an important role in the cleanup of
intake valves, fuel injectors and combustion chambers, resulting in fuel economy savings and
emissions reductions.
Commenter: Motor & Equipment Manufacturers Association (MEMA)
Control of fuel sulfur content is a critical aspect of Tier 3. Specifically for additives, MEMA
agrees with the proposed treatment of sulfur content in downstream additives that are used at less
than one percent (1.0%) by volume of the resultant additive/fuel mixture. When used at the
5-62
-------
Tier 3 Summary and Analysis of Comments
maximum recommended treatment rate, aftermarket gasoline additives contribute less than the
proposed limit of three (3) parts-per-million sulfur to the finished gasoline. Since many of these
products are recommended for use every 3,000 to 10,000 miles, the overall net increase in sulfur
levels is further diluted. For downstream additive manufacturers, the proposed recordkeeping
requirements demonstrating compliance is reasonable and manageable.
Commenter: New York State Department of Environmental Conservation
Control of the sulfur content of gasoline additives is necessary. The proposed sulfur limits are
low enough that the sulfur content of fuel additives can have a material impact on the finished
gasoline sulfur content, and therefore the environmental performance. We support EPA's
proposal to indirectly control the sulfur content of these additives by limiting the increase from
an additive (at its maximum treat rate) to 3 ppm. We also encourage EPA to promulgate limits on
the combined sulfur contribution for all additives blended into a batch of gasoline, not just for
each individual additive.
We also support EPA's requirement that any blender blending an additive at a rate of 1.0 percent
by volume or higher comply with the gasoline sulfur requirements applicable to refineries.
Our Response:
The Tier 3 FRM finalizes the requirement that manufacturers of gasoline additives used
downstream of the refinery at less than 1 volume percent must limit the sulfur contribution to the
finished gasoline from the use of their additive to no more than 3 ppm when the additive is used
at the maximum recommended treatment rate. The additive manufacturer will be required to
maintain records of its additive production quality control activities which demonstrate that the
sulfur content of additive production batches is such that when the additive is used at its
maximum recommended treatment rate it will add no more than 3 ppm to sulfur content of the
finished gasoline.
These requirements are designed to prevent the potential dumping of high sulfur
materials into gasoline under the guise of the addition of gasoline additives. We continue to
believe that all current gasoline additives contribute less than 3 ppm to the sulfur content of the
finished fuel when used at the maximum recommended treatment rate (with 3 ppm being the
extreme). Normal additive production quality control practices already have had to consider the
sulfur contribution of the additive to finished gasoline as a result of the Tier 2 gasoline sulfur
requirements. The maximum recommended treatment rate is already stated on product transfer
document or packaging for the additive. Therefore, the Tier 3 requirements for gasoline
additives will not constrain the use of genuine gasoline additives or result in significant
additional costs to gasoline additive manufacturers. Parties that introduce additives to gasoline at
over 1.0 volume percent are required to satisfy all of the obligations of a refiner and fuel
manufacturer including demonstration that the finished blend meets the applicable sulfur
specification.
We believe that additional controls to limit the combined sulfur contribution for all
additives blended into a batch of gasoline in addition to controlling the sulfur contribution from
5-63
-------
Chapter 5: Tier 3 Fuel Sulfur Program
individual additives are not necessary at this time, would add an unwarranted additional
compliance burden, and could interfere with the use of necessary downstream additives. Certain
additives that provide critical fuel performance characteristics (e.g., corrosion control,
demulsifiers) contain sulfur-containing compounds as an essential functional component. Such
additives are used to remedy specific instances of gasoline quality problems, and their treatment
rate is governed by the desire to limit the added cost from their use.
5.1.4. Leadtime
5.1.4.1. Environmental Benefit of 2017 Start Date
What Commenters Said:
Commenter: Advanced Engine Systems Institute (AESI)
Unfortunately, as occurred during the debate on EPA's Tier 2 regulations, the petroleum industry
is again opposing new requirements to reduce sulfur in transportation fuels. This is happening
despite the significant pollution reduction benefits that would be achieved immediately from the
use of lower sulfur fuel by the existing Tier 2 fleet, and those benefits that would result from
phasing in Tier 3 starting in model year 2017.
Commenter: Union of Concerned Scientists (UCS)
We concur with EPA's assessment that these lower levels are achievable and support the
timelines that EPA has outlined.
Our Response:
As discussed in more detail in Section III of the preamble to the final rule, the Tier 3
program will significantly reduce emissions of VOC, NOx (including NC)2), direct PM2.5, CO, SO2,
and air toxics nationwide, with immediate benefits due to the reduction in gasoline sulfur content
from 30 to 10 ppm starting in 2017. The 10 ppm sulfur standard will provide significant
reductions in harmful emissions independent of the vehicle standards and these reductions are
significant and contribute to the total monetized health benefits. The vehicle standards will also begin
to reduce emissions as the cleaner cars and trucks begin to enter the fleet in model year 2017 and
model year 2018, respectively. The magnitude of reduction will grow as more Tier 3 vehicles enter
the fleet. These reductions will help state and local agencies in their effort to attain and maintain
health-based NAAQS "as expeditiously as practicable." (Sections 172(a)(2) and 181 of the
CAA). The Tier 3 emissions reductions and air quality improvements occur in a timeframe that
is relevant for 2008 ozone nonattainment areas classified as moderate, serious, severe and
extreme, and beginning the program in 2017 is specifically relevant for attainment of the
NAAQS for moderate areas in 2018. Without the Tier 3 program, some areas would have to
adopt other, less cost-effective measures to reduce emissions from other sources under their state
or local authority. In the absence of additional controls, many areas would continue to have
ambient ozone concentrations exceeding the NAAQS in the future. Further, by 2030, the Tier 3
5-64
-------
Tier 3 Summary and Analysis of Comments
gasoline sulfur and vehicle emission standards will result in emissions reductions of 21 percent
for NOx and VOC emissions from on-highway vehicles, and 24 percent for CO emissions.
National emissions of many air toxics from on-highway vehicles will also be reduced by 10 to
nearly 30 percent.
5.1.4.2. Vehicle Need for 2017 Start Date
We received comments in support of and against our January 1, 2017 compliance date.
Commenters that opposed this date argued that it was not needed in light of the implementation
dates for vehicle standards. Commenters in support argued that it had to be synchronized with
the 2017 compliance date for certain light duty vehicles.
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM)
Implementation on January 1, 2017 is not necessary to enabling heavy duty vehicles to meet the
emission standards, for the development of lower cost technologies to improve fuel economy; or
to reduce emissions from the in-use vehicle fleet.
In the proposal, EPA asserts that reducing the sulfur content of gasoline to 10 ppm has three
primary benefits: a) enabling heavy duty vehicles to meet NMOG and NOx standards throughout
their useful life; b) enabling the development of lower cost technologies to improve fuel
economy; and, c) reducing emissions from the in-use vehicle fleet. Each of these claims are
discussed below. None of these claimed benefits justify a January 1, 2017 effective date for the
10 ppm gasoline sulfur standard.
EPA asserts that lowering gasoline sulfur to 10 ppm is necessary to enable larger vehicles and
trucks to reduce the NMOG+NOx low enough to comply with the 30 mg/mi fleet-average
standard vehicle emission standards over the useful life of the vehicles. Even if reducing gasoline
sulfur to 10 ppm does help such vehicles comply with the standard over their useful life, as
EPA's own proposal makes clear, that does not justify January 1, 2017 implementation. The
vehicle standards for vehicles and trucks exceeding 6000 Ibs. GVWR is not even effective until
the 2018 model year. Thus, clearly, January 1, 2017 implementation is not necessary.
EPA proposes to allow vehicle manufacturers to generate early federal credits against the current
Tier 2 Bin 5 requirement in MYs 2015 and 2016 for vehicles under 6,000 Ibs. GVWR and MYs
2016 and 2017 for vehicles greater than 6,000 Ibs. GVWR. These early federal credits can be
used without limitation for MY 2017. In other words, the vehicles are not required to meet the
Tier 3 vehicle emission standards at the start of the program. See 78 Federal Register 29867-68.
As such, it is clearly not necessary to reduce the gasoline sulfur level to 10 ppm on January 1,
2017.
5-65
-------
Chapter 5: Tier 3 Fuel Sulfur Program
Numerous other provisions of the proposed rule support the conclusion that implementation on
January 1, 2017 is not necessary. In particular, the provisions for generation of early credits for
compliance with the vehicle standards, the schedule for phase in of the vehicle standards, the
small refiner/refinery exemption until 2020, the per-gallon gasoline sulfur cap, and the early
credit program for the gasoline sulfur program all demonstrate that EPA is unnecessarily rushing
implementation of this rule by proposing a January 1, 2017 implementation. Furthermore,
because the impact of sulfur on vehicle catalysts is reversible, providing more lead time will not
prevent vehicles from meeting the vehicle emission standards over their useful life. Each of these
issues is discussed below.
In addition to the ability to delay implementation of the vehicle standards through the generation
of early credits, the program for vehicle manufacturers also contains phase-in schedules that
make clear that it is not necessary to implement the 10 ppm gasoline sulfur requirement on
January 1, 2017. As noted above, EPA maintains that reducing gasoline sulfur to 10 ppm is
necessary to help heavy duty vehicles meet the NMOG + NOx standards over their useful life.
Even if that is true, as noted above, the vehicles standards do not even apply to vehicles above
6000 Ibs. GVWR until MY 2018. Moreover, EPA is considering not even requiring vehicle
manufactures to meet the standards until MY 2019. See 78 Federal Register 29876.
Commenter: Johnson Matthey
Adoption of this part of the Tier 3 proposal would bring the U.S. in line with Europe and Japan,
who already have these low sulfur gasoline limits, and places like China who have committed to
bringing in low sulfur gasoline by the end of 2017.
Commenter: Marathon Petroleum Company LP (MFC)
In the proposal, EPA asserts that reducing the sulfur content of gasoline to 10 ppm has three
primary benefits: a) enabling heavy duty vehicles to meet NMOG and NOx standards throughout
their useful life; b) enabling the development of lower cost technologies to improve fuel
economy; and, c) reducing emissions from the in-use vehicle fleet. Each of these claims are
discussed below. None of these claimed benefits justify a January 1, 2017 effective date for the
10 ppm gasoline sulfur standard.
EPA asserts that lowering gasoline sulfur to 10 ppm is necessary to enable larger vehicles and
trucks to reduce the NMOG+NOx low enough to comply with the 30 mg/mi fleet-average
standard vehicle emission standards over the useful life of the vehicles. Even if reducing gasoline
sulfur to 10 ppm does help such vehicles comply with the standard over their useful life, as
EPA's own proposal makes clear, that does not justify January 1, 2017 implementation. The
vehicle standards for vehicles and trucks exceeding 6000 Ibs GVWR is not even effective until
the 2018 model year. Thus, clearly, January 1, 2017 implementation is not necessary.
Numerous other provisions of the proposed rule support the conclusion that implementation on
January 1, 2017 is not necessary. In particular, the provisions for generation of early credits for
compliance with the vehicle standards, the schedule for phase in of the vehicle standards, the
small refiner/refinery exemption until 2020, the per-gallon gasoline sulfur cap, and the early
5-66
-------
Tier 3 Summary and Analysis of Comments
credit program for the gasoline sulfur program all demonstrate that EPA is unnecessarily rushing
implementation of this rule by proposing a January 1, 2017 implementation. Furthermore,
because the impact of sulfur on vehicle catalysts is reversible, providing more lead time will not
prevent vehicles from meeting the vehicle emission standards over their useful life. Each of these
issues is discussed below.
In addition to the ability to delay implementation of the vehicle standards through the generation
of early credits, the program for vehicle manufacturers also contains phase-in schedules that
make clear that it is not necessary to implement the 10 ppm gasoline sulfur requirement on
January 1, 2017. As noted above, EPA maintains that reducing gasoline sulfur to 10 ppm is
necessary to help heavy duty vehicles meet the NMOG + NOx standards over their useful life.
Even if that is true, as noted above, the vehicles standards do not even apply to vehicles above
6000 Ibs GVWR until MY 2018. Moreover, EPA is considering not even requiring vehicle
manufactures to meet the standards until MY 2019. See 78 Federal Register 29876.
EPA proposes to allow vehicle manufacturers to generate early federal credits against the current
Tier 2 Bin 5 requirement in MYs 2015 and 2016 for vehicles under 6,000 Ibs GVWR and MYs
2016 and 2017 for vehicles greater than 6,000 Ibs GVWR. These early federal credits can be
used without limitation for MY 2017. In other words, the vehicles are not required to meet the
Tier 3 vehicle emission standards at the start of the program. See 78 Federal Register 29867-68.
As such, it is clearly not necessary to reduce the gasoline sulfur level to 10 ppm on January 1,
2017.
Numerous aspects of the proposed gasoline sulfur rule make clear that January 1, 2017
implementation is not necessary to enable vehicles to meet the emission standards and that EPA
is unnecessarily rushing implementation of the rule.
Although EPA proposes to make the 10 ppm gasoline sulfur standard generally effective January
1, 2017, EPA proposes to allow small refiners and small refineries to continue to produce
gasoline under the existing Tier 2 rules - i.e., 30 ppm annual average standard/80 ppm per-gallon
cap - until January 1, 2020.
Similarly, EPA proposes to maintain the existing Tier 2 per-gallon cap of 80 ppm, under the Tier
3 rules, at least until 2020. This too demonstrates that EPA does not believe that January 1, 2017
implementation of the 10 ppm sulfur standard is necessary for vehicles to meet the Tier 3
standards.
The early credit provisions under the gasoline sulfur program demonstrate the same thing. As
EPA explains in the proposal, the effect of the early credit program is to delay the
implementation of the 10 ppm annual average standard. We believe the early credit provision is
an important part of the proposal and we support it. However, it does demonstrate that January 1,
2017 of the 10 ppm sulfur standard is not necessary for vehicles to meet the Tier 3 standards.
In all of the above situations, EPA places no restrictions on the use of such Tier 2 gasoline to
prohibit its use in newer vehicles that are required to comply with the new vehicle standards,
demonstrating that January 1, 2017 implementation is not necessary.
5-67
-------
Chapter 5: Tier 3 Fuel Sulfur Program
As explained above, January 1, 2017 implementation of the 10 ppm gasoline sulfur requirement
is not necessary to enable vehicles to meet the NMOG + NOx standards. But, even if it is true as
EPA claims that sulfur will negatively impact catalyst performance on such vehicles, that does
not justify the January 1, 2017 implementation date, because the effect of sulfur on catalysts is
reversible. A new study (described in detail in Section II.B of these comments) demonstrates that
exposure to gasoline fuels containing sulfur levels of 80 ppm sulfur has no lasting impact on the
performance of exhaust emission control systems on modern vehicles operated on 10 ppm sulfur
gasoline.
Commenter: New York State Department of Environmental Conservation
The 2017 gasoline sulfur compliance date for most refineries must not be delayed.
Significant immediate benefits accrue when the gasoline sulfur content is reduced, as sulfur
catalyst poisoning of existing catalytic converters is (at least partly) reversed. Although any large
chemical process construction project is a substantial undertaking, the processes and equipment
necessary to produce low sulfur gasoline are well known. Tier 3 has been under active
discussion, with petroleum industry participation, for several years.
Commenter: Shell Oil Products for Shell and Motiva
Furthermore, as explained in more detail in the API/AFPM comments, a January 1, 2017
implementation date is not necessary to achieve EPA's stated objectives. EPA asserts that the
lower sulfur levels are necessary to help heavy duty vehicles meet the vehicle emission standards
over their useful life, yet those vehicle standards are not even effective until the 2018 model
year, and various other provisions from the phase-in of the vehicle standards to the small
refiner/refinery exemption demonstrate that the January 1, 2017 implementation date is not
necessary.
Commenter: Sierra Club, Clean Air Watch, Respiratory Health Association
Tier 3 Standards complement existing vehicle standards: Finalizing the proposed Tier 3
standards by December 31, 2013 is critical to ensuring that existing vehicle standards are fully
implemented. Automakers are innovating and employing advanced vehicle technologies needed
to meet the historic light-duty vehicle efficiency standards that will increase fleet efficiency to an
average of 54.5 miles per gallon by 2025. In order to fully utilize advanced vehicle technologies,
automakers need lower-sulfur gasoline.
For lower-sulfur gasoline to reach the market in 2017, in time to synchronize with the 2017-2025
light-duty vehicle efficiency standards, the proposed Tier 3 standards must be finalized no later
than December 31, 2013.
Our Response:
5-68
-------
Tier 3 Summary and Analysis of Comments
We disagree with the comments that the January 1, 2017 start date is not necessary given
that the heavy duty emissions standards are effective with the 2018 model year. We also
disagree with the commenters that argue that the phase-in schedule and flexibilities seem to
suggest that there is no need for the gasoline sulfur standards to begin in 2017. As explained in
Section IV of the preamble to the Tier 3 final rule, vehicle standards begin to phase in for
vehicles less than 60001bs GWVR with the 2017 model year. As also explained in Section IV of
the preamble to the final rule, such vehicle manufacturers may comply with today's standards by
using early credits but that does not mean they are not required to meet the standards in 2017.
Thus, we expect that the vehicle standards will begin to reduce emissions as the cleaner cars the
fleet in model year 2017. As also previously discussed, immediate benefits will result in 2017, as
the average sulfur standard of 10 ppm will facilitate emission reductions from vehicles that are
on the road today. Thus, as explained in Section III of the preamble, and as some of the
commenters correctly noted, the Tier 3 sulfur standards will reduce emissions from the on-road fleet
immediately upon implementation in calendar year 2017.
We also disagree with the comments that the January 1, 2017 implementation date is
unnecessary given that sulfur impacts on catalyst is reversible. As discussed in more detail in
Chapter 4.1.4.3 of this document and as explained in Section IV.A.6 of the preamble, operating a
catalyst at a sufficiently high temperature under net reducing conditions (e.g., air-to-fuel equivalence
that is net fuel-rich of stoichiometry) can effectively release the sulfur oxides from the catalyst
components. Thus, regular operation at sufficiently high temperatures at rich air-to-fuel ratios can
minimize the effects of fuel sulfur levels on catalyst active materials and catalyst efficiency. It
cannot, however, completely eliminate the effects of sulfur poisoning. In current vehicles,
desulfurization conditions occur typically at high loads when there is a degree of commanded
enrichment (i.e., fuel enrichment commanded by the engine management system primarily for
protection of engine and/or exhaust system components). A study of Tier 2 vehicles in the in-use
fleet recently completed by EPA shows that emission levels immediately following high speed/load
operation is still a function of fuel sulfur level, suggesting that lower gasoline sulfur levels will bring
emission benefits unachievable by catalyst desulfurization procedures alone. For example, if a
vehicle operates on gasoline with less than 10 ppm sulfur, exhaust emissions stabilize over repeat
FTP tests at emissions near those of the first FTP that follows the high speed/load operation and
catalyst desulfurization. Therefore, continued operation on gasoline with 10 ppm average sulfur
content or lower is necessary after catalyst desulfurization in order to achieve emissions reductions
with the current in-use fleet. We also explained that to the extent such conditions do not occur
today as part of normal driving for vehicles in the in-use fleet, there is no practical way to
modify such existing vehicles to do so. Thus, reducing fuel sulfur levels has been the primary
regulatory mechanism to minimize sulfur contamination of the catalyst and ensure optimum
emissions performance over the useful life of a vehicle and is the only effective means for the in-
use fleet.
For Tier 3 vehicles, there are several reasons why regular operation at the high exhaust
temperatures and rich air-to-fuel ratios necessary for catalyst desulfurization is not desirable and may
not be possible. The temperatures necessary to release sulfur oxides are high enough to lead to
thermal degradation of the catalyst over time through thermal sintering of active materials. Sintering
reduces the surface area available to participate in reactions. Additionally, it is not always possible to
maintain these catalyst temperatures (because of cold weather, idle conditions, light load operation).
Also, the air-to-fuel ratios necessary for sulfur removal can result in increased PM, NMOG and CO
emissions.
5-69
-------
Chapter 5: Tier 3 Fuel Sulfur Program
While the fuel ABT provisions and small refiner/small volume refinery provisions
already allow this to some degree - in order to provide a smooth phase-in of the 10 ppm average
sulfur standard - delaying the standard would exacerbate this further. Since, as discussed in
Section V.L of the preamble to the final rule, we believe there is sufficient lead time provided to
refiners, there seems to be no reason to delay implementation of the fuel standards further.
The Tier 3 light-duty vehicle emissions standards will phase in beginning with model
year (MY) 2017. This will enable auto manufacturers to be able to design a 50-state fleet of
vehicles as a 2017 start date will harmonize the Federal standards with the California LEV III
program. Further, we also have designed the Tier 3 program to be implemented in the same
timeframe as the GHG emissions and fuel economy standards for model years 2017-2025. The
2017 start date of the Tier 3 program addresses interactions with the 2017 LD GHG rule in a
manner that aligns implementation of the two programs, to achieve significant criteria pollutant
and GHG emissions reductions while providing regulatory certainty and compliance efficiency.
Thus, vehicle manufacturers will be able to design new vehicle platforms for compliance with
both the GHG and Tier 3 standards at the same time.
Delaying the program to 2018 would not only negate the benefits of harmonizing with
LEV III and the LD GHG program, it would also delay important health benefits that come from
the systems approach of beginning the Tier 3 vehicle and fuel standards in tandem. The vehicle
and fuel ABT programs, small refiner/small refinery provisions, and vehicle phase-in allow for
the Tier 3 program to begin in 2017, while offering manufacturers and refiners the ability to
spread out their compliance costs in the manner that best works for their specific situations.
5.1.4.3. Refinery Lead Time
5.1.4.3.1. Overall Lead Time Required
What Commenters Said:
Commenter: American Fuel & Petrochemical Manufacturers (AFPM)
Organization Type: Fuels Industry
We believe that there should be at least five years between the promulgation of a Tier 3 rule that
includes reduction in the sulfur content, gasoline, and its effective date. Refiners need this lead
time to schedule equipment modifications or the installation of new equipment during a
turnaround. Otherwise refiners may be temporarily shut down with the associated supply
impacts.
Tier 3 is not required by law, and the selection of 2017 is arbitrary. If the EPA promulgates a
reduction in the sulfur content of gasoline in late 2013, then the effective date should not be
earlier than 2019.
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
5-70
-------
Tier 3 Summary and Analysis of Comments
Manufacturers (AFPM)
Rather than the 3 years EPA proposes, if EPA decides to proceed with the Tier 3 rulemaking,
EPA should provide 5 years of lead time from the date of publication of the final rule in the
federal register in order to implement any changes to the average gasoline sulfur requirement.
Five year lead time from publication of Final Rule is required for implementation.
EPA asserts that 3 years of lead time is more than sufficient for the petroleum industry to
implement the 10 ppm gasoline sulfur requirement. 78 Federal Register 29923. We disagree.
EPA needs to provide 5 years of lead time. As discussed above, rushing implementation of the
gasoline sulfur requirement is not necessary. Rushing implementation will unnecessarily raise
the cost to implement the program. Rushing implementation undermines the ability of the
industry to adequately plan projects, secure contractors and equipment at competitive rates,
optimize solutions, and align construction projects with existing maintenance turn-around
schedules. In addition, air permits will be necessary at many refineries, and EPA has been
challenged in issuing permits in a timely fashion (as is discussed in further detail in section VII.F
of these comments.) All of these factors will tend to increase compliance costs.
In the proposal, EPA puts forth Table V-3 with a timeline that the Agency believes illustrates the
time needed to implement the 10 ppm gasoline sulfur requirement at a refinery. 78 Federal
Register 29925. We disagree with EPA's suggested timeline. We believe that a more accurate
amount of time needed to implement major new rules like the 10 ppm gasoline sulfur
requirement is at least five years. The Tier 2 rules were finalized in February 2000, with
compliance required to phase in from 2004-06 (with compliance flexibilities that had a more
meaningful impact, as discussed below). Ultra Low Sulfur Diesel (ULSD) rules were finalized in
January 2001; with a compliance phase-in period requiring that 15 ppm sulfur diesel constitute
80% of the highway pool in mid-2006. Compliance lead times need to be longer than these two
significant rulemakings to reduce sulfur because of new GHG requirements at refiners, and the
increased reliability of refining equipment has extended refinery turn-around time to
approximately 5 years. These recent changes for refiners add complexities that warrant
additional Tier 3 compliance lead time.
In the proposal, EPA suggests that 3 years of lead time is feasible, in part, because EPA began
talking to refiners about the possibility of regulating gasoline sulfur a couple of years ago. This is
inappropriate because EPA decisions concerning key aspects of this rule are not yet final. EPA
understands that there can be many changes made from the time a rule is proposed until it is
finalized in response to comments submitted by the public, and that therefore there is too much
uncertainty to make detailed implementation plans until the final rule is issued. In this case, for
example, we do not yet know whether EPA will actually require a reduction in gasoline sulfur
given the high costs and de minimis environmental benefits, we do not know when the rule will
be effective, we do not know whether EPA will leave the per-gallon cap at 80 ppm, etc.
EPA's three-year lead time is grossly inadequate. Six years is workable. Five years is the
absolute minimum. Three years is insufficient to ensure against market disruptions. Refiners are
able to operate for about five years between turnarounds, which is the most efficient time to
5-71
-------
Chapter 5: Tier 3 Fuel Sulfur Program
make major changes. A three-year lead time means that normal maintenance schedules are
likely to be disrupted, potentially impacting gasoline supply and increasing the overall cost.
Commenter: BP Products North America Inc.
Achieving the proposed January 1, 2017 effective date raises major concerns across BP. The
availability of sufficient sulfur credits under the proposed Averaging, Banking and Trading
program also raises major concerns. Even the availability of credits from 2014 will not be known
until early 2015, less than 2 years from EPA's proposed effective date. Completing the necessary
engineering, permitting, procurement, construction and startup by January 1, 2017 is highly
unlikely. Pursuing completion of the investments on a compressed schedule also leads to
increased costs. In order to complete the necessary investments and minimize unnecessary costs,
BP proposes that EPA provide five years of lead time from the date the final rule is published.
BP also agrees with API and AFPM's conclusion that the 10 ppm annual average gasoline sulfur
standard is not necessary on January 1, 2017 as asserted by EPA.
Commenter: Chevron Products Company
We are opposed to the implementation schedule proposed by the EPA. The 2017 compliance
deadline does not provide sufficient time to make necessary refinery modifications and would
significantly increase the cost of the required upgrades. If the agency proceeds with the gasoline
sulfur reduction requirement, the program start date should be no sooner than five years after the
date of the final regulation.
We do not agree with the program start date of January 1, 2017 as this date only provides 3 years
for implementation, assuming the rule is finalized by the end of 2013. This is insufficient for a
refinery that is required to invest and chooses to do so by the start of the program. By
comparison, the Tier 2 program was finalized in 1999, with implementation phased in from 2004
to 2006. The Tier 3 program start date should be extended to allow a minimum of five years for
implementation. We would propose a start date of January 1, 2019, assuming that the Tier 3
regulations are finalized by January 1, 2014.
We agree that the technology for sulfur reduction to the 10 ppm level is available and has been
demonstrated at certain refineries in operation today. However, we believe that EPA has
underestimated the cost and complexity of the changes which will be required in the refining
industry to achieve this reduction.
The time required for permitting is becoming longer due to local regulations and GHG programs.
We do not agree that the analysis of permit times should be limited to the estimates provided by
OAQPS. We are seeing other government agencies placing additional requirements on refinery
permits for a variety of GHG, safety, and environmental concerns, all of which increase the time
required to obtain a permit. California refineries producing gasoline subject to the Tier 3
regulations may encounter even longer permitting times and increased scrutiny, taking them well
beyond the proposed implementation period.
Commenter: ExxonMobil
5-72
-------
Tier 3 Summary and Analysis of Comments
If the Agency decides to proceed with this rulemaking, we recommend that EPA provide a five
year lead time from the publication date of the final rule in the Federal Register before
implementation of a new fuel standard;
Commenter: Independent Fuel Terminal Operators Association (IFTOA)
The Association recognizes that EPA is proposing to require a significant reduction in the sulfur
content of gasoline, beginning in 2017, as part of a systems approach to addressing the impacts
of motor vehicles and fuels on air quality and public health. However, Members have concerns
that the proposed rule is very aggressive and far more burdensome and costly than the Agency
envisions. Therefore, the Association's recommendations set out below would more properly
balance the goal of better air quality and the costs and burdens associated with implementation.
If a better balance is achieved, the benefits to the consuming public would be enhanced.
A primary concern of Association Members is the proposed January 1, 2017 implementation date
for the Tier 3 annual average sulfur standard of 10 ppm. Assuming that the final Tier 3 rule is
issued by EPA at the end of 2013, refiners and importers would have only three years in which to
comply. Three years is simply insufficient to make such major modifications.
Fourth, approval for funding for such major changes is a time-consuming and often difficult
process. The estimated schedule put forth by EPA does not appear to include the time needed to
obtain or allocate such funding.
The EPA appears to assume that there would be ample early credits available to assist refiners
(provide flexibility) in meeting the three-year implementation date by deferring investments. To
date, there is insufficient information available to EPA and the market about the volume of early
credits that would or could be generated before the implementation date. Therefore, industry
could not, during the planning stages, rely on the fact that there would be an adequate supply of
such credits.
Recommendation: Accordingly, the Association recommends that EPA amend its proposed rule
and make the annual average sulfur standard of 10 ppm effective no earlier than January 1, 2019.
Delay implementation of the proposed Tier 3 standard until at least 2019 to give the regulated
community sufficient time to make necessary refinery changes in an efficient and cost-effective
manner;
Commenter: Irving Oil Terminals Inc.
Irving Oil commends EPA's efforts to improve air quality through the reduction of sulfur content
in motor gasoline; however, as outlined in our attached comments, the proposed 2017 Effective
Date causes great concern because the 10 ppm annual average sulfur content standard cannot be
achieved as quickly as EPA desires. Typically, the modifications required to comply with the
proposed standard would be conducted during a five-year turnaround cycle at a refinery during
which all major renovations, new units and reconfigurations are implemented. A three-year
implementation period will increase operational risk to refiners, create inefficiencies, increase
5-73
-------
Chapter 5: Tier 3 Fuel Sulfur Program
expense for many entities and increase the risk of supply disruptions. Industry needs adequate
time to plan, finance and implement the significant changes required to meet the proposed Tier 3
standards. A 2019 Effective Date will provide the minimal five-year timeframe for these
complex and costly adjustments to be implemented.
Moreover, when the EPA initiated discussions on the Tier 3 rule with the regulated community
in 2011, it anticipated that the final rule would be issued by 2012, and provided refiners with a
five-year timeframe to make the necessary adjustments by creating a 2017 Effective Date. Delay
in issuing the final rule —expected now to be by the end of 2013 — warrants an extension of the
Effective Date to at least 2019 — thereby preserving the original five-year implementation
schedule. This proposed timeline is consistent with the previous implementation of both the Tier
2 and Ultra Low Sulfur Diesel (ULSD) rules.
Even though the proposed rule has been highly anticipated by the regulated community, design
approval for major capital investments can only proceed with 'regulatory clarity' which will not
exist until the final rule is issued.
Accordingly, adoption of the 2019 Effective Date for the 10 ppm annual average sulfur standard,
along with incentives encouraging the generation of early credits at the earliest possible date,
will allow industry to achieve compliance in the most efficient and cost-effective manner while
meeting EPA's objectives for cleaner air.
Irving Oil fully supports the Agency's desire to reduce sulfur in motor gasoline as a means of
minimizing air pollution as soon as possible. However, as discussed below, the proposed
effective date— 2017 — will not provide the regulated community with sufficient time to
implement the changes required and could result in substantial supply chain challenges, supply
disruptions and unnecessary additional costs. Ensuring compliance with the proposed change in
sulfur content is a significant undertaking. Accordingly, the EPA should make the effective date,
January 1, 2019. Assuming the final rule will be issued at the end of 2013, this revised effective
date would reduce business risk to refiners, allowing time to make the necessary adjustments in
an efficient and cost-effective manner while continuing to produce adequate supplies of motor
gasoline to meet consumer demand.
In the Preamble that accompanied the Proposed Rule, EPA recognizes that gasoline refining is a
complex operation, and refiners would have to take many steps to comply with the proposed Tier
3 sulfur standards. These steps include scoping studies, financing, process design for new or
revamped refinery units or subunits, permitting, detailed engineering based on the process
design, field construction of the sulfur reduction units, and start-up and shakedown of the newly-
installed desulfurization equipment. The Agency states that it conducted a lead time analysis and
spoke with refiners to determine the amount of time that a typical refiner would take to complete
renovations and meet the 10 ppm sulfur standard. EPA has proposed that a lead time of three
years is sufficient. The details of the Agency analysis are set out in Table V-3 — 'Anticipated
Compliance Timelines.'
Irving Oil has reviewed EPA's information on 'lead times' and believes that it is overly
optimistic. Also, some of the data on which the determination was made are now out of date due
5-74
-------
Tier 3 Summary and Analysis of Comments
to major changes that are occurring in the petroleum industry. The following are several
significant reasons why a three-year lead time is insufficient:
1. Those companies that license the technology to desulfurize motor gasoline have limited
resources and likely would be unable to support many simultaneous scoping studies and process
designs. Thus, some refiners would not be able to begin these activities as early as needed to
support a three-year lead time.
2. Irving Oil agrees with EPA's assessment that most refiners would revamp existing FCC
gasoline post-treaters rather than constructing grassroots units. However, Irving Oil has often
found that revamping units, particularly during the early project phases (the scoping study and
process design) is more complex and time consuming than building and installing an entirely
new, grassroots unit. Therefore, EPA's assumption that refiners can revamp existing units more
rapidly is not supported by Irving Oil's many years of experience.
3. EPA has indicated that some of the initial phases of the revamping project can overlap,
thereby saving additional time. Again, this assumption is contrary to Irving Oil's experience and
belief. Irving Oil has found that the process design phase cannot be started until after the scoping
study is complete. Thus, it is more reasonable to assume that the time needed to make the
appropriate changes to the refinery would be the same for both a revamping of units and the
installation of new units.
4. The EPA believes that there would be sufficient resources (technology licensors, engineering
and construction services, skilled laborers, and equipment) available to the refining industry.
However, EPA must recognize that the U.S is undergoing a domestic petroleum renaissance, and
demand for these resources in new areas of production such as North Dakota, Montana, and
Texas is great. Those projects would compete for resources directly with refiners trying to meet
the EPA three-year time schedule. Shortages of these resources would, therefore, likely inhibit
the ability of refiners to comply in a timely manner.
5. While the early credit provision of the Tier 3 program may provide refiners with some
flexibility to meet the proposed January 1 2017 compliance deadline, the number of such credits
available on the market cannot be predicted with any certainty. Refiners would carefully balance
the costs associated with the production of lower sulfur motor gasoline in 2014 to 2016 with the
value derived from selling such early credits. Thus, EPA and the industry cannot assume that
early credits would provide refiners with enough assistance to comply with the effective dates as
proposed.
Recommendations: Change Effective Date to 2019. Irving Oil understands EPA's desire to
reduce air pollution by requiring the use of lower sulfur fuels. However, it is essential for the
EPA to promulgate this new standard in a manner and timeframe that does not disadvantage
members of the regulated community and provides sufficient time for refiners to plan and
execute capital investments efficiently.
Therefore, Irving Oil strongly recommends that EPA change the 'effective date' of the proposed
Tier 3 rule to 2019 — thereby providing at least five years for implementation of this complex
and costly program. In this way, EPA would achieve its clean air objectives and minimize
potential supply disruptions and costs associated with compliance that would ultimately be borne
by consumers. Adoption of 2019 as the effective date and extension of early credits from 2013
until three years following the effective date would properly balance these important objectives.
5-75
-------
Chapter 5: Tier 3 Fuel Sulfur Program
Commenter: Marathon Petroleum Company LP (MFC)
The Agency must extend the implementation date until January 1, 2019 or five years after the
final rule is issued.
EPA asserts that it is necessary to implement the 10 ppm annual average gasoline sulfur standard
on January 1, 2017 to help heavy duty vehicles meet the Tier 3 vehicle emission standards
throughout their useful life, to enable new vehicle technologies to improve fuel efficiency, and to
reduce emissions from the current vehicle fleet. EPA also asserts that 3 years of lead time
(assuming this rule is finalized by January 1, 2014) provides more than sufficient lead time for
the refining industry to make the changes necessary to reduce gasoline sulfur levels to 10 ppm on
average. However, as explained in more detail below, implementation on January 1, 2017 is not
necessary, and does not provide sufficient lead time. Rather than the 3 years EPA proposes, if
EPA decides to proceed with the Tier 3 rulemaking, EPA should provide 5 years of lead time to
implement any changes to the average gasoline sulfur requirement.
EPA asserts that 3 years of lead time is more than sufficient for the petroleum industry to
implement the 10 ppm gasoline sulfur requirement. 78 Federal Register 29923. We disagree. We
suggest that EPA provide 5 years of lead time. As discussed above, rushing implementation of
the gasoline sulfur requirement is not necessary. Not only is it unnecessary, rushing
implementation will unnecessarily raise the cost to implement the program. Rushing
implementation undermines the ability of the industry to adequately plan projects, secure
contractors and equipment at competitive rates, optimize solutions, and align construction
projects with existing maintenance turn-around schedules. All of these factors will tend to
increase compliance costs.
In the proposal, EPA puts forth Table V-3 with a timeline that the Agency believes illustrates the
time needed to implement the 10 ppm gasoline sulfur requirement at a refinery. 78 Federal
Register 29925. We disagree with EPA's suggested timeline. We believe that a more accurate
amount of time needed to implement major new rules like the 10 ppm gasoline sulfur
requirement is at least five years. The Tier 2 rules were finalized in February 2000, with
compliance required to phase in from 2004-06 (with compliance flexibilities that had a more
meaningful impact, as discussed below). Ultra Low Sulfur Diesel (ULSD) rules were finalized in
January 2001; with a compliance phase-in period requiring that 15 ppm sulfur diesel constitute
80% of the highway pool in mid-2006. Compliance lead times need to be longer than these two
significant rulemakings to reduce sulfur because of new GHG requirements at refiners, and the
increased reliability of refining equipment has extended refinery turn-around time to
approximately 5 years. These recent changes for refiners add complexities that warrant
additional Tier 3 compliance lead time.
In the proposal, EPA suggests that 3 years of lead time is feasible, in part, because EPA began
talking to refiners about the possibility of regulating gasoline sulfur a couple of years ago. This is
inappropriate because EPA decisions concerning key aspects of this rule have not already been
decided before the Agency even issued this notice of proposed rulemaking. EPA understands that
there can be many changes made from the time a rule is proposed until it is finalized in response
to comments submitted by the public, and that therefore there is too much uncertainty to make
5-76
-------
Tier 3 Summary and Analysis of Comments
detailed implementation plans until the final rule is issued. In this case, for example, we do not
yet know whether EPA will actually require a reduction in gasoline sulfur given the high costs
and de minimis environmental benefits, we do not know when the rule will be effective, we do
not know whether EPA will leave the per-gallon cap at 80 ppm, etc. Based on this, it is
unreasonable for EPA to assert that refiners had sufficient notice of this rule's requirements two
years before it was even proposed to begin implementing the rule. In this rulemaking, as with all
rules, EPA must consider whether lead time is adequate from the date of promulgation of a final
rule, not from the time that EPA starts thinking about possibly issuing a proposed rule.
Commenter: Michigan Department of Environmental Quality (MDEQ)
The deadline proposed should be more than enough time given that the technology and methods
are already available.
Commenter: Mid-Atlantic Regional Air Management Association Inc. (MARAMA)
MARAMA states recognize that the proposed rules may increase costs for some refineries
located in this region, and raise the cost of gasoline. MARAMA supports EPA's efforts to
mitigate those costs by providing time to install needed equipment at refineries.
Commenter: Monroe Energy, LLC
In addition, the controls will take at least 2 years to design and engineer and will necessitate a
shutdown of the refinery for tie-in. Like most refineries in the United States, Monroe operates on
a 5 year turnaround cycle. Monroe's next turnaround is scheduled for 2018. Therefore, in order
to comply in a timely manner with the Tier 3 gasoline sulfur standard, Monroe would be forced
to take an out of sequence maintenance turnaround, shutting off its production and costing the
refinery millions of dollars of lost operating revenues, or it will be forced to buy credits from
other refiners in order to comply. EPA's rules should provide sufficient lead time so that refiners
have time to install controls and are not required to rely on purchasing credits from other refiners
in order to comply. Therefore, the rule should give refiners no less than six years to comply from
the time the rule is issued to ensure that all refiners have adequate lead time to design, engineer,
and install controls without taking unscheduled or out of sequence turnarounds or buying credits
from other refiners.
Commenter: New York State Department of Environmental Conservation
EPA estimates (Chapter 4 of the RIA) that three years will be necessary for the design,
permitting, construction and commissioning of grass roots Fluid Catalytic Cracking (FCC)
gasoline hydrotreaters. Revamps are expected to require two years. The Tier 3 proposal provides
more than three years lead time, six for smaller refineries. Even if there are cases where three
years are insufficient, the Averaging, Banking, and Trading provisions of the proposed rule
provide additional compliance flexibility.
5-77
-------
Chapter 5: Tier 3 Fuel Sulfur Program
While industry-wide compliance significantly earlier than 2017 may be difficult to achieve, the
proposed compliance dates provide ample time for industry to make the changes needed to
reduce gasoline sulfur content, and should be promulgated unchanged.
Commenter: PBF Energy Inc.
PBF conducted a preliminary facility review to evaluate the modifications needed to comply
with the new proposed fuel standards. We found that all three PBF refineries will require
significant investments to comply with the proposed new regulation to meet the 10 ppm annual
average gasoline sulfur standard.
Additionally, PBF is concerned that the timing of this regulation when coupled with the impact
of other fuels regulatory programs RFS2, Ozone NAAQs, and Northeast States switching to 15
ppm sulfur home heating oil places a significant regulatory related economic burden at the same
time the EPA is proposing to enact the new Tier3 gasoline sulfur standards. The Northeast
gasoline market is highly competitive and the imposition of the added regulatory compliance
costs associated with each program will have a significant impact on business outlook, regional
investment decisions, and PBF's competitiveness versus imported gasoline products.
PBF requests that the EPA review the timing of all regulatory impacts and defer the
implementation of the Tier 3 gasoline sulfur standard until it is actually needed by the new Tier 3
motor vehicles being rolled out with the most stringent fuel requirements at the end of the
vehicle early credit program.
Commenter: Phillips 66 Company
Implementation date - The January 1, 2017 implementation date does not provide sufficient lead
time for refineries to complete needed capital projects.
The API and AFPM comments provide detailed analysis of EPA's justification of the proposed
10 ppm average sulfur standard. These comments address the areas of the technical need for a 10
ppm sulfur level, the environmental impact, and the cost-benefit analysis. Phillips 66 participated
in the development of these comments, agrees with the content, and questions the justification
for the revised sulfur standards. We are not specifically addressing these aspects of the
rulemaking in our comments but are relying on the API/AFPM comments to convey our views
on these issues.
Implementation Date - January 1, 2017 does not provide sufficient lead time for refineries to
complete projects.
The proposed implementation date of January 1, 2017 does not provide adequate lead time for
refineries to implement needed capital projects. EPA has proposed this aggressive timeline and
uses two basic arguments to support their proposal. Firstly, EPA predicts that most projects will
be of moderate scope and therefore refiners can complete needed capital projects within a 3-year
time frame. Based on our initial analysis, we expect our company (and likely industry) will be
faced with projects of significant scope requiring more than 3 years to complete.
5-78
-------
Tier 3 Summary and Analysis of Comments
Project Scope: EPA provided two project timeline estimates in the proposed rule - one for unit
revamps (24 months) and one for new grass roots units (36 months). We believe EPA has
underestimated the time requirement for every phase of a project - whether a revamp or a new
unit.
In its investigations, EPA concludes that most refineries will require only projects of moderate
revamp scope and that a relatively few will require grass roots desulfurization investments. The
DRIA is quoted, "For companies that already have an FCC naphtha postreater we assumed that
all that would be necessary to meet the proposed sulfur standards was to revamp their existing
FCC postreating units." Also in its analysis, EPA uses a PADD average approach which may
neglect individual refinery configurations.
To reduce sulfur in gasoline, it is common for refiners to separate the FCC naphtha stream into
several streams with differing characteristics requiring different treatment technologies. Some of
these separated streams, but not all, may currently be undergoing sulfur reduction. The likely
approach to achieving Tier 3 will be to treat the remaining FCC streams plus additional light
streams from other process units. Due to the differences in characterization, the in-place
technology likely will not be appropriate to treat the remaining untreated stream(s) required for
Tier 3 and therefore revamp(s) would not be chosen. For example, heavy FCC naphtha contains
a significantly higher amount of sulfur and uses a technology that is not appropriate for light
FCC naphtha. In cases where a refinery is currently only treating the heavy FCC naphtha,
refiners will require new grass roots units to treat other portions of the FCC naphtha, tailored to
the individual characteristics of the untreated streams.
These extensive projects and the fact that industry will be competing for many of the same
engineering, procurement and field resources to complete them, requires significantly more time
than EPA is providing in its proposal.
Project time requirements: Our experience with both grassroots projects and significant revamps
of this magnitude is that they require well in excess of three years to complete. As such, EPA
should extend the compliance date to coincide with 5 years from issuance of the final rule.
Below are two projects time lines, for a grassroots and revamp, provided by Phillips 66 internal
project management staff. As shown the grass roots projects will need almost 5 years to
complete. By comparison, the revamp project can be completed in only 6 months less time. As
such, we believe that EPA has underestimated the time requirement and uncertainty for every
phase of a project.
In particular, a critical phase of any project is in the initial scoping. It is imperative that refineries
take the requisite time needed to fully evaluate various technology options and operating
configurations. It may be necessary to incorporate other facets into the project, such as
alleviating existing bottlenecks to allow full operational optimization and flexibility with any
new or revamped units. Technology vendors continue to work on improvements especially in the
area of catalyst technology. It is a very time consuming effort to work with multiple technology
vendors to evaluate various feed streams with their technologies and then do economic
evaluations on the different options. In order to meet a January 1, 2017 implementation date, a
5-79
-------
Chapter 5: Tier 3 Fuel Sulfur Program
refinery would need to have already completed the project scoping phase, well in advance of the
final rule. Since there is uncertainty in when the final rule will be issued, and in the provisions it
will contain, refiners cannot set a final design basis. They therefore cannot complete scoping
until the final rule is issued.
The industry is requesting a lead time of 5 years from final rule issue, should EPA proceed with
a modification of the standard. This timing would provide sufficient lead time and would fit
better with the 5-year refinery turnaround cycles.
The proposed changes in gasoline sulfur level represent a significant change and one that will
require wide scale refinery investment. Although EPA did initially start talking about potential
changes to the gasoline sulfur standards almost two years ago, it did not promulgate a notice of
proposed rulemaking until May 21, 2013. EPA has proposed a very aggressive implementation
date of January 1, 2017 for the proposed new standard to take effect. EPAs justification for this
short implementation timing is a combination of estimated project timelines and ability to
generate early credits to help forestall capital investment. Phillips 66 believes that the Agency
has been overly optimistic in both their estimate of required time for various project phases as
well as the refining industry's ability to generate early credits. Refineries that require
modifications be complete in order to meet the new standard need more time than what is
proposed to complete their projects. Existing units are limited in their capability to further reduce
sulfur content and the collective cost to generate early credits may be very high (lost octane,
shortened run length that would necessitate additional shut down, etc.). Phillips 66, along with
others in the industry, is requesting EPA provide a 5 year lead time to allow needed project
completion.
Commenter: Shell Oil Products for Shell and Motiva
EPA is proposing to make the 10 ppm annual average gasoline sulfur standard effective on
January 1, 2017. Assuming that EPA issues the final rule by the end of 2013, that leaves only
three years of lead time. We urge EPA to provide additional lead time. Three years is simply not
enough time to allow obligated parties to adequately plan and implement the changes that will be
necessary to reduce sulfur from the current 30 ppm annual average standard to the 10 ppm
standard. It also does not allow refiners to optimize their plans and to integrate the necessary
changes into existing turn-around and maintenance plans. The result of such a rushed
implementation is likely to be higher compliance costs, which could translate in to higher prices
for consumers.
Our Response:
We received several requests for extension of time from 3years to Syears. These requests
justified the additional time on grounds of compliance costs and state/federal permitting
requirements resulting from new regulations such as the GHG rule. Several commenters also
stated that we had underestimated the project time lines for revamps and installation of
grassroots units.
5-80
-------
Tier 3 Summary and Analysis of Comments
We disagree with the comments that a 3 year lead time is insufficient for refiners to comply with
today's sulfur standards. We acknowledge that an adequate amount of lead time is required for
the implementation of any rulemaking. Depending on the level of effort required to comply,
more or less lead time is also required. Section 21 l(c) of the Act is silent on the compliance date
for standards promulgated there under. We are therefore exercising our discretion in setting an
appropriate compliance period for today's standards. In doing so we have considered various
factors such as the time needed to select the technology and the vendor that will provide the
technology with which they will comply with the fuels standard; time needed to arrange an
engineering and construction (E & C) contractor which will design and oversee the construction
of the refinery unit and the time needed to obtain the necessary permits and procure the
necessary hardware. Further, we have considered the time needed to construct the unit and the
time needed to make the necessary unit tie-ins of the unit with the rest of the refinery and then
startup the unit. In response to these time considerations, we have provided a generous ABT
program as well as provisions allowing for the phase-in of sulfur standards beginning in 2017
and through 2019. As also previously explained, EPA is also adopting Tier 3 emissions standards
for motor vehicles under section 202. These vehicle standards begin to phase-in for vehicles less
than 60001bs GWVR with the 2017 model year. As discussed in Section IV.A.6 of the preamble,
for new vehicles, the use of 10 ppm average sulfur fuel is an essential part of achieving Tier 3 levels
while applying an array of advancements in emissions control technology to the light-duty fleet.
The commenters supporting more lead time, focused solely on the January 1, 2017 start
date and failed to take into account that the standard is actually phased-in through the flexibilities
provided through 2019. Focusing solely on the January 1, 2017 start data for larger refineries
does not reflect the significant flexibilities provided by the Tier 3 averaging, banking, and
trading provisions, the delayed start for small refiners and small volume refineries, the ability to
carry over credits from Tier 2, and the availability of hardship relief as a failsafe. Were the
program to have been designed without these flexibilities, then their comments would have been
more relevant. With these flexibilities, the final Tier 3 program provides ample lead time.
As discussed in Section V.D of the preamble to the final rule, we are finalizing an
averaging, banking and trading (ABT) program that would significantly help refiners comply
with the January 1, 2017 start date. There are three provisions of the ABT program which help
with respect to leadtime and we have concluded that when these ABT provisions are taken into
account, refiners will essentially have the equivalent of 6 years of leadtime. The ABT program
allows for ongoing intra-company and inter-company trading nationwide. This will allow some
refineries to over-comply with the 10 ppm gasoline sulfur standard (in our analysis, we modeled
these refineries bringing their gasoline down to 5 ppm), allowing other refineries that would
otherwise need to install grassroots units to not invest and purchase credits instead. This aspect
of the ABT program is very important because our analysis estimates that only one refinery
would need to install a grassroots hydrotreater whereas without the ABT provisions, there could
be as many as 20 grassroots units. This one aspect has important implications for leadtime
because as discussed in the above, revamps require two years or less whereas grassroots FCC
postreater units require approximately three years to install. We are convinced that this aspect of
the ABT program will be utilized to the maximum extent possible because refineries revamping
their postreaters in lieu of installing grassroots postreaters results in the most cost-effective
mechanism for meeting the 10 ppm annual average standard.
5-81
-------
Chapter 5: Tier 3 Fuel Sulfur Program
An important question is whether refiners will not invest in a grassroots unit trusting that
the credits will be freely available. For the NPRM, we conservatively assumed that refiners
would only rely on credits if they could generate them internal to the company. As discussed in
Section V.L. of the preamble and Chapter 4.4 of the RIA, we assessed how the sulfur credits
were being traded under Tier 2 and we found that over half the sulfur credits were freely traded
between companies (as opposed to only being used within companies), and many single-refinery
companies had sulfur levels above 30 ppm (single-refinery companies must purchase credits
from other companies). Because we set up the Tier 3 credit trading program to work just like the
Tier 2 credit trading program, we are confident that there will be widespread trading within and
between refining companies which means that few grassroots units will be need to be built for
Tier3.
A second aspect of the ABT program that helps with leadtime is the provision for
generating early sulfur credits and banking them for later use. This provision allows refineries to
reduce their gasoline sulfur to less than 30 ppm prior to January 1, 2017 and bank the credits for
later use. Based on comments that we received on the proposed rule, we are allowing Tier 2
credits which are generated during the years 2012 and 2013 to also be used to show compliance
for Tier 3. This effectively extends the early credit generation period for Tier 3 to encompass the
years 2012 to 2016, which is 5 years. Analyzing the 2012 gasoline quality data that refiners
reported to EPA, we found that gasoline sulfur levels in the U.S. averaged 26.7 ppm. Thus,
refiners have already begun overcomplying with Tier 2 by 3.3 ppm, and are therefore already
generating early credits for Tier 3. If refiners do nothing more but continue to overcomply with
Tier 2 by 3.3 ppm over the 5 years of early credit generation, refiners will have generated enough
credits to delay the completion of their capital projects by more than one year. Furthermore
since those credits generated in 2012 and 2013 will expire in 2017 and 2018 respectively,
refiners will have an incentive to either use them themselves or trade them in 2017 and 2018.
Thus refiners that may need to count on them to delay their capital investment are likely to be
able to have access to them.
We believe that refiners will generate a lot more early credits with their existing gasoline
sulfur control units than the 3.3 ppm we observed in 2012. As we discussed in our cost analysis,
to produce more diesel fuel in response to a greater demand for diesel fuel relative to gasoline,
refiners are undercutting the swingcut portion of FCC naphtha at their refineries.l This action to
shift what historically was blended into the gasoline pool to the diesel fuel pool, also
dramatically reduces the sulfur content of the gasoline pool. If the entire swingcut portion of
FCC naphtha is undercut to the diesel fuel pool, the amount of sulfur in the gasoline pool is
reduced by about 50 percent. Our cost analysis estimates that at almost one quarter of U.S.
refineries, refiners are fully undercutting the FCC naphtha to diesel fuel today. At many other
refineries, our cost analysis estimates that refiners are partially undercutting their FCC naphtha.
These refineries will be able to reduce the sulfur of their gasoline well below their current levels
and generate a large number of early credits for Tier 3. Even for the subset of refineries where
FCC naphtha is not being undercut, refiners can assess how much activity or catalyst life is left
in its FCC postreater catalyst and compare this time with the time to the next turnaround when
the FCC postreater catalyst is scheduled to be replaced. If there is spare catalyst life, the refiner
1 The term swingcut means that this portion of the FCC product pool can be blended into gasoline or diesel fuel
while still meeting the fuel quality specifications for either fuel regardless of where this swingcut is blended.
5-82
-------
Tier 3 Summary and Analysis of Comments
could elect to increase the severity of their postreaters to reduce their gasoline sulfur levels to
under 30 ppm. With this strategy, the refiner would generate early sulfur credits. Also, when the
refiner replaces the catalyst in its Tier 2 postreater, it can elect to do so with a more active
catalyst which would allow the refinery to produce gasoline at sulfur levels below 30 ppm and
generate more early credits for Tier 3.
Based on the early actions refiners are either already taking, or could take, to reduce their
gasoline sulfur levels, we believe that refiners would be able to reduce their gasoline sulfur to as
low as 20 ppm, on average, without making any capital investments. By averaging 20 ppm for
2.5 years prior to 2017, refiners would be able to delay completion of all capital investments for
Tier 3 until mid 2019. If we add the 3.3 ppm of credits during 2012, 2013 and first part of 2014,
refiners would be able to delay completion of all capital investments in Tier 3 until 2020. Thus,
the early credit provisions in-effect can provide nearly 6 years of leadtime for full compliance
with the fuels program. This will allow ample time for refiners to complete their investment and
schedule their tie-ins during normal shutdown activities. It effectively provides even more lead
time than 5 years that the refining industry requested in their comments. The delay in the
program implementation will also help to distribute the demand on the E & C industry over more
years ensuring that the E & C industry would not be overwhelmed. Thus, the Tier 3 program
with a very flexible ABT program provides ample leadtime. In chapter 4 of the RIA we provide
multiple examples of how the refining industry can comply with the January 1, 2017 start date
which accounts for the types of projects that will need to be installed, a range of the credit
market size that will exist and the timing associated with tying-in the revamp or grassroots unit
with the rest of the refinery.
As discussed in more detail in Section V.E.I of the preamble, the provisions for small
refiners and small volume refineries is a third aspect of the Tier 3 ABT provisions that helps
with leadtime. Small refiners and small volume refineries (i.e., refineries processing less than or
equal to 75,000 net barrels per day of crude oil) are exempted from complying with the 10 ppm
average sulfur standard until 2020. This provides an estimated 36 refineries, of the total 108
refineries, nearly 6 years of lead time; again more than the 5 years that the refining industry
requested in their comments. As a group, we believe that these refiners and refineries are
disproportionally impacted when it comes to their cost of compliance and ability to rationalize
investment costs in today's gasoline market. Giving these refiners and refineries additional lead
time provides more time to invest in desulfurization technology, take advantage of advancements
in technology, develop confidence in a Tier 3 credit market as a means of compliance, and avoid
competition for capital, engineering, and construction resources with the larger refineries. The
small refiner and small volume refinery exemption until 2020 reduces the number of refineries
which will need to make a significant capital investment to comply with Tier 3 prior to 2020 to a
total of 49 non-small refineries (48 revamps and 1 grassroots unit), thus 15 refineries could wait
to take action until 2020. Although the small refiners and small volume refineries are not
required to comply with Tier 3 until 2020, they can still generate early credits (from January 1,
2017 through December 31, 2019) relative to 30 ppm for sale to other small refiners/small
volume refineries, and relative to 10 ppm for sale to non-small refiners. Such credits generated
relative to 10 ppm could provide another pool of early credits for Tier 3.
5-83
-------
Chapter 5: Tier 3 Fuel Sulfur Program
In summary, the ABT program provides ample flexibility for complying with Tier 3. The
averaging provisions will allow refiners that only need to revamp their Tier 2 postreaters to
overcomply and generate credits which will allow refineries that otherwise need to install
grassroots units to comply solely through the purchasing of credits. The banking provisions,
which allow refiners to generate early credits, effectively delays investments for compliance to
potentially as late as the year 2020. Finally, the small refiner and small refinery provisions delay
compliance for approximately 30 refiners until 2020. The provisions also allow them to generate
and sell credits during this period if they so choose. All these ABT provisions effectively
address the leadtime concerns. Furthermore, were we to shift the start date back another 2 years
as the refinery industry suggests in their comments, it would provide nearly 8 years of leadtime
for refinery changes that require just 2 or 3 years to complete. Refiners would not have to even
begin taking action for Tier 3 for a couple of years. Given that the lead time and associated
programmatic flexibility we are finalizing is sufficient to allow industry to readily comply, we do
not expect that a delay in the start date of the fuel standards would change the cost of compliance
discussed in Chapter 5. Any further delay in the program start date would simply delay the
actions to comply. Furthermore, delaying the start of the program would forego significant
emissions, air quality, and health benefits.
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM)
In the proposal, EPA suggests that 3 years of lead time is feasible, in part, because EPA began
talking to refiners about the possibility of regulating gasoline sulfur a couple of years ago. This is
inappropriate because EPA decisions concerning key aspects of this rule are not yet final. EPA
understands that there can be many changes made from the time a rule is proposed until it is
finalized in response to comments submitted by the public, and that therefore there is too much
uncertainty to make detailed implementation plans until the final rule is issued. In this case, for
example, we do not yet know whether EPA will actually require a reduction in gasoline sulfur
given the high costs and de minimis environmental benefits, we do not know when the rule will
be effective, we do not know whether EPA will leave the per-gallon cap at 80 ppm, etc. Based on
this, it is unreasonable, and arguably a violation of the Administrative Procedures Act, for EPA
to assert that refiners had sufficient notice of this rule's requirements two years before it was
even proposed to begin implementing the rule. In this rulemaking, as with all rules, EPA must
consider whether lead time is adequate from the date of promulgation of a final rule, not from the
time that EPA starts thinking about possibly issuing a proposed rule.
Our Response:
Our leadtime assessment is based on complying after the Tier 3 rulemaking is
promulgated. The conversations we held with refining companies only figured into the means
for compliance.
5.1.4.3.2. Impact of Turnaround Timing on Leadtime
5-84
-------
Tier 3 Summary and Analysis of Comments
What Commenters Said:
Commenter: Phillips 66 Company
Additionally, industry turnaround Intervals are typically every 4 to 5 years. An implementation
date of January 1, 2017, about 3 1/2 years from publication of the proposed rule does not sync
well with the refinery cycles. For example, if a refinery's normal turnaround date is 2014 and
again in 2019, there is not enough time to design, plan and execute the needed modifications
during the 2014 shutdown. In this case, refiners would be forced either to use credits in 2017,
2018 and part of 2019 until they could complete projects or to take a shut downs outside of
normal turnaround intervals. We question the availability of adequate credits available to
industry and believe that many refineries will require process unit shutdowns ahead of schedule.
Commenter: Chevron Products Company
The 2017 start date may have a negative impact on gasoline supply during the implementation
phase of the program. Ideally, refiners would be able to schedule investment to correspond with
their existing maintenance schedules, which often occur on 4 to 5 year intervals. Requiring
investments and construction to occur in a limited three-year window may compel refineries to
schedule additional off-cycle turnarounds which can reduce refinery utilization rates and also
reduce the supply of products to the market.
Commenter: Irving Oil Terminals Inc.
Irving Oil commends EPA's efforts to improve air quality through the reduction of sulfur content
in motor gasoline; however, as outlined in our attached comments, the proposed 2017 Effective
Date causes great concern because the 10 ppm annual average sulfur content standard cannot be
achieved as quickly as EPA desires. Typically, the modifications required to comply with the
proposed standard would be conducted during a five-year turnaround cycle at a refinery during
which all major renovations, new units and reconfigurations are implemented. A three-year
implementation period will increase operational risk to refiners, create inefficiencies, increase
expense for many entities and increase the risk of supply disruptions. Industry needs adequate
time to plan, finance and implement the significant changes required to meet the proposed Tier 3
standards. A 2019 Effective Date will provide the minimal five-year timeframe for these
complex and costly adjustments to be implemented.
Every year, refineries shut down select units in their facilities and perform regular inspections
and maintenance. These planned 'turnarounds' last about four weeks and are time-consuming
and disruptive, but they provide refiners the opportunity to make small adjustments to capacity,
improve fuel quality to some degree, repair, clean and update units. Typically, most maior
processing units undergo such turnarounds approximately every five years on a staggered
schedule. It is also during these unit turnarounds that any required revamp upgrades are
performed, which can extend turnaround durations beyond four weeks. These revamps can be
minor or major in magnitude. Units can be modified or replaced, capacity may be increased, and
there may be overall reconfiguration. The changes associated with this type of turnaround can be
significant. They typically take years to plan, design, purchase equipment and install, and such
5-85
-------
Chapter 5: Tier 3 Fuel Sulfur Program
changes can cost hundreds of millions of dollars. The refinery changes necessary to meet the
proposed annual average sulfur standard of 10 ppm may fall within these latter types of
modifications and would be most efficiently implemented as part of a comprehensive refinery
turnaround.
Unfortunately, the proposed rule does not give consideration to these normal refinery turnaround
cycles. It simply proposes an effective date that is approximately three years from the anticipated
date when the final rule will be issued. Three years is not sufficient. Moreover, Irving Oil
understands that when EPA first began work on the Tier 3 rule in 2010 and thought that a final
rule would be issued in 2012, an effective date of 2017 was appropriate. It would have provided
industry with five years to meet the proposed standard. However, because the proposed and final
rules were delayed, the effective date for the 10 ppm sulfur standard should also be adjusted
accordingly. An implementation date that provides five years would generally align with the
turnaround cycle for most refiners. A shorter date would inadvertently place some at a significant
financial and competitive disadvantage. Such refiners would either be forced to buy credits to
meet the new standard or perform an additional turnaround — an inefficient and costly measure.
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM)
The increased reliability of refining equipment has extended refinery turn-around time to
approximately 5 years.
Commenter: Marathon Petroleum Company LP (MFC)
The increased reliability of refining equipment has extended refinery turn-around time to
approximately 5 years.
Our Response:
We also received comments indicating that the January 1, 2017 does not complement
refinery turnaround schedules, which often occur in 4-5 year intervals.
We agree that the need to make tie-ins, which must occur during a refinery turnaround,
must be considered when assessing the feasibility of leadtime, and even when we factor the time
needed to do so, our analysis shows that refiners can comply with Tier 3 with the leadtime
provided. This is true because the final rule effectively provides nearly 6 years of leadtime to
complete capital projects, and also because the capital projects do not have to be completed prior
to installing the necessary tie-ins for new Tier 3 units.
As explained in Section V.L of the preamble to today's final rule, when a refiner builds a
grassroots unit or some sort of revamp that involves a new reactor or perhaps an added
distillation column, the new vessels and associated equipment must be "tied-in" to the rest of the
refinery. The tie-in usually involves connecting a pipe from the existing unit to the new unit
installed. However, a pipe cannot simply be added while the refinery is operating. Instead, the
refiner will add the necessary pipe for making the tie-in when the refinery, or at least that
5-86
-------
Tier 3 Summary and Analysis of Comments
refinery unit, is shutdown for regular maintenance. The revamp or grassroots unit does not have
to be started up at that time, thus, the tie-in piping can be added well before the revamp or
grassroots unit is completed. Instead, the connection pipe just needs to be added and blocked off
with a sealing-type valve and a blind flange (essentially a flat piece of steel) is bolted on as a
precaution against a leaky valve. This is a very simple process that would take several pipefitters
a half a day of work to complete including completing all the necessary safety protocols.2 Once
this piping has been added, the refiner can restart its refinery. Then when the refiner is ready to
complete the tie-in to the completed revamp or grassroots unit, the refiner would remove the
blind flange and connect a pipe that connects the existing part of the refinery to the newly
installed grassroots postreater unit or revamp postreater subunit. This last step can either occur
when the refinery is shutdown or still operating. At that point the refiner would only need to
open the block valve to complete the tie-in of the grassroots unit or revamp to the existing
refinery. One refiner who owns a number of refineries informed us that it installed the tie-ins for
a possible Tier 3 rule when it installed its Tier 2 units.
Refiners commented that FCC turnarounds occur on a 4 to 5 year basis, or a 5 year basis.
However, this amount of time between FCC turnarounds is contradicted based on information
provided on the American Petroleum Institute's webpage, which reports that the average time
between major FCC unit turnarounds is 4 years. Also, an Energy Information Administration
(EIA) study makes a similar finding, which is that refiners target 3-5 years, or 4 years on
average, between refinery turnarounds. The EIA study also reported the results of an Octane
Week survey of about 20 refineries which found that refiners target 4-5 years between
turnarounds of the FCC units, but over 25% of the time refiners need to conduct turnarounds
earlier than targeted because a maintenance issue forces the earlier turnaround. However, the
EIA study did not estimate how much earlier the turnaround occurred so it was not possible to
estimate an actual average turnaround schedule, which may be less than 4 years. This means that
on average, 25% of U.S. refineries shutdown to perform maintenance on its FCC units each year.
In Chapter 4 of the RIA, we presented our analysis of several different scenarios for
credit availability versus demand for complying with the Tier 3 sulfur standard, assuming that
credits will be traded nationwide as it has under Tier 2. In one scenario, if refiners reduce their
gasoline sulfur to 20 ppm immediately after the Tier 3 rule is finalized, as we anticipate they are
capable of, then the substantial number of early credits generated would allow refiners to delay
their unit start-ups sufficiently for refiners to not only make the tie-ins, but even complete their
capital projects as necessary. However, we also conducted a more conservative analysis which
assesses the ability of refiners to comply with Tier 3 if they generate no or few early credits.
Using the estimate that turnarounds are scheduled at refineries every 4 years, we also found that
refiners would be able to comply with the Tier 3 program start date.
2 Since most refiners have already completed their scoping studies with the vendor companies which license the
desulfurization technologies, they likely already understand what steps would need to be taken to tie-in their
revamps and grassroots units with their existing refinery. Installation of tie-ins is relatively simple refinery change
that can be engineered and installed in a short period of time. For this reason, we believe that refiners can begin
making their tie-ins as soon as the spring of 2014
340CFR80.385(e).
5-87
-------
Chapter 5: Tier 3 Fuel Sulfur Program
5.1.4.3.3. Amount of Time Required to Install Grassroots Units and Revamp Existing
Units
What Commenters Said:
Commenter: Phillips 66 Company
The proposed implementation date of January 1, 2017 does not provide adequate lead time for
refineries to implement needed capital projects. EPA has proposed this aggressive timeline and
uses two basic arguments to support their proposal. Firstly, EPA predicts that most projects will
be of moderate scope and therefore refiners can complete needed capital projects within a 3-year
time frame. Based on our initial analysis, we expect our company (and likely industry) will be
faced with projects of significant scope requiring more than 3 years to complete.
Project Scope: EPA provided two project timeline estimates in the proposed rule - one for unit
revamps (24 months) and one for new grass roots units (36 months). We believe EPA has
underestimated the time requirement for every phase of a project - whether a revamp or a new
unit.
In its investigations, EPA concludes that most refineries will require only projects of moderate
revamp scope and that a relatively few will require grass roots desulfurization investments. The
DRIA is quoted, "For companies that already have an FCC naphtha postreater we assumed that
all that would be necessary to meet the proposed sulfur standards was to revamp their existing
FCC postreating units." Also in its analysis, EPA uses a PADD-average approach which may
neglect individual refinery configurations.
To reduce sulfur in gasoline, it is common for refiners to separate the FCC naphtha stream into
several streams with differing characteristics requiring different treatment technologies. Some of
these separated streams, but not all, may currently be undergoing sulfur reduction. The likely
approach to achieving Tier 3 will be to treat the remaining FCC streams plus additional light
streams from other process units. Due to the differences in characterization, the in-place
technology likely will not be appropriate to treat the remaining untreated stream(s) required for
Tier 3 and therefore revamp(s) would not be chosen. For example, heavy FCC naphtha contains
a significantly higher amount of sulfur and uses a technology that is not appropriate for light
FCC naphtha. In cases where a refinery is currently only treating the heavy FCC naphtha,
refiners will require new grass roots units to treat other portions of the FCC naphtha, tailored to
the individual characteristics of the untreated streams.
These extensive projects and the fact that industry will be competing for many of the same
engineering, procurement and field resources to complete them, requires significantly more time
than EPA is providing in its proposal.
Project time requirements: Our experience with both grassroots projects and significant revamps
of this magnitude is that they require well in excess of three years to complete. As such, EPA
should extend the compliance date to coincide with 5 years from issuance of the final rule.
5-S
-------
Tier 3 Summary and Analysis of Comments
Below are two projects time lines [EPA-HQ-OAR-2011-0135-4463-Al], for a grassroots and
revamp, provided by Phillips 66 internal project management staff. As shown the grass roots
projects will need almost 5 years to complete. By comparison, the revamp project can be
completed in only 6 months less time. As such, we believe that EPA has underestimated the time
requirement and uncertainty for every phase of a project.
In particular, a critical phase of any project is in the initial scoping. It is imperative that refineries
take the requisite time needed to fully evaluate various technology options and operating
configurations. It may be necessary to incorporate other facets into the project, such as
alleviating existing bottlenecks to allow full operational optimization and flexibility with any
new or revamped units. Technology vendors continue to work on improvements especially in the
area of catalyst technology. It is a very time consuming effort to work with multiple technology
vendors to evaluate various feed streams with their technologies and then do economic
evaluations on the different options. In order to meet a January 1, 2017 implementation date, a
refinery would need to have already completed the project scoping phase, well in advance of the
final rule. Since there is uncertainty in when the final rule will be issued, and in the provisions it
will contain, refiners cannot set a final design basis. They therefore cannot complete scoping
until the final rule is issued.
Commenter: Monroe Energy, LLC
The controls will take at least 2 years to design and engineer.
Commenter: Chevron Products Company
We are opposed to the implementation schedule proposed by the EPA. The 2017 compliance
deadline does not provide sufficient time to make necessary refinery modifications and would
significantly increase the cost of the required upgrades. If the agency proceeds with the gasoline
sulfur reduction requirement, the program start date should be no sooner than five years after the
date of the final regulation.
We agree that the technology for sulfur reduction to the 10 ppm level is available and has been
demonstrated at certain refineries in operation today. However, we believe that EPA has
underestimated the cost and complexity of the changes which will be required in the refining
industry to achieve this reduction.
The time required for permitting is becoming longer due to local regulations and GHG programs.
We do not agree that the analysis of permit times should be limited to the estimates provided by
OAQPS. We are seeing other government agencies placing additional requirements on refinery
permits for a variety of GHG, safety, and environmental concerns, all of which increase the time
required to obtain a permit. California refineries producing gasoline subject to the Tier 3
regulations may encounter even longer permitting times and increased scrutiny, taking them well
beyond the proposed implementation period.
We are also concerned about EPA's statement that many refineries have already started scoping
work on investments in advance of the Tier 3 rulemaking. While it is true that refineries have
5-89
-------
Chapter 5: Tier 3 Fuel Sulfur Program
been aware of EPA's intention to conduct a Tier 3 rulemaking for over two years, the NPRM
was only published in the Federal Register on May 21, 2013, and will not be finalized until late
2013. It is unlikely that refineries have conducted significant planning or engineering based on
speculation about Tier 3 when there has been no certainty about the timing or the specific
requirements of the rule until very recently. The timeline for program implementation should be
based on realistic estimates of the time required for planning, investment, permitting,
construction, and operation, and refineries should be able to conduct these activities to fully
comply on day one of the new program. To suggest that the timeline could be shortened based on
actions that refiners may have taken prior to even a proposed rulemaking is inappropriate.
Chevron is concerned about the global availability of long lead time equipment, like reactors and
compressors, which will be required for many of the industry projects. Refinery investments
require specialized labor for both design and construction. These industries are experiencing a
great demand for services due to other domestic energy investment projects currently underway.
EPA's analysis of the capability of the labor force does not fully consider the competition for
these resources and the impact this may have on the timelines required for Tier 3 investments.
Commenter: Irving Oil Terminals Inc.
Even though the proposed rule has been highly anticipated by the regulated community, design
approval for major capital investments can only proceed with 'regulatory clarity' which will not
exist until the final rule is issued.
Every year, refineries shut down select units in their facilities and perform regular inspections
and maintenance. These planned 'turnarounds' last about four weeks and are time-consuming
and disruptive, but they provide refiners the opportunity to make small adjustments to capacity,
improve fuel quality to some degree, repair, clean and update units. Typically, most maior
processing units undergo such turnarounds approximately every five years on a staggered
schedule. It is also during these unit turnarounds that any required revamp upgrades are
performed, which can extend turnaround durations beyond four weeks. These revamps can be
minor or major in magnitude. Units can be modified or replaced, capacity may be increased, and
there may be overall reconfiguration. The changes associated with this type of turnaround can be
significant. They typically take years to plan, design, purchase equipment and install, and such
changes can cost hundreds of millions of dollars. The refinery changes necessary to meet the
proposed annual average sulfur standard of 10 ppm may fall within these latter types of
modifications and would be most efficiently implemented as part of a comprehensive refinery
turnaround.
In the Preamble that accompanied the Proposed Rule, EPA recognizes that gasoline refining is a
complex operation, and refiners would have to take many steps to comply with the proposed Tier
3 sulfur standards. These steps include scoping studies, financing, process design for new or
revamped refinery units or subunits, permitting, detailed engineering based on the process
design, field construction of the sulfur reduction units, and start-up and shakedown of the newly-
installed desulfurization equipment. The Agency states that it conducted a lead time analysis and
spoke with refiners to determine the amount of time that a typical refiner would take to complete
renovations and meet the 10 ppm sulfur standard. EPA has proposed that a lead time of three
5-90
-------
Tier 3 Summary and Analysis of Comments
years is sufficient. The details of the Agency analysis are set out in Table V-3 — 'Anticipated
Compliance Timelines.'
Irving Oil has reviewed EPA's information on 'lead times' and believes that it is overly
optimistic. Also, some of the data on which the determination was made are now out of date due
to major changes that are occurring in the petroleum industry. The following are several
significant reasons why a three-year lead time is insufficient:
1. Those companies that license the technology to desulfurize motor gasoline have limited
resources and likely would be unable to support many simultaneous scoping studies and process
designs. Thus, some refiners would not be able to begin these activities as early as needed to
support a three-year lead time.
2. Irving Oil agrees with EPA's assessment that most refiners would revamp existing FCC
gasoline post-treaters rather than constructing grassroots units. However, Irving Oil has often
found that revamping units, particularly during the early project phases (the scoping study and
process design) is more complex and time consuming than building and installing an entirely
new, grassroots unit. Therefore, EPA's assumption that refiners can revamp existing units more
rapidly is not supported by Irving Oil's many years of experience.
3. EPA has indicated that some of the initial phases of the revamping project can overlap,
thereby saving additional time. Again, this assumption is contrary to Irving Oil's experience and
belief. Irving Oil has found that the process design phase cannot be started until after the scoping
study is complete. Thus, it is more reasonable to assume that the time needed to make the
appropriate changes to the refinery would be the same for both a revamping of units and the
installation of new units.
4. The EPA believes that there would be sufficient resources (technology licensors, engineering
and construction services, skilled laborers, and equipment) available to the refining industry.
However, EPA must recognize that the U.S is undergoing a domestic petroleum renaissance, and
demand for these resources in new areas of production such as North Dakota, Montana, and
Texas is great. Those projects would compete for resources directly with refiners trying to meet
the EPA three-year time schedule. Shortages of these resources would, therefore, likely inhibit
the ability of refiners to comply in a timely manner.
Commenter: Independent Fuel Terminal Operators Association (IFTOA)
In the Preamble that accompanied the proposed rule, EPA includes a discussion on the
"Adequacy of Proposed Refinery Lead Time." It provides that the Agency has sequenced the
estimated time to complete scoping studies (6 to 9 months), process design (6 to 9 months),
permitting (9 months), detailed engineering (6 to 9 months), field construction (6 to 12 months),
and start-up and shakedown in advance of production (6 to 9 months). In addition, EPA believes
that there would be substantial overlap among many of these steps. Therefore, the Agency states
that three years is adequate to implement the required refinery changes, and early credits would
add some flexibility. This estimate is far too optimistic. It fails to understand the process and
timing refineries (both domestic and foreign) employ when they make decisions about refinery
modifications, how financial planning for major modifications is handled, and that longer lead
times are needed when the regulated community is directed to meet a new standard - not a single
refiner/importer.
5-91
-------
Chapter 5: Tier 3 Fuel Sulfur Program
First, refiners generally would not undertake to make major modifications until the final
regulation is issued by EPA. It would not be prudent to begin modifications down one path
under a proposed rule and subsequently have to alter those modifications if the Agency finalizes
a rule taking another approach.
Second, once the rule is adopted, there would be a great demand for consulting and engineering
firms to assist the refining industry in its efforts to meet this new standard. As EPA is aware,
there is already a great demand for the services of such firms due to increased production of
crude oil in PADDs II and III. Thus, some refiners would have to wait for assistance.
Third, EPA envisions overlapping of some of the enumerated steps such as moving forward with
design processing while the scoping study is underway. Based on the experience and expertise
of a number of our Members, we have been told that it is unwise to move forward with design
processing until the scoping study is complete. Again, substantial man-power, time and money
could be wasted if design processing anticipates what would emerge from the scoping process.
Fourth, approval for funding for such major changes is a time-consuming and often difficult
process. The estimated schedule put forth by EPA does not appear to include the time needed to
obtain or allocate such funding.
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM)
Rushing implementation undermines the ability of the industry to adequately plan projects,
secure contractors and equipment at competitive rates, optimize solutions, and align construction
projects with existing maintenance turn-around schedules. In addition, air permits will be
necessary at many refineries, and EPA has been challenged in issuing permits in a timely fashion
(as is discussed in further detail in section VII.F of these comments.) All of these factors will
tend to increase compliance costs.
In the proposal, EPA puts forth Table V-3 with a timeline that the Agency believes illustrates the
time needed to implement the 10 ppm gasoline sulfur requirement at a refinery. 78 Federal
Register 29925. We disagree with EPA's suggested timeline. We believe that a more accurate
amount of time needed to implement major new rules like the 10 ppm gasoline sulfur
requirement is at least five years. The Tier 2 rules were finalized in February 2000, with
compliance required to phase in from 2004-06 (with compliance flexibilities that had a more
meaningful impact, as discussed below). Ultra Low Sulfur Diesel (ULSD) rules were finalized in
January 2001; with a compliance phase-in period requiring that 15 ppm sulfur diesel constitute
80% of the highway pool in mid-2006. Compliance lead times need to be longer than these two
significant rulemakings to reduce sulfur because of new GHG requirements at refiners, and the
increased reliability of refining equipment has extended refinery turn-around time to
approximately 5 years.
Commenter: Marathon Petroleum Company LP (MFC)
5-92
-------
Tier 3 Summary and Analysis of Comments
The Agency must extend the implementation date until January 1, 2019 or five years after the
final rule is issued.
Rushing implementation undermines the ability of the industry to adequately plan projects,
secure contractors and equipment at competitive rates, optimize solutions, and align construction
projects with existing maintenance turn-around schedules. All of these factors will tend to
increase compliance costs.
In the proposal, EPA puts forth Table V-3 with a timeline that the Agency believes illustrates the
time needed to implement the 10 ppm gasoline sulfur requirement at a refinery. 78 Federal
Register 29925. We disagree with EPA's suggested timeline. We believe that a more accurate
amount of time needed to implement major new rules like the 10 ppm gasoline sulfur
requirement is at least five years. The Tier 2 rules were finalized in February 2000, with
compliance required to phase in from 2004-06 (with compliance flexibilities that had a more
meaningful impact, as discussed below). Ultra Low Sulfur Diesel (ULSD) rules were finalized in
January 2001; with a compliance phase-in period requiring that 15 ppm sulfur diesel constitute
80% of the highway pool in mid-2006. Compliance lead times need to be longer than these two
significant rulemakings to reduce sulfur because of new GHG requirements at refiners, and the
increased reliability of refining equipment has extended refinery turn-around time to
approximately 5 years. These recent changes for refiners add complexities that warrant
additional Tier 3 compliance lead time.
Our Response:
Several commenters also stated that we had underestimated the project time lines for
revamps and installation of grassroots units. Some commenters compared the time required for
installing units for Tier 3 with previous rulemakings such as the Tier 2 or highway diesel fuel
rules.
We disagree with comments that we underestimated project time lines for revamps and
installation of grass root units. As explained in Section V.E. of the preamble to today's final
rule, we believe that the situation for Tier 3 is different than previous rulemakings. The
technologies for complying with Tier 3, which are the same as those used for complying with
Tier 2, are well known and well proven. Refiners that complied with Tier 2 using FCC naphtha
desulfurization technologies installed the following units, all of which were grassroot
installations: Axens Prime G+, CDTech's CDHydro and CDFIDS, UOP's ISAL, Phillip's
(Sinopec's) S-Zorb and Exxon's Scanfming. Prior to choosing a technology, refiners needed to
evaluate each these different technologies and choose among them, all of which were largely
untested at the time, which required us to provide more lead time for Tier 2. Since it has been 10
years since the Tier 2 sulfur standard began to be phased in, refiners now have direct experience
with the installation and operation of these technologies, and the vendor companies that license
them continue to support their installations onsite. This fact will allow refiners to reach a
decision very quickly when complying with Tier 3, particularly, because in most cases the
refiners will solely be revamping the units installed for Tier 2 when complying with Tier 3.
5-93
-------
Chapter 5: Tier 3 Fuel Sulfur Program
Based on our discussions with refiners, construction companies, vendor companies and
from published literature, we estimated the time it takes to revamp existing postreaters and install
grassroots postreaters. Revamping an existing postreater is expected to require up to two years.
Installing a grassroots postreater is estimated to require three years.
We believe that the revamping of postreaters could take even less time than 2 years
because many of the Tier 3 revamps are expected to be very modest (e.g., change out a reboiler
or heat exchanger). The availability of a more active catalyst and the expanding practice of
undercutting the FCC naphtha to the distillate pool (which reduces the FCC volume by 15% and
sulfur content by 50%) permit the existing FCC postreaters to achieve much further sulfur
control.
Furthermore, we noted at proposal that since EPA held discussions with many refiners in
2011 about EPA's plan to pursue additional sulfur control post-Tier 2 (Tier 3), refiners began the
process of assessing how they would comply. The Tier 3 proposal was delayed for about a year
and it is our understanding from recent discussions with vendor companies and some refiners
that many refiners have already completed their scoping studies. By completing their scoping
studies, refiners have chosen their technologies, and in the case of grassroots units, vendor
selection as well (refiners with a particular postreater technology in most cases are expected to
simply revamp the same vendor's technology, so there is no need to select a vendor). If refiners
have already completed their scoping studies, we estimate that installation of the revamps or
grassroots units would be about 3 months shorter than the 2 and 3 years, respectively, than our
estimated timelines.
These project timelines are reasonable in light of past industry experiences that show
FCC postreaters being installed in refineries in less time than what we estimate. At the Motiva
refinery in Port Arthur, TX, a grassroots CDTech postreater was designed, constructed and
started up in less than 2 years. At two refineries in Germany, two Prime G+ units were designed,
constructed and started up - one of them in two years, and the other in 18 months. As an
extreme example, the $3.6 billion dollar, 180 kbbl/day crude oil volume expansion at Marathon's
Garyville, LA refinery was designed, constructed and started up in less than 4 years. This single
project involved the construction of 10 major refinery units. Since these may be best case
examples, we continue to believe that our construction projections are reasonable.
One commenter stated that our FCC installation timeline did not include the time
required to obtain funding for the postreater installation. First, most all FCC postreater projects
are revamps which are relatively inexpensive installations and thus, funding should be relatively
easy to obtain. Second, funding for design and construction of the FCC postreater installation
can be obtained when the scoping study is being completed. As discussed in more detail in
Section V.E.2.b, if a refiner is experiencing financial hardship and is having trouble obtaining the
necessary funding which leads to a delay in complying with Tier 3, the refiner may apply for an
extreme circumstances hardship waiver, which is a hardship waiver based on severe economic or
physical lead time limitations of the refinery to comply with the Tier 3 standards at the start of
the program. A refiner seeking such hardship relief under this provision must demonstrate
criteria, and we expect that a refiner would apply for such hardship prior to the start of the Tier 3
5-94
-------
Tier 3 Summary and Analysis of Comments
program (as a refiner should know during the planning stages if there will be extreme financial
hardship).
In response to one commenter's concern about permitting and its impact on completing
projects, as we discuss in Chapter 5.2, we expect that only a few permits are expected to be
needed. Because of that, the few refineries needing to file a permit are expected to experience
construction times about as long as we estimated, and the rest of the refineries, which are not
expected to file a permit, could experience shorter construction timelines than what we
estimated.
One commenter noted that we show overlapping time between important elements that
comprise the planning, design and construction schedule, and commented that the overlap is not
possible. The commenter stated that one element must be completed before the next one can
commence. We show overlap between the different elements because some aspects of each
element can begin before the next element begins. For example, before the scoping study is
completed, the refiner will initiate the process to set up a contract to begin the process design.
Also, before the detailed design is completed, the refiner is expected to begin site preparation
aspect of the construction which may include demolition of nonused equipment and soil testing.
Finally, the refiner is likely to test the individual pieces of equipment (such as pumps and
activated valves) before trying to start up the entire unit which is why startup can begin before
construction is complete. If refiners overlap the different design, construction and plant start up
elements as depicted, the refinery would be able to meet the grassroots and revamp timelines we
laid out. As cited above, the very fast FCC postreater installations (CDTech at Port Arthur and
two German refineries) and very fast Garyville refinery expansion were likely made possible
because there was overlap between several of the design and construction elements.
While we acknowledge that some refineries may require more time than we have
assumed to complete their refinery modifications, others would be able to do so with less time.
If the Tier 3 program were designed in a way that had a firm fixed date for compliance with no
flexibility, then perhaps the much more conservative assumptions by the commenters would be
more appropriate, as the worst case refinery situation would be important in determining the
appropriate lead-time and start date. However, this is not the case here. As described above, the
program is designed with considerable flexibility. Thus, it is more appropriate to consider more
typical project timelines when evaluating lead time. Furthermore, the program, with its
flexibilities, allows nearly 6 years of lead time for the industry to complete their projects, long
enough for even the worst case project timelines highlighted by the commenters.
5.2. Refinery GHG Emissions and Air Permitting Interactions (NSR/NSPS/NESHAP)
5.2.1. GHG Emissions and Refineries
What Commenters Said:
Commenter: Association of Fuel & Petrochemical Manufacturers (AFPM)
5-95
-------
Chapter 5: Tier 3 Fuel Sulfur Program
One oil industry commenter said that the new regulations will create conflicts with existing
regulations and could jeopardize a refiner's ability to comply with federal formulation
regulations since modifications to operations may be necessary to make cleaner fuels which
could trigger greenhouse gas emissions regulations, putting these projects in jeopardy.
Specifically, this commenter noted that mandating lower sulfur fuels under the Clean Air Act
will require facilities to install advanced technologies that increase energy use for the
formulation of increasingly complex motor fuels and this additional energy consumption will
result in greenhouse gases and other emissions. Therefore, the proposed gasoline sulfur reduction
standard will increase the carbon footprint at refineries.
Our Response:
EPA acknowledges that reducing sulfur from gasoline will result in a small increase in
CC>2 emissions associated with higher energy use required in the process of removing sulfur
within the refinery. As an extension of our refinery-by-refinery cost modeling described in the
Preamble, Section VII.B., we calculated the CO2 emission impacts of Tier 3 gasoline sulfur
control. We estimated refinery-specific changes in process energy and then applied emission
factors that correspond to those changes, on a refinery-by-refinery basis. As described in
Chapter 4.5 of the RIA, the results showed an increase of up to 1.9 MMTCC^e in 2018 and 1.6
MMTCO26 in 2030 for all U.S. refineries complying with the lower sulfur standards assuming
that the sulfur standards are fully phased-in. In 2018, the combined impact of CH4 and N2O
emission reductions from the vehicles and CC>2 emission increases from the refineries shows a
slight net decrease on a CC>2 equivalent basis. While still small, this net decrease grows to a
range between 2.5 to 2.7 MMTCO2e by 2030.
In our updated analysis for the final rule, we adjusted our refmery-by refinery analysis to
reflect the existence of a nationwide average, banking, and trading (ABT) program and refined
our estimates regarding the physical and operational changes that will be required at each
refinery (as described in the final RIA). The modifications at a given refinery could include
revamps to existing FCC pre- or post-treatment unit(s) or the installation of a new grassroots
post-treatment unit for sulfur reduction. Based on the updated projections of refinery-specific
changes, we re-estimated the increased demand for energy (i.e., fuel to generate process heat,
steam, and electricity), hydrogen, and sulfur recovery associated with meeting the final Tier 3
standards. Having received no comments suggesting that they should be changed, we re-applied
the representative industry emission factors for NAAQS pollutants, their precursors, and GHGs
for each emitting process and combined them with estimates of incremental activity to estimate
the emissions changes at each equipment unit (or group of similar units) at each refinery.
We found that under the high case (worst case scenario), nine refineries appeared likely
to have significant emissions increases for one or more pollutants and thus would trigger major
NSR. This estimate equates to approximately eight percent of the 108 refineries projected to sell
gasoline that will be subject to the Tier 3 standards. Of these nine refineries, we predicted that
three refineries would need major source permits for NAAQS-related pollutants and their
precursors (PSD and/or Nonattainment NSR) and for GHGs, while six refineries would need
PSD permits to address GHG emissions. This number could be lower if those refineries apply
5-96
-------
Tier 3 Summary and Analysis of Comments
pollution controls, such as SCR for NOx, to sufficiently reduce the emissions increases to levels
that are below the applicable pollutant significance level, or if the refineries can achieve
emissions reductions elsewhere at the facility to "net out" of major NSR.
5.2.2. Issuing Permits
What Commenters Said:
Commenters: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM); Concerned Citizens Around Murphy; Marathon
Petroleum Company LP (MFC)
Several oil industry commenters stated that EPA needed to issue permits expeditiously and that
EPA should adopt the recommendations offered by the Clean Air Act Advisory Committee
(CAAAC) in the September 2012 report from the Committee's GHG Permit Streamlining
Workgroup.
One citizens' organization commented that clean fuels are an essential part of the Ozone and
particulate matter solutions. However, many of our fenceline communities do not have
appropriate monitoring for the 2010 sulfur dioxide ambient air standards. Extra air monitoring
could ensure the air permit limits are maintained, and air monitoring may even help the refineries
become more efficient.
Our Response:
EPA is continuing to assess possible approaches to streamlining GHG permitting,
including those offered in the referenced CAAAC report. As discussed below in Chapter 5.2.2,
EPA believes that the permits that may be required as a result of the Tier 3 fuel sulfur standards
will be issued in a timely manner.
EPA is planning to propose a rule soon that will describe requirements for air agencies to
characterize sulfur dioxide concentrations through ambient monitoring or air quality modeling
techniques in targeted areas around the country in which the largest sources of emissions are
located. The air quality information collected by the air agencies will then be used to inform
future rounds of area designations under the Clean Air Act. The rule will reference appropriate
guidance on monitoring and modeling techniques, and it will include timelines for air agencies to
conduct the required analyses. Comments regarding which sources should be subject to this
requirement will be welcome as part of that rulemaking.
5.2.3 GHG Permitting
What Commenters Said:
Commenters: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM); Marathon Petroleum Company LP (MFC)
5-97
-------
Chapter 5: Tier 3 Fuel Sulfur Program
Several oil industry commenters cited recent EPA-provided information on how many GHG
permits have been issued by EPA and other permitting authorities since the GHG permitting
requirement became effective in January 2011. These commenters stated that the small number
of permits issued brings into question whether EPA's forecast of how quickly GHG permits will
be issued when needed in connection with the Tier 3 fuel program is realistic.
Several oil industry commenters also noted that if the emissions thresholds that trigger
permitting for GHG are lowered from the values assumed in EPA's preliminary assessment of
permitting implications, the number of refineries that would need GHG permits could be higher
than estimated in that assessment.
Our Response:
The commenters have cited EPA statements about the number of permits issued, without
the context of when they were found to be complete. While some permits may have taken longer
than 12 months to be issued once they were complete during this start-up period of the GHG
permitting program, EPA continues to believe that in the future time frame in which a small
number of refineries will be seeking GHG permits, those permits can be issued in a timely
manner that will assure compliance with the requirements of the final Tier 3 fuel program. The
experience gained by permitting agencies and the processing approaches developed during the
start-up period should allow more expeditious processing in the future. Many of the permit
applications cited by the commenters as not having reached the point of the final permit having
been issued (as of the May 2012 date of the EPA-provided information) were for sources in
Texas, where EPA has been the permitting authority for GHG. EPA has recently proposed to
approve Texas' plan for assuming the role of that authority (79 FR 9123, February 18, 2014),
with a final action expected around the middle of 2014. For permits to refineries in Texas issued
by Texas during the implementation of the Tier 3 program, a number of requirements in federal
laws that are applicable to EPA permitting actions will no longer apply, for example the federal
Endangered Species Act. It should be possible both for sources to prepare complete applications
more expeditiously since they will not need to include all the information needed to ensure
compliance with these federal laws, and for the Texas permitting authority to review permit
applications in less time than has been the case for EPA. Since May 2012, EPA has also
proposed or taken final action to transfer GHG permitting authority to Arkansas, Florida, Idaho,
Kentucky, Clark County in Nevada and Wyoming. In addition to those states for which we
approved GHG PSD authority, EPA has approved PSD authority (including GHG) to the
following air pollution districts in California: Eastern Kern, Imperial County, Placer County, San
Joaquin Valley Unified, and Yolo-Solano. EPA has also approved the South Coast Air Quality
Management District's ability to issue GHG PSD Permits.
Regarding the comments that the number of refiners needing GHG permits could be
higher than estimated if the thresholds are lowered, EPA finalized GHG thresholds for Step 3 of
the GHG permitting program that are the same as in Steps 1 and 2 on July 12, 2012. These are
the thresholds assumed in the updated assessment of permitting implications of the final Tier 3
fuel program. EPA will complete a further rulemaking by April 2016 to establish permanent
5-98
-------
Tier 3 Summary and Analysis of Comments
thresholds (the Step 4 rulemaking). In the Step 4 rulemaking, EPA may consider the possible
need for the permanent thresholds to be phased in over some period after that final rule.
5.2.4. New Source Review
What Commenters Said:
Commenters: Countrymark; Environmental Defense Fund (EDF); Northeast States for
Coordinated Air Use Management (NESCAUM); Sierra Club; Small Business
Refiners
A number of refiners requested that the final rule explicitly affirm that any investment required
for compliance with Tier 3 requirements not be subject to NSR permitting review, provided that
capacity is not increased by more than 10 percent of prior capacity.
Two environmental groups commented that they supported EPA's not allowing vehicle emission
reductions to be used as part of "netting out" under the NSR program, while one state air board
supported the use of vehicle emission reductions for purposes of "netting out" under the NSR
program.
An environmental group recommended that, to the extent that EPA moves forward with
developing streamlining options for NSR, EPA must reach out to all stakeholders to identify
methods that are wholly consistent with the CAA and the goals of the PSD program. Any
committee of EPA employees with refinery expertise that oversees streamlining for permits
related to the Tier 3 fuels program must be shielded against industry lobby interests and any
conflicts of interest in implementing NSR regulations.
Our Response:
EPA believes such a blanket exemption based on a percentage increase in refinery
capacity would not be consistent with CAA requirements. Under the Clean Air Act as
interpreted by the courts, new source review applies to any physical modification that causes
more than a de minimis net emissions increase. (See New York v. EPA, 413 F.3d 3 (D.C. Cir.
2005.) The final Tier 3 rule does not include such an exemption.
The final rule does not include any provision to allow the use of vehicle emission
reductions for purposes of "netting out" of the NSR program. EPA believes that the use of
vehicle emissions reductions for netting purposes is not permitted by the NSR regulations, since
the creditable emissions reductions used for netting purposes must result from reductions
occurring at the modified source. For example, the definition of net emissions increase in 40
CFR 52.21(b)(3)(i) includes only increases and decreases occurring "at the source". We believe
this provision in the NSR rules is consistent with the CAA. With regard to the commenter
supporting the use of "netting out," the commenter did not provide any grounds for concluding
that allowing the use of vehicle emission reductions for netting out purposes would be consistent
with the CAA.
5-99
-------
Chapter 5: Tier 3 Fuel Sulfur Program
Regarding the comment on streamlining, EPA has concluded that only a small number of
refineries appear to have the potential of triggering major source permitting as a result of
modifications needed to meet the Tier 3 fuel program. Furthermore, given the flexibilities
provided by the final program in terms of when modifications must be in place in order to
achieve compliance, even the limited number of permits will likely be spread out over a period
of time. There were no industry comments that demonstrated a specific reason for concern about
permitting implications on a broad scale. Therefore, EPA intends to address any permitting
complications that do arise on a case-by-case basis with the appropriate EPA Regional Office in
the lead, rather than convene an agency-wide committee of experts.
5.2.5. GHG Emissions and Offsets
What Commenters Said:
Commenters: American Petroleum Institute (API); Association of Fuel & Petrochemical
Manufacturers (AFPM)
One commenter stated that since EPA claims that the refinery GHG emissions will be offset by
other parts of this rule, EPA needs to establish a methodology to allow each refinery that
participates in Tier 3 to increase its future GHG baselines for any permitting purposes by the
amount that the Tier 3 final regulations will increase its GHG emissions.
Our Response:
EPA interprets this comment to be a recommendation that refineries making GHG-
increasing modifications in order to comply with new sulfur limits should, for PSD permitting
purposes, be considered not to be increasing their GHG emissions since Tier 3 standards will
result in GHG-reducing emissions of N2O and CH4 from light-duty vehicles and trucks. EPA
disagrees with this suggested approach and the final rule does not include changes in the
definition of baseline GHG emissions for refineries. The definition of baseline emissions for
PSD purposes is contained in the PSD regulations, which EPA did not propose to amend, or take
comment on, in this action. Nor did EPA propose any changes to the definition of stationary
source that would allow a stationary source to take credit for reductions occurring at offsite
mobile sources. Moreover, EPA's analysis of the possible permitting implications of the final
rule requirements for sulfur content indicated that the need for GHG permits will be limited to
relatively few refineries, and that these refineries and the relevant permitting authorities will be
able to issue the necessary permits in a timely manner. Also, we expect that GHG Best Available
Control Technology (BACT) requirements determined for this permitting of GHG emissions will
generally be the use of energy efficiency measures that refineries will naturally include in their
modifications anyway for economic reasons.
5.2.6. Flexibilities
5-100
-------
Tier 3 Summary and Analysis of Comments
What Commenter Said:
Commenters: Natural Resources Defense Council (NRDC)
An environmental group commented that due to the absence in the proposed Tier 3 rulemaking
of any changes to EPA's permitting regulations, it would be unwarranted for the final action to
create new flexibilities. The same commenter indicated its support for EPA efforts to help
ensure that air permits meeting current permitting requirements are issued in a timely manner.
Another environmental group also commented in opposition to any scheme to side-step any NSR
requirements. This commenter expressed particular concern about streamlining options that
might allow refineries to avoid the applicability of major NSR entirely and about any reliance on
presumptive BACT limits or general permits.
Our Response:
The final rule does not establish any new flexibilities or exceptions to current permitting
regulations. On an ongoing basis, EPA continues to consider ways to streamline the permitting
process consistent with CAA requirements and goals.
5.3. Standards for Denatured Fuel Ethanol and Other Oxygenates
5.3.1. Sulfur Standard
What Commenters Said:
Commenter: American Coalition for Ethanol (ACE)
ACE supports 10 ppm sulfur for denatured fuel ethanol as developed by ASTM with input from
oil companies and automakers.
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM)
In the event that EPA finalizes a more stringent average sulfur standard, API and AFPM support
EPA's proposal that producers of fuel ethanol, or other oxygenates, be required to also meet a
10-ppm sulfur cap.
In the event that EPA finalizes a more stringent 10-ppm refinery average sulfur standard, API
and AFPM support EPA's proposal within new §80.1610(a)(l) that producers of denatured fuel
ethanol (DFE), or other oxygenates, for use by oxygenate blenders be required to meet a 10-ppm
sulfur cap, as determined in accordance with the test requirements for refiners and importers.
Commenter: Chevron Products Company
5-101
-------
Chapter 5: Tier 3 Fuel Sulfur Program
Standards for Denatured Fuel Ethanol and Other Oxygenates: We support the Agency's proposal
that manufacturers of Denatured Fuel Ethanol (DFE) for use by oxygenate blenders would be
required to meet a 10-ppm sulfur cap. We note the Agency's comments that the California
regulations (Title 13, Section 2262.9) already require this maximum level of sulfur (among other
specifications) and that manufacturers of DFE have already been producing to the California
specification because of logistical difficulties in segregating ethanol destined for California from
other destinations. We, therefore, concur with the Agency's view that implementation of a 10-
ppm sulfur cap would not result in an additional burden on the ethanol industry.
Commenter: Growth Energy
EPA should change the proposed DFE 10 ppm sulfur cap to a 10 ppm sulfur average to put DFE
on a level playing field with unleaded.
In the past, EPA has regulated DFE with a sulfur limit that "reflects that 30-ppm refinery
average." EPA says that "consistent with" its new 10-ppm refinery sulfur standard, EPA
proposes "that manufacturers of DFE for use by oxygenate blenders would be required to meet a
10-ppm sulfur cap."
If DFE will have the same batch testing and registration requirements as gasoline, requiring a 10
ppm sulfur cap versus an average could disadvantage DFE versus gasoline, which can generate
and profit from sulfur credits. DFE having the same sulfur limits as gasoline, along with a
compliant petroleum blendstock, will result in ethanol blends having a sulfur limit that is
identical to the proposed finished gasoline specifications.
Furthermore, if the burden of implementing batch testing and reporting is imposed on ethanol
producers, they should be allowed to generate sulfur credits. EPA states that "While certain
batches of ethanol could theoretically be low enough in sulfur to generate credits, it is our desire
to limit credit generation to companies required to comply with the proposed Tier 3 sulfur
standards, i.e., refiners and importers.'
Growth Energy believes that ethanol producers should not be required to undertake batch testing;
however, if DFE producers are required to complete the batch testing, reporting, and record
keeping on DFE similar to refiner requirements on gasoline, they should have the same ability to
generate or use credits.
EPA mistakenly states that "since many refiners currently comply with our standards taking into
consideration the fuel property changes expected as a result of downstream ethanol blending,
providing ethanol blenders with sulfur credit would result in double counting the effects of
ethanol." However, double counting would not result if refiners are assuming that DFE has a
sulfur level of 10 ppm. Credits would only be generated at levels that are below this 10 ppm
level.
EPA attempts to further reject ethanol generating sulfur credits by stating that "Over compliance
with the per-gallon cap would not be a valid basis for credit generation, as you would expect that
in all cases the DFE would be below the cap. To allow credit generation, we would need to
5-102
-------
Tier 3 Summary and Analysis of Comments
propose an additional annual average sulfur standard for DFE at some level below 10 ppm, and
allow credits to be generated for over compliance with that standard." Ethanol producers should
have the same ability to generate credits if they have the same testing and reporting standards. In
order to remain price-competitive with gasoline, ethanol should be allowed to generate credits in
a similar manner as gasoline. Furthermore, such credit trading could allow ethanol producers to
offset the cost of [new] DFE testing and reporting requirements.
Commenter: Independent Fuel Terminal Operators Association (IFTOA)
Current gasoline requirements include the prohibition on blending gasoline with denatured
ethanol that has a sulfur content higher than 30 ppm. EPA is proposing to amend that standard
and not allow blending with denatured ethanol that has sulfur content higher than 10 ppm. The
Association supports this proposed change. If denatured ethanol contained 30 ppm sulfur, it
could be difficult for refiners and importers to meet the annual average standard of 10 ppm.
Recommendation: The Association recommends that EPA adopt its proposal to lower the sulfur
content of denatured ethanol. Reduce the sulfur content of denatured ethanol to 10 ppm to
facilitate meeting the proposed Tier 3 standard.
Commenter: Irving Oil Terminals Inc.
Under current gasoline regulations, gasoline may not be blended with denatured fuel ethanol that
has a sulfur content greater than 30 ppm. Under the Tier 3 proposal, the ethanol sulfur content
would be reduced to 10 ppm consistent with the proposed standard for gasoline. Irving Oil
supports this proposed amendment and believes that such a reduction would be essential if the 10
ppm sulfur annual average is adopted for gasoline. If the sulfur standard for denatured fuel
ethanol were left at its current level of 30 ppm, refiners would have to further assess and
determine if greater capital investments would have to be made to their FCC gasoline post-treater
units or additional investments be made to adjust and lower the sulfur content of other gasoline
blending streams to prevent violation of the ultimate 10 ppm annual average standard.
Commenter: Marathon Petroleum Company LP (MFC)
In the event that EPA finalizes a more stringent 10-ppm refinery average sulfur standard, MFC
supports EPA's proposal within new §80.1610(a)(l) that producers of denatured fuel ethanol
(DFE), or other oxygenates, for use by oxygenate blenders be required to meet a 10-ppm sulfur
cap, as determined in accordance with the test requirements for refiners and importers.
Commenter: National Corn Growers Association (NCGA)
EPA proposed that manufacturers of DFE for use by oxygenate blenders meet a 10 ppm sulfur
cap, and that only natural gasoline, gasoline, and gasoline blendstocks for oxygenate blending
(BOB) be used as ethanol denaturants. NCGA supports EPA's proposal for DFE standards. We
believe it would be unnecessary and burdensome to adopt California's additional specifications
for DFE.
5-103
-------
Chapter 5: Tier 3 Fuel Sulfur Program
Commenter: New York State Department of Environmental Conservation
The proposed limits on the sulfur content of blending streams such as denatured fuel ethanol are
necessary to ensure that the program benefits are actually realized.
Commenter: POET, LLC
EPA should change the proposed DFE 10 ppm sulfur cap to a 10 ppm sulfur average to put DFE
on a level playing field with unleaded.
EPA has in the past regulated DFE with a sulfur limit that 'reflects the current 30-ppm refinery
average. (67) EPA says that 'consistent with' its new 10-ppm refinery sulfur standard, EPA
proposes 'that manufacturers of DFE for use by oxygenate blenders would be required to meet a
10-ppm sulfur cap. (68) EPA also proposes that DFE manufacturers would be subject to batch
testing (including batch volume, sulfur content, and denaturant concentration as applicable). (69)
While the final ethanol blends (e.g., E30, E85, or some other blend ratio from E16 to E85) would
not necessarily be subject to batch testing (under the Blender Option, below), EPA here proposes
that the DFE would be tested.
If DFE will have the same batch testing and registration requirements as gasoline, requiring a 10
ppm sulfur cap versus an average could disadvantage DFE versus gasoline which can generate
and profit from sulfur credits. DFE having the same sulfur limits as gasoline, along with a
compliant petroleum blendstock, will result in ethanol blends (whether MLEBs or HLEBs)
having a sulfur limit that is identical to the proposed finished gasoline specifications.
Furthermore, if the burden of implementing batch testing and reporting is imposed on DFE
producers, they should be allowed to generate sulfur credits. EPA states that 'While certain
batches of ethanol could theoretically be low enough in sulfur to generate credits, it is our desire
to limit credit generation to companies required to comply with the proposed Tier 3 sulfur
standards, i.e., refiners and importers.' (70) If DFE producers are required to complete the batch
testing, reporting, and record keeping on DFE similar to refiner requirements on gasoline, they
should have the same ability to generate or use credits.
EPA mistakenly states that 'since many refiners currently comply with our standards taking into
consideration the fuel property changes expected as a result of downstream ethanol blending,
providing ethanol blenders with sulfur credit would result in double counting the effects of
ethanol.' (71) However, double counting would not result if refiners are assuming that DFE has a
sulfur level of 10 ppm. Credits would only be generated at levels that are below this 10 ppm
level.
EPA attempts to further reject ethanol generating sulfur credits by stating that 'Over compliance
with the per-gallon cap would not be a valid basis for credit generation, as you would expect that
in all cases the DFE would be below the cap. To allow credit generation, we would need to
propose an additional annual average sulfur standard for DFE at some level below 10 ppm, and
allow credits to be generated for over compliance with that standard.' (72) Ethanol producers
should have the same ability as refiners to generate credits if they have the same testing and
5-104
-------
Tier 3 Summary and Analysis of Comments
reporting standards. In order to remain price-competitive with gasoline, ethanol should be
allowed to generate credits in a similar manner as gasoline. Furthermore, such credit trading
could allow ethanol producers to offset the cost of new DFE testing and reporting requirements.
Commenter: Renewable Fuels Association (RFA)
This proposed rulemaking adds the regulatory burden of a 10 ppm sulfur maximum on a batch
basis for Denatured Fuel Ethanol (DFE). However, DFE does not have a sulfur issue. The
hydrocarbon denaturant is typically the sole source of sulfur in DFE.
Although current gasoline requirements include the prohibition on blending gasoline with DFE
that has sulfur content higher than 30 ppm, the State of California has long required that DFE
meet a maximum 10 ppm sulfur content standard. In 2002, RFA conducted an industry survey
that demonstrated DFE manufacturers were meeting the new California requirement of 10 ppm
sulfur content. Since then, RFA has recommended to the ethanol industry that all DFE meet this
California requirement and all indications are that ethanol producers are adhering to this
recommendation.
Because DFE already meets the proposed 10 ppm sulfur standard in nearly all cases, EPA
believes that the implementation of a 10-ppm sulfur cap for DFE would not result in an increased
regulatory or administrative burden for ethanol producers. We agree that there would be no new
burden in terms of ensuring DFE complies with the 10 ppm maximum sulfur limit.
Commenter: Sutherland Asbill & Brennan LLP (Sutherland)
Producers of denatured fuel ethanol and other oxygenates also would be required to meet the
annual average of 10 ppm sulfur.
Our Response:
The Tier 2 gasoline requirements include the prohibition on blending gasoline with DFE
that has sulfur content higher than 30 ppm.3 This requirement reflects the 30 ppm refinery
gasoline average sulfur requirement under the Tier 2 program. Since the sulfur content of DFE
was typically below 30 ppm, addition of DFE to gasoline would typically lower the sulfur
content of the mixture relative to gasoline. Hence, additional regulatory requirements regarding
the sulfur content of DFE were not necessary under the Tier 2 program to ensure DFE addition
did not increase gasoline sulfur levels. With the introduction of the 10 ppm average sulfur
standard for gasoline under the Tier 3 program, additional control of DFE sulfur content is
needed to ensure that that the addition of DFE to gasoline does not result in an increased sulfur
level for the mixture. Hence, consistent with the approach under the Tier 2 program and the 10
ppm refinery average sulfur standard for gasoline finalized under the Tier 3 program, the Tier 3
FRM requires that producers of DFE and other oxygenates for use by oxygenate blenders meet a
10 ppm sulfur cap beginning January 1, 2017.
! 40 CFR 80.385(e).
5-105
-------
Chapter 5: Tier 3 Fuel Sulfur Program
In demonstrating compliance with the Tier 3 program's 10 ppm average gasoline sulfur
standard, gasoline refiners and importers may adjust the sulfur levels in the gasoline and BOBs
that they produce/import to account for the downstream addition of ethanol. Therefore, the
sulfur level of DFE has direct effect on the extent of the desulfurization measures that a
refiner/importer will have to undertake to comply with the gasoline sulfur standards finalized
today.
EPA believes that requiring DFE to comply with a 10 ppm sulfur cap is the most
appropriate means of ensuring that finished gasoline blends attain the sulfur control goals of the
Tier 3 program. Controlling the sulfur content of DFE can be readily accomplished by limiting
the sulfur content of the denaturant that is used. Neat ethanol produced with standard quality
control practices should have negligible sulfur content. Denaturants with sufficiently low-sulfur
content to facilitate compliance with the 10 ppm sulfur cap for DFE are widely available.
Allowing DFE to exceed 10 ppm would result in the use of higher-sulfur denaturants, thereby
increasing gasoline refiner capital costs to install desulfurization equipment. As discussed in
Chapter 5.5.1 of this Summary and Analysis of Comments document, there are several reasons
why we do not believe that it is appropriate to expand the ABT provisions to include ethanol
producers and importers. Furthermore, as discussed in Section V.G.4. of the preamble to the
Tier 3 FRM and Chapter 5.3.3 of this Summary and Analysis of Comments document, sulfur
testing on each batch of DFE will not be required provided that the DFE producer or importer
demonstrates compliance with the 10 ppm sulfur cap for DFE with volumetric blending records,
whereas an average standard would require testing of every batch. We anticipate that DFE
producers and importers will typically choose to demonstrate compliance with the 10 ppm sulfur
cap using volumetric blending records rather than per-batch sulfur testing. Therefore, we do not
anticipate that DFE producers and importers will need to install additional sulfur testing
equipment as a result of the Tier 3 final rule. The registration, reporting, and recordkeeping
requirements for DFE producers that we are finalizing will not necessitate substantial capital
investments. Hence, there is no need to extend the flexibility of meeting an annual average
sulfur standard and participation in the ABT program to ethanol producers and importers to help
facilitate their compliance as exists for gasoline refiners.
5.3.2. Requirements to Address the Potential Emissions Effects of Fuel Parameters Other
than Sulfur in DFE
5.3.2.1. Specifications on the Benzene, Olefins, and Aromatics Content of DFE and
the Denaturants Used in the Manufacture of DFE
What Commenters Said:
Commenter: American Coalition for Ethanol (ACE)
4 Accounting for the effect of oxygenate added downstream of the refinery or import facility in demonstrating
compliance with the average gasoline sulfur standard is addressed in regulations finalized today at §80.1603(d). See
Section V.C. in the preamble to the Tier 3 FRM regarding the sulfur level in DFE that must be used by refiners and
importers in making this compliance determination.
5-106
-------
Tier 3 Summary and Analysis of Comments
Standards for Denatured Fuel Ethanol: Again, in keeping with EPA's desire to test vehicles on
fuel available in the marketplace, we believe it is unnecessary to adopt California's specialized
requirements.
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM)
We agree EPA should not set limits for benzene, olefms, and aromatics content of ethanol.
API and AFPM agree with EPA's plans to not propose limits for benzene, olefms, and aromatics
content of DFE. We agree that ASTM International Specification D4806 in combination with
TTB denaturant requirements and low denaturant concentrations (typically 2 volume
percent) would limit the benzene, olefms, and aromatics content of DFE to very low levels. We
also do not support the adoption of the State of California's benzene, olefm and aromatics
specifications for DFE. Limits imposed by the California Air Resources Board (CARB) address
concerns specific to California, and hence should not be applied to the remainder of the country.
Commenter: Chevron Products Company
The Agency also requests comment on whether it should adopt California's limits on benzene,
olefms and aromatics for DFE. The Agency is not proposing any limits on these parameters. Our
view is that the Agency should adopt California's requirements in their entirety. Since many
ethanol producers are already manufacturing to a single national 'specification' to meet the
California requirements, such action by the Agency would not impose any additional burden on
them while offering the benefit of formalizing the 'de facto' application of a single DFE
specification nationwide.
Many of the questions that the Agency seeks input on, can be simply resolved via the adoption of
the California DFE specifications, including whether additional denaturants should be allowed
(beyond, natural gasoline, gasoline, and gasoline blendstocks) and the range of allowable
denaturant content.
Commenter: Growth Energy
It is unnecessary to establish benzene, aromatics and olefm limits on DFE.
The Proposed Rule states that limiting denaturants (e.g., to natural gasoline, gasoline, and BOBs
"along with Internal Revenue Service ethanol denaturant requirements would limit benzene,
olefms, and aromatics content of DFE to very low levels." Growth Energy agrees with EPA's
"not proposing any limits on these parameters in DFE."
We further agree that the California DFE standards for benzene, aromatics, and olefms not be
adopted. Not only would this impose a testing and record keeping burden on DFE producers, but
it would not be meaningful given that there are no aromatics or olefm limitations for gasoline
components (outside of base and certification fuel standards). (10)
5-107
-------
Chapter 5: Tier 3 Fuel Sulfur Program
10 The California regulations are also inappropriately stringent. California requires ethanol benzene
levels to be 0.06 percent or lower and the denaturant to contain 1.1 percent benzene maximum. This
benzene level in the DFE is 10 percent of the 0.62 percent benzene limit for unleaded gasoline. Ethanol
should not bear a burden so strict compared to gasoline. Also, many natural gasoline streams do not meet
this 1.1 percent benzene specification, as the benzene content is related to the natural gasoline source
stream and processing. Many locations need to use denaturants with non-California-compliant benzene
content. Using California benzene limits would increase costs as many companies would be forced to use
far more expensive denaturants.
Commenter: Marathon Petroleum Company LP (MFC)
MFC agrees with EPA's plans to not propose limits for benzene, olefins, and aromatics content
of DFE. We agree that ASTM International Specification D4806 in combination with TTB
denaturant requirements and low denaturant concentrations (typically 2 volume percent) would
limit the benzene, olefins, and aromatics content of DFE to very low levels. We also do not
support the adoption of the State of California's benzene, olefm and aromatics specifications for
DFE. Limits imposed by the California Air Resources Board (CARB) address concerns specific
to California, and hence should not be applied to the remainder of the country.
Commenter: National Corn Growers Association (NCGA)
Standards for Denatured Fuel Ethanol (DFE)
NCGA supports EPA's proposal for DFE standards. We believe it would be unnecessary and
burdensome to adopt California's additional specifications for DFE.
Commenter: POET, LLC
POET agrees with EPA that it is unnecessary to establish benzene, aromatics and olefm limits on
DFE.
The Proposed Rule states that limiting denaturants (e.g., to natural gasoline, gasoline, and BOBs)
'along with Internal Revenue Service ethanol denaturant requirements would limit benzene,
olefins, and aromatics content of DFE to very low levels.' (73) POET agrees with EPA's 'not
proposing any limits on these parameters in DFE.' (74)
POET further agrees with EPA's assessment that it should not adopt the California DFE
standards for benzene, aromatics, and olefins. (75) Not only would this impose a testing and
record keeping burden on DFE producers, but it would not be meaningful given that there are no
aromatics or olefm limitations for gasoline components (outside of base and certification fuel
standards). (76)
76 - The California regulations are also inappropriately stringent. California requires ethanol benzene
levels to be 0.06% or lower and the denaturant to contain 1.1% benzene maximum. This benzene level in
the DFE is 10% of the 0.62% benzene limit for unleaded gasoline. Ethanol should not bear a burden so
strict compared to gasoline. Also, many natural gasoline streams do not meet this 1.1% benzene
specification, as the benzene content is related to the natural gasoline source stream and processing. Many
locations need to use denaturants with non-California-compliant benzene content. Using California
5-108
-------
Tier 3 Summary and Analysis of Comments
benzene limits would increase costs as many companies would be forced to use far more expensive
denaturants.
Commenter: Renewable Fuels Association (RFA)
We agree that parametric limits on benzene, olefins, and aromatics in DFE are unnecessary, as
these substances are typically absent from DFE or are present at extremely low levels as a result
of denaturing. RFA agrees no redundancy is needed in restricting the benzene, aromatics and
olefins allowed in denatured fuel ethanol. Benzene, aromatics and olefins found in denatured fuel
ethanol are strictly a product of the mandatory TTB denaturant addition.
Our Response:
We requested comment on adopting the additional fuel specifications currently in force
for DFE in the State of California. The California requirements for DFE include maximum
specifications for benzene, olefins, and aromatics as well as sulfur.5 We believe that limitation
on denaturant concentration and other requirements for DFE that we are finalizing as described
in Section V.G. of the Tier 3 final rule preamble, and Chapters 5.3.2.2 and 5.3.2.3 of this
Summary and Analysis of Comments document, are sufficient to limit the potential emissions
effects of fuel parameters other than sulfur in DFE at this time. Therefore, we are not finalizing
benzene, aromatics, and olefins specifications for DFE (or certified ethanol denaturant) at this
time.
5.3.2.2. Limitations on the Types of Denaturants Used in the Manufacture of DFE
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel &
Petrochemical Manufacturers (AFPM)
API and AFPM do not support limiting the products that can be used as alcohol denaturants, and
recommend that if EPA must act, the latest version of ASTM International Specification D4806
should be adopted, and EPA should not further narrow the list of available denaturants.
API and AFPM do not support EPA's proposal to restrict the number of available denaturants
within new §80.1610(a)(3) to gasoline, RBOB, CBOB or natural gas liquids. As EPA points-out
in the preamble, ASTM International Specification D4806 provides for the use of natural
gasoline, gasoline blendstocks, and gasoline, as denaturants. Furthermore, the State of California
also approves the use of gasoline components, as well as natural gasoline and unleaded gasoline,
as DFE denaturants. Contrary to the preamble, EPA's proposal does not adopt the same
specification as ASTM D4806 by restricting the scope of available gasoline blendstocks to
RBOB and CBOB. This narrowing will restrict the potential supply of available denaturants for
producers of denatured fuel ethanol, which are also required to meet U.S. Department of
Treasury Alcohol and Tobacco Tax and Trade Bureau (TTB) denaturant requirements and State
: California Code of Regulations 13 CCR section 2262.9.
5-109
-------
Chapter 5: Tier 3 Fuel Sulfur Program
of California DFE requirements, when DFE is supplied to California. Although EPA expresses
concern that denaturant limitations are needed to prevent the use of other components that
"might adversely impact vehicle emission performance", data supporting this presumption is not
presented. API and AFPM believe that with the combination of low denaturant concentrations
(typically 2 volume percent) and TTB denaturants restrictions, any components of concern
would be at very low levels and would unlikely impact vehicle emissions performance. For
simplicity, API and AFPM recommend that if EPA must act, the latest version of ASTM
International Specification D4806 should be adopted, and EPA should not further narrow the list
of available denaturants.
Commenter: Chevron Products Company
Many of the questions that the Agency seeks input on, can be simply resolved via the adoption of
the California DFE specifications, including whether additional denaturants should be allowed
(beyond, natural gasoline, gasoline, and gasoline blendstocks) and the range of allowable
denaturant content. We believe there is sufficient flexibility in the slate of available potential
denaturants that ethanol producers can utilize while adhering to ASTM D4806 requirements and
do not see a benefit in broadening the list to include additional materials. The current list
represents the material that would be most commonly available to ethanol producers and, thus,
there is little (if any) incremental flexibility in allowing additional ones as potential denaturant
'options.' By the same token, maintaining the list as it currently exists would not present a new
burden on DFE producers. Our recommendation to the Agency is to simply follow the
requirements of ASTM in this regard since that is the right technical forum where any proposed
addition of new denaturants to the specification should be entertained.
Commenter: Flint Hills Resources, LP (FHR)
In summary, should a more stringent standard be finalized as proposed by EPA, FHR believes
that other gasoline blendstocks should be allowed to be used as denaturants for DFE.
Other Gasoline Blendstocks Should be Allowed for Use as Denaturants: FHR does not support
EPA's proposal to restrict the number of available denaturants within new §80.1610(a)(3) to only
gasoline, RBOB, CBOB or natural gas liquids. FHR recommends that if EPA must act, the latest
version of ASTM Standard Specification D4806 for Denatured Fuel Ethanol (DFE) should be
adopted, similar to the certification test fuels requirement proposed within §80.177(b)(i).
As EPA points-out in the preamble, ASTM D4806 provides for the use of natural gasoline,
gasoline blendstocks or unleaded gasoline, as denaturants. In addition, the State of California
also approves the use of gasoline components, as well as natural gasoline and unleaded gasoline.
Contrary to the preamble, EPA's proposal does not adopt the same specification as ASTM
D4806 by restricting the scope of available gasoline blendstocks to RBOB and CBOB. This
narrowing will restrict the potential supply of available denaturants to DFE producers, which
must also meet U.S. Department of Treasury Alcohol and Tobacco Tax and Trade Bureau (TTB)
denaturant requirements and State of California DFE requirements, when supplied to California.
Although EPA expresses concern that denaturant limitations are needed to prevent the use of
other components that 'might adversely impact vehicle emission performance', data supporting
5-110
-------
Tier 3 Summary and Analysis of Comments
this presumption is not presented. FHR believes that with the combination of low denaturant
concentrations (typically 2 volume percent) and ASTM D4806 specifications and TTB
denaturant restrictions, any potential components of concern would be at very low levels and
would unlikely impact vehicle emissions performance.
Commenter: Growth Energy
EPA should not limit denaturants to only certified BOBs, certified gasoline and natural gasoline.
EPA limiting denaturants to only certified BOBs, certified gasoline, and natural gasoline would
unnecessarily reduce denaturant flexibility to prevent use ofhydrotreatedor refinery produced
streams that are readily available for gasoline blending or may be available for future use.
Several denaturants are in use today that do not meet the Alcohol Tobacco Tax and Trade Bureau
(TTB) or EPA definition of natural gasoline.
Historically DFE producers have used refinery gasoline blendstocks such as "light straight run"
and "light naphtha" as denaturants. Removal of this flexibility will increase costs for DFE
producers. Especially if the sulfur limit for DFE is reduced (i.e., from 30 ppm to 10 ppm) and the
industry is forced to look at hydrotreated streams not covered in the natural gasoline definitions.
Importantly, "gasoline blendstocks" and "gasoline blendstocks for oxygenate blending (BOB)"
are very different and adopting the stricter BOB requirement will have negative economic effects
on the ethanol industry.
Instead, EPA should allow a wide range of hydrocarbons for denaturant as long as they meet the
TTB requirements and the final DFE meets the required benzene and sulfur specifications. The
TTB requirements at 27 CFR § 19.747 allow for substantial flexibility in denaturant use.28 EPA
should follow similar provisions as the TTB, or incorporate them by cross-reference.
The proposed definition ofNGLs in 40 CFR 80.2 is too restrictive. The definition should not
include an unduly narrow listing of the types of facilities that produce NGLs and should include,
for instance, NGLs that may be produced through refineries and hydro-treating. The definition
should be revised as follows (with additions underlined): "(zzz) Natural Gas Liquids (NGL)
means the components of natural gas (primarily propane, butane, pentane, hexane, and heptane)
that are separated from the gas state in the form of liquids in facilities such as a natural gas
liquids fractionation production facility, or in a natural gas processing plant, or in a natural gas
pipeline, or refinery or similar facility. The higher temperature boiling components of NGL are
sometimes also referred to as "natural gasoline"."
28 In particular, the TTB requirements at The TTB requirements at 27 CFR § 19.747 provide that "If a
proprietor wishes to use a material to render spirits unfit for beverage use that is not authorized under §
19.746 or that is not on the published list of materials, the proprietor may submit an application for
approval to the appropriate TTB officer. The application must include the name of the material and the
quantity of material that the proprietor proposes to add to each 100 gallons of spirits. The appropriate
TTB officer may require the proprietor to submit an 8-ounce sample of such material. The proprietor may
not use any proposed material until the appropriate TTB officer approves its use. Any material that
impairs the quality of the spirits for fuel use will not be approved. The proprietor must retain as part of the
5-111
-------
Chapter 5: Tier 3 Fuel Sulfur Program
records available for inspection by appropriate TTB officers any application approved by the appropriate
TTB officer under this section."
Commenter: Marathon Petroleum Company LP (MFC)
MFC does not support EPA's proposal to restrict the number of available denaturants within new
§80.1610(a)(3) to gasoline, RBOB, CBOB or natural gas liquids. As EPA points-out in the
preamble, ASTM International Specification D4806 provides for the use of natural gasoline,
gasoline blendstocks, and gasoline, as denaturants. Furthermore, the State of California also
approves the use of gasoline components, as well as natural gasoline and unleaded gasoline, as
DFE denaturants. Contrary to the preamble, EPA's proposal does not adopt the same
specification as ASTM D4806 by restricting the scope of available gasoline blendstocks to
RBOB and CBOB. This narrowing will restrict the potential supply of available denaturants for
producers of denatured fuel ethanol, which are also required to meet U.S. Department of
Treasury Alcohol and Tobacco Tax and Trade Bureau (TTB) denaturant requirements and State
of California DFE requirements, when DFE is supplied to California. Although EPA expresses
concern that denaturant limitations are needed to prevent the use of other components that
"might adversely impact vehicle emission performance", data supporting this presumption is not
presented. MFC believes that with the combination of low denaturant concentrations (typically 2
volume percent) and TTB denaturants restrictions, any components of concern would be at very
low levels and would unlikely impact vehicle emissions performance. For simplicity, MFC
recommends that if EPA must act, the latest version of ASTM International Specification D4806
should be adopted, and EPA should not further narrow the list of available denaturants.
Commenter: National Corn Growers Association (NCGA)
EPA proposed that manufacturers of DFE for use by oxygenate blenders meet a 10 ppm sulfur
cap, and that only natural gasoline, gasoline, and gasoline blendstocks for oxygenate blending
(BOB) be used as ethanol denaturants. NCGA supports EPA's proposal for DFE standards.
Commenter: POET, LLC
EPA should change the proposed DFE 10 ppm sulfur cap to a 10 ppm sulfur average to put DFE
on a level playing field with unleaded.
EPA has in the past regulated DFE with a sulfur limit that 'reflects the current 30-ppm refinery
average. (67) EPA says that 'consistent with' its new 10-ppm refinery sulfur standard, EPA
proposes 'that manufacturers of DFE for use by oxygenate blenders would be required to meet a
10-ppm sulfur cap. (68) EPA also proposes that DFE manufacturers would be subject to batch
testing (including batch volume, sulfur content, and denaturant concentration as applicable). (69)
While the final ethanol blends (e.g., E30, E85, or some other blend ratio from E16 to E85)
would not necessarily be subject to batch testing (under the Blender Option, below), EPA here
proposes that the DFE would be tested.
If DFE will have the same batch testing and registration requirements as gasoline, requiring a 10
ppm sulfur cap versus an average could disadvantage DFE versus gasoline which can generate
and profit from sulfur credits. DFE having the same sulfur limits as gasoline, along with a
5-112
-------
Tier 3 Summary and Analysis of Comments
compliant petroleum blendstock, will result in ethanol blends (whether MLEBs or HLEBs)
having a sulfur limit that is identical to the proposed finished gasoline specifications.
Furthermore, if the burden of implementing batch testing and reporting is imposed on DFE
producers, they should be allowed to generate sulfur credits. EPA states that 'While certain
batches of ethanol could theoretically be low enough in sulfur to generate credits, it is our desire
to limit credit generation to companies required to comply with the proposed Tier 3 sulfur
standards, i.e., refiners and importers.' (70) If DFE producers are required to complete the batch
testing, reporting, and record keeping on DFE similar to refiner requirements on gasoline, they
should have the same ability to generate or use credits.
EPA mistakenly states that 'since many refiners currently comply with our standards taking into
consideration the fuel property changes expected as a result of downstream ethanol blending,
providing ethanol blenders with sulfur credit would result in double counting the effects of
ethanol.' (71) However, double counting would not result if refiners are assuming that DFE has a
sulfur level of 10 ppm. Credits would only be generated at levels that are below this 10 ppm
level.
EPA attempts to further reject ethanol generating sulfur credits by stating that 'Over compliance
with the per-gallon cap would not be a valid basis for credit generation, as you would expect that
in all cases the DFE would be below the cap. To allow credit generation, we would need to
propose an additional annual average sulfur standard for DFE at some level below 10 ppm, and
allow credits to be generated for over compliance with that standard.' (72) Ethanol producers
should have the same ability as refiners to generate credits if they have the same testing and
reporting standards. In order to remain price-competitive with gasoline, ethanol should be
allowed to generate credits in a similar manner as gasoline. Furthermore, such credit trading
could allow ethanol producers to offset the cost of new DFE testing and reporting requirements.
POET agrees with EPA that it is unnecessary to establish benzene, aromatics and olefm limits on
DFE.
The Proposed Rule states that limiting denaturants (e.g., to natural gasoline, gasoline, and BOBs)
'along with Internal Revenue Service ethanol denaturant requirements would limit benzene,
olefms, and aromatics content of DFE to very low levels.' (73) POET agrees with EPA's 'not
proposing any limits on these parameters in DFE.' (74)
POET further agrees with EPA's assessment that it should not adopt the California DFE
standards for benzene, aromatics, and olefms. (75) Not only would this impose a testing and
record keeping burden on DFE producers, but it would not be meaningful given that there are no
aromatics or olefm limitations for gasoline components (outside of base and certification fuel
standards). (76)
EPA should not limit denaturants to only certified BOBS, certified gasoline, and natural
gasoline.
5-113
-------
Chapter 5: Tier 3 Fuel Sulfur Program
EPA limiting denaturants to only certified BOBs, certified gasoline, and natural gasoline would
unnecessarily reduce denaturant flexibility to prevent use of hydrotreated or refinery produced
streams that are readily available for gasoline blending or may be available for future use.
Several denaturants are in use today that do not meet the Alcohol Tobacco Tax and Trade Bureau
(TTB) or EPA definition of natural gasoline.
Historically DFE producers have used refinery gasoline blendstocks such as 'light straight run'
and 'light naphtha' as denaturants. Removal of this flexibility will increase costs for DFE
producers, especially if the sulfur limit for DFE is reduced (i.e., from 30 ppm to 10 ppm) and the
industry is forced to look at hydrotreated streams not covered in the natural gasoline definitions.
Importantly, 'gasoline blendstocks' and 'gasoline blendstocks for oxygenate blending (BOB)'
are very different and adopting the stricter BOB requirement will have negative economic effects
on the ethanol industry.
Instead, EPA should allow a wide range of hydrocarbons for denaturant as long as they meet the
TTB requirements and the final DFE meets the required benzene and sulfur specifications. The
TTB requirements at 27 CFR § 19.747 allow for substantial flexibility in denaturant use. (94)
EPA should follow similar provisions as the TTB, or incorporate them by cross-reference.
94 - In particular, the TTB requirements at The TTB requirements at 27 CFR § 19.747 provide that 'If a
proprietor wishes to use a material to render spirits unfit for beverage use that is not authorized under §
19.746 or that is not on the published list of materials, the proprietor may submit an application for
approval to the appropriate TTB officer. The application must include the name of the material and the
quantity of material that the proprietor proposes to add to each 100 gallons of spirits. The appropriate
TTB officer may require the proprietor to submit an 8-ounce sample of such material. The proprietor may
not use any proposed material until the appropriate TTB officer approves its use. Any material that
impairs the quality of the spirits for fuel use will not be approved. The proprietor must retain as part of the
records available for inspection by appropriate TTB officers any application approved by the appropriate
TTB officer under this section.'
We support the proposal to limit approved denaturants for denatured fuel ethanol to natural
gasoline, gasoline, and gasoline blendstocks for oxygenate blending (BOBs), however we would
like to see an option for proposing a new denaturant in the future.
Commenter: Renewable Fuels Association (RFA)
As for allowable denaturants for denatured fuel ethanol, RFA recommends that EPA recognize
that the Alcohol and Tobacco Tax and Trade Bureau (TTB) and ASTM carefully restrict the
allowable materials that can be used as denaturant. RFA also recommends that EPA recognize
future options for denaturant and allow for the possibility of approving non-hydrocarbon origin
denaturants. Regulatory schemes such as the California Low Carbon Fuels Standard promote the
use of low carbon intensity fuels and thus nontraditional denaturants may be needed in the future.
The ethanol industry should be granted as much flexibility as possible to meet the regulatory
burden of the denaturing requirements of the TTB while continuously improving ethanol's
performance in a spark ignition engine. EPA adding identical effectual denaturant requirements
is not helpful for the industry or an improvement of air quality.
5-114
-------
Tier 3 Summary and Analysis of Comments
Commenter: Private Citizen
Approve denaturants that are non-petroleum based for Alcohol Fuel.
Our Response:
To limit the variability in DFE composition and the associated potential impact on
vehicle emissions, we proposed to allow the use of only certified gasoline, gasoline blendstocks
for oxygenate blending (BOBs), and natural gasoline as denaturants.
The ASTM specification for DFE requires that the only denaturants that can be used are
gasoline, gasoline blendstocks, and natural gasoline. The State of California incorporated the
ASTM limits on allowable denaturants into its regulations for DFE by reference.6 Thus, both
ASTM and the State of California allow the use of gasoline blendstocks other than BOBs as
denaturants.
While industry standards are typically focused on performance concerns, EPA continues
to believe that it is appropriate to implement additional controls to address the potential impact
of fuel components in denaturants other than sulfur on vehicle emissions. In particular, EPA is
concerned about the use of denaturants that might have a high concentration of benzene or other
aromatics. We believe that limitation on denaturant concentration and other requirements for
DFE that we are finalizing as described in Section V.G. of the Tier 3 final rule preamble, and
Chapter 5.3.2.3 of this Summary and Analysis of Comments document, are sufficient to limit the
potential emissions effects of fuel parameters other than sulfur in DFE at this time. Therefore,
we are not finalizing additional controls on the types of denaturants that may be used beyond
those currently adopted by ASTM and the State of California at this time. In response to
comments, the Tier 3 FRM includes the requirement that only gasoline, gasoline blendstocks,
and natural gasoline liquids may be used to denature DFE.7 This requirement is essentially the
same as the current ASTM and State of California specifications for the type of denaturants that
may be used. In response to the comment that the proposed definition of natural gasoline liquids
was too restrictive, we are finalizing the following definition: Natural gas liquids (NGL) means
the components of natural gas (primarily propane, butane, pentane, hexane, and heptane) that are
separated from the gas state in the form of liquids in facilities such as a natural gas production
facility, a gas processing plant, a natural gas pipeline, or a refinery or similar facility. The higher
temperature boiling components of NGL are sometimes referred to as "natural gasoline".
We will continue to monitor the need for additional denaturants, and when appropriate
EPA may undertake a future rulemaking to evaluate the suitability of additional denaturants and
consider allowing their use.
5.3.2.3. Limitation of Denaturant Concentration
6 The California Code of Regulations references ASTM D 4806-99 which limits the allowed denaturants to gasoline,
gasoline components, and natural gasoline.
7 Finished gasoline used as denaturant must be compliant with the applicable EPA requirements.
5-115
-------
Chapter 5: Tier 3 Fuel Sulfur Program
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM)
And, we contend that EPA should not limit the concentration of denaturant to 2%.
Although EPA expresses concern that denaturant limitations are needed to prevent the use of
other components that "might adversely impact vehicle emission performance", data supporting
this presumption is not presented. API and AFPM believe that with the combination of low
denaturant concentrations (typically 2 volume percent) and TTB denaturants restrictions, any
components of concern would be at very low levels and would unlikely impact vehicle emissions
performance. For simplicity, API and AFPM recommend that if EPA must act, the latest version
of ASTM International Specification D4806 should be adopted,
API and AFPM contend that EPA does not need to limit the concentration of denaturant in DFE
to 2. volume percent within new §80.1610(a)(4). Under the RFS2 regulations, the denaturant
level of DFE must be limited to 2 volume percent, in order for a producer to generate Renewable
Identification Numbers (RINs) for each gallon of renewable fuel produced. API and AFPM fully
understand that DFE producers may decide to produce DFE and not generate RINs. However,
considering the mandate for RINs, it is very unlikely that a DFE producer would not generate
RINs. Furthermore, the TTB restricts the denaturant concentration to 5 volume percent, which
would also limit potential components of concern to very low levels and would unlikely impact
vehicle emissions performance. As a result, the inclusion of a 2 volume percent requirement
within the Tier 3 rules is unnecessary.
API and AFPM contend that EPA does not need to limit the concentration of denaturant in DFE.
Commenter: Chevron Products Company
Adoption of the California DFE requirements would also address the issue of allowable
denaturant content since the state's allowable range is 1.96 volume percent minimum to 4.76
volume percent maximum. The corresponding ASTM (D4806) range is 1.96-5 volume percent.
We would like the Agency to reconsider its proposal to limit maximum ethanol denaturant
content to 2 volume percent (even though we understand the Agency's recognizes that this 'limit'
would actually entail allowing up to 2.5 volume percent denaturant content). We see no value in
introducing yet a more restrictive requirement in the allowable denaturant range and dispute the
Agency's assertion that this is in any way a facilitation vis-a-vis the RFS2 requirements. The
minimum value is effectively set by the IRS' requirements at 1.96 volume percent and California
mirrors that value. The Agency's setting the maximum at 2 volume percent unnecessarily results
in nearly identical practical maximum and minimum values with only the rounding tolerance
remaining for some operating flexibility.
There is no reason that we can see why denaturant content in excess of 2 volume percent cannot
be 'backed out' of RFS2 REST calculations and IRS calculations, nor do we consider the process of
making those adjustments particularly onerous or burdensome. We disagree with the Agency's
5-116
-------
Tier 3 Summary and Analysis of Comments
assertion that denaturant producer registration and demonstration of compliance with the sulfur,
aromatics, benzene and olefm content requirements would be necessary if the maximum
denaturant content is not set at 2 volume percent. In our view, adoption of the California DFE
requirements in their entirety resolves this issue and obviates the need for such burdensome
administrative requirements on denaturant producers.
Commenter: Flint Hills Resources, LP (FFIR)
Although EPA expresses concern that denaturant limitations are needed to prevent the use of
other components that 'might adversely impact vehicle emission performance', data supporting
this presumption is not presented. FHR believes that with the combination of low denaturant
concentrations (typically 2 volume percent) and ASTM D4806 specifications and TTB
denaturant restrictions, any potential components of concern would be at very low levels and
would unlikely impact vehicle emissions performance.
Commenter: Growth Energy
EPA should not limit the denaturant content to 2 percent of DFE by volume. Denaturant limits
exist in other regulations and should not be limited to 2 percent.
EPA notes regarding this issue that it is "proposing to limit the maximum concentration of
denaturant that can be used in DFE to 2 volume percent."
Both ASTM and California allow for up to 5 percent denaturant, and EPA should not adopt a
more stringent limit. Ethanol producers are strongly incentivized by the RFS to keep denaturant
under 2 percent.30 However, under certain conditions the flexibility to have a slightly higher
denaturant content can be very important to ethanol producers. Accordingly, EPA should adopt
no more stringent than a 5 percent denaturant limit.
30 The Proposed Rule notes that "Under the RFS2 regulations, if the denaturant level is 2 volume percent or less
(effectively less than 2.5 volume percent considering rounding) the entire volume of denatured fuel ethanol can be
used for determining compliance with the RFS2 renewable fuel volume requirements." Id.
Commenter: Marathon Petroleum Company LP (MFC)
MFC contends that EPA does not need to limit the concentration of denaturant in DFE to 2.
volume percent within new §80.1610(a)(4). Under the RFS2 regulations, the denaturant level of
DFE must be limited to 2 volume percent, in order for a producer to generate Renewable
Identification Numbers (RINs) for each gallon of renewable fuel produced. MFC fully
understands that DFE producers may decide to produce DFE and not generate RINs. However,
considering the mandate for RINs, it is very unlikely that a DFE producer would not generate
RINs. Furthermore, the TTB restricts the denaturant concentration to 5 volume percent, which
would also limit potential components of concern to very low levels and would unlikely impact
vehicle emissions performance. As a result, the inclusion of a 2 volume percent requirement
within the Tier 3 rules is unnecessary.
Commenter: POET, LLC
5-117
-------
Chapter 5: Tier 3 Fuel Sulfur Program
Denaturant limits exist in other regulations and EPA limiting denaturant content to 2% of DFE
by volume is conflicting and unnecessary.
EPA notes regarding this issue that it is 'proposing to limit the maximum concentration of
denaturant that can be used in DFE to 2 volume percent.' (95)
Both ASTM and California allow for up to 5% denaturant, and EPA should not adopt a more
stringent limit. Ethanol producers are strongly incentivized by the RFS to keep denaturant under
2%. (96) However, under certain conditions the flexibility to have a slightly higher denaturant
content can be very important to ethanol producers. Accordingly, EPA should adopt no more
stringent than a 5% denaturant limit.
96 - The Proposed Rule notes that 'Under the RFS2 regulations, if the denaturant level is 2
volume percent or less (effectively less than 2.5 volume percent considering rounding) the entire
volume of denatured fuel ethanol can be used for determining compliance with the RFS2
renewable fuel volume requirements.' Id.
Commenter: Renewable Fuels Association (RFA)
Several existing regulations and specifications already govern this 2.% volume of denaturant in
our denatured fuel ethanol (e.g., ASTM, State of California, Alcohol Tax & Trade Bureau
(TTB)) that confine ethanol producers on their choices of denaturant.
RFA does not support EPA limiting the amount of denaturant. RFA does not think the EPA
should restrict the amount of denaturant content in DFE in this regulation. RFA thinks the
current ASTM specifications for denaturant content are adequate. The DFE specification ASTM
D4806 allows 5% denaturant content maximum. Further, the industry already has a regulatory
restriction under the RFS2 for denaturant content; the RFS2 requires that DFE contain no more
than 2. % denaturant for RIN generation. However, the option to use more denaturant when
economical has proven effective and manageable by the marketplace.
Commenter: Sutherland Asbill & Brennan LLP (Sutherland)
The Proposed Rule would also limit the maximum concentration of denaturant that can be used
in denatured fuel ethanol to 2 percent by volume. The Proposed Rule provides little tolerance in
light of the 1.96 percent minimum level of denaturant required by Alcohol and Tobacco Tax and
Trade Bureau regulations. A reasonable tolerance should be allowed so as to avoid inadvertent
compliance issues.
Our Response:
To limit the potential impact on vehicle emissions of fuel parameters in ethanol
denaturants other than sulfur, we proposed to limit denaturant concentration in DFE to a
maximum of 2 volume percent, which translates to 2.5 volume percent considering rounding.
5-118
-------
Tier 3 Summary and Analysis of Comments
California allows a maximum denaturant content of 5 volume percent consistent with the
r-M-, r-M-, -r^ &
industry consensus ASTM International (ASTM) specification for DFE.
Since we did not finalize benzene, olefm, or aromatics specifications for ethanol
denaturants similar to those in California as part of the Tier 3 FRM, we continue to believe that it
is appropriate to implement a more stringent limit on maximum denaturant concentration to
address concerns about the potential impact on vehicle emissions of fuel parameters in ethanol
denaturants other than sulfur. Denaturants with high benzene content could have a substantial
impact on the final benzene content of the finished ethanol-gasoline blend. Setting a more
stringent limit on denaturant concentration will ensure that harmful components such as benzene
potentially present in ethanol denaturants are adequately diluted in the finished fuel blend. We
agree with comments that it is appropriate to provide additional flexibility for the allowable
denaturant levels that may be used. Therefore, we are finalizing a 3.0 volume percent maximum
on ethanol denaturant concentration. This approach provides sufficient flexibility to DFE
producers while avoiding the need to impose additional testing burdens on denaturant and DFE
producers (e.g. for benzene and aromatics). Given the comments that DFE typically contains
approximately 2 volume percent denaturant, we expect that the 3.0 volume limit that we are
finalizing will have a minimal impact on the operations of DFE producers and importers.
5.3.3. Compliance Demonstration Including Registration, Testing, Product
Transfer Document, and Reporting Requirements
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM)
In the event that EPA finalizes a more stringent 10-ppm refinery average sulfur standard, API
and AFPM support EPA's proposal within new §80.1610(a)(l) that producers of denatured fuel
ethanol (DFE), or other oxygenates, for use by oxygenate blenders be required to meet a 10-ppm
sulfur cap, as determined in accordance with the test requirements for refiners and importers.
API and AFPM believe that cumbersome individual batch reporting for all fuels has little value.
API and AFPM contend that EPA does not need to limit the concentration of denaturant in DFE
API and AFPM believe that it is unnecessary for EPA to require manufacturers of denaturants to
register with EPA. We believe that requiring producers of DFE to meet the latest version of
ASTM International Specification D4806 and TTB denaturant requirements, along with very low
denaturant concentrations (typically 2 volume percent) would sufficiently limit the risk that
denaturants might adversely impact vehicle emissions. Therefore, the proposed registration of
denaturant manufacturers would not provide for any meaningful purpose.
8 ASTM International D4806-13a, "Standard Specification for Denatured Fuel Ethanol for Blending with Gasolines
for Use as Automotive Spark-Ignition Engine Fuel".
5-119
-------
Chapter 5: Tier 3 Fuel Sulfur Program
DFE has become a significant component within the gasoline fuel pool and should be subject to
requirements similar to gasoline refiners. However, API and AFPM believe that individual batch
reporting for all fuels has little value. We propose that batch records be part of the
recordkeeping requirements for both refiners and DFE producers but only aggregated reports be
submitted to EPA. Refiners and DFE producers should be required to keep batch records that
demonstrate compliance with per-gallons standards and support their annual compliance
reporting, and should be required to provide records to EPA on demand, similar to how the diesel
sulfur program is currently structured. However, if EPA insists on batch-level reporting of
benzene and sulfur for DFE, EPA should require those batch properties be entered into EMTS at
the same time RINs are generated rather than establishing an entirely new and duplicative batch
reporting mechanism for DFE producers.
Commenter: Chevron Products Company
We disagree with the Agency's assertion that denaturant producer registration and demonstration
of compliance with the sulfur, aromatics, benzene and olefm content requirements would be
necessary if the maximum denaturant content is not set at 2 volume percent. In our view,
adoption of the California DFE requirements in their entirety resolves this issue and obviates the
need for such burdensome administrative requirements on denaturant producers.
We concur with the Agency's proposal that producers of DFE be required to provide annual
batch reports to the Agency to demonstrate that their product meets the proposed quality
requirements. DFE manufacturers are required to register their product with the Agency prior to
introducing DFE into commerce and they routinely test each batch of their product to provide the
necessary documentation to their customers (i.e., fuel blenders). While we do not necessarily
agree that the incorporation of DFE batch reporting would 'facilitate unfettered downstream
ethanol blending,' we also do not see it as adding any significant burden to DFE producers. The
single most important component in facilitating downstream blending is the adoption of uniform
national specifications for ethanol and, thus, our recommendation of adoption of the California
requirements.
Commenter: Flint Hills Resources, LP (FHR)
In addition, individual batch reporting of gasoline or other fuels should not be required.
Individual Batch Reporting of Gasoline and Other Fuels Should Not Be Required
FHR believes that reporting of each batch of gasoline under §80.1652 has little value, and EPA
should not consider individual batch reporting for DFE producers. FHR agrees that batch records
should continue to be part of the recordkeeping requirements within §80.1653, but only
aggregated reports be submitted to EPA. Similar to how the diesel sulfur program is currently
structured, fuel producers should be required to keep batch records that demonstrate compliance
with standards and support annual reporting. Individual batch records would be made readily
available to EPA upon request or during inspections.
Commenter: Growth Energy
5-120
-------
Tier 3 Summary and Analysis of Comments
Denaturant manufacturers should not be required to register with EPA.
EPA requests comment "on whether to require manufacturers of denaturants for use in DFE to
register with EPA, and demonstrate compliance with the maximum sulfur, benzene, olefms, and
aromatics specifications enforced in the State of California based on the anticipated dilution with
ethanol."
EPA should not require denaturant manufacturers to register with the EPA. Doing so would limit
supply and unfairly increase production costs on the ethanol industry. Gasoline component
manufacturers are not required to register, e.g., manufacturers of pyrolysis gasoline, natural
gasoline, and chemical byproducts. We believe a registration requirement would limit the
number of denaturant suppliers willing to supply the ethanol industry, versus the gasoline or
crude blending markets where they would not need to register. It is inappropriate that a natural
gasoline producer would need to register in order to supply product for DFE, but not need to
supply the same product to a gasoline blender. The requirements should be similar to keep the
market competitive and provide a level playing field to both gasoline and ethanol.
EPA also proposes that DFE manufacturers would be subject to batch testing (including batch
volume, sulfur content, and denaturant concentration as applicable). While the final ethanol
blends (e.g., E30, E85, or some other blend ratio from E16 to E85) would not necessarily be
subject to batching testing, EPA here proposes that the DFE would be tested.
Commenter: Marathon Petroleum Company LP (MFC)
In the event that EPA finalizes a more stringent 10-ppm refinery average sulfur standard, MFC
supports EPA's proposal within new §80.1610(a)(l) that producers of denatured fuel ethanol
(DFE), or other oxygenates, for use by oxygenate blenders be required to meet a 10-ppm sulfur
cap, as determined in accordance with the test requirements for refiners and importers.
MFC believes that it is unnecessary for EPA to require manufacturers of denaturants to register
with EPA. We believe that requiring producers of DFE to meet the latest version of ASTM
International Specification D4806 and TTB denaturant requirements, along with very low
denaturant concentrations (typically 2 volume percent) would sufficiently limit the risk that
denaturants might adversely impact vehicle emissions. Therefore, the proposed registration of
denaturant manufacturers would not provide for any meaningful purpose.
DFE has become a significant component within the gasoline fuel pool and should be subject to
requirements similar to gasoline refiners. However, MFC believes that individual batch reporting
for all fuels has little value. We propose that batch records be part of the recordkeeping
requirements for both refiners and DFE producers but only aggregated reports be submitted to
EPA. Refiners and DFE producers should be required to keep batch records that demonstrate
compliance with per-gallons standards and support their annual compliance reporting, and should
be required to provide records to EPA on demand, similar to how the diesel sulfur program is
currently structured. However, if EPA insists on batch-level reporting of benzene and sulfur for
DFE, EPA should require those batch properties be entered into EMTS at the same time RINs
5-121
-------
Chapter 5: Tier 3 Fuel Sulfur Program
are generated rather than establishing an entirely new and duplicative batch reporting mechanism
for DFE producers.
Commenter: POET, LLC
EPA should not require denaturant manufacturers to register with the EPA. Doing so would limit
supply and unfairly increase production costs on the ethanol industry. Gasoline component
manufacturers are not required to register, e.g., manufacturers of pyrolysis gasoline, natural
gasoline, and chemical byproducts. POET believes a registration requirement would limit the
number of denaturant suppliers willing to supply the ethanol industry, versus the gasoline or
crude blending markets where they would not need to register. It is inappropriate that a natural
gasoline producer would need to register in order to supply product for DFE, but not need to
register to supply the same product to a gasoline blender. The requirements should be similar to
keep the market competitive and provide a level playing field to both gasoline and ethanol.
Commenter: Renewable Fuels Association (RFA)
Because denaturants are the sole source of sulfur in denatured fuel ethanol (DFE), the EPA
should not require sulfur content batch reporting by DFE producers. Current practices for
monitoring the sulfur content of denaturant and DFE are adequate.
However, if EPA finalizes the currently proposed regulatory language pertaining to batch
reporting, it would create a substantial and unnecessary cost burden for DFE producers.
The EPA may not know that ethanol production facilities do not have the laboratory capability to
analyze for sulfur content. This proposed regulation as written would be an expensive endeavor
for ethanol producers.
Since the California sulfur specifications were adopted, ethanol facilities have worked closely
with their denaturant suppliers (mostly natural gasoline marketers) to supply a California-
compliant denaturant that results in a finished DFE with less than 10 ppm sulfur. Ethanol
producers require denaturant suppliers to provide a low sulfur product that also meets the other
attributes of the California regulations pertaining to benzene, olefins and aromatics content. A
certificate of analysis is received with each denaturant delivery. Ethanol production facilities
reference the sulfur content on the certificate of analysis, which is typically a quarterly reported
number recorded by third party commercial laboratories using the latest version of ASTM Test
Method D5453.
What EPA is proposing via §80.1610 is that producers of DFE be required to meet the proposed
fuel quality requirements for their product and sample and test each batch produced and provide
batch reports to EPA. This is similar to the requirements for gasoline refiners. This would require
all ethanol production facilities to purchase lab equipment to analyze for sulfur Using ASTM
D5453; the most widely used and accepted method for DFE. Just the initial equipment
investment for this test method is expected to be $60,000 at a minimum for each location, not to
mention training and personnel time to perform this work. If finalized, this proposed regulation
would require the ethanol industry to initially invest over $12 million dollars in lab equipment to
5-122
-------
Tier 3 Summary and Analysis of Comments
analyze daily for sulfur content of less than 10 ppm that the industry has been in voluntary
compliance since 2002.
Results from the ASTM Interlaboratory Crosscheck Program (ILCP) for DFE further underscore
the superfluous nature of the proposed batch reporting. RFA participates in many areas within
ASTM including the test methods committee and ILCP for DFE. The ILCP program for DFE has
fifty laboratories that participate in the blind analytical testing of DFE using the ASTM D5453
test method for sulfur content. The collection of data has been statistically analyzed and shows
DFE typically has sulfur content in the range of 1 - 5 ppm. In many cases, sulfur levels are so
low that precision problems have arisen in attempting to analyze sulfur content.
For these reasons, RFA is requesting some relief from the batch reporting requirements proposed
in § 80.1610. We believe EPA should allow continuation of current practices for denaturant
blending/sulfur limit compliance instead of following the proposed §80.1630 sampling and
testing requirements for refiners and importers. The sulfur content of the DFE can be monitored
mathematically, based on the incoming certificate of analysis supplied by the denaturant
manufacturer and confirmed via a third party commercial laboratory by sampling and testing
DFE on a defined periodic basis utilizing standard test methods. Additional recordkeeping
requirements for DFE producers (beyond what is specified in the proposal) could be part of this
procedure. Producers would need to maintain the certificate of analysis showing denaturant
sulfur content and the calculations used to determine DFE sulfur content as evidence that DFE
meets the 10 ppm sulfur requirements. They would also have to perform standard quality
assurance activities with the denaturant supplier by sampling incoming shipments of denaturant
for sulfur content against the specified certificate of analysis on a frequent basis.
This would be similar to the way butane is treated in the proposed §80.82 pertaining to butane
blending. The current regulations reduce the burden of compliance with EPA gasoline quality
requirements for parties that blend butane into conventional gasoline, reformulated gasoline
(RFG), or reformulated blendstock for oxygenate blending (RBOB) downstream of a refinery.
Parties that conduct such butane blending are considered refiners; however, they are not subject
to sampling and testing requirements that would otherwise apply to a refiner provided they use
butane of known quality and meet certain other requirements. Proposed language for butane
blending states that any refinery that blends butane for which the refinery has documents from
the butane supplier which demonstrate that the butane is commercial grade may demonstrate
compliance with the standards based on the properties specified by the butane supplier. We are
simply seeking the same treatment for blending of denaturant with ethanol.
EPA should adopt the current California regulatory practice of certifying the sulfur content of
denatured fuel ethanol as this certification system has proven extremely effective. Recently,
California has taken steps to preserve the sulfur content certification which confirms the ethanol
industry's standard practice as accurate.
The above suggested changes to the current proposed language regarding validation of sulfur
content in DFE, which mimic the California regulations, will not compromise the intentions of
the gasoline sulfur proposed limit of 10 ppm maximum.
5-123
-------
Chapter 5: Tier 3 Fuel Sulfur Program
§80.1651 Product transfer document requirements. The product transfer document from Subpart
M § 80.1453 product transfer document requirements for the ethanol producers under the
Renewable Fuels Standard contains the same information and should be allowed to suffice for
§80.1651 with the additional statement as required by §80.1610(c)(l) "Meets EPA standards for
denatured fuel ethanol for use in gasoline." Under the RFS program's EMTS platform, the
ethanol industry already has a batch numbering system and PTD requirements. Requiring
another unique batch number for the sulfur program by §80.1610(d) is redundant and not
necessary for tracking batches of product.
§80.1652 Reporting requirements. Ethanol producers are required to annually report production
volumes under fuels and fuel additives (FFARs) (40 CFR 79 Subparts A, B, C, D, and F). We
suggest simply adding a check box to form 3520-13 A Fuel Additive Manufacturer Report to
include the §80.1652(c)(3) attestation.
In lieu of requiring sulfur content batch reporting for DFE producers, we recommend the
following changes to 80.1653 to ensure that DFE producers can demonstrate, upon request, that
the sulfur content of the denaturant used for DFE blending resulted in a final DFE blend with
less than 10 ppm sulfur:
Denaturant manufacturers should be required to demonstrate compliance with the sulfur
specifications enforced in the State of California. We feel compelled to point out again that the
primary source of sulfur in DFE is the hydrocarbon denaturant itself. Very little sulfur is
contained in the feedstocks and processing aids used for ethanol production. Thus, the primary
concern for ethanol producers regarding proposed sulfur limits on ethanol is continued access to
low sulfur hydrocarbon denaturants. As such, EPA could consider a sulfur maximum for
denaturant, similar to California's regulations.
Overall, RFA is not opposed to the DFE sulfur content standards, provided that sufficient
supplies of low sulfur denaturant continue to be available.
Our Response:
To demonstrate compliance with the Tier 3 program 10 ppm sulfur cap for gasoline
oxygenates, we proposed that producers and importers of oxygenates would be required to test
the sulfur content of each batch of oxygenate they produce. We also requested comment on
whether to require producers of denaturants for use in denatured fuel ethanol DFE to register
with EPA and to demonstrate compliance with a maximum sulfur specification based on the
anticipated dilution with ethanol. With DFE now comprising essentially 10 percent of the
gasoline supply, it is important that EPA have effective means of enforcing gasoline quality
standards, particularly given that the majority of DFE is blended into gasoline at a large number
of diverse locations downstream of the refinery gate. We appreciate the desire of the industry to
avoid regulation. However, important emissions and air quality improvements resulting from
our fuel standards would be at risk without EPA's ability to readily enforce gasoline quality
standards.
5-124
-------
Tier 3 Summary and Analysis of Comments
DFE producers currently use certificates of sulfur analysis from denaturant producers and
volumetric DFE blending records to assure themselves that when a sample of DFE is tested by
the State of California that it will be found to be compliant with a 10 ppm sulfur cap.9 Under this
approach, DFE producers would maintain records regarding the denaturant sulfur content and the
calculations used to determine the sulfur content of the finished DFE. Because the sulfur content
of neat (un-denatured) ethanol manufactured using industry standard quality control practices
should be negligible, the sulfur content of DFE is effectively determined by the sulfur content of
denaturant used.10
EPA agrees that is appropriate to finalize requirements for the demonstration of
compliance with the 10 ppm sulfur cap for DFE that are based on current industry practices
rather than requiring sulfur testing on each batch of DFE.11 We agree that that potential
contribution to the sulfur content of DFE other than from the addition of denaturants can be
adequately addressed by the retention of production quality control records by the ethanol
manufacturer. We also agree that the denaturant typically contributes the majority of the sulfur
to the finished DFE. Therefore, the Tier 3 FRM finalized streamlined provisions that DFE
producers and importers may use in demonstrating compliance with the 10 ppm sulfur cap for
DFE as an alternative to testing each batch of DFE for its sulfur content. These streamlined
provisions are based on the use of denaturant batch sulfur testing conducted by denaturant
producers who have registered with EPA. Denaturants from unregistered denaturant producers
may be used by DFE producers provided that they test each batch of DFE to demonstrate
1 9
compliance with the 10 ppm sulfur cap for DFE.
DFE manufacturers that use denaturants that have been designated on the denaturant
product transfer document (PTD) as a certified ethanol denaturant will be able use PTDs for the
denaturants used and volumetric blending records which show that the denaturant was added at
3.0 volume percent or less in demonstrating compliance with the 10 ppm sulfur cap for DFE in
lieu of per-batch sulfur testing of DFE. The sulfur content of "neat" (i.e. un-denatured) ethanol
may be assumed to be negligible for the purposes of demonstration of compliance using
volumetric blending records provide that the DFE manufacturer maintains quality control records
that demonstrate this assumption is justified. The Tier 3 FRM rule also requires that DFE
manufacturers conduct quality assurance to demonstrate affirmative defenses to presumptive
liability.13 Producers and importers of DFE must initiate a PTD to accompany each batch of
DFE which states that it meets federal standards.
9 The State of California does not rely on industry testing and recordkeeping to help establish compliance with their
sulfur requirements for DFE, instead choosing to focus on direct testing that they conduct. EPA relies on the review
of industry records including those associated with tesing conducted by industry as well as direct testing conducted
by EPA for compliance assurance.
10 Ethanol manufacturers conduct periodic sulfate testing on neat ethanol to ensure that sulfur contamination from
the manufacturing process is negligible.
11 Per-batch sulfur testing for other potential oxygenates will be required to demonstrate compliance with the 10
ppm sulfur cap. EPA may consider amending the requirements for other potential oxygenates in a later rulemaking
based on additional information that we might receive.
12 The limitation on the types of denaturants that may be used and the maximum concentration at which a denaturant
may be used (discussed in section V.G.2 and 3 of the Tier 3 FRM) applies regardless of whether a denaturant from a
registered or non-registered denaturant manufacture are used.
13 Such quality assurance practices include periodic calibration of the denaturant blending equipment to ensure that
denaturants are not added in excess of 5 volume percent.
5-125
-------
Chapter 5: Tier 3 Fuel Sulfur Program
Denaturant manufacturers are accustomed to providing certificates of sulfur analysis to
DFE manufacturers. Therefore, we believe that the requirements finalized today for denaturant
manufacturers to conduct per-batch sulfur testing, initiate a product transfer document stating
that denaturant is suitable for use in manufacturing DFE that meets federal requirements, retain
records, and register with EPA will not represent a substantial new burden. To facilitate
compliance with the requirement that DFE manufacturers who use the mathematical method to
demonstrate compliance with DFE sulfur requirements must only use denaturants from
registered denaturant manufacturers. Registering with EPA will be a one-time act, as will be the
necessary modifications to denaturant product transfer documents. Thus, the requirement to
register with EPA should not be a serious impediment for a manufacturer to enter the denaturant
supply market. We believe that it is necessary to require denaturant manufacturers to register
with EPA in order to facilitate compliance oversight. EPA needs to be able identify all
manufacturers of denaturants in order to periodically audit their records, and to recognize
potential denaturants in the system that are incorrectly designated as appropriate for use in
manufacturing DFE that meets federal sulfur requirements. Denaturant manufacturers that
supply their product to refiners for use in the manufacture of gasoline are not required to register
with EPA because, unlike the DFE manufacturers, refiners are responsible for testing the final
gasoline they produce.
As is current practice today, we anticipate that ethanol manufacturers will negotiate the
specific sulfur level they require from denaturant manufacturers to facilitate compliance with the
10 ppm sulfur cap for DFE taking into consideration what level of compliance margin a given
manufacturer feels is necessary. We believe that it is appropriate to allow this practice to
continue. We understand that ethanol manufacturers currently require denaturant manufacturers
to provide a product with a sulfur content of 120 ppm or less in order to ensure that DFE that
contains 5 volume percent denaturant can comply with California's 10 ppm sulfur cap for DFE.
Thus, we expect that denaturant manufacturers will not need to change the sulfur content of the
denaturant they manufacture in order to comply with the Tier 3 requirements. Manufacturers of
denaturants used by DFE producers that employ the volumetric blending record method in
demonstrating compliance with the sulfur requirements for DFE must retain per batch sulfur test
data on the denaturants they produce to demonstrate that the sulfur content of the denaturant will
not cause the sulfur content of DFE to exceed 10 ppm when added to neat ethanol at 3.0 volume
percent. Any sample of denaturant which is designated as appropriate for use in manufacturing
DFE that meets federal requirements, that is found by EPA to have sulfur content above 330 ppm
will be deemed to be noncompliant, and the denaturant manufacturer may be liable for the
associated penalties. A denaturant with a sulfur content of 330 ppm when used at 3.0 volume
percent would result in a sulfur content of the finished DFE of slightly less than 10 ppm.
We continue to believe that annual reports from oxygenate producers (as well as gasoline
refiners and importers) are important enforcement and compliance assurance tool. Therefore, we
are finalizing the requirement that producers and importers of DFE and other oxygenates must
submit annual reports to EPA that include the total volume of DFE/oxygenate produced and an
attestation that all batches met the proposed fuel quality requirements. The flexibilities provided
in the Tier 3 rule will allow DFE producers and importers to avoid testing DFE production
batches, thereby avoiding substantial capital costs to comply with the Tier 3 requirements. We
5-126
-------
Tier 3 Summary and Analysis of Comments
do not believe that it appropriate to provide additional regulatory relief to DFE producers and
importers from registration, reporting, and recordkeeping requirements. As noted above, these
requirements are essential to ensure the quality of a substantial fraction of the total volume of
motor vehicle fuel.
In response to comments, EPA will work with stakeholders to evaluate the potential for use of
EMTS and other existing EPA mechanisms to meet some of the reporting requirements for DFE
producers and importers.
5.3.4. Other
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel &
Petrochemical Manufacturers (AFPM)
Ethanol has become a significant component within the gasoline fuel pool and should be subject
to requirements similar to gasoline refiners.
Commenter: Growth Energy
We promote expanding the use of ethanol in gasoline, decreasing our dependence on foreign oil,
improving our environment and creating American jobs. Ethanol is a home-grown, renewable
fuel that provides significant benefits to our nation's air quality. We vigorously support the
Renewable Fuel Standard (RFS) and the continued use of ethanol in our nation's fuel supply as it
significantly reduces carbon monoxide, particulate matter and greenhouse gas emissions. As
such, we are pleased to have this opportunity to comment on the Agency's proposed "Tier 3"
Emission and Fuel Standards as there are several key issues that, when addressed, will continue
to promote the expanded use of biofuels in our nation's transportation fuel system.
We also have several concerns about some of the specific standards for Denatured Fuel Ethanol
(DFE).
EPA must not burden the ethanol industry with additional, unnecessary production restrictions:
EPA must not burden the ethanol industry with additional, unnecessary production restrictions.
These restrictions include both the specifications for DFE (denatured fuel ethanol) as well as the
process of blending gasoline with DFE to create ethanol-blended fuels.
Commenter: Chevron Products Company
We concur with the Agency's opinion that the proposed fuel specifications for DFE also apply to
other oxygenates used in gasoline. The January 1, 2017 effective date for oxygenate
producers/importers to meet the new quality requirements seems reasonable in that it appears to
provide sufficient lead time for any manufacturing process modifications that oxygenate
producers may need to make to comply. However, should EPA decide to extend the overall
5-127
-------
Chapter 5: Tier 3 Fuel Sulfur Program
program start date to a date that occurs after January 1, 2017, the effective dates for new DFE
requirements should be extended to be consistent with the overall Tier 3 program.
Commenter: Marathon Petroleum Company LP (MFC)
DFE has become a significant component within the gasoline fuel pool and should be subject to
requirements similar to gasoline refiners.
Commenter: Sutherland Asbill & Brennan LLP (Sutherland)
It appears that an importer of ethanol that is denatured upon entry or already denatured would
qualify as a 'producer' of denatured ethanol. Sutherland believes that EPA should clarify this
point.
Our Response:
We agree with the comment that additional requirements for denatured fuel ethanol
(DFE) are necessary to ensure DFE quality. With DFE now comprising essentially 10 percent of
the gasoline supply, it is important that EPA have effective means of enforcing gasoline quality
standards, particularly given that the majority of DFE is blended into gasoline at a large number
of diverse locations downstream of the refinery gate. We appreciate the desire of the ethanol
industry to avoid regulation. However, important emissions and air quality improvements
resulting from our fuel standards would be at risk without EPA's ability to readily enforce
gasoline quality standards. As discussed in chapter 5.3., we have included flexibilities in the
requirements for denatured fuel ethanol that will minimize the burden on DFE producers.
We have amended the regulatory text to further clarify that denatured fuel ethanol
importers as well as producers are subject to DFE quality requirements.
We agree with the comment that the specifications for DFE should also apply to other
gasoline oxygenates. This is reflected in the oxygenate requirement that we are finalizing.
We agree with the comment that the January 1, 2017 effective date for oxygenate
producers/importers to meet the new quality requirements that we are finalizing in the Tier 3
final rule provides sufficient time for the industry to prepare to comply
5.4. Standards for Fuel Used in Flexible Fueled Vehicles (E51-83 and Mid-Level (E16-
50) Blends)
What Commenters Said:
Commenter: Alliance of Automobile Manufacturers (Alliance) and Association of
Global Automakers (Global)
5-128
-------
Tier 3 Summary and Analysis of Comments
Recommendations for E51-83 and E16-50 Market Fuel Specifications - E51-E83: As EPA notes,
(86) the standards for E51-83, "Ethanol Fuel Blends for Flexible-Fuel Automotive Spark Ignition
Engines" have been changing rapidly in recent years. Indeed, a revised specification, ASTM
D5798-13, issued in May, 2013. The most significant driver for the recent revisions to the E51-
83 specification was the inability of terminals to blend E85 that met the requirements of the E85
specification using current gasoline and BOBs (blend stock for oxygenate blending). The low
vapor pressures of modern gasoline made it difficult to meet the vapor pressure minima
prescribed by the specification, even at the maximum rate of gasoline addition. After running a
test program to determine that the fuel vapor pressure requirements could not be changed, the
ASTM task group decreased the minimum allowable ethanol percentage to 51% as the best way
to enable E51-83 blends with compliant RVPs. The higher allowed level of gasoline is required
not just to meet cold weather vapor pressure requirements but also to meet summer requirements
when using the low vapor pressure RBOBs (Reformulated Blend-stocks for Oxygenate
Blending) used in many areas of the U.S.
We agree with EPA's proposal to subject E51-83 to the same sulfur, benzene and RVP
requirements as gasoline. It is important that producers of E51-83, where they are merely
oxygenate blenders, be exempt from the regulatory burden that falls on those EPA classifies as
"refiners." The goal of this regulation is to ensure that blenders produce E51-83 blends that are
compliant to the gasoline sulfur, RVP and benzene standards, while avoiding all the sampling,
testing, and reporting obligations of refiners. To this end, requiring blenders to maintain records
including the certificates of analysis, product transfer documents, and the ratios used to blend the
fuel, would be sufficient to document that the resultant blends were compliant with EPA
standards.
E16-50
We recommend that EPA treat El6-50 as an alternative fuel for use in FFVs or other vehicles
specifically certified for use with these fuels (87). These fuels should be subject to the same
sulfur, benzene and RVP requirements as gasoline. It is important that producers of E16-50,
when they are merely oxygenate blenders, also be exempt from the regulatory burden that falls
on those EPA classifies as "refiners" as noted in the preceding section.
86 - 78 Fed. Reg. 29908 at 29937 (May 21, 2013).
87 - 78 Fed. Reg. 29908 at 29937-29938 (May 21, 2013).
Commenter: American Coalition for Ethanol (ACE)
Standards for E51-E83 (new ASTM spec): In keeping with EPA intent to provide test fuels that
approximate those actually used by vehicle operators, and in the interest of providing ethanol
producers with similar flexibility given to refiners for decades, ACE supports specifications
consistent with ASTM D5798-13 and would recommend allowing the use of blendstocks such as
natural gasoline and butane in certification fuels.
Standards for Mid-level Ethanol Blends (E16-E50): As mentioned earlier, unless and until a
definitive new ethanol blend is established by auto manufacturers, ACE believes other ethanol
5-129
-------
Chapter 5: Tier 3 Fuel Sulfur Program
certification fuels should be created using E10 or E15 blendstocks along with Denatured Fuel
Ethanol or ethanol flex fuel, since that is the way such fuels would likely be created in the
marketplace
Commenter: American Petroleum Institute (API) and the Association of Fuel &
Petrochemical Manufacturers (AFPM)
API and AFPM agree that El6-50 ethanol blends for use in FFV should meet the same sulfur,
RVP, and benzene standards, and minimum deposit control requirements otherwise applicable to
gasoline. API and AFPM do not agree that EPA should treat El6-50 as gasoline under current
regulations. In API's and AFPM's view, this action would require formal rulemaking with a
waiver from the substantially similar requirements.
While API and AFPM may support EPA's effort to develop regulations for E51-E83, the process
EPA used to inform the regulated community runs afoul of the federal Administrative
Procedures Act (APA) and the federal Clean Air Act (CAA). These additional regulations
include prescribing new requirements for E51-E83 for key fuel properties such as sulfur, RVP,
and benzene, although the benzene limits proposed would preclude the use of some denaturants,
and EPA should set E51-83 benzene standards similar to gasoline standards.
Standards for E16-50: API and AFPM agree that E16-50 ethanol blends for use in FFV should
meet the same sulfur, RVP, and benzene standards otherwise applicable to gasoline. API and
AFPM do not agree that EPA should treat El6-50 as gasoline under current regulations. In API's
and AFPM's view, this action would require formal rulemaking with a waiver from the
substantially similar requirements in CAA section 21 l(f).
EPA is proposing that gasoline deposit control requirements be removed from the El6-50 and
E51-83 fuels. API and AFPM believe that it is in the best interest of the consumer to maintain a
deposit control requirement for all spark-ignition fuels at a minimum level to protect the engines
of consumers.
Standards for E51-83: EPA's preamble to the proposed Tier 3 regulations references a separate
memorandum (38) that includes additional proposed regulations for which EPA seeks comments.
These additional regulations include prescribing new requirements for E51- E83 for key fuel
properties such as sulfur, RVP, and benzene. While API and AFPM may support EPA's effort to
develop regulations for E51-E83, the process EPA used to inform the regulated community runs
afoul of the federal Administrative Procedures Act (APA) and the federal Clean Air Act (CAA).
The APA is clear that EPA's forum for informing the public of substantive rules, amendments,
and revisions is through the Federal Register. Further, the APA states "except to the extent that a
person has actual and timely notice of the terms thereof, a person may not in any manner be
required to resort to, or be adversely affected by, a matter required to be published in the Federal
Register and not so published."
Section 307(d)(3) of the CAA echoes the requirement for proposed rules to be published in the
Federal Register. Among other things, the notice must "specify the period available for public
5-130
-------
Tier 3 Summary and Analysis of Comments
comment." In addition, at the time of publication, there must be a publicly available docket that
includes: (1) the factual data on which the proposal is based; (2) an explanation of how EPA
collected and analyzed the data; and (3) an explanation of the major legal interpretations and
policy considerations underlying the proposal.
In the instant matter, EPA describes and seeks comment on very significant potential changes to
the regulations. EPA is required, by law, to publish such a proposal in the Federal Register.
Posting on EPA website or including within the rulemaking docket is not an option - publication
in the Federal Register means publication in the Federal Register. EPA must include all proposed
changes in the proposed rule. References to documents external to the proposed rule do not give
the regulated community and public the opportunity to which it is entitled to generate informed
and timely comments regarding the proposed rule.
Notwithstanding the APA requirement to publish proposed rules within the Federal Register,
API and AFPM would like to provide the following comments to the Jeff Herzog Memorandum
dated April 8, 2013 that is posted within the docket (ID: EPA-HQ-OAR-2011-0135-0529):
API and AFPM agree that E51-83 for use in Flexible Fueled Vehicles (FFV) should be required
to meet the same sulfur, maximum Reid Vapor Pressure (RVP), and benzene standards otherwise
applicable to gasoline, as well as the Clean Air Act Section 21 l(f)(l) substantially similar
provision. EPA should also allow for the use of butane or other natural gas liquids (e.g. pentane,
natural gasoline, etc.) to manufacture E51-83 with sufficient volatility to meet the ASTM D5798
specifications. API and AFPM do not support limits on specific components, but do support
adoption of reasonable standards to facilitate blending.
However, API and AFPM have identified a serious concern with the suggested 0.20 volume
percent benzene cap for E51-83. Requiring E51-83 to meet a 0.20 volume percent benzene cap
would effectively preclude the blending of gasoline blendstock and natural gasoline to
manufacture E51-83. As EPA states, refiners and importers of gasoline are subject to a 0.62
volume percent annual average and 1.3 volume percent maximum average benzene standard.
Furthermore, natural gasoline typically contains in excess of 1 volume percent benzene. As an
example, E51-83 composed of 49 percent gasoline blendstock with 1.3 volume percent benzene
and 0.01 percent natural gasoline denaturant with 1 volume percent benzene could have a
resulting benzene content exceeding 0.65 volume percent. In another example, E51-83 composed
of 17 percent gasoline blendstock with 1.3 volume percent benzene and 1.7 percent natural
gasoline denaturant with 1 volume percent benzene could have a resulting benzene content
exceeding 0.24 volume percent. Therefore, the setting of a 0.20 volume percent benzene cap
would interfere with the expansion of E51-83 into the marketplace by preventing blends at the
both the upper and lower range of the allowable ethanol content. To resolve this issue, API and
AFPM recommend that E51-83 be required to meet the same benzene standards applicable to
gasoline.
API and AFPM support EPA's suggested requirement that compliance with the Section 21 l(f)(l)
substantially similar provision would be effectively achieved by ensuring that E51-83 be
composed solely of carbon, hydrogen, nitrogen, oxygen, and sulfur.
5-131
-------
Chapter 5: Tier 3 Fuel Sulfur Program
API and AFPM agree that manufacturers of E51-83 for use in FFV should also be subject to the
full responsibilities of a refiner (e.g. registration; batch sampling and testing; and reporting
obligations), which is outlined as Option 1 within the Herzog Memorandum. However, if EPA
wishes, as a second option, to allow oxygenate blenders to be exempt from the sampling, testing
and reporting obligations of refiners, oxygenate blenders should be required to only use gasoline
blendstocks, denatured fuel ethanol, natural gas liquids (NGL) and butane that meet certain
specifications. API and AFPM agree with this two-option approach as a means to not restrict
refiners or oxygenate blenders from choosing maximum blending flexibility by meeting the more
stringent requirements, as outlined within Option 1. API and AFPM also agree that any resulting
E51-83 should also be required to meet the downstream sulfur, RVP and benzene per-gallon cap
standard.
API and AFPM agree that it may be possible to develop a calculative approach for E51-83
blenders to determine RVP and avoid sampling and testing. We agree that testing may be
required for some period of time when a blender receives a new batch of gasoline blendstock.
We also agree that RVP of different batches of DFE would not vary significantly due to TTB
requirements related to denaturants and denaturant percentages, along with RFS2 requirements
related to denaturant concentrations for RIN generation.
API and AFPM agree that butane used for E51-83 blending should meet the same standards as
specified for butane blended into gasoline at a terminal. However, the limits on other
components in E51-83 should be restricted to their total contribution toward regulated properties
on the final fuel, not the same limits imposed on the components. Sources for NGLs should not
be limited, provided the NGL conforms to the requirements.
Commenter: BP Products North America Inc.
Standards for E51-83- Meeting the ultimate goals of the Renewable Fuel Standard is highly
dependent upon the expansion of E51-83 into the marketplace. BP, therefore, supports EPA's
proposal to allow the use of butane and natural gas liquids to manufacture E51-83. BP also
agrees that the final blend should meet the same sulfur, RVP, and benzene standards otherwise
applicable to gasoline. BP does not support imposing tighter standards on E51-83, such as EPA's
proposed 0.2 wt. % benzene.
EPA has proposed alternate conformance demonstrations with RVP requirements for E51- 83.
Under option 2, the blender could use a blending model and measure the RVPs of the blending
components. Alternatively the blender could use reduced frequency testing to verify vapor
pressure conformance. BP is supportive of these options which will provide greater flexibility for
E51-83 blending.
Standards for Butane and Natural Gas Liquids (NGLs)- BP supports the limits on butane
composition that match those used now for butane blending of gasoline at the terminal. BP's
experience has shown that only relatively low levels (one to few percent) of butane are needed to
bring ethanol blends into vapor pressure conformance with ASTM D 5798 specification. In
general, one volume percent butane raises vapor pressure of flex fuel about 1 psi. Hence an upper
limit on butane is self-determined by the vapor pressure limits of D 5798 specification. Limits on
5-132
-------
Tier 3 Summary and Analysis of Comments
other components in E51-83, including NGLs, should be restricted to their total contribution
toward regulated properties on the final E51-83 blend. Imposing the same limits on individual
blending components, would severely limit the blending of such components. BP also
recommends that EPA not restrict the use of NGLs to natural gas processing plants. If NGLs are
used in the blending of E51-83 and the final blend conforms to the necessary requirements,
restricting the source of NGLs is unnecessary.
BP further recommends that EPA not impose upper limits on the concentration for butane and
NGL's in E51-83 blending, however, does support adoption of reasonable standards to facilitate
such blending.
For additional flexibility, BP also encourages EPA to allow blending of pentanes as a means of
increasing flex fuel vapor pressure to allow conformance with ASTM D 5798 specification for
flex fuel blends.
Establish Gasoline Deposit Control Requirements for E51-83- BP recommends that gasoline
deposit control requirements be imposed upon E51-83 and mid-level ethanol blends. It is in
consumers' best interest to establish a deposit control requirement for all spark-ignition fuels, in
order to protect their engines.
Pump Dispenser Label for E51-83 -BP supports the use of the following pump dispenser label
for E51-83 and a consumer education program on flexible fuel vehicles, E51-83, E16-50 and the
terms 'flex fuel' and E85.
Commenter: California Air Resources Board (CARB)
C ARB staff also supports the proposed requirements for midlevel ethanol blends, as well as
promoting the power plant design advantages of an E30 blend.
Commenter: Chevron Products Company
Standards for Flexible-Fuel Vehicle Fuel and Mid-Level Ethanol Blends: Chevron actively
participated in the development of the API/AFPM comments on these topics. We refer you to
those comments as our detailed feedback on this section.
Commenter: Energy Future Coalition and Urban Air Initiative
EPA Should Enable a Market for Optimized, Dedicated, and Flex- Fuel Vehicle Production
through Recognition of Ethanol's Life-Cycle GHG Benefits. As was the case in the transition
from leaded to unleaded gasoline, a mid-level ethanol blend cannot be made available
nationwide, and market forces will not prompt any greater availability, until vehicles designed to
run on this fuel are in widespread use. Meanwhile, legacy vehicles designed for conventional
gasoline (EO) will not perform adequately on a mid-level ethanol blend, while new optimized or
dedicated vehicles will not perform adequately on much lower-octane, conventional E10 fuel.
EPA, however, can break through this impasse by encouraging automobile manufacturers to
equip new vehicles (beginning in model year 2017) with FFV capabilities that enable the
5-133
-------
Chapter 5: Tier 3 Fuel Sulfur Program
vehicles to run on both conventional fuel and the new mid-level ethanol blend (as well as higher
ethanol blends such as E85), thereby opening the door to a new generation of engines optimized
for the new fuel as it becomes available (and eventually standard). (141)
FFV technology is both widely available and relatively inexpensive, and it can build a bridge to a
cleaner fuel standard by solving the transitional problem of engine and fuel compatibility that
must precede an aromatics phase-down. Manufacturers can equip any new vehicle with FFV
technology that senses the alcohol content of the fuel being used and adjusts the engine's timing
accordingly. With the right incentives, this technology could be installed in every new vehicle in
the country. This would avoid potential legacy fleet issues when automobile manufacturers begin
to optimize or dedicate all vehicles to run on cleaner fuel and when EPA begins to require
aromatics reductions in standard market fuel. Legacy vehicles would, at that point, be mostly
FFVs capable of running on the new fuel.
In order to begin to phase in a cleaner, higher-ethanol fuel that significantly reduces air toxics
and other dangerous pollutants, while at the same time helping manufacturers design more
efficient engines to reduce GHG pollution, EPA should allow all vehicles capable of running on
a mid-level ethanol blend (dedicated, optimized, or FFV) to be certified for purposes of GHG
emissions and fuel economy on the new high-octane, mid-level ethanol blend.
EPA Should Grant the New Mid-Level Ethanol Blend the Same One-Pound RVP Waiver
Congress Granted to E10. As gasoline evaporates, VOCs enter the atmosphere, producing PM
and contributing to pollution. (151) The Reid vapor pressure (RVP) of gasoline refers to the
tendency of gasoline to evaporate and is measured in pounds per square inch (psi). In order to
prevent VOCs from forming ozone and PM pollution in the presence of sunlight
("photochemical smog"), Congress required gasoline to be limited to a maximum RVP of 9.0 psi
during the summer ozone season as part of the 1990 Clean Air Act Amendments. (152) In
addition, Congress authorized EPA to set more stringent standards for certain non-attainment
areas or former non-attainment areas. (153) EPA limits "high-volatility non-attainment" areas to
a maximum RVP of 7.8 psi during the summer ozone season. (154) During the summer ozone
season, about 75% of gasoline is 9.0 psi.
In its neat form (when not blended with gasoline), ethanol's RVP is much lower than that of
conventional gasoline (pure ethanol's RVP is approximately 2.0 psi). However, due to an
azeotropic effect when blended in small percentages with gasoline, ethanol can increase the
overall blend's volatility. (155) As a result, ethanol splash-blended with gasoline at 10% has a
tendency to raise vapor pressure by approximately one pound. (156) Because of this effect,
Congress also stipulated in the Clean Air Act Amendments that gasoline containing 10% ethanol
would be permitted to exceed the applicable maximum RVP by one psi (the "one-pound
waiver"). (157) In its regulations implementing this section of the statute, EPA specified that the
waiver only applies to blends containing "at least 9% and no more than 10%" ethanol (158)—a
distinction that was not particularly relevant at that time. In the Proposed Rule, however, EPA
asks whether it should "allow El6 to E50 blends to have higher RVP levels than otherwise
required by our regulations for gasoline." (159) Our answer is 'yes.' It is important that EPA,
consistent with the terms of the statute, treat the new mid-level ethanol blend the same as E10,
5-134
-------
Tier 3 Summary and Analysis of Comments
granting it a one-pound waiver of the RVP limit, but requiring that the gasoline portion of the
fuel meet the applicable RVP standard.
Because ethanol's effect on RVP is azeotropic, and not due to a higher RVP of ethanol itself,
adding ethanol beyond 10% is likely to lower, rather than raise, RVP. (160) Nevertheless,
allowing the one-pound waiver would remove uncertainty for distributors and also ensure
compliance with the statutory mandate that "fuel blends containing gasoline and 10 percent
denatured anhydrous ethanol" (which would include any mid-level ethanol blend) receive the
one-pound waiver. (161) The tailpipe emissions reductions achieved by a mid-level ethanol
blend will be greater than those achieved by E10, and will more than compensate for any slight
increase (over EO) in evaporative emissions that may occur because of the higher RVP of the
blended fuel. (162) As a result, a mid-level ethanol splash blend would be superior to both EO
and E10 on all counts: Its toxic, hydrocarbon, and NOx tailpipe emissions would be lower, and,
due to the higher concentration of ethanol, its evaporative emissions would be lower as well.
(163)
The Agency's requests for comment demonstrate its awareness of the benefits of a mid-level
ethanol blend. However, approving a new certification fuel is not enough. We urge EPA to take
the initiative to shepherd a nationwide transition to this cleaner, more efficient, renewable fuel in
order to realize its benefits to air quality and public health, automotive performance, and
consumer costs, while meeting the Agency's obligation to regulate air toxics in motor vehicle
fuel. Although critically important, a new certification fuel should be only the first step in a
comprehensive plan to make a mid-level ethanol blend available at pumps across the country and
to enable the production of vehicles designed to maximize its benefits. We appreciate the
opportunity to offer such a plan, and we look forward to providing any additional information or
assistance that might be useful to EPA as it strives to protect air quality and public health for all
Americans.
141 - Again, we reiterate that manufacturers themselves are powerless to make a fuel commercially
available. EPA, in contrast, has wide authority to regulate toxic compounds in motor vehicle fuel, thereby
ensuring that cleaner fuels are available, just as it did in the lead phase-out. In this case, EPA has
authority—and indeed the responsibility—to mandate aromatics reductions under sections 202(1) and 211.
151 - See supra Parts III.B.1-B.2.
152-42U.S.C. §7545(h)(l).
153-Id. § 7545(h)(l)-(2).
154-40C.F.R. §80.27(a)(2).
155 - See Proposed Tier 3 Rule, 78 Fed. Reg. at 29938 ("As the ethanol level increases, the volatility
increase caused by blending ethanol with gasoline begins to decline, such that at E30 there is only about a
0.5-psi RVP increase.").
156 - See id.
157 - 42 U.S.C. § 7545(h)(4) ("Ethanol Waiver. For fuel blends containing gasoline and 10 percent
denatured anhydrous ethanol, the Reid vapor pressure limitation under this subsection shall be one pound
per square inch (psi) greater than the applicable Reid vapor pressure limitations established under
paragraph (1).").
158 - 40 C.F.R. § 80.27(d)(2); see also id. § 80.28(g)(8) (providing as a defense for a noncompliant E10
blend that its blendstock met the RVP standard).
159 - Proposed Tier 3 Rule, 78 Fed. Reg. at 29938.
160 - See id.
161 -42 U.S.C. §7545(h)(4).
5-135
-------
Chapter 5: Tier 3 Fuel Sulfur Program
162 - See Fann, supra note 58, at 176, fig. 4 (ranking PM and VOCs according to the dollar/ton benefit of
their respective reductions).
163 - Proposed Tier 3 Rule, 78 Fed. Reg. at 29938 ("The evaporative emission increase caused by E30
would ... be less than for E10.").
Commenter: Flint Hills Resources, LP (FFfR)
Regarding E51-83 for use in flexible fuel vehicles (FFV), this fuel should be defined to meet
ASTM 5798-13, including the, maximum RVP specifications, and be required to meet the same
sulfur and benzene standards otherwise applicable to gasoline. Also, an effective date should be
established for the E51-83 standards and regulations. Lastly, FFfR believes that formal rules-
should be developed for El6-50 use in FFVs.
Comments to the Jeff Herzog Memorandum dated April 8, 2013: Notwithstanding the APA and
CAA requirements to publish proposed rules within the Federal Register, FUR provides the
following comments to the Jeff Herzog Memorandum dated April 8, 2013 posted within the
docket (ID: EPA—HQ—OAR-2011-0135-0529):
E51-83 Should be Defined to Meet ASTM D5798-13: FHR agrees that E51-83 for use in
Flexible Fueled Vehicles (FFV) should be required to meet the same sulfur and benzene
standards otherwise applicable to gasoline, as well as the Clean Air Act Section 21 l(f)(l)
substantially similar provision. In addition to sulfur and benzene standards, FHR recommends
that the definition within the draft regulatory text in §80.2(cccc) be revised to require E51-83 to
meet the requirements of ASTM Standard Specification for Ethanol Blends for Flexible-Fuel
Automotive Spark-Ignition Engines D5798-13.
FHR believes that the adoption of ASTM 05798-13 would be a key supporting element to
achieve EPA's E51-83 objectives. ASTM D5798-13 presents performance requirements for
ethanol fuel blends of DFE and hydrocarbon blendstock. Within ASTM D5798-13, DFE must
meet the requirements of ASTM D4806, and the hydrocarbon blendstock may be composed of
unleaded gasoline, gasoline blendstock for oxygenate blending (BOB), natural gasoline or other
hydrocarbons in the gasoline boiling range. ASTM 05798-13 also sets further requirements for
hydrocarbon blendstock properties, including maximum distillation end point, oxidation stability,
copper strip corrosion, and silver strip corrosion.
Different Requirements Should be Provided for E51-83 Refiners and Blenders: FHR agrees that
E51-83 refiners should be subject to the full responsibilities of a refiner (e.g. registration, batch
sampling, testing, and reporting obligations), as described in Option 1 within the Herzog
Memorandum. FHR also agrees that E51-83 blenders should be exempt from the sampling,
testing, and reporting obligations of refiners, when using gasoline, gasoline blendstocks for
oxygenate blending, DFE, natural gas liquids (NGL) and butane that meet certain specifications,
as outlined in Option 2 within the Herzog Memorandum.
FHR agrees with this two-option approach, as a means to not restrict refiners or downstream
blenders from choosing maximum blending flexibility by meeting the more stringent refiner
requirements, as outlined in Option 1. FHR also believes that E51-83 blenders should be required
to meet the 95 ppm downstream per-gallon cap sulfur standard included within §80.1604(b),
5-136
-------
Tier 3 Summary and Analysis of Comments
instead of the 80 ppm per-gallon cap standard sulfur of §80.1603 (a)(2), as incorporated within
§80.1608(b)(l).
As means of further streamlining the requirements of Option 1, FHR recommends that EPA
consider adopting batch sampling, testing, and reporting obligations for batches of E51-83
hydrocarbon blendstock, as defined within ASTM 05798-13, that include the test results obtained
from testing a hand blend with DFE, similar to the RBOB/RFG requirements of §80.69(a).
Under these circumstances, a refiner could manufacture, sample and test any batch quantity of
E51-83 blendstock , blend with DFE into multiple tank trucks at a refinery loading rack, and
avoid the requirement to sample, test and report each tank truck.
E51-83 Should Meet the Reid Vapor Pressure (RVP) Specifications within ASTM D5798-13:
ASTM D5798-13 sets maximum RVP requirements for ethanol fuel blends for FFV that vary
only slightly from EPA's maximum RVP requirements for gasoline within §80.27(a)(2). As
such, FFIR believes that EPA should adopt the maximum RVP requirements of ASTM D5798-13
within the draft regulatory text for Option 1 E51-83 refiners in §80.1608(c)(6) and Option 2 E51-
83 blenders in §80.1608(b)(4).
In addition, FFIR believes that EPA does not need to impose minimum volatility standards for
E51-83. By defining E51-83 to meet the requirements of ASTMD5798-13, EPA would be
adopting the minimum RVP specifications within ASTM D5798-13 by reference.
E51-83 Should Meet the Benzene Standards Otherwise Applicable to Gasoline: FUR has
identified a concern with the suggested 0.20 volume percent benzene cap for E51-83 refiners and
E51-83 blenders within the draft regulatory text in §80.1608(c)(3) and §80.1608(b)(2),
respectively. To resolve this issue, FUR recommends that E51-83 refiners and E51-83 blenders
be required to meet the same benzene standards otherwise applicable to gasoline within
§80.1230.
Requiring E51-83 to meet a 0.20 volume percent benzene cap would effectively preclude the
blending of gasoline blendstock to manufacture E51-83. As EPA states, refiners and importers of
gasoline are subject to a 0.62 volume percent annual average and 1.3 volume percent maximum
average benzene standards. Furthermore, natural gasoline, used as a DFE denaturant, typically
contains in excess of 1 volume percent benzene. As an example, E51-83 composed of 49 percent
gasoline blendstock with 1.3 volume percent benzene and 0.01 percent natural gasoline
denaturant with 1 volume percent benzene could have a resulting benzene content exceeding
0.65 volume percent. In another example, E51-83 composed of 17 percent gasoline blendstock
with 1.3 volume percent benzene and 1.7 percent natural gasoline denaturant with 1 volume
percent benzene could have a resulting benzene content exceeding 0.24 volume percent.
Therefore, the setting of a 0.20 volume percent benzene cap could interfere with the expansion of
E51-83 into the marketplace by preventing blends at the both the upper and lower range of the
allowable ethanol content with gasoline blendstock meeting the requirements of §80.1230.
Furthermore, requiring natural gasoline for use by E51-83 blenders to meet a 0.03 volume
percent benzene cap within §80.1612(a)(4), or the 0.20 volume percent stated within the
memorandum, would effectively preclude the use of natural gasoline, as a hydrocarbon
5-137
-------
Chapter 5: Tier 3 Fuel Sulfur Program
blendstock, to manufacture E51-83. As previously stated, natural gasoline typically contains 1
volume percent benzene, and as stated within the memorandum, could be as high as 2 volume
percent benzene. FHR believes that it is reasonable for natural gasoline blended into E51-83 to
be required to meet a maximum benzene standard of 1.1 volume percent within §80.1612(a)(4)
and a maximum sulfur standard of 20 ppm within §80.1612(a)(3).
An Effective Date Should be Set for E51-83: Apart from the gasoline sulfur implementation
dates via §80.1603(a), sections 80.1608 and 80.1611 do not specify an effective date for the
standards and requirements for E51-83. FHR infers the effective date to be sometime in 2017,
but requests this clarification be added into the regulation.
Formal Rules Should be Developed for El6-50 Use in Flexible Fuel Vehicles: FHR agrees that
E16-50 ethanol blends for use only in FFV should meet the same sulfur, RVP, and benzene
standards otherwise applicable to gasoline. FHR also agrees that EPA should treat E16-50 as
gasoline under current regulations. However, in FHR's opinion, this action would require formal
rulemaking, including a waiver from the substantially similar requirements, explicitly listing
E16-50 in §80.1601(a), and providing a definition within §80.2.
1- Herzog, J. (January 2012). Possible Approach to Fuel Quality Standards for Fuels Used in Flexible-
Fuel Automotive Spark-Ignition Vehicles (FFVs), Memorandum to the docket. The memorandum is
referenced in footnote 388 of the pre-publication proposed rule; however, the footnote reference is to a
memorandum dated January 2012 but the memorandum in the docket is dated April 8, 2013. Further, this
memorandum was not in the docket at the time the proposed rules were announced on EPA's website.
2 - United States Code, Title 5, Chapter 5
3 - 5 USC 552(a)(l)
Commenter: Growth Energy
In particular, Growth Energy fully supports the treatment of midlevel ethanol blends (El6-50) as
alternative fuels.
Higher Ethanol Blends Improve Air Quality and Warrant Consideration for Higher Reid
Vapor Pressure Limits: In the proposal, EPA seeks comment on whether it is appropriate to
allow higher RVP limits for midlevel in-use ethanol blends (E16-E50). Currently, EPA is
required to limit gasoline to 9.0 pounds per square inch (psi) except in areas where further
emission control is required, and E10 currently has a 1 psi waiver allowing it to rise to 10.0 psi.
Any review of RVP and ethanol blends shows that RVP decreases as ethanol is added and that
the highest RVP occurs with an E10 blend (estimated RVP curve below) [Figure can be found on
p. 8 of Docket number EPA-HQ-OAR-2011-0135-4681-A3]. As ethanol increases, RVP
decreases and better air quality is achieved. Therefore it is certainly appropriate and Growth
Energy would fully support higher RVP limits for all ethanol blends above E10. It would be
counter to the intent of this proposed rule to continue to provide a waiver for E10 while requiring
blends above E10 which demonstrate better air quality not to enjoy the same treatment.
Specifically, EPA sought comment on the regulation of midlevel and highlevel ethanol blends
(E16-E85) as an alternative fuel family as discussed in the Herzog memo to the docket (Jeff
Herzog, "Possible Approach to Fuel Quality Standards for Fuel Used in Flexible-Fuel
5-138
-------
Tier 3 Summary and Analysis of Comments
Automotive Spark-Ignition Vehicles (FFVs)"). Growth Energy fully supports the treatment of
ethanol blends above El5 as an alternative fuel family rather than as gasoline and feels as though
EPA can use its authority under 211(C) to take ethanol blends' lower RVP relative to E10
gasoline to provide similar RVP leeway as that given to gasoline. Additionally, because
evaporative emissions from the current vehicle fleet are already well controlled and will be
further controlled through the Tier 3 evaporative emissions standards, Growth Energy believes
that a similar 1 psi waiver should also be extended to gasoline blends above E10 as well as
midlevel ethanol blends.
Benzene limits for ethanol blends must take into account the gasoline blended with ethanol. EPA
believes "it may be possible to set the benzene standard for E51-83 at a 0.20 volume percent cap
or even lower without imposing any burden rather than implementing an average annual benzene
standard similar to that for gasoline." (Herzog, FQS for FFVs, p. 5) Gasoline has 0.62 volume
percent average maximum benzene content.
A benzene cap of 0.20 percent on ethanol blends would create a tremendous burden on blenders
and could render certain production impossible. For instance, an E51 blend composed of 51
percent DFE and 49 percent certified gasoline would likely have a benzene content of 0.31
percent as the gasoline would bring 0.62 percent benzene to the mixture. This mixture would not
meet the regulatory requirements.
Under the proposed approach, E51-83 could only exist as E75-83 with DFE at a 0.06 percent
benzene and gasoline at 0.62 percent benzene. Higher benzene levels in the gasoline due to
averaging would further limit the hydrocarbon content of this fuel as well as the RVP and cold
start capability in cold weather. It is imperative that EPA provide a higher benzene limit to
account for the gasoline blended with ethanol. Growth Energy feels that EPA should set this
benzene standard at 0.62 volume percent, identical to the gasoline limit.
We agree with EPA that addressing RVP in blends does not require batch testing. EPA states that
"For blenders that use only gasoline, BOB's, and DFE to produce E51-83, demonstration of
compliance with the maximum RVP standard under consideration could be accomplished by
retention of Product Transfer Documents for the blendstocks."
Additionally, EPA raises the issue of whether the SAE "Model for Estimating Vapor Pressures
of Commingled Ethanol Fuels" is adequate to avoid testing for NGLs and refinery naphthas for
RVP. 17 Growth Energy agrees that the SAE model referenced in the FQS for FFVs memo
should be adequate to avoid testing for NGLs and refinery naphthas for RVP. 18
EPA states that "Given the proposal in the Tier 3 NPRM to limit the denaturants that can be used
to manufacture DFE to NGL, gasoline, and BOBs, and the narrow range in denaturant
concentration, we believe that the RVP of different batches DFE would not vary substantially.
Therefore, we currently do not believe that deliveries of new batches of DFE into the storage
tank used to produce E51-83 would necessitate a return to per batch testing." We agree that the
variability of the ethanol RVP would not vary substantially, and batch testing should not be
required, with a 5 percent denaturant cap.20
5-139
-------
Chapter 5: Tier 3 Fuel Sulfur Program
EPA's streamlined "Blender Option" requirements should be the default regulations for E16-E85
production (as opposed to full refiner responsibilities). The Tier 3 rule appropriately recognizes
that ethanol blending is of central importance to our nation's transportation fuel supply and
accomplishing key environmental and energy security goals. To achieve these goals and
maintain a robust domestic transportation fuel supply, it is critical that EPA design its regulations
to facilitate the production and blending of ethanol.
EPA discusses its possible approaches to the regulation of ethanol (and vehicles that use ethanol)
in the Proposed Rule, including the section "Standards for Fuel Used in Flexible Fueled
Vehicles." This section then references a detailed memo to the docket, the FQS for FFVs Memo.
The FQS for FFVs memo outlines two options for regulating manufacturers of E51-E83. Growth
Energy strongly supports "Option 2: Treatment as a Blender" (the "Blender Option") as the
default option. Under this Blender Option, EPA states that E51-E83 producers can reduce their
regulatory burden if they only use blendstocks that meet certain specifications that would ensure
sulfur and benzene does not exceed appropriate amounts.12
More specifically, under this Blender Option, EPA proposes generally to limit blendstocks for
ethanol blends to gasoline, BOBs, and DFE, as well as natural gas liquids (NGLs) and butane if
they are produced to certain standards. Under this Blender Option, the finished E51-83 would be
subject to the proposed downstream sulfur cap for gasoline of either 65 ppm or 95 ppm. A
benzene limit would also apply to the finished blend (as discussed above).
Growth Energy agrees that testing of final blends would not be needed as the DFE and
blendstocks are regulated individually for sulfur and benzene (and RVP issues can be handled in
a streamlined fashion as well).
Growth Energy also believes that the Blender Option requirements could apply equally to E16-
E50 as well as E51-E83. The FQS for FFVs memo also notes that "the draft regulations written
to address E5183 might be modified to cover E16-50 blends." Growth Energy believes that this
Blender Option can be applied across all ethanol blended fuels from E16-E85. If EPA considers
it necessary to further delineate midlevel ethanol blending, we suggest that applicable ASTM
standards already provide a suitable approach. In particular, ASTM Standard D7794 allows
midlevel ethanol blends only to be created from a mixture of gasoline and DFE, or gasoline and
E85.15 Thus, the constituents of midlevel ethanol blends (gasoline, DFE, and E85) are already
controlled. This is consistent with the streamlined "Blender Option" that avoids batch testing for
blends when the constituents of those blends are already controlled.
Under the Blender Option, gasoline blendstocks should be allowed for blending (in addition to
NGLs and butane) as long as they meet adequate specifications. Light straight run naphtha,
hydrotreated naphthas, and hydrotreated natural gasoline are potential streams to consider. These
gasoline blendstocks could be subject to the same sulfur and benzene specifications as NGLs and
butane and should be allowed as blendstocks for ethanol blends.
Growth Energy supports facilitating the use of NGLs under the Blender Option in the production
of ethanol blends. We do not believe that any EPA limits on the volume of NGL used as a
5-140
-------
Tier 3 Summary and Analysis of Comments
blendstock are necessary for ethanol blends (E51-E83). For midlevel ethanol blends (E16-E50),
if EPA considers that some NGL limits are necessary, we would suggest that applicable ASTM
standards already effectively limit the NGL content in midlevel ethanol blends. In particular, as
noted above, ASTM Standard D7794 allows midlevel ethanol blends only to be created from a
mixture of gasoline and DFE, or gasoline and high level ethanol blends (E51-E83). The amount
of NGLs allowed in these constituent products are already controlled. Therefore, the volume of
NGL in midlevel ethanol blends can effectively be no greater than what is already contained in
approved components of a midlevel ethanol blend.
Furthermore, we respond to the following statements by EPA regarding NGLs under the Blender
Option:
• Production "using NGL blendstocks would require that the NGL blendstock, like butane,
meet a 10-ppm per-gallon sulfur cap."
o We agree that NGL (or gasoline blendstocks such as hydrotreated refinery
streams) used for blending E16-E8 3 should meet the proposed 10 ppm sulfur cap
under the Blender Option and therefore avoid burdensome batch testing
requirements for blending.
• EPA notes that "Similar to the envisioned 0.20 volume percent benzene cap for butane,
we would look to implement a 0.20 volume percent benzene cap standard for NGL rather
than an average standard in order to simplify the regulatory requirements."
o A benzene cap of 0.20 per cent for NGLs is too restrictive and may be impractical
to meet. The benzene standard for NGLs should be 0.62 percent benzene similar
to unleaded gasoline. Therefore, this NGL stream or gasoline could be used in the
blend with no detrimental effect to the finished fuel since the DFE in which it is
blended will have little to no benzene.
• "Since NGL properties from different sources can vary widely ... we may also need to
consider restricting the sources of NGL to natural gas liquids fractionation facilities and
natural gas processing plants to ensure an appropriate level of quality control.'
• EPA should maintain a provision for an alternative NGL source to allow for
hydrotreated streams or refinery streams that meet the same criteria [i.e., benzene
and sulfur limits]. Refineries (including where hydrotreating takes place) can provide the
same or superior quality control as NGL fractionation facilities and natural gas
processing plants.
"Stakeholders have stated that as additive concentration diminishes due to dilution with DFE,
there is a point where the presence of a deposit control additive ceases to be beneficial and can
actually contribute to deposit formation. In light of this, it may be appropriate to remove the
requirement that the gasoline portion of E51-83 and perhaps E16-50 must contain a deposit
control additive until the specific deposit control needs of these blends can be evaluated."
o We support removal of the requirement that the gasoline portion of E51-E83 must contain a
deposit control additive until the specific deposit control needs of these blends can be evaluated.
A similar approach may be warranted for E16-E50.
5-141
-------
Chapter 5: Tier 3 Fuel Sulfur Program
EPA should not unduly regulate E16-E83 blendstocks. EPA states that "Currently, only ...
RBOBs ... and gasoline ... are used to manufacture E51-83." This is inaccurate. Gasoline
blendstocks and natural gasoline can also be used in E51-E85 today.26
Some ethanol producers may wish to use non-standard blendstocks, which they should maintain
the option to do as long as the end product meets appropriate specifications. Ethanol producers
should have the option to do so, if they opt to undertake broader "refiner" responsibilities. These
"refiner" responsibilities could require batch testing of the end product, and some ethanol
producers may consider this extra cost and burden appropriate. 7
EPA's proposed regulations unduly restrict using NGL as a denaturant or blendstock. A
requirement to "Certify that the NGL was derived solely of from [sic] natural gas liquids
fractionation facilities and natural gas processing plants" is too narrow. As noted above, EPA
should include refineries (including hydrotreating) as potential hydrocarbon stream sources as
long as blendstocks meet certain requirements (i.e., benzene and sulfur limits).
A requirement to demonstrate "that the benzene content of each production batch of NGL is not
greater than 0.03 volume percent" is too restrictive. This specification is unrealistic, and it is
unclear whether a product that meets this specification even exists in any significant commercial
quantities. For instance, this specification is 20 times more stringent than gasoline and wouldn't
apply to NGLs blended directly into unleaded gasoline. [Even California ethanol specifications
allow benzene content up to 0.06 volume percent in DFE. Therefore, the resultant specification
allows NGL with benzene levels up to 2.4 volume percent.]
A prohibition that "NGL and E51-83 made from NGL cannot be sold as gasoline or El6 or be
commingled with gasoline or El6-50 and subsequently sold as gasoline or El6-50" is unduly
restrictive. This prohibition on commingling E51-83 made with NGL with MLEBs would unduly
limit the ability for retail stations to provide E16-E83 fuels, especially if they employ blender
pumps. Alternatively, EPA should follow ASTM requirements as stated in standards D4814,
D4806, D5798 and standard practice D7794.
12 Some ethanol producers may wish to use non-standard blendstocks, which they should maintain the
option to do as long as the end product meets appropriate specifications. Ethanol producers should have
the option to do so, if they opt to undertake broader "refiner" responsibilities. These "refiner"
responsibilities could require batch testing of the end product, and some ethanol producers may consider
this extra cost and burden appropriate.
15 See ASTM D7794-12 (D7794) "Standard Practice for Blending Mid-Level Ethanol Fuel Blends for
Flexible-Fuel Vehicles with Automotive Spark-Ignition Engines." This standard incorporates by reference
ASTM standards governing the permissible constituents of MLEBs including one for gasoline (D4814),
as well as ASTM D4806-12 ("Standard Specification for Denatured Fuel Ethanol for Blending with
Gasolines for Use as Automotive Spark-Ignition Fuel") (DFE) and ASTM D5798-12 ("Standard
Specification for Ethanol Fuel Blends for Flexible-Fuel Automotive Spark-Ignition Engines") (HLEBs).
17 By way of background, EPA stated "If NGL and/or butane were used to manufacture E51-83,
demonstration of compliance with a maximum RVP requirement could not be accomplished merely by
retention of [product transfer documents (PTDs)] indicating that blendstocks were produced subject to
applicable standards.... However, it may be possible to avoid sampling and testing for RVP as well. RVP
models exist that could perhaps be developed further to avoid the need for sampling and testing for RVP
when NGL or butane were used, but they are not yet applicable to these blends." Id.
5-142
-------
Tier 3 Summary and Analysis of Comments
18 See id. at 7 n. 22 (referencing SAE paper 2007-01-4006, "A Model for Estimating the Vapor Pressures
of Commingled Ethanol Fuels.").
20 [By way of example, 5 percent denaturant would create a .75 psi difference between a 9 psi and 14.5
psi denaturant in DFE. Either blend would change the E85 RVP by 0.5 psi or less according to the model
in SAE paper 2007-01-4006. This minor impact would not cause any concerns for E85 (which already
have low volatility), and the impact on MLEBs from the RVP of denaturant would be even smaller due to
the overall smaller proportional share of denaturant in the overall blend (i.e., 5 percent denaturant in an
E30 blend would be only 1.5 percent of the total fuel mixture).]
26 See e.g., ASTM standard for HLEBs, ASTM D5798-12.
27 Under either the Blender Option or the more stringent refiner option (which would regulate ethanol
producers similar to gasoline refiners), the resulting E16-E83 "produced would be required to meet the
downstream sulfur, RVP, and a benzene per-gallon cap standard in order to ensure emission performance
over the life of the vehicles that use it." FQS for FFVs Memo, supra, at 4.
Commenter: Independent Fuel Terminal Operators Association (IFTOA)
Currently, only RFG, conventional gasoline, and RBOB may be used to produce gasoline with
high concentrations of ethanol - E51 to E83. EPA recognizes that there will be a greater need
for outlets for ethanol as the Renewable Fuel Standard Mandates increase. Therefore, EPA is
encouraging the use of E51 to E83 fuels. EPA would like to remove regulatory obstacles to its
production while ensuring that the product meets the fuel quality requirements needed to
maintain vehicle emission performance.
EPA has stated in the Preamble to the proposed regulations that the exclusive use of RFG,
conventional gasoline, and RBOB in the production of such higher ethanol-concentrated blends
may interfere with the expansion of the E51 to E83 market. EPA notes that with higher ethanol -
concentration fuels, blenders cannot meet volatility specifications set by ASTM for cold start
performance. Therefore, EPA is proposing to apply to E51 to E83 blends, the same sulfur, RVP,
and benzene standards applicable to gasoline as well as the "substantially similar" rule
requirements. Application of such regulations would allow the production of E51 to E83 using
butane and natural gasoline liquids, which, in turn, would allow the blends to meet ASTM
volatility specifications. In addition, because butane is a relatively inexpensive component, its
use should reduce the cost of these higher ethanol-concentration blends, making the final product
- E51 to E83 - more attractive to consumers.
While Members of the Association understand EPA's desire to expand the E51 to E83 blend
market, it is essential that EPA ensure that whatever actions are taken, the authority for use of a
one pound RVP waiver during the summer months be retained. As more and more ethanol is
added to blends, the waiver becomes more important. If it were eliminated, refiners would have
great difficulty meeting RVP standards, and would have to use more costly components with
lower RVP (e.g., alkylates) in their gasoline production. Such actions could substantially
increase the ultimate fuel price to consumers.
Recommendation: Accordingly, if EPA decides to amend the regulations to expand the E51 to
E83 blend market, it should also ensure that the RVP waiver for summer months is preserved.
5-143
-------
Chapter 5: Tier 3 Fuel Sulfur Program
While encouraging the use of higher ethanol-concentration blends - E51 to E83 - preserve the
one pound RVP waiver during the summer months;
Commenter: Magellan Midstream Partners, L.P.
Magellan Comment — we have developed an Ethanol Flex Fuel (EFF) program which
exclusively utilizes gasoline as the hydrocarbon portion of the EFF mixture. All EFF distributed
from Magellan's terminals meet ASTM-5798-11 everyday of the year. It is indeed feasible to
utilize gasoline as the sole hydrocarbon component in producing EFF. Our EFF program meets
the quality expectations of our customers, the automobile industry and others.
Generally, we applaud EPA for taking the step to regulate higher level ethanol blends such as
EFF. This regulatory action is overdue. Magellan agrees that EFF should also be required to
meet the same sulfur, Reid Vapor Pressure and benzene standards which are applicable to
gasoline and the interpretive rule defining the phrase 'substantially similar' in CAA section
211(0(1) (i.e., contain only carbon, hydrogen, oxygen, nitrogen, and sulfur) that apply to any fuel
used in an FFV requirement. While we appreciate EPA's proposals to expand the use of natural
gas liquids (NGLs) as the hydrocarbon portion of the EFF blend, this increases the potential for
the development of EFF which may not meet stakeholder expectations unless stringent standards
are established. However, we do not believe the use of NGLs should be permitted in EFF until
adequate testing is complete and appropriate standards are established.
Although NGLs have been used in gasoline, they have not been used for the production of EFF,
or have not at least been tested under controlled conditions to assess performance and establish
specifications. We do not believe EPA should proceed with rules which allow an untested fuel in
the market place. The use of very light hydrocarbons should be fully tested and incorporated into
specifications and regulations only after they have been demonstrated to provide satisfactory
performance. In our view, the use of NGLs should not be allowed as the hydrocarbon piece of
EFF solely to reduce the cost of the blend when compared to the utilization of gasoline.
We have reviewed EPA's options as explained in the April 8, 2013 memorandum to the docket.
Option 1 requires the EFF producer (i.e. denatured fuel ethanol producers) to assume the full
responsibilities of a refiner. While we believe the EFF producer should be required to register
and report as a refiner, EFF producers who may be allowed to blend NGLs in the future and
refiners currently blending NGLs into gasoline should be allowed to utilize the additional
flexibility provided in Option 2 as long as the respective hydrocarbon components meet well
established specifications.
Compliance burden when blending NGLs into Gasoline / Blendstock Representation: Magellan
Comment — as stated above, we do not believe EPA should permit the use of NGLs as the
hydrocarbon portion of EFF until adequate testing is complete. Once testing is complete and
satisfactory standards have been established, we would support the use of NGLs in EFF. We
applaud EPA's options raised in the EFF preamble regarding the compliance burden normally
associated with refiner sampling and reporting. We believe this approach should promptly be
made available to refiners producing gasoline using NGL blendstocks in gasoline.
5-144
-------
Tier 3 Summary and Analysis of Comments
To this end, we propose the following Blendstock Specifications for gasoline blenders: We
believe adoption of these specifications will allow additional flexibility for blending which
would lower the overall cost impact of the Tier 3 fuel requirements in general.
Lastly we agree that it is important to reduce the compliance burden for blending natural
gasoline into ethanol as a denaturant, just as it is for gasoline blenders. As EPA considers
regulatory compliance requirements for ethanol and flex fuels, as a matter of fairness, we
encourage the agency to provide similar compliance requirements for all fuel production from
gasoline range blend stocks.
Commenter: Manufacturers of Emission Controls Association (MECA)
MECA also supports defining fuel sulfur average limits or caps on any alternative transportation
fuels (e.g., blends of alcohols with gasoline, natural gas) that are consistent with the proposed 10
ppm national average for gasoline or the existing 15 ppm national sulfur cap on diesel fuel.
Commenter: Marathon Petroleum Company LP (MFC)
Standards for Mid-Level Ethanol Blends (E16-50): MFC agrees that Mid-Level Ethanol Blends
(El6-50) for use in FFV should meet the same sulfur, RVP, and benzene standards otherwise
applicable to gasoline. MFC does not agree that EPA should treat E16-50 as gasoline under
current regulations. In MFC's view, this action would require formal rulemaking with a waiver
from the substantially similar requirements.
EPA is proposing that gasoline deposit control requirements be removed from the El6-50 and
E51-83 fuels. MFC believes that it is in the best interest of the consumer to maintain a deposit
control requirement for all spark-ignition fuels at a minimum level to protect the engines of
consumers and to reduce air emissions.
Standards for E51-83: EPA's preamble to the proposed Tier 3 regulations references a separate
memorandum25 that includes additional proposed regulations for which EPA seeks comments.
These additional regulations include prescribing new requirements for E51- E83 for key fuel
properties such as sulfur, RVP, and benzene. While MFC may support EPA's effort to develop
regulations for E51-E83, the process EPA used to inform the regulated community runs afoul of
the federal Administrative Procedures Act (APA) and the federal Clean Air Act (CAA).
The APA is clear that EPA's forum for informing the public of substantive rules, amendments,
and revisions is through the Federal Register. Further, the APA states "except to the extent that a
person has actual and timely notice of the terms thereof, a person may not in any manner be
required to resort to, or be adversely affected by, a matter required to be published in the Federal
Register and not so published."
Section 307(d)(3) of the CAA echoes the requirement for proposed rules to be published in the
Federal Register. Among other things, the notice must "specify the period available for public
comment." In addition, at the time of publication, there must be a publicly available docket that
includes: (1) the factual data on which the proposal is based; (2) an explanation of how EPA
5-145
-------
Chapter 5: Tier 3 Fuel Sulfur Program
collected and analyzed the data; and (3) an explanation of the major legal interpretations and
policy considerations underlying the proposal.
In the instant matter, EPA describes and seeks comment on very significant potential changes to
the regulations. EPA is required, by law, to publish such a proposal in the Federal Register.
Posting on EPA website or including within the rulemaking docket is not an option - publication
in the Federal Register means publication in the Federal Register. EPA must include all proposed
changes in the proposed rule. References to documents external to the proposed rule do not give
the regulated community and public the opportunity to which it is entitled to generate informed
and timely comments regarding the proposed rule.
Notwithstanding the APA requirement to publish proposed rules within the Federal Register,
MFC would like to provide the following comments to the Jeff Herzog Memorandum dated
April 8, 2013 that is posted within the docket (ID: EPA-HQ-OAR-2011-0135-0529):
MFC agrees that E51-83 for use in Flexible Fueled Vehicles (FFV) should be required to meet
the same sulfur, maximum Reid Vapor Pressure (RVP), and benzene standards otherwise
applicable to gasoline, as well as the Clean Air Act Section 21 l(f)(l) substantially similar
provision. EPA should also allow for the use of butane or other natural gas liquids (e.g. pentane,
natural gasoline, etc.) to manufacture E51-83 with sufficient volatility to meet the ASTM D5798
specifications. MFC does not support limits on specific components, but does support adoption
of reasonable standards to facilitate blending.
However MFC has identified a serious concern with the suggested 0.20 volume percent benzene
cap for E51-83. Requiring E51-83 to meet a 0.20 volume percent benzene cap would effectively
preclude the blending of gasoline blendstock and natural gasoline to manufacture E51-83. As
EPA states, refiners and importers of gasoline are subject to a 0.62 volume percent annual
average and 1.3 volume percent maximum average benzene standard. Furthermore, natural
gasoline typically contains in excess of 1 volume percent benzene. As an example, E5183
composed of 49 percent gasoline blendstock with 1.3 volume percent benzene and 0.01 percent
natural gasoline denaturant with 1 volume percent benzene could have a resulting benzene
content exceeding 0.65 volume percent. In another example, E51-83 composed of 17 percent
gasoline blendstock with 1.3 volume percent benzene and 1.7 percent natural gasoline denaturant
with 1 volume percent benzene could have a resulting benzene content exceeding 0.24 volume
percent. Therefore, the setting of a 0.20 volume percent benzene cap would interfere with the
expansion of E51-83 into the marketplace by preventing blends at the both the upper and lower
range of the allowable ethanol content. To resolve this issue, MFC recommends that E51-83 be
required to meet the same benzene standards applicable to gasoline.
MFC support s EPA's suggested requirement that compliance with the Section 21 l(f)(l)
substantially similar provision would be effectively achieved by ensuring that E51-83 be
composed solely of carbon, hydrogen, nitrogen, oxygen, and sulfur.
MFC agrees that manufacturers of E51-83 for use in FFV should also be subject to the full
responsibilities of a refiner (e.g. registration; batch sampling and testing; and reporting
obligations), which is outlined as Option 1 within the Herzog Memorandum. However, if EPA
5-146
-------
Tier 3 Summary and Analysis of Comments
wishes, as a second option, to allow oxygenate blenders to be exempt from the sampling, testing
and reporting obligations of refiners, oxygenate blenders should be required to only use gasoline
blendstocks, denatured fuel ethanol, natural gas liquids (NGL) and butane that meet certain
specifications. MFC agrees with this two-option approach as a means to not restrict refiners or
oxygenate blenders from choosing maximum blending flexibility by meeting the more stringent
requirements, as outlined within Option 1. MFC also agrees that any resulting E51-83 should
also be required to meet the downstream sulfur, RVP and benzene per-gallon cap standard.
MFC agrees that it may be possible to develop a calculative approach for E51-83 blenders to
determine RVP and avoid sampling and testing. We agree that testing may be required for some
period of time when a blender receives a new batch of gasoline blendstock. We also agree that
RVP of different batches of DFE would not vary significantly due to TTB requirements related
to denaturants and denaturant percentages, along with RFS2 requirements related to denaturant
concentrations for RIN generation.
MFC agrees that butane used for E51-83 blending should meet the same standards as specified
for butane blended into gasoline at a terminal. However, the limits on other components in E51-
83 should be restricted to their total contribution toward regulated properties on the final fuel, not
the same limits imposed on the components. Sources for NGLs should not be limited, provided
the NGL conforms to the requirements.
Although MFC does not believe that the Herzog Memorandum is a part of the proposed Tier 3
regulations since it has not been published in the Federal Register, MFC provides the following
specific comments to the draft regulatory text:
80.2 Definitions: New paragraph 80.2(cccc), which will provide the definition for E51-83,
includes the reference to the ASTM D5798-11 standard for ethanol fuel blends for FFV's. This
reference should be updated to reflect the currently published version ASTM D5798-13 and
should be updated further if a newer is adopted by ASTM.
80.1601 Fuels subject to the provisions of this subpart: EPA proposes adding item 80.1601(a)(4)
to explicitly include E51-83 as being subject to the provisions of Subpart O regarding gasoline
sulfur. However, we do not see an explicit addition to also include E16-50. Assuming EPA
believes Subpart O regarding gasoline sulfur should also apply to E16-50, then E16-50 should
explicitly also be listed in 80.1601 (a).
Under, 80.1601(b)(2), "California gasoline as defined in §80.1600 subject to the provisions of
§80.1654" is not subject to the standards and requirements of Subpart O. It is unclear if and how
the Subpart O requirements apply to E51-83 and/or E16-50 sold in California.
80.1608 Standards and requirements for E51-83: Apart from the sulfur dates via the requirements
in 80.1603(a)(2), paragraph 80.1608 does not specify an effective date for the standards and
requirements for E51-83. EPA needs to insert a clear effective date for all of the standards and
requirements for E5183 in 80.1608. We infer this is sometime in 2017, but seek this clarity be
added into the regulation.
5-147
-------
Chapter 5: Tier 3 Fuel Sulfur Program
80.1608(b)(2) as proposed sets the maximum benzene content of each batch of E51-83 at 0.20
volume percent maximum. This benzene limit is too low, because it is inconsistent with EPA's
clear intent to allow E51-83 blenders to manufacture E51-83 using gasoline, gasoline
blendstocks for oxygenate blending, denatured fuel ethanol, natural gas liquids, and/or butane
per 80.1611. EPA needs to eliminate the explicit benzene limit in E51-83, when a blender
manufactures this fuel under the provisions of 80.1611. Otherwise, EPA needs to increase the
explicit benzene limit to reflect EPA's intent to allow E85 production using compliant RFG,
RBOB, CG, and/or CBOB.
Section 80.1608(b)(4) listing the RVP requirements for E51-83 contains the typographical error
"80.1627(a)(2)" when "80.27(a)(2) is intended by EPA.
EPA should clarify in 80.1608(b)(4) where it requires that the Reid Vapor Pressure of "each
batch" of E51-83 must comply with same standards applicable for gasoline in 80.27(a)(2) that
being a "batch requirement" this requirement only applies to E51-83 blenders. However, if EPA
also intends for the RVP requirements to be applicable during the "high ozone season" from June
1st through September 15* of each year at retail sites, then EPA has to recognize and address that
the potentially far slower turnover of E51-83 inventory at retail sites (relative to gasoline) may
make an RVP requirement at retail sites infeasible and dissuade retailers from offering E51-83.
EPA should note that ASTM recognized and addressed this issue of potentially slow moving
E51-83 at retail sites by exempting the E51-83 at a retail site from ASTM's seasonal RVP
requirements in ASTM's "E51-83" standard D5798 Table 2 with notel as follows (emphasis
added),
"This schedule, subject to agreement between the purchaser and the seller, denotes the vapor
pressure class of the fuel at the time and place of bulk delivery to fuel-dispensing facilities for the
end user. Shipments should anticipate this schedule."
80.1608(c)(3) provides the benzene standard for E51-83 when it is manufactured by a party that
does not qualify as an E51-83 blender:
1. Similar to our comments regarding the 0.20 vol% benzene limit for E51 -83 blenders in
80.1608(b)(2) being too low. 0.20 vol% benzene listed in 80.1608(c)(3) for E51-83
"refiners" is also too low and should be revised consistent with our recommendations for
E51-83 blenders.
2. We believe "80.1632" is a typographical error with regard to the test requirements for
benzene in E51-83 produced by a "refiner". We believe EPA means and should correct
this to "80.1630" which addresses test requirements.
80.1611 Blended E51-83 requirements: EPA needs to insert a clear effective date for the blended
E51-83 requirements in 80.1611. We infer this is sometime in 2017, but seek this clarity be
added into the regulation.
As proposed by EPA, 80.161 l(a) and (b) require that only RFG or RBOB be used as the gasoline
or gasoline blendstock for oxygenate blending in E51-83 intended for sale in RFG areas. EPA
has not provided a basis for this limitation based on air quality impact. Absent EPA providing a
5-148
-------
Tier 3 Summary and Analysis of Comments
reasonable basis, these requirements should be deleted from EPA's proposed rules, allowing
conventional gasoline and/or CBOB to be used as well for E51-83 in RFG areas.
80.1612 Standards and requirements for natural gas liquids blendstock used by E51-83 blenders:
80.1612(b) as proposed includes a prohibition on using E51-83 made from NGL in a blend with
gasoline to make E16-50. We do not understand EPA's logic in this prohibition and ask that EPA
explain its reasoning further.
80.1662 Liability for violations under subpart 0: For a sulfur cap standard violation in
80.1660(b), paragraph 80.1662(a)(5) includes a branded refiner/importer liability as follows
(emphasis added),
"Branded refiner/importer liability. Any refiner or importer whose corporate, trade, or brand
name, or whose marketing subsidiary's corporate, trade, or brand name appeared at a facility
where a violation of §80.1660(b) occurred, is deemed in violation of §80.1660(b)."
However, this branded refiner/importer liability is too broad when EPA is now proposing to
further expand the sulfur cap requirements in 80.1660(b) to include El6-50 and E5183 fuels sold
at retail sites. At retail, the branded refinery has no control on the retailer who might
independently make these product offerings from source of his/her own choosing. Branded
refiners should not be liable in these cases simply because their brand name "appeared at a
facility", especially in the case where the branded refiner contractually required the retailer to
identify at the dispenser that the E16-50 or E51-83 in question was not a branded product. State
laws and the markets have changed. At present jobbers can do their own blending. Thus the rules
dealing with presumptive liability no longer make little sense since refiners no longer control the
actions of many jobbers.
25 Herzog, J. (January 2012). Possible Approach to Fuel Quality Standards for Fuels Used in Flexible-
Fuel Automotive Spark-Ignition Vehicles (FFVs), Memorandum to the docket. The memorandum is
referenced in footnote 388 of the pre-publication proposed rule; however, the footnote reference is to a
memorandum dated January 2012 but the memorandum in the docket is dated April 8, 2013. Further, this
memorandum was not in the docket at the time the proposed rules were announced on EPA's website.
Commenter: Mid-Continental Energy (MCE)
EPA is proposing the definition of any fuel that has less than 50% ethanol content be classified
as gasoline. MCE supports that definition. For clarification in this document, any ethanol fuel
that contains more than 51% ethanol content will be referred to as a Flex Fuel.
In addition, EPA is proposing Flex Fuel have a sulfur content not to exceed the gasoline sulfur
mandate. MCE agrees with EPA in their proposal.
MCE supports the EPA premise of using RVP models to calculate and certify the final RVP of
Flex Fuel produced from butane, natural gasoline, and denatured fuel ethanol ("DFE")
blendstocks which meet EPA specifications.
5-149
-------
Chapter 5: Tier 3 Fuel Sulfur Program
EPA is proposing allowing natural gasoline and butane as blendstocks used to produce E51 to
E83 Flex Fuel. Our understanding is that natural gasoline specification standards will be defined
by the EPA as butane was defined in Federal Register Title 40 Part §80.82. MCE requests that
EPA specification standards for natural gasoline should: 1) be reasonably consistent with current
industry specifications and standards, 2) not limit origin of the natural gasoline so long as
specifications are met, 3) be mindful of final, blended Flex Fuel properties including vapor
pressure, octane, benzene, and sulfur. MCE will comment on specific natural gasoline
specifications within 90 days of this letter. MCE is conducting studies on natural gasoline quality
and blending, and MCE can provide the study findings to the EPA on a confidential basis.
MCE requests that a blender of natural gasoline to produce Flex Fuel would be allowed to blend
approved butane into the natural gasoline (blendstock) before the natural gasoline is blended
with DFE to meet the seasonal RVP requirements as allowed in EPA specifications for Flex
Fuel. The blender will use an RVP Model approved by EPA to calculate the finished RVP of the
Flex Fuel instead of having to test each batch of Flex Fuel leaving the terminal facility. The
terminal blender would not have to recertify the blendstock provided the two components met all
specifications approved by EPA before blending the two components together.
MCE requests the EPA provide clarification on how Flex Fuel can be produced and blended. Can
DFE producers blend Flex Fuel with existing ethanol and denaturant infrastructure? If a DFE
producer blended approved blendstock with DFE to produce a Flex Fuel and then shipped the
Flex Fuel to a petroleum terminal, would the terminal blender be allowed to blend butane into
the Flex Fuel to increase the RVP to the maximum allowable RVP specifications per EPA
seasonal specifications for Flex Fuel before the Flex Fuel is loaded into a transport truck? Could
a petroleum terminal use Flex Fuel to blend El 0-15 gasoline at the terminal in place of DFE?
MCE would favor allowing the terminal blender to use a process analyzer to determine the RVP
of the Flex Fuel stream in the process of blending butane into the Flex Fuel to increase the RVP
to the maximum allowable for the seasonal variations. This is the same process presently being
used by terminal blenders to blend butane into gasoline under the guidance in the Federal
Register Title 40 Part §80.82.
During the VOC period, MCE requests the EPA consider allowing blenders to blend butane and
natural gasoline with DFE to produce Flex Fuel in RFG markets as long as the maximum RVP of
the Flex Fuel does not exceed maximum RVP specifications as calculated by EPA approved
RVP models. During the non-VOC season, the maximum RVP during would be under States
rulemaking for Flex Fuel specifications.
Under the present definition of a 'batch' in Federal Register Title 40 Part §80.82, EPA looks at
each transport truck load leaving a petroleum terminal with gasoline for a service station as a
'batch'. EPA is proposing to allow a terminal blender to blend natural gasoline with DFE to
produce Flex Fuel during the VOC period using an approved EPA model to calculate the
finished blend RVP. This is a change to what is allowed under Part §80.82 for a butane blender.
To make the rules like-minded, MCE is proposing EPA remove the language in §80.82(e)(l)
which states 'When butane is blended with conventional gasoline under this section during the
period May 1 through September 15, the refiner shall demonstrate through sampling and testing,
5-150
-------
Tier 3 Summary and Analysis of Comments
using the test method for Reid vapor pressure in § 80.46, that each batch of conventional
gasoline blended with butane meets the volatility standards specified in § 80.27.'
The terminal blender can use the model provided by EPA to calculate the RVP of the blended
gasoline, install a downstream process analyzer to use as an oversight reporting analyzer for
determining the RVP of the blended gasoline. RVP process analyzers on the market today take
five to six minutes to produce results and there are several manufacturers of process RVP
analyzers making downstream RVP a demonstrable option. MCE is not requesting to make the
allowance applicable to gasoline marketed in RFG areas.
MCE suggests the EPA consider new definitions for NGL, natural gasoline liquids, and/or
pentanes. The proposed wording is potentially confusing and may not be consistent with industry
standards. The proposed regulation defines NGL as: "the components of natural gas (primarily
propane, butane, pentane, hexane, and heptane) that are separated from the gas state in the form
of liquids in a natural gas production facility, in a gas processing plant, or in a natural gas
pipeline. NGL is sometimes also referred to as "natural gasoline".
Natural gasoline is a product of the NGL fractionation process consisting primarily of pentane
and heavier components. MCE believes the EPA's intends to allow Flex Fuel production using
natural gasoline and not NGL as technically defined in the proposed regulations.
Commenter: National Corn Growers Association (NCGA)
EPA divided its proposal for in-use fuels for FFVs into two levels by ethanol concentration: E51-
83 andE16-50 (orMLEBs).
For E51-83, EPA proposed that these blends meet the same sulfur, vapor pressure, and benzene
standards otherwise applicable to gasoline as well as the Clean Air Act "substantially similar"
requirements. EPA expressed the belief that establishing such requirements could allow the use
of butane and natural gasoline liquids to manufacture E51-83 with sufficient volatility to meet
the ASTM specifications. We support specifications that are consistent with ASTM D5798-13,
and allowing the use of blendstocks such as NGL, butane and others that would permit
compliance with ASTM D5798-13.
For E16-E50, EPA proposed that these blends meet the same standards that apply to gasoline.
EPA also requested comments on "...whether it might be an appropriate reading of our regulatory
and statutory authority to allow El6 to E50 blends to have higher RVP levels than otherwise
required by our regulations for gasoline." EPA went on to say: "As the ethanol level increases
the volatility increase caused by blending ethanol with gasoline begins to decline, such that at
E30 there is only about a 0.5 psi vapor pressure increase. While still an increase compared to the
standards that apply to gasoline other than E9-E10, it is considerably less than the full 1 psi
vapor pressure increase that results at E10. The evaporative emission increase caused by E30
would be then be less than that for E10."
NCGA believes that EPA should use its regulatory and statutory authority to allow El6-50
blends to have higher vapor pressures, so that gasoline blendstocks used currently for E10 could
5-151
-------
Chapter 5: Tier 3 Fuel Sulfur Program
also be used for E16-E50, and these MLEBs could be used in current FFVs. This is critical to the
success of high octane MLEBs, since it would facilitate distribution of a high octane MLEB that
could likely be requested by automakers to meet the 2017 and later GHG and CAFE standards.
At least two major automakers have reported research programs evaluating advanced technology
gasoline direct engines with high octane MLEBs (18, 19).
18 - McCarthy, T "Fuel Octane and Ethanol Impacts on Petroleum Displacement, CO2, and the
Consumer, Ford Motor Company, presentation at SAE High Octane Fuel Symposium, January 29, 2013.
19 - Woebkenberg, W, "Advanced Powertrain Technology Coupled with Octane and Ethanol, Benefits
and Opportunities", SAE Fligh Octane Fuels Symposium, January 29, 2103.
Commenter: New York State Department of Environmental Conservation
The proposed limits on the sulfur content of blending streams such as denatured fuel ethanol are
necessary to ensure that the program benefits are actually realized.
Due to ethanol blending and an increase in trade of intermediate gasoline blending components
over the last several decades, the blending of finished gasoline is no longer limited to crude oil
refineries. In general these blenders have become refiners in the regulatory sense, with all the
compliance documentation that entails. EPA has made efforts to allow some downstream
blending activities, particularly ethanol blending, to be accomplished with a reduced level of
documentation.
The Department supports this reduced level of documentation for downstream blending,
provided that the environmental benefits promised by Tier 3 are not compromised. This requires
that the materials blended be sufficiently controlled to ensure that the finished gasoline will
comply with applicable regulations. These streams must, as EPA proposes, comply with a 10
ppm sulfur cap on their own. We believe that other properties, particularly benzene and olefms
content, may also need to be capped to prevent backsliding.
...and only RFG compliant gasoline or blendstock for oxygenate blending (BOB) may be used to
produce 'E51-83' for use in RFG areas. This latter requirement makes it difficult to produce fuel
that has both high ethanol content and meets the minimum ASTM volatility requirements.
It is also necessary to regulate commercial E51-83 to ensure that emissions controls are effective
in the field.
The EPA staff memo discusses two possible approaches, in one approach the blender of the fuel
is treated as a refiner, with substantial compliance documentation requirements in exchange for
production freedom. The other approach, closer to what is done today for ethanol blending into
BOBs to produce finished gasoline, reduces the compliance burden but limits blenders to
blendstocks with known and regulated composition.
Clearly the 'ethanol blender' approach has advantages for all concerned as long as environmental
and fitness-for-use requirements can be reliably met. Properties such as sulfur content, benzene
5-152
-------
Tier 3 Summary and Analysis of Comments
content, aromatics content, and olefms content can be controlled by ensuring that all allowable
blendstocks meet standards that ensure the final product will meet its standards.
If high volatility blendstocks such as butane and natural gasoline are used as blendstocks it
becomes possible to produce fuel that exceeds gasoline DVPE limits. EPA appears to be
searching for a mechanism to allow this without any need to sample the resulting product and
conduct laboratory tests. We believe that sampling and testing for DVPE can't be avoided,
although it may not be necessary to obtain sample results prior to releasing each truck load of
product. Assuming that blending occurs during delivery truck loading, it may be sufficient to
prepare and test a hand blend of the components to be used, and periodically sample truck
compartments as part of a documented quality control program to ensure that the planned blend
is being achieved in practice (without holding the truck for the test results).
Commenter: Phillips 66 Company
- E51-83 - The memorandum included in the docket should not suffice as a proposed rule. We
have reviewed the concept and have some comments.
EPA included a memorandum in the rulemaking docket entitled "Possible Approach to Fuel
Quality Standard for Fuel Used in Flexible-Fuel Automotive Spark-Ignition Vehicles (FFVs)".
We take this to be informational concerning work EPA has done to evaluate potential regulatory
changes to provide specifications and guidance for these fuels. We do not believe this constitutes
a proposed rule even though EPA has included draft regulatory language. In order for this to
constitute a NPRM, the proposed language and preamble would need to be published in the
Federal Register.
We offer the following comments on some of the concepts that EPA discusses in the memo:
- Fuel name should be consistent on product transfer documents and labels EPA has coined the
term E51-83 to describe what historically has been referred to as E85. While we agree that E85 is
now an outdated term given the recent changes in the ASTM specification, there may be other
options for naming convention of this fuel. EPA included in its conceptual PTD language a
requirement to put a statement identifying the fuel as E51-83. Any name or reference on the
PTDs should align with potential label nomenclature. The industry is still wrestling with what to
call this fuel now that ASTM has modified its specification. However, consumers could be
confused by a label that calls the fuel E51-83. We would suggest there are other options such as
Flex Fuel (typically what is now shown on new vehicles and their gas caps) or Ethanol Fuel or
Ethanol Flex Fuel. We suggest EPA not rush to create a designation for this fuel and require it on
a PTD that may not be in alignment with pump labels.
- Agree that Flex Fuel (E51-83) should be required to meet the same sulfur, benzene and
maximum RVP standards as gasoline
- The suggested 0.20 benzene cap for E51-83 is problematic
Some hydrocarbon blendstocks that could be used to produce flex fuel (such as natural gasoline)
could result in a benzene content greater than 0.20. We feel that the E51-83 should meet the
same standards as gasoline.
- Providing two options for sampling, testing, and reporting is a good approach
5-153
-------
Chapter 5: Tier 3 Fuel Sulfur Program
Commenter: POET, LLC
Also, mid-level ethanol blends that contain 16-50% ethanol (MLEBs) are slated to be an
important 'fuel of the future.' MLEBs have been assessed by auto manufacturers as providing a
'sweet spot' in terms of a cost-effective fuel that can provide octane benefits (enabling greater
engine efficiency) and lower overall vehicle emissions.
Additionally, EPA must not burden the ethanol industry with additional, unnecessary production
restrictions.
The Proposed Rule takes comment on 'whether it might be an appropriate reading of our
regulatory and statutory authority to allow E16 to E50 blends to have higher RVP levels than
otherwise required by our regulations for gasoline.' (42) POET suggests that doing so is not only
appropriate, but essential (with this regulatory relief including blends of El 1 and higher).
This RVP issue can be handled for ethanol blends of E16 and higher by, as described later in
section 4 of these comments, by creating a separate 'Ethanol Fuel Family' for all blends of El 6
and higher (e.g., E16-E85). The RVP provisions in CAA Section 21 l(h) by their terms only
apply to 'gasoline,' and thus should not apply to a blend in the 'ethanol' fuel family. (43)
EPA must ensure that the various clean-burning attributes of ethanol are not prevented from
entering into commerce by arbitrary RVP limits when in fact these MLEB and higher-level
ethanol blends have a lower (environmentally-preferable) RVP. EPA states that it has an ability
to broadly craft RVP standards for the ethanol fuel family per CAA Section 21 l(c). (44) If EPA
can meet the statutory criteria under 21 l(c) for regulating RVP from higher-level ethanol blends
(i.e., if such regulations are necessary, taking into account the inherently lower RVP of ethanol),
this regulatory authority should be pursued to allow the equivalent of the 1 psi waiver for blends
of E16 and above, including all MLEBs.
E16-E85 should be regulated as a single, alternative 'ethanol' fuel family that covers MLEBs
and ULEBs. In the Proposed Rule, EPA requests comment on how to regulate and promote
MLEBs and 'aspects of today's proposed standards that might need to be modified to provide an
incentive for, or remove obstacles to, the development of highly efficient vehicles optimized for
use on higher level ethanol blends.' (50)
EPA further asks 'should the Agency simply treat E16-50 as gasoline under our regulations' or
'Alternatively, should the Agency treat E16-50 as an alternative fuel'? (51) Similarly, the 'FQS
for FFVs Memo' in the docket raises the possibility of regulating E16-E85 vehicles as a single
alternative fuel family. (52) POET supports this approach. This 'fuel family' would include both
MLEBs and HLEBs. Importantly, an ethanol fuel family provides a means to appropriately
structure RVP limits for MLEBs.
Under EPA regulations, a 'fuel family' is defined as 'a set of fuels and fuel additives which share
basic chemical and physical formulation characteristics and can be used in the same engine or
vehicle.' (53) Given that ethanol blends from E16 through E85 today 'can be used in the same
engine or vehicle' (i.e., an FFV), it is more appropriate that MLEBs be regulated within the
5-154
-------
Tier 3 Summary and Analysis of Comments
existing ethanol fuel family (which includes HLEBs), as opposed to being considered gasoline.
(54) By comparison, MLEBs cannot be used in an ordinary 'gasoline' vehicle.
An E16-E85 'Ethanol Fuel Family' can readily meet core fuel quality standards, including
appropriate chemical composition, benzene, and sulfur; a separate fuel family also provides a
means to appropriately structure RVP limits for MLEBs.
EPA believes that it is important that fuel quality requirements for chemical composition, sulfur
and benzene 'apply to any fuel used in an FFV.' (55) Regarding chemical composition, EPA
believes that it is appropriate to explicitly apply to ethanol blends 'substantially similar'
requirements that 'E51-83 must be composed solely of carbon, hydrogen, nitrogen, oxygen, and
sulfur.' (56) POET supports 'substantially similar' chemical composition requirements, and they
can be readily applied across a single Ethanol Fuel Family comprised of E16-E85 blends. (57)
Similarly, sulfur and benzene limits must be properly structured, but should not be a problem for
E16-E85 blends, given that ethanol contains no benzene and is nearly sulfur-free. (58)
Finally, as discussed above in Section 3 of these comments, it is absolutely imperative that EPA
address unnecessary and environmentally counter-productive RVP limits that prohibit the use of
MLEBs. An Ethanol Fuel Family provides a ready means to have MLEBs subject to an
appropriately designed RVP limit that is different than that for 'gasoline' but nonetheless
adequately protective. (59) HLEBs could be subject to this standard as well, although an
alternative RVP standard is less critical for HLEBs because ethanol blends greater than 50% can
already meet existing 'standard' RVP limits, due to the inherently environmentally-friendly RVP
of unblended ethanol. (60)
Benzene limits for ethanol blends must take into account the gasoline blended with ethanol.
EPA believes 'it may be possible to set the benzene standard for E51-83 at a 0.20 volume percent
cap or even lower without imposing any burden rather than implementing an average annual
benzene standard similar to that for gasoline.' (77) Gasoline has a 0.62 volume % average
maximum benzene content.
A benzene cap of 0.20% on ethanol blends would create a tremendous burden on blenders and
could render certain production impossible. For instance, an E51 blend composed of 51% DFE
and 49% certified gasoline would likely have a benzene content of 0.31% as the gasoline would
bring 0.62% benzene to the mixture. This mixture would not meet the regulatory requirements.
Under the proposed approach, E51-83 could only exist as E75-83 with DFE at a 0.06% benzene
and gasoline at 0.62% benzene. Higher benzene levels in the gasoline due to averaging would
further limit the hydrocarbon content of this fuel as well as the RVP and cold start capability in
cold weather. It is imperative that EPA provide a higher benzene limit to account for the gasoline
blended with ethanol. POET believes EPA should set this benzene standard at 0.62 volume %,
identical to the gasoline limit and the directly competitive fuel.
5-155
-------
Chapter 5: Tier 3 Fuel Sulfur Program
EPA's streamlined 'Blender Option' requirements should be the default regulations for E16-E85
production (as opposed to full refiner responsibilities).
The Tier 3 rulemaking appropriately recognizes that ethanol blending is of central importance to
our nation's transportation fuel supply and accomplishing key environmental and energy security
goals. To achieve these goals and maintain a robust domestic transportation fuel supply, it is
critical that EPA design its regulations to facilitate the production and blending of ethanol.
EPA discusses its possible approaches to the regulation of ethanol (and vehicles that use ethanol)
in the Proposed Rule, including the section 'Standards for Fuel Used in Flexible Fueled
Vehicles.' (78) This section then references a detailed memo to the docket, the FQS for FFVs
Memo.
The FQS for FFVs Memo outlines two options for regulating manufacturers of E51-E83. POET
strongly supports 'Option 2: Treatment as a Blender' (the 'Blender Option') as the default
option. Under this Blender Option, EPA states that E51-E83 producers can reduce their
regulatory burden if they only use blendstocks that meet certain specifications that would ensure
sulfur and benzene do not exceed appropriate amounts. (79)
More specifically, under this Blender Option, EPA proposes generally to limit blendstocks for
ethanol blends to gasoline, BOBs, and DFE, as well as natural gas liquids (NGLs) and butane if
they are produced to certain standards. Under this Blender Option, the finished E51-83 would be
subject to the proposed downstream sulfur cap for gasoline of either 65 ppm or 95 ppm. (80) A
benzene limit would also apply to the finished blend (discussed above).
POET agrees that testing of final blends would not be needed as the DFE and blendstocks are
regulated individually for sulfur and benzene (and RVP issues can be handled in a streamlined
fashion as well).
POET also believes that the less-stringent Blender Option requirements could apply equally to
E16-E50 as well as E51-E83. The FQS for FFVs Memo also notes that 'the draft regulations
written to address E51-83 might be modified to cover E16-50 blends.' (81) POET believes that
this Blender Option can be applied across an Ethanol Fuel Family from E16-E85 (i.e., across
both MLEBs and ULEBs). If EPA considers it necessary to further delineate MLEB blending,
POET suggests that applicable ASTM standards already provide a suitable approach. In
particular, ASTM Standard D7794 allows MLEBs only to be created from a mixture of gasoline
and DFE, or gasoline and HLEBs. (82) Thus, the constituents of MLEBs (gasoline, DFE, and
HLEBs) are already controlled. This is consistent with the streamlined 'Blender Option' that
avoids batch testing for blends when the constituents of those blends are already controlled.
While POET supports EPA's Blender Option approach, blendstocks for ethanol blends (i.e.,
those substances blended with DFE to achieve a certain ethanol/'gasoline' blend) should not be
regulated more than is necessary to accomplish EPA's emission-control goals, as further
discussed immediately below.
5-156
-------
Tier 3 Summary and Analysis of Comments
Under the Blender Option, gasoline blendstocks should be allowed for blending (in addition to
NGLs and butane) as long as they meet adequate specifications.
Light straight run naphtha, hydrotreated naphthas, and hydrotreated natural gasoline are potential
streams to consider. These gasoline blendstocks could be subject to the same sulfur and benzene
specifications as NGLs and butane and should be allowed as blendstocks for ethanol blends.
POET agrees with EPA that addressing RVP in blends does not require batch testing.
EPA states that 'For blenders that use only gasoline, BOB's, and DFE to produce E51-83,
demonstration of compliance with the maximum RVP standard under consideration could be
accomplished by retention of PTDs [Product Transfer Documents] for the blendstocks.' (83)
Additionally, EPA raises the issue of whether the SAE 'Model for Estimating Vapor Pressures of
Commingled Ethanol Fuels' is adequate to avoid testing for NGLs and refinery naphthas for
RVP. (84) POET agrees that the SAE model referenced in the FQS for FFVs memo should be
adequate to avoid testing for NGLs and refinery naphthas for RVP. (85)
EPA states that 'Given the proposal in the Tier 3 NPRM to limit the denaturants that can be used
to manufacture DFE to NGL, gasoline, and BOBs, and the narrow range in denaturant
concentration, we believe that the RVP of different batches DFE would not vary substantially.
Therefore, we currently do not believe that deliveries of new batches of DFE into the storage
tank used to produce E51-83 would necessitate a return to per batch testing.' (86) POET agrees
that the variability of the ethanol RVP would not vary substantially, and batch testing should not
be required, with a 5% denaturant cap.
POET supports facilitating the use of NGLs under the Blender Option in the production of
ethanol blends.
POET does not believe that any EPA limits on the volume of NGL used as a blendstock are
necessary for HLEBs (E51-E83). For MLEBs (E16-E50), if EPA considers that some NGL
limits are necessary, POET suggests that applicable ASTM standards already effectively limit
the NGL content in MLEBs. In particular, as noted above, ASTM Standard D7794 allows
MLEBs only to be created from a mixture of gasoline and DFE, or gasoline and HLEBs. The
amount of NGLs allowed in these constituent products (gasoline, DFE, and HLEBs) is already
controlled. Therefore, the volume of NGL in MLEBs can effectively be no greater than what is
already contained in approved components of an MLEB.
Furthermore, POET responds to the following statements by EPA regarding NGLs under the
Blender Option:
- Production 'using NGL blendstocks would require that the NGL blendstock, like butane, meet a
10-ppm per-gallon sulfur cap.' (87)
- POET agrees that NGL (or gasoline blendstocks such as hydrotreated refinery streams) used for
blending E16-E83 should meet the proposed 10 ppm sulfur cap under the Blender Option and
therefore avoid burdensome batch testing requirements for blending.
5-157
-------
Chapter 5: Tier 3 Fuel Sulfur Program
- EPA notes that 'Similar to the envisioned 0.20 volume percent benzene cap for butane, we
would look to implement a 0.20 volume percent benzene cap standard for NGL rather than an
average standard in order to simplify the regulatory requirements.' (88)
- A benzene cap of 0.20% for NGLs is too restrictive and may be impractical to meet. The
benzene standard for NGLs should be 0.62% benzene similar to unleaded gasoline. Therefore,
this NGL stream or gasoline could be used in the blend with no detrimental effect to the finished
fuel since the DFE in which it is blended will have little to no benzene.
- 'Since NGL properties from different sources can vary widely ... we may also need to consider
restricting the sources of NGL to natural gas liquids fractionation facilities and natural gas
processing plants to ensure an appropriate level of quality control.' (89)
- EPA should maintain a provision for an alternative NGL source to allow for hydrotreated
streams or refinery streams that meet the same criteria (i.e., benzene and sulfur limits). Refineries
(including facilities where hydrotreating takes place) can provide the same or superior quality
control as NGL fractionation facilities and natural gas processing plants.
- 'Stakeholders have stated that as additive concentration diminishes due to dilution with DFE,
there is a point where the presence of a deposit control additive ceases to be beneficial and can
actually contribute to deposit formation. In light of this, it may be appropriate to remove the
requirement that the gasoline portion of E51-83 and perhaps El6-50 must contain a deposit
control additive until the specific deposit control needs of these blends can be evaluated.' (90)
- POET supports removing the requirement that the gasoline portion of E51-E83 must contain a
deposit control additive until the specific deposit control needs of these blends scan be evaluated.
A similar approach may be warranted for E16-E50.
EPA should not unduly regulate E16-E83 blendstocks. EPA states that 'Currently, only ...
RBOBs ... and gasoline ... are used to manufacture E51-83.' (91) This is inaccurate. Gasoline
blendstocks and natural gasoline can also be used in E51-E85 today. (92)
Some ethanol producers may wish to use non-standard blendstocks, which they should maintain
the option to do as long as the end product meets appropriate specifications. Ethanol producers
should have the option to do so, if they opt to undertake broader 'refiner' responsibilities. These
'refiner' responsibilities could require batch testing of the end product, and some ethanol
producers may consider this extra cost and burden appropriate. (93)
The proposed definition of NGLs in 40 CFR 80.2 is too restrictive. (97) The definition should
not include an unduly narrow listing of the types of facilities that produce NGLs and should
include, for instance, NGLs that may be produced through refineries and hydro-treating. The
definition should be revised as follows (with additions underlined): '(zzz) Natural Gas Liquids
(NGL) means the components of natural gas (primarily propane, butane, pentane, hexane, and
heptane) that are separated from the gas state in the form of liquids in facilities such as a natural
gas liquids fractionation production facility, or in a natural gas processing plant, or in a natural
gas pipeline, or refinery or similar facility. The higher temperature boiling components of NGL
are sometimes also referred to as 'natural gasoline'.'
A requirement to 'Certify that the NGL was derived solely of from [sic] natural gas liquids
fractionation facilities and natural gas processing plants' is too narrow. (98) As noted above,
5-158
-------
Tier 3 Summary and Analysis of Comments
EPA should include refineries (including hydrotreating) as potential hydrocarbon stream sources
as long as blendstocks meet certain requirements (i.e., benzene and sulfur limits).
A requirement to demonstrate 'that the benzene content of each production batch of NGL is not
greater than 0.03 volume percent' is too restrictive. (99) This specification is unrealistic, and it is
unclear whether a product that meets this specification even exists in any significant commercial
quantities. For instance, this specification is 20 times more stringent than gasoline and wouldn't
apply to NGLs blended directly into unleaded gasoline.
A prohibition that 'NGL and E51-83 made from NGL cannot be sold as gasoline or E16 or be
commingled with gasoline or El6-50 and subsequently sold as gasoline or El6-50' is unduly
restrictive. (100) This prohibition on commingling E51-83 made with NGL with MLEBs would
unduly limit the ability for retail stations to provide E16-E83 fuels, especially if they employ
blender pumps. Alternatively, EPA should be guided by ASTM requirements as stated in
standards D4814, D4806, D5798 and standard practice D7794.
51.-Id. at29,938.
52 - See FQS for FFV Memo, supra, at 10 (stating that, instead of regulating MLEBs as 'gasoline,' EPA
'could treat El6-50 as an alternative fuel used in FFVs.').
53 - See 40 CFR 79.50 (emphasis added). EPA has designated six fuel families per 40 CFR 79.56,
including gasoline, diesel, methanol, ethanol, methane, and propane.
54 - In any possible re-categorization of MLEBs, EPA should ensure that no testing burdens are
increased, and that all other applicable regulations are appropriate for ethanol blends.
55 - 78 Fed. Reg. at 29,937.
56 - FQS for FFVs Memo, at 6.
57 - However, these 'substantially similar' requirements for ethanol should also allow the presence of
chlorine and sulfur, per the current provisions for the ethanol fuel family at 40 CFR 79.55(e).
58 - However, gasoline that is blended with ethanol does contain benzene; appropriate benzene and sulfur
limits for blends in the Ethanol Fuel Family are discussed in the section 5 of these comments.
59 - The 'FQS for FFVs Memo' recognizes that categorizing MLEBs in an ethanol fuel family would
allow RVP regulations different from those applicable to 'gasoline' to apply. Id. at 11.
60 - As noted above, it is only when ethanol is blended with significant concentrations of gasoline does
the mixture of the two substances exceed current RVP limits in certain circumstances.
77 - FQS for FFVs Memo, at 5.
78 - 78 Fed. Reg. at 29,936-38.
79 - Some ethanol producers may wish to use non-standard blendstocks, which they should maintain the
option to do as long as the end product meets appropriate specifications. Ethanol producers should have
the option to do so, if they opt to undertake broader 'refiner' responsibilities. These 'refiner'
responsibilities could require batch testing of the end product, and some ethanol producers may consider
this extra cost and burden appropriate.
80 - FQS for FFVs Memo, supra, at 7.
81 -Id. at 11.
82 - See ASTM D7794-12 (D7794) 'Standard Practice for Blending Mid-Level Ethanol Fuel Blends for
Flexible-Fuel Vehicles with Automotive Spark-Ignition Engines.' This standard incorporates by reference
ASTM standards governing the permissible constituents of MLEBs including one for gasoline (D4814),
as well as ASTM D4806-12 ('Standard Specification for Denatured Fuel Ethanol for Blending with
Gasolines for Use as Automotive Spark-Ignition Fuel') (DFE), and ASTM D5798-12 ('Standard
Specification for Ethanol Fuel Blends for Flexible-Fuel Automotive Spark-Ignition Engines') (HLEBs).
83 - FQS for FFVs Memo, at 7.
5-159
-------
Chapter 5: Tier 3 Fuel Sulfur Program
84 - By way of background, EPA stated 'If NGL and/or butane were used to manufacture E51-83,
demonstration of compliance with a maximum RVP requirement could not be accomplished merely by
retention of PTDs.... However, it may be possible to avoid sampling and testing for RVP as well. RVP
models exist that could perhaps be developed further to avoid the need for sampling and testing for RVP
when NGL or butane were used, but they are not yet applicable to these blends.' Id.
85 - See id. at 7 n. 22 (referencing SAE paper 2007-01-4006, 'A Model for Estimating the Vapor
Pressures of Commingled Ethanol Fuels.').
86-Id. at 8.
87 - Id. at 9.
88 - Id.
89-Id. at 10.
90-Id. at 11.
91 - 78 Fed. Reg. at 29,937.
92 - See, e.g., ASTM standard for H LEBs, ASTM D5798-12.
93 - Under either the Blender Option or the more stringent refiner option (which would regulate ethanol
producers similar to gasoline refiners), the resulting E16-E83 'produced would be required to meet the
downstream sulfur, RVP, and a benzene per-gallon cap standard in order to ensure emission performance
over the life of the vehicles that use it.' FQS for FFVs Memo, supra, at 4.
97 - See FQS for FFVs Memo, supra, at 13; also, 78 Fed. Reg. at 30,003.
98 - See FQS for FFVs Memo, at 20. ' Id.
100-Id.
Commenter: Renewable Fuels Association (RFA)
Existing standards and specifications for "flex fuels" (i.e., 51-83%vol. ethanol) are sufficient and
EPA should not subject these fuels to the same sulfur, RVP, and benzene standards applicable to
gasoline
RFA agrees with EPA that increased volumes of higher ethanol blends will be used in flexible-
fuel vehicles (FFVs) with implementation of the expanded Renewable Fuels Standard (RFS2).
RFA has dedicated resources to develop both the proliferation of the manufacture of FFVs and
the retail fuel offering for "flex fuels," ethanol blended fuels containing 51 to 85% ethanol
(volume.) There are effective fuel quality specifications already in existence for these flex fuels;
ASTM International publishes the fuel specification D5798-13 "Standard Specification for
Ethanol Fuel Blends for Flexible-Fuel Spark Ignition Engines." RFA supports the continued
responsibility of ASTM International, of which EPA is a member of the D5798 Task Force, to
further describe the characteristics needed for flex fuels. The ASTM D5798-13 specification
includes regulatory restrictions and has a track record of timely and frequent content update.
Recently, the D5798 specification has undergone significant improvements to better characterize
and require appropriate blending components and blending practices. Where EPA regulations
may have appeared silent on requirements for flex fuels, ASTM International has stepped in to
address these issues with robust restrictions for flex fuels. The ASTM D5798-13 flex fuel
specification does not allow inappropriate blending components. The ASTM standards process is
designed for industry and government driven consensus standards with a thorough vetting
process. RFA supports the use of NGLs, butane and other ASTM D5798-13 suitable
hydrocarbons as the balance of the fuel blend with ethanol. EPA must avoid prescribing onerous
and unprecedented requirements, such as a minimum and maximum vapor pressure, for flex
fuels. RFA supports the continued reliance on ASTM D5798-13 to describe ethanol fuel blends
5-160
-------
Tier 3 Summary and Analysis of Comments
containing 51-83% ethanol and to allow blending of flex fuels with ASTM D5798-13 allowed
hydrocarbons in all markets including RFG areas.
E16-E50 blends should be treated as alternative fuels.
RFA supports EPA's continued treatment of El 6-50 fuels as alternative fuels; this range of fuel
blends is only legally allowed for use in FFVs. E16-50 fuel blends are not gasoline. In fact, a fuel
blend being predominantly gasoline does not effectuate the use of that fuel blend in gasoline-
certified vehicles.
RFA agrees with EPA that data gaps exist on the effectiveness of detergents in E16-83 ethanol
fuel blends. Currently, there is insufficient data to render a decision of the effectiveness across
the broad range of FFVs in the marketplace. RFA continues to investigate the most efficient and
effective composition of these fuel blends and will continue to play a role in addressing data
gaps.
RFA supports the EPA requirement for "substantially similar" for all fuels: gasoline and
alternative, which restricts the composition to only carbon, hydrogen, oxygen, nitrogen and
sulfur.
Commenter: Shell Oil Products for Shell and Motiva
In the preamble and in a separate document entered in to the docket (April 8, 2013 memorandum
from Jeff Herzog, "Possible Approach to Fuel Quality Standards for Fuel Used in Flexible-Fuel
Automotive Spark-Ignition Vehicles (FFVs)") EPA discusses potentially regulating - 83 and
El5-50. In general, we agree that these fuels should be regulated similar to gasoline. As it is,
these fuels are essentially exempt from the sulfur, benzene, and detergent additive requirements
that apply to gasoline.
The following identifies numerous specific issues in the provisions included in the Herzog
memo, and makes recommendations to address those issues.
Of the issues discussed below, one is particularly noteworthy. The suggested benzene level for
E51-85 of 0.2 vol% is too low and is not consistent with the existing benzene specifications for
gasoline. If EPA proceeds with this, it will essentially require specialized blendstocks with lower
benzene levels to produce E51-85.
80.2 Definitions: New paragraph 80.2(cccc), which will provide the definition for E51-83,
includes the reference to the ASTM D5798-11 standard for ethanol fuel blends for FFVs. This
reference should be updated to reflect the currently published version ASTM 05798-13 and
should be updated further if a newer version is adopted by ASTM.
80.1601 Fuels subject to the provisions of this Subpart: EPA proposes adding item 80.1601(a)(4)
to explicitly include E51-83 as being subject to the provisions of Subpart O regarding gasoline
sulfur. However, we do not see an explicit addition to also include El6-50. Assuming EPA
5-161
-------
Chapter 5: Tier 3 Fuel Sulfur Program
believes Subpart O regarding gasoline sulfur should also apply to E16-50, then E16-50 should
explicitly also be listed in 80. 1601 (a).
- Under, 80.1601(b)(2), "California gasoline as defined in §80.1600 subject to the provisions of
§80. 1654" is not subject to the standards and requirements of Subpart O. Do the requirements of
Subpart O apply to E5 1-83 and/or El 6-50 sold in California? And if so, how do they apply?
80. 1608 Standards and requirements for E5 1-83 : Apart from the compliance dates for the new
sulfur levels via the requirements in 80.1603(a)(2), paragraph 80.1608 does not specify an
effective date for the standards and requirements for E5 1-83. EPA needs to insert a clear
effective date for all of the standards and requirements for E5 1-83 in 80. 1608. We infer this is
sometime in 2017, but ask that EPA specify the effective date in the regulation.
- 80.1608(b)(2) as proposed sets the maximum benzene content of each batch of ESl-'SS at 0.20
volume percent maximum. This benzene limit is too low, because it is inconsistent with EPA's
clear intent to allow E51-83 blenders to manufacture E51-83 using gasoline, gasoline
blendstocks for oxygenate blending, denatured fuel ethanol, natural gas liquids, and/or butane
per 80.1611.
- As an example, per 80.1611, E51-83 can be made from compliant RFG and compliant
denatured fuel ethanol. In this case with:
- the benzene content of E10 RFG being up to 1.30 vol% benzene and
- the amount of E10 RFG in the blend potentially being as much as -53 vol% of the total fuel
recipe (resulting in a final blend containing 51% pure ethanol),
- the final E51-83 made using RFG may contain up to -0.69 vol% benzene. However, this would
conflict with EPA's proposed 0.20 vol% benzene limit for E51-83. The benzene content of the
E51-83 may be even higher when conventional gasoline is used in the E51-83 recipe instead of
RFG (or RBOB). EPA needs to eliminate the explicit benzene limit in E51-83 when a blender
manufactures this fuel under the provisions of 80. 161 1 . Otherwise, EPA needs to increase the
explicit benzene limit to reflect EPA's intent to allow E85 production using compliant RFG,
RBOB, CG, and/or CBOB.
- Section 80.1608(b)(4) listing the RVP requirements for E51-83 contains the typographical error
"80.1627(a)(2)" when "80.27(a)(2) is intended by EPA. The EPA intended language should be:
"The Reid vapor pressure of each batch of E5 1-83 must comply with the same standards
applicable for gasoline in §80.-l-627(a)(2).
- EPA should clarify that section 80.1608(b)(4), which would require that "each batch" of E51-
83 must comply with same standards applicable for gasoline in 80.27(a)(2), only applies to E51-
83 blenders (and refiners, importers, distributors, resellers, and carriers). If EPA also intends for
the RVP requirements to be applicable during the "high ozone season" from June 1st through
September 15th of each year at retail sites as well, then EPA has to recognize and address the
5-162
-------
Tier 3 Summary and Analysis of Comments
potentially far slower turnover of E51-83 inventory at retail sites (relative to gasoline) may make
an RVP requirement at retail sites infeasible and dissuade retailers from offering E51-83. EPA
should note that ASTM recognized and addressed this issue of potentially slow moving E51-83
at retail sites by exempting the E51-83 at a retail site from ASTM's seasonal RVP requirements
in ASTM's "E51-83" standard D5798 Table 2 with notel as follows (emphasis added), "This
schedule, subject to agreement between the purchaser and the seller, denotes the vapor pressure
class of the fuel at the time and place of bulk delivery to fuel-dispensing facilities for the end
user. Shipments should anticipate this schedule."
- 80.1608(c)(3) provides the benzene standard for E51-83 when it is manufactured by a party that
does not qualify as an E51-83 blender:
- Similar to our comments regarding the 0.20 vol% benzene limit for E51-83 blenders in
80.1608(b)(2) being too low. 0.20 vol% benzene listed in 80.1608(c)(3) for E51-83 "refiners" is
also too low and should be revised consistent with our recommendations for E51-83 blenders.
- We believe "80.1632" is a typographical error with regard to the test requirements for benzene
in E51-83 produced by a "refiner". We believe EPA means and should correct this to "80.1630"
which deals with test requirements. At this time no section 80.1632 appears proposed or intended
by EPA.
80.1611 Blended E51-83 requirements:
- EPA needs to insert a clear effective date for the blended E51-83 requirements in 80.1611. We
infer this is sometime in 2017, but request that EPA specify the effective date.
- As proposed by EPA, 80.161 l(a) and (b) require that only RFG or RBOB be used as the
gasoline or gasoline blendstock for oxygenate blending in E51-83 intended for sale in RFG
areas. EPA has not provided a basis for this limitation based on air quality impact. Absent EPA
providing a reasonable basis, these requirement should be deleted from EPA's proposed rules,
allowing conventional gasoline and/or CBOB to be used as well for E51-83 in RFG areas.
80.1612 Standards and requirements for natural gas liquids blendstock used by E5 l^SS blenders:
- 80.1612(b) as proposed includes a prohibition on using E51-83 made from NGL in a blend with
gasoline to make E16-50. We do not understand EPA's logic in this prohibition and ask that EPA
explain its reasoning further.
80.1662 Liability for violations under Subpart 0:
- For a sulfur cap standard violation in 80.1660(b), paragraph 80.1662(a)(5) includes a branded
refiner/importer liability as follows (emphasis added):
" Branded refiner/importer liability. Any refiner or importer whose corporate, trade, or brand
name, or whose marketing subsidiary's corporate, trade, or brand name appeared at a facility
where a violation of §80.1660(b) occurred, is deemed in violation of §80.1660(b)."
However, this branded refiner/importer liability is too broad when EPA is now proposing to
further expanded the sulfur cap requirements in 80.1660(b) to include E16-50 and E51-83 fuels
sold at retail sites where the branded refinery has no control on the retailer who might
independently make these product offerings from sources of his/her own choosing. Branded
refiners should not be liable in these cases simply because their brand name "appeared at a
5-163
-------
Chapter 5: Tier 3 Fuel Sulfur Program
facility", especially in the case where the branded refiner contractually required the retailer to
identify at the dispenser that the E16-50 or E51-83 in question was not a branded product.
The Herzog memo discusses detergent additives for E51-85. EPA is seeking comment on the
amount (if any) of detergent additive needed.
The EPA proposed regulation for E85 (51% to 83% pure ethanol) and Mid-Level Ethanol blends
(16% to 50% pure ethanol), for use only in Flex Fuel Vehicles states:
"Stakeholders have stated that as additive concentration diminishes due to dilution with DFE,
there is a point where the presence of a deposit control additive ceases to be beneficial and can
actually contribute to deposit formation. In light of this, it may be appropriate to remove the
requirement that the gasoline portion of E51-83 and perhaps E16-50 must contain a deposit
control additive until the specific deposit control needs of these blends can be evaluated".
There is evidence that unadditized E85 generates deposits on inlet valves. There is also evidence
that the use of additives in E85 helps to remove inlet valve deposits. Therefore, we recommend
that EPA maintain the approach taken in the current deposit control regulations which require
that the gasoline portion of E51-83 and E16-50 must contain a certified deposit control additive
at a concentration at least as great as that used during gasoline deposit control additive
certification testing (referred to as the Lowest Additive Concentration or LAC). If EPA should
decide to pursue changes to the current approach, we recommend that EPA initiate a stakeholder
process to vet the issues.
Commenter: Sunoco Logistics Partners L.P.
As the owner/operator of approximately 40 refined product terminals, nearly all of which,
inventory and blend ethanol, Sunoco Logistics supports expanding the use of blend stocks like
butane and natural gasoline to manufacture E51-83, provided the finished E51-83 blend meets
existing industry standard ASTM 5798. At the lower fuel ethanol ranges (i.e., E51), the use of
butane and natural gasoline as blend stocks would likely require the addition of other lower
volatility components to meet all volatility standards in ASTM 5798.
To meet the envisioned flex fuel quality standards and comply with the applicable sections of the
CAA, we support providing maximum flexibility by allowing E51-83 producers to choose
between option 1 (refiner) or option 2 (blender) as outlined in the 4/8/13 memorandum authored
by Jeff Herzog. The option 2 approach requiring blend stock specifications, quality control
testing and reporting should parallel the requirements under 40CFR80 for blending butane into
conventional gasoline.
Our Response:
As explained in Section V.H. of the preamble to today's final rule, we are deferring final
action not on these provisions at this time. We appreciate the valuable input provided by the
commenters and will take this into consideration in our final actions on these standards. We will
continue to work with stakeholders in developing in-use fuel quality standards for higher level
5-164
-------
Tier 3 Summary and Analysis of Comments
ethanol blends. We may issue a supplementary proposal prior to issuing a final rule if the
additional information we receive from stakeholders warrants such an action.
5.5. Flexibilities
5.5.1 General
What Commenters Said:
Commenter: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
The main concern with the protracted compliance flexibility in the proposed rule is that it
enables on-going unpredictable and unidentified "hot spot" issues where chronically high sulfur
fuels may be sold, forcing OEMs to manufacture vehicles that can perform at higher sulfur level
to account for such variability. The flexibility also generally precludes uniformity at the retail
pump and therefore makes it more difficult to achieve emissions reductions across the country.
The Alliance and Global Automaker support a reasonable date for 100% compliance with sulfur
standards, and urges EPA to further limit the period for early crediting and trading that will delay
full compliance with the rule.
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM)
The so-called flexibilities proposed in the rule do little to offset the burden of compliance.
Commenter: Manufacturers of Emission Controls Association (MECA)
MECA also appreciates the importance of providing some compliance flexibilities to the oil and
refining industries (as EPA has done in past fuel sulfur regulations) to facilitate cost-effective
compliance with the standards.
Commenter: Michigan Department of Environmental Quality (MDEQ)
The USEPA is proposing to use phase-in flexibilities, credit and allowance programs, more lead
time for small business and manufacturers, and hardship provisions as has been historically done.
The MDEQ, Air Quality Division supports this flexibility. A reduced number of United States
refineries in recent years has resulted in temporary fuel price spikes when events take them 'off-
line.' However, pollutants emitted due to high sulfur content in fuel have a cost to the
environment and the public's health that can be much greater and longer duration. With the
multitude of complex issues regarding cost to citizens, to refineries, to the environment and to
health, these allowances for special cases seem prudent.
Commenter: Ozone Transport Commission (OTC)
5-165
-------
Chapter 5: Tier 3 Fuel Sulfur Program
We anticipate that EPA will ensure flexibility for automobile manufacturers and refiners of
gasoline through averaging, banking, trading programs, or other mechanisms as appropriate
without diminishing the environmental benefits of the rule.
Additionally, refineries should be extended flexibility to avoid 2 turnarounds (shutdowns and
startups for major maintenance or equipment installation) in a relatively short time period. To the
extent reasonable, the equipment installation for low sulfur fuel equipment should be
accommodated during a normally scheduled maintenance turnaround. This avoids the excess
emissions that occur during multiple shutdowns and startups of a refinery.
Commenter: Pennsylvania Department of Environmental Protection (DEP)
DEP supports flexibility for.. .refiners of gasoline through averaging, banking, trading programs,
or other mechanisms and efforts to mitigate those costs by providing time to install needed
equipment at refineries.
Commenter: State of Utah
Utah supports including mitigation provisions in the final rule to help refineries as they transition
to Tier 3 gasoline, including providing an averaging, banking, and trading (ABT) program to
allow refineries to spread out the cost of these investments among their facilities and hardship
and flexibility provisions (e.g., delayed implementation) for small volume refineries such as
those in Utah.
Commenter: Union of Concerned Scientists (UCS)
We also support the flexibilities in the Tier 3 standard that were also used in the Tier 2 program.
Commenter: Utah Air Quality Board
The Board further recognizes that the low-sulfur gasoline component of the proposed Tier 3
Program will require significant investments by Utah's petroleum refineries. The Board supports
EPA's efforts to mitigate the economic impacts of the Tier 3 Gasoline on this industry,
including, but not limited to, providing an averaging, banking, and trading (ABT) program to
spread out the cost of these investments and proposing hardship provisions and flexibility
provisions (e.g., delayed implementation) for small volume refineries such as those found in
Utah.
Our Response:
As discussed more below in Chapters 5.5.2 through 5.5.4 (and in Section V of the
preamble to the final rule), we are in fact finalizing a number of flexibilities in the Tier 3
gasoline sulfur program including: a nationwide ABT program, with the ability to use Tier 2
"banked" credits; a three-year delay for approved small refiners and small refineries; and
hardship provisions. We believe that these flexibilities address concerns of the potential for
5-166
-------
Tier 3 Summary and Analysis of Comments
having off-cycle turnarounds (which could in turn lead to supply and/or pricing issues) and will
help to reduce the burden of compliance on regulated entities. In response to comments, we have
also made changes to the final ABT program that we believe will allow for a smooth transition
between the Tier 2 and Tier 3 ABT programs, and provide more certainty for refiners of the
availability of credits—thus allowing refiners to spread out the cost of their investments among
their facilities and in a manner that best works for them.
With regard to the concern that flexibilities preclude uniformity of the gasoline sulfur
standards, we note that the January 1, 2020 compliance date for small refiners and small
refineries is also the date that all refiners and importers can no longer use credits that were
generated with respect to the 30 ppm sulfur standard. Therefore, all production and credit
generation/use will be relative to the 10 ppm average sulfur standard at this point. As discussed
above in Chapter 5.1.2, we do not anticipate any such "hot spot issues" that the commenter
mentioned.
5.5.2. Averaging, Banking, and Trading (ABT) Program
5.5.2.1 General
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM)
API and AFPM support the inclusion of an ABT program as part of the Tier 3 regulations that
allows refiners (including blender refiners) and importers to generate early and standard credits.
We generally agree with the structure of the standard credit generation program, allowing credits
to be generated for gasoline that over-complies with the 10 ppm annual average standard. Some
changes are needed to make the system more effective, including changes to early credit life,
treatment of remaining Tier 2 credits, trading restrictions, and the deficit carryover provisions.
Commenter: Chevron Products Company
If EPA proceeds with the gasoline sulfur reduction, we do support several of the proposed
elements including: averaging, banking and trading program.
EPA's refinery modeling assumes very optimistic refinery conversion timing and also assumes
that refineries would target a 10 ppm level when in fact refineries would have to target a lower
level (e.g., 8 ppm or lower) to ensure that the annual average met the 10 ppm standard. Also, the
proposed Averaging, Banking and Trading (ABT) program is required to be widely used by the
participants in order that the costs be kept down and to meet the proposed implementation
timeline — this represents the ideal situation. There is uncertainty about the number of credits
that will be available in the market to allow parties to meet the 10 ppm annual average standard.
We believe that use of the ABT should be a 'relief valve' for compliance and not a critical
component.
5-167
-------
Chapter 5: Tier 3 Fuel Sulfur Program
While the ABT program does provide flexibility for refiners, the short three-year implementation
time period removes a great portion of the benefit of this flexibility by requiring refiners to use
the ABT program to manage their investment program and stay in compliance. Refiners who
choose to invest for compliance on time for a 2017 start may be unable to do so due to the time
required for capital investment, permitting, and construction. Reliance on the ABT program to
obtain credits for compliance should be an option for refiners, not a requirement. A refiner
should have a clear path to meet the new sulfur standards by investing early and being able to
meet the start date without having to use the ABT program.
Commenter: Countrymark
CountryMark believes that the ABT program will provide an easier way for the refining industry
to transition from Tier 2 fuels requirements to Tier 3 fuels requirements. This program provides
opportunities for obligated parties to meet compliance through a sulfur credit program.
Commenter: Small Business Refiners (SBR)
Sulfur Average, Banking, and Trading (ABT) Program: The Small Business Refiners believe that
the ABT program will provide way for the refining industry to transition from Tier 2 fuels
requirements to Tier 3 fuels requirements. This program provides opportunities for obligated
parties to meet compliance through a sulfur credit program.
Commenter: Environmental Defense Fund (EDF)
The averaging, banking, and trading (ABT) program will help achieve these low costs by
allowing refiners to optimize their investments. And that low cost estimate assumes that less than
a quarter of the 111 existing refiners will participate in the ABT program.
Commenter: ExxonMobil
If the Agency decides to proceed with this rulemaking, we recommend that EPA retains the
flexibility of the Averaging, Trading, and Banking program with the additional modifications
submitted by API and AFPM;
Commenter: Marathon Petroleum Company LP (MFC)
In general, MFC supports the sulfur Averaging, Banking, and Trading program concept but has
some specific recommendations to improve the proposal.
MFC supports the inclusion of an ABT program as part of the Tier 3 regulations.
Commenter: Natural Resources Defense Council (NRDC)
EPA's proposal provides generous flexibilities for the meeting the sulfur requirements. The
proposed averaging, banking and trading (ABT) program minimizes cost of compliance by
5-168
-------
Tier 3 Summary and Analysis of Comments
allowing for investments in the most efficient locations and through early credit generation
opportunities that allow refiners to spread out investments.
Commenter: Phillips 66 Company
Credit, Banking and Trading - We support an early credit generation provision.
Our Response:
Consistent with our proposal, we are finalizing a nationwide ABT program as part of the
final Tier 3 program. We believe that the ABT program provides flexibility will help enable the
January 1, 2017 start date for the gasoline sulfur standards. Details of the ABT program can be
found in Section V.D of the preamble to the final rule, and specific elements of the ABT
program that we received comments on are also discussed below in Chapters 5.5.2.2 through
5.5.2.5.
Regarding the commenter's concern that our modeling is optimistic, based on our
analysis of the Tier 2 credit trading market (see RIA Chapter 4.3.2) and current refinery sulfur
levels (RIA Chapter 4.4), we believe that there will be sufficient credits in the marketplace and
that there will be both intra- and inter-company trading. As discussed in more detail below, the
Tier 3 ABT program will flow seamlessly from the Tier 2 ABT program. Tier 2 "banked"
credits will be able to receive their full credit life, and credits generated prior to January 1, 2017
will be relative to the current (Tier 2) 30 ppm sulfur standard. Further, these credits can be used
for compliance with either the Tier 2 or Tier 3 programs. Current refinery sulfur levels indicate
that there are refineries near, and even below, 10 ppm; thus, we anticipate that these refineries
will all be able to generate credits for overcompliance with the 10 ppm sulfur standard beginning
January 1, 2017. Considering all of these factors, we believe that there will be sufficient credits
available for those needing (or choosing) to defer investment to better align with their existing
turnaround schedules or financial needs.
We received comments both expressing concerns that the ABT program will hinder the
efforts of the Tier 3 program, and expressing concern that the Agency should not rely on the
ABT program - it should be an option available to refiners, not a requirement. We note that the
success of the Tier 2 gasoline sulfur program occurred with a nationwide ABT program. The
Tier 3 program will simply continue this ABT program, just relative to a 10 ppm standard
instead of 30 ppm. We considered, but quickly rejected, other compliance options for the Tier 3
program that included for example, a fixed date for all refineries with no ABT or other
flexibilities. Without this flexible ABT program, a January 1, 2017 start date of the program
may not have been feasible and would have caused impacts to the fuel refining industry that
could have negative supply or pricing impacts. However, this would have gone counter to our
successful approaches in implementing prior fuel standards, and counter to refiner support for an
ABT program. We could also have designed it with a declining sulfur standard each year
beginning in 2014. However, that would have required a rigid phase-down of the sulfur levels
between now and 2020. In contrast, the program as finalized allows the industry to determine
the most cost-effective way to phase-down sulfur levels between now and 2020. Finally, we
believe that it is entirely appropriate to consider the flexibilities when considering the start date
5-169
-------
Chapter 5: Tier 3 Fuel Sulfur Program
for the fuel standards. The Agency has always taken into consideration the flexibility afforded
by the ABT programs in setting the start date for its standards, and in fact has argued that it is
legally required to in the context of some of vehicle and engine programs where we are obligated
to set the most stringent standards feasible considering cost and other factors. Incorporating
ABT provisions allows the Agency to set standards that are less costly and can achieve benefits
sooner than might otherwise be possible.
5.5.2.2 Who Can Generate Credits
What Commenters Said:
Commenter: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
EPA has also requested comment on its proposal that sulfur credits only be generated for
gasoline that is subject to the proposed sulfur requirements as described at §80.1603. This
approach would exclude gasoline designated for export and California, but would include
gasoline produced by California refineries for use outside the state. California gasoline is not
covered by EPA's proposed rule. The Agency has requested comment on whether to include
California gasoline in the averaging, banking and trading program under Tier 3 (not the early
credit program) (57).
We agree with EPA that sulfur credits may only be generated for gasoline that is subject to the
proposed sulfur requirements as described at §80.1603. However, California gasoline should not
be included in the ongoing ABT program under Tier 3. California gasoline has a 20 ppm per
gallon cap at retail and market fuels have averaged 10 or less for several years (58). Allowing
federal credits for improvements done long ago to meet CARB requirements serves only to dilute
the federal program. In addition, refiners should only be granted credits if EPA also reduces the
refinery gate and retail per gallon caps well below Tier 2 levels.
57 - 78 Fed. Reg. 29908 at 29873 (May 21, 2013). To be consistent, if EPA were to grant the oil industry
sulfur credits for California gasoline sold, it should also grant OEMS federal NMOG credit for CA
vehicles sold in CA and 177 states.
58 - 78 Fed. Reg. 29908 at 29820 (May 21, 2013).
Commenter: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
Furthermore, we also agree with EPA that over-compliance with the DFE per-gallon cap is not a
valid basis for credit generation (59). EPA should only allow credit generation were it to impose
an additional annual average sulfur standard for DFE (denatured fuel alcohol) at some level
below 10 ppm, and allow credits to be generated for over-compliance with that standard.
59 - 78 Fed. Reg. 29908 at 29820 (May 21, 2013).
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
5-170
-------
Tier 3 Summary and Analysis of Comments
Manufacturers (AFPM)
We do not agree with EPA's statement in the preamble that credit generation should be limited to
refiners who produce gasoline from crude oil. Instead, any refiner that meets the definition of
80.2(i), which includes parties who combine blendstocks to produce gasoline, should be eligible
to generate both early and standard Tier 3 credits. We note that the regulatory text at
80.1615(a)(l) does not appear to restrict credit generation to refiners who produce gasoline from
crude oil. Therefore, we believe the preamble should be changed to remove the reference to
producing gasoline by processing crude oil. Refiners who produce gasoline by combining
components are allowed to generate credits under Tier 2. The Tier 3 program should not remove
this provision since component blenders should be incentivized to produce gasoline that over-
complies with the Tier 3 standard.
We do agree that U.S. importers of gasoline should be eligible to generate both early and
standard credits. Other parties who are not refiners or importers of gasoline should not be eligible
to generate credits. We believe that including other parties, like ethanol producers and oxygenate
blenders, will add needless complexity to the program without generating any benefits. All of the
gasoline entering the U.S. market can be covered under the proposed program structure.
Similarly, we support the prohibition against ownership of credits by any party who is not a
registered refiner or importer.
We agree with the proposal that gasoline designated for export and California gasoline should
both be excluded from the Tier 3 program and should not be allowed to generate credits. While
some California gasoline could generate Tier 3 credits and provide some compliance flexibility,
imposing a 10 ppm sulfur average on California gasoline adds an additional blending constraint
to supplying the challenging California market. We do not believe that the additional Tier 3
flexibility justifies this additional burden for the producers of California gasoline. Gasoline
refined in California and shipped to another state under Tier 3 should be subject to the same
regulatory framework that currently exists today.
Commenter: Chevron Products Company
We agree that both U.S. refiners and U.S. importers of gasoline should be eligible to generate
both early and standard credits. Other parties who are not refiners or importers of gasoline should
not be eligible to generate credits. We believe that including other parties, like ethanol producers,
will add needless complexity to the program without generating any benefits. All of the gasoline
entering the U.S. market can be covered under the proposed program structure. Similarly, we
support the prohibition against ownership of credits by any party who is not a registered refiner
or importer.
It is important to maintain the current rules for participation in the program. Entities who are not
producers or importers of gasoline and who are not regulated under Tier 3 should not be allowed
to participate in the ABT program.
We agree with the proposal that gasoline designated for export and California gasoline should
both be excluded from the Tier 3 program and should not be allowed to generate credits. While
5-171
-------
Chapter 5: Tier 3 Fuel Sulfur Program
some California gasoline could generate Tier 3 credits and provide some compliance flexibility,
imposing a 10 ppm sulfur average on California gasoline adds an additional blending constraint
to supplying the already challenging California market. We do not believe that the additional
Tier 3 flexibility justifies this additional burden for producers of California gasoline. Gasoline
produced in California for sale outside the state should continue to be included in the program.
Commenter: Independent Fuel Terminal Operators Association (IFTOA)
In the Preamble that accompanied the proposed regulations, EPA explained that it is the
Agency's plan to limit both credit generation and ability to hold title to generated credits to
companies required to comply with the proposed Tier 3 sulfur standards - refiners and importers.
These limitations are designed to prevent abuse and fraud. The Association agrees with EPA and
believes that such an approach would promote the integrity of (i) credit generation and use, and
(ii) the overall program.
EPA is proposing to encourage early gasoline desulfurization and to give members of the
refining industry flexibility to stagger their investments. Therefore, under the proposed rule, EPA
would allow refiners and importers to generate early credits for "over-complying" with the
current 30 ppm gasoline sulfur standard on a volume-weighed annual average basis from January
1, 2014 through December 31, 2016. These early credits would have a limited credit life; they
could be used to demonstrate compliance for three years - 2017, 2018, and 2019. EPA has asked
for comments on whether importers should be eligible to generate early credits because they do
not make investments in refinery modifications to meet the proposed Tier 3 standard.
Members of the Association strongly support EPA's proposal to allow importers, as well as
refiners, to generate early credits for several reasons:
First, eligibility for such credits would encourage importers to maximize their efforts to bring
lower sulfur gasoline into the United States. These efforts would result in greater volumes of
lower sulfur gasoline on the market sooner and help EPA meet its objective of better air quality
at the earliest possible date;
Second, the greater the number of early credits available, the greater the flexibility U.S. refiners
would have in making the needed investments in their facilities; and
Third, EPA is correct that importers do not, themselves, make financial investments in
desulfurization equipment. However, when importers purchase lower sulfur gasoline abroad,
they pay a premium for this fuel, reflecting the financial investment made by the foreign refiner.
Thus, importers are essentially making an investment, similar to domestic refiners, when they
import lower sulfur fuel into U.S. commerce.
Commenter: Irving Oil Terminals Inc.
Irving Oil strongly supports the generation of early credits by both refiners and importers.
Commenter: Magellan Midstream Partners, L.P.
5-172
-------
Tier 3 Summary and Analysis of Comments
Early Sulfur Credits — EPA has proposed program flexibility through an average, banking and
trading program.
Magellan Comment - EPA has limited its proposal to allow only crude oil refiners and importers
to generate early credits. We believe any refiner that meets the definition of 80.2(i), which
includes parties who combine blendstocks to produce gasoline, should be eligible to generate
both early and standard credits under the proposed rule. For example, importers are not required
to produce gasoline from crude oil refining and could be producing it by blending, and yet they
are allowed to generate early credits. In addition, the primary reason for allowing importers to
generate early credits because there is no baseline required. This is also applicable to a domestic
gasoline blender.
The preamble also mentions the inclusion of importers in the generation of early credits allows
domestic refiners more time to make capital investments to reduce sulfur in gasoline. Allowing
domestic blenders to generate early credits would have the same impact.
Commenter: Marathon Petroleum Company LP (MFC)
We do not agree with EPA's statement in the preamble that credit generation should be limited to
refiners who produce gasoline from crude oil. Instead, any refiner that meets the definition of
80.2(i), which includes parties who combine blendstocks to produce gasoline, should be eligible
to generate both early and standard Tier 3 credits. We note that the regulatory text at
80.1615(a)(l) does not appear to restrict credit generation to refiners who produce gasoline from
crude oil. Therefore, we believe the preamble should be changed to remove the reference to
producing gasoline by processing crude oil. Refiners who produce gasoline by combining
components are allowed to generate credits under Tier 2. The Tier 3 program should not remove
this provision since component blenders should be incentivized to produce gasoline that over-
complies with the Tier 3 standard.
We do agree that U.S. importers of gasoline should be eligible to generate both early and
standard credits. Other parties who are not refiners or importers of gasoline should not be eligible
to generate credits. We believe that including other parties, like ethanol producers and oxygenate
blenders, will add needless complexity to the program without generating any benefits. All of the
gasoline entering the U.S. market can be covered under the proposed program structure.
Similarly, we support the prohibition against ownership of credits by any party who is not a
registered refiner or importer.
We agree with the proposal that gasoline designated for export and California gasoline should
both be excluded from the Tier 3 program and should not be allowed to generate credits. While
some California gasoline could generate Tier 3 credits and provide some compliance flexibility,
imposing a 10 ppm sulfur average on California gasoline adds an additional blending constraint
to supplying the challenging California market. We do not believe that the additional Tier 3
flexibility justifies this additional burden for the producers of California gasoline. Gasoline
refined in California and shipped to another state under Tier 3 should be subject to the same
regulatory framework that currently exists today.
5-173
-------
Chapter 5: Tier 3 Fuel Sulfur Program
Commenter: Northeast States for Coordinated Air Use Management (NESCAUM)
It is not appropriate to allow ethanol producers or blenders to generate sulfur credits under the
proposed gasoline sulfur program. This would require expansion of the program's sampling,
testing, recordkeeping, and reporting requirements and require ethanol producers and blenders to
be treated as refiners. This expansion of the program to include non-obligated parties would
unnecessarily complicate the program with no corresponding benefit.
Commenter: Phillips 66 Company
We support refineries and importers being able to generate credits. This would include entities
that meet the definition of a refinery through creation of finished gasoline via blendstock
blending. These "refineries" have to meet the standards, sampling and testing, and reporting
requirements. In order to generate credits, they would need to secure very low sulfur blendstocks
and should be allowed to generate credits. We also agree that other entities (e.g., oxygenate
blenders) should not be allowed to generate credits.
The proposed rule continues to exempt California gasoline from the federal gasoline sulfur
regulation, however, EPA asked for comment on whether to include California gasoline in the
ongoing ABT program under Tier 3. Phillips 66 agrees that the California gasoline volumes
(CARB gasoline) should not be included under the Tier 3 provisions but we provide an
alternative option for consideration. We ask EPA to consider an opt-in provision whereby a
California refinery could chose to have its CARB gasoline volumes subject to the Tier 3
standards and requirements. If a refinery opted in, all of their CARB gasoline would be subject to
the 10 ppm average sulfur standard and they would be able to generate credits against the 10
ppm standard.
Commenter: Renewable Fuels Association (RFA)
As detailed above, RFA believes DFE producers should be exempt from onerous sulfur content
batch reporting requirements. However, if EPA's final rule requires ethanol producers to
report sulfur content by batch (as proposed), they should be allowed to participate in sulfur credit
averaging, banking, and trading (ABT) program. Ethanol has value for its sulfur reduction
ability. As larger volumes of ethanol enter the marketplace, EPA should allow ethanol
manufacturers the option to capture the value to the marketplace through the dilution of sulfur
from ethanol that contains below lOppm sulfur. The option to bank and trade sulfur credits
would come with the regulatory reporting requirements of the Sulfur banking and trading
program imposed on oil refiners.
Commenter: Shell Oil Products for Shell and Motiva
EPA should allow all "refiners," Including Gasoline Blenders, To Participate in the Credit
Generation Program.
5-174
-------
Tier 3 Summary and Analysis of Comments
We generally support EPA's proposed averaging, banking and trading provisions. These
provisions provide important flexibility that should help the industry reduce compliance costs.
We do not agree, however, with one of the restrictions that EPA proposes - only refineries that
process crude oil are eligible to generate credits. That is an unnecessary limitation that unfairly
excludes gasoline blenders. If a gasoline blender - classified as a refiner under EPA's rules - is
able to secure low sulfur blend stocks and supply lower sulfur gasoline to the market, EPA
should encourage it. By not allowing gasoline blenders to generate credits, EPA is unnecessarily
foregoing the emission reductions that could result from such parties supplying lower sulfur
gasoline and it deprives the entire industry of credits that could help the entire industry meet
EPA's aggressive compliance schedule.
Commenter: Sutherland Asbill & Brennan LLP (Sutherland)
EPA proposes that refiners and importers be able to generate credits through early compliance
and over-compliance. Only 'Obligated Parties' (refiners, blenders and importers) may generate,
use, transfer or own early or standard credits. Though we acknowledge that the same entities are
implicated under the RFS, we encourage EPA to use a term other than 'Obligated Parties' to
describe the parties capable of generating credits to avoid confusion. The use of such term is
unnecessarily confusing as Obligated Parties cannot generate credits under the RFS, and have
entirely different obligations under the RFS than they would under the Proposed Rule.
Commenter: Weaver and Tidwell LLP
Under this section [80.1615(a)], only refiners and importers may earn credits under Subpart O.
Weaver believes that butane blenders and transmix processors should be included in this list.
Commenter: Countrymark
the rule. This is important to the CountryMark because we own and operate the refining and
terminal assets.
Commenter: Small Business Refiners (SBR)
Credit generation should be permitted only for the obligated party as defined in the rule. This is
important to the SBRs because we own and operate the refining and terminal assets.
Commenter: American Coalition for Ethanol (ACE)
On a related note, ACE also believes that ethanol producers should be eligible to generate credits
under the gasoline sulfur program. EPA states that 'since many refiners currently comply with
our standards taking into consideration the fuel property changes expected as a result of
downstream ethanol blending, providing ethanol blenders with sulfur credit would result in
double counting the effects of ethanol.' Actually, giving sulfur credit to ethanol producers would
apply credit to the fuel supplier that is actually reducing sulfur, as opposed to forcing ethanol
producers to give that benefit away to refiners, who will use it to delay reducing sulfur in
5-175
-------
Chapter 5: Tier 3 Fuel Sulfur Program
gasoline. Regardless of the mechanism used, ACE does not believe EPA should provide refiners
with pathways that allow delayed adoption of cleaner fuel standards, and we particularly object
to the suggestion that the ethanol industry should enable that delay at our own expense.
Commenter: Growth Energy
If DFE will have the same batch testing and registration requirements as gasoline, requiring alO
ppm sulfur cap versus an average could disadvantage DFE versus gasoline, which can generate
and profit from sulfur credits. DFE having the same sulfur limits as gasoline, along with a
compliant petroleum blendstock, will result in ethanol blends having a sulfur limit that is
identical to the proposed finished gasoline specifications.
Furthermore, if the burden of implementing batch testing and reporting is imposed on ethanol
producers, they should be allowed to generate sulfur credits. EPA states that "While certain
batches of ethanol could theoretically be low enough in sulfur to generate credits, it is our desire
to limit credit generation to companies required to comply with the proposed Tier 3 sulfur
standards, i.e., refiners and importers.'
If DFE producers are required to complete the batch testing, reporting, and record keeping on
DFE similar to refiner requirements on gasoline, they should have the same ability to generate or
use credits.
EPA mistakenly states that "since many refiners currently comply with our standards taking into
consideration the fuel property changes expected as a result of downstream ethanol blending,
providing ethanol blenders with sulfur credit would result in double counting the effects of
ethanol." However, double counting would not result if refiners are assuming that DFE has a
sulfur level of 10 ppm. Credits would only be generated at levels that are below this 10 ppm
level.
EPA attempts to further reject ethanol generating sulfur credits by stating that "Over compliance
with the per-gallon cap would not be a valid basis for credit generation, as you would expect that
in all cases the DFE would be below the cap. To allow credit generation, we would need to
propose an additional annual average sulfur standard for DFE at some level below 10 ppm, and
allow credits to be generated for over compliance with that standard." Ethanol producers should
have the same ability to generate credits if they have the same testing and reporting standards. In
order to remain price-competitive with gasoline, ethanol should be allowed to generate credits in
a similar manner as gasoline. Furthermore, such credit trading could allow ethanol producers to
offset the cost of [new] DFE testing and reporting requirements.
Commenter: POET, LLC
If DFE will have the same batch testing and registration requirements as gasoline, requiring a 10
ppm sulfur cap versus an average could disadvantage DFE versus gasoline which can generate
and profit from sulfur credits. DFE having the same sulfur limits as gasoline, along with a
compliant petroleum blendstock, will result in ethanol blends (whether MLEBs or HLEBs)
having a sulfur limit that is identical to the proposed finished gasoline specifications.
5-176
-------
Tier 3 Summary and Analysis of Comments
Furthermore, if the burden of implementing batch testing and reporting is imposed on DFE
producers, they should be allowed to generate sulfur credits. EPA states that 'While certain
batches of ethanol could theoretically be low enough in sulfur to generate credits, it is our desire
to limit credit generation to companies required to comply with the proposed Tier 3 sulfur
standards, i.e., refiners and importers.' (70) If DFE producers are required to complete the batch
testing, reporting, and record keeping on DFE similar to refiner requirements on gasoline, they
should have the same ability to generate or use credits.
EPA mistakenly states that 'since many refiners currently comply with our standards taking into
consideration the fuel property changes expected as a result of downstream ethanol blending,
providing ethanol blenders with sulfur credit would result in double counting the effects of
ethanol.' (71) However, double counting would not result if refiners are assuming that DFE has a
sulfur level of 10 ppm. Credits would only be generated at levels that are below this 10 ppm
level.
EPA attempts to further reject ethanol generating sulfur credits by stating that 'Over compliance
with the per-gallon cap would not be a valid basis for credit generation, as you would expect that
in all cases the DFE would be below the cap. To allow credit generation, we would need to
propose an additional annual average sulfur standard for DFE at some level below 10 ppm, and
allow credits to be generated for over compliance with that standard.' (72) Ethanol producers
should have the same ability as refiners to generate credits if they have the same testing and
reporting standards. In order to remain price-competitive with gasoline, ethanol should be
allowed to generate credits in a similar manner as gasoline. Furthermore, such credit trading
could allow ethanol producers to offset the cost of new DFE testing and reporting requirements.
70-Id. at 29,93 8.
71-Id.
72-Id. at 29,938-39.
Our Response:
Use of the Term "Obligated Parties"
With respect to comments requesting that EPA not use the term "obligated party" in the
Tier 3 regulations, we agree with these comments and are using the terms "refiner" and
"importer" in the final Tier 3 regulations. The Tier 2 rule was finalized in 2000, and used this
term as a general way to refer to regulated parties such as refiners and importers. However, this
term now has a very specific meaning under the RFS program. The use of "obligated party" in
the Tier 3 proposed regulations was to remain consistent with the Tier 2 regulations; however,
we agree that this could cause confusion with the RFS program.
Participation in the AST Program
Consistent with our proposal, gasoline sulfur credits may be generated by U.S. refiners,
and importers of gasoline into the U.S., only for gasoline that is subject to the sulfur
requirements as described in the regulations at §80.1603. This provision excludes gasoline
produced or imported for use in California ("California gasoline") and gasoline designated for
5-177
-------
Chapter 5: Tier 3 Fuel Sulfur Program
export, but includes gasoline produced by California refineries for use outside the state. Further,
importers will also be able to participate fully in ABT program, so they will be able to generate
credits for sending over-compliant gasoline to the U.S. This is unlike other previous and existing
fuel programs that precluded their participation in the early credit portion of an ABT program.
In response to comments, we have revised the regulations by removing the terms
"produce gasoline from crude" from §80.2(1). Under the final Tier 3 program, refiner-blenders
who comply with the full suite of sampling, testing, and reporting requirements can participate in
the Tier 3 ABT program. Additionally, we are extending these provisions to butane and pentane
blenders (per Section V.I of the preamble to the final rule) if they likewise comply with the full
suite of sampling, testing, and reporting requirements. However, any refiner-blender who uses
the reduced sampling, testing, and reporting provisions for blending butane or pentane into
gasoline will not be allowed to generate Tier 3 credits.
Consistent with our proposal, ethanol and oxygenate producers are not allowed to
participate in the Tier 3 ABT program. As discussed in Section V.G.4 of the preamble to the
final rule, and consistent with existing EPA fuel programs, we continue to believe that it is not
appropriate expand the ABT provisions to cover ethanol producers and oxygenate blenders for
several reasons. First, expanding the ABT program beyond refiners and importers could greatly
increase the number of parties participating thereby potentially complicating EPA compliance
assurance activities while having little overall impact on the sulfur credit pool. Second, the
current ABT program under the Tier 2 gasoline sulfur program, which is limited to gasoline
refiners and importers, has functioned effectively with few compliance irregularities. Third,
experience with the unleaded gasoline program suggests widespread abuse and fraud when
credits have been allowed to be generated or sold by parties other than refiners or importers
subject to the regulations. Fourth, it would require a considerably more complicated compliance
structure, including the application of all refiner responsibilities to ethanol producers and
blenders. Fifth, there is no need for denatured fuel ethanol (DFE) producers to generate credits
in order to recoup the value for any lower sulfur content of their product. The value of any lower
sulfur content will be reflected in the market price of DFE, similar to the octane value to refiners.
Sixth, the sulfur ABT provisions were included to ease the burden of compliance for refiners
who have to make capital changes to their facilities to meet the more stringent Tier 3 sulfur
standards. In addition to reducing the cost of the Tier 3 gasoline sulfur program, the ABT
provisions allow for an earlier effective date of the sulfur standards than would otherwise be
possible. Such considerations are not applicable to ethanol producers since capital expenditures
for desulfurization equipment or other equipment will not be needed at their facilities to comply
with the final Tier 3 sulfur standards.14 Finally, overcompliance with the per-gallon sulfur cap
for DFE is not a valid basis for credit generation. We expect that in all cases, the DFE sulfur
level will be below 10 ppm. To allow credit generation for these parties, we would need to set
an additional annual average sulfur standard for DFE at some level below 10 ppm and allow
credits to be generated for overcompliance with that standard. Accordingly, we do not believe it
is appropriate to allow ethanol producers or blenders to generate sulfur credits under the Tier 3
gasoline sulfur program, and as such, we did not finalize such a provision. The Tier 3 rule
prohibits any person downstream of the refinery or importer that produced or imported gasoline,
5-178
-------
Tier 3 Summary and Analysis of Comments
CBOB, or RBOB who adds oxygenate to such product from including the volume and sulfur
content of the oxygenate in any compliance calculations for credit generation.
5.5.2.3 Design of Credit Program
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM)
We generally agree with the structure of the standard credit generation program, allowing credits
to be generated for gasoline that over-complies with the 10 ppm annual average standard. Credits
should be generated annually on a calendar year compliance period.
We support the inclusion of a special period of credit generation from 2014 to 2016 to facilitate
the transition from Tier 2 to Tier 3. We agree that the program should be available to refiners and
importers, and agree with the structure of generating credits against the current 30 ppm standard
with no individual refinery baseline. However, the ability to generate credits during this period
does not guarantee that sufficient credits will be available to allow refiners to broadly defer
investments past the proposed January 1, 2017 implementation date. EPA has described a very
optimistic scenario for refineries to generate credits by operating existing pre-treaters and post-
treaters at sub-30 ppm gasoline levels in 2014-16. If refineries fail to meet these optimistic
predictions, credit availability might be limited and will hinder the industry transition to the 10
ppm standard in 2017.
During the 2014-2016 period, we do not agree with the proposed structure of the early credit
program and instead recommend a different program structure to transition from the current Tier
2 program to the start of the Tier 3 program. We believe that any refinery should be able to
generate credits against the 30 ppm standard during the 2014-2016 period, and that these credits
should have a 5 year life. Credit generators should not have to designate these credits as either
Tier 2 or Tier 3 credits. The party who holds a credit generated in 2014-2016 should be able to
use the credit for compliance with Tier 2 standards or bank the credit for future use with Tier 3.
By granting these credits a 5 year life and by allowing a party to bank these credits for future
Tier 3 use, the refiner or importer is incentivized to reduce the sulfur content of their gasoline
pool. This could maximize the removal of sulfur from the pool even in advance of the Tier 3
program. No transition from Tier 2 Standard Credits to Tier 3 Early Credits to Tier 3 Standard
Credits would be required. This single credit program would be easier for EPA to manage and
enforce compared to implementation of an early credit program.
If EPA finalizes the early credit program as proposed, we do not agree that refiners should have
to designate credits generated from 2014 to 2016 as either Tier 2 credits or Tier 3 early credits.
We believe that these credits, generated against the 30 ppm standard, should be allowed to be
used for either Tier 2 or Tier 3 compliance without prior designation by the generating party.
Generated credits could accrue undesignated in the party's account and should be available to be
freely used for Tier 2 compliance, Tier 3 early compliance, or traded to other parties who could
5-179
-------
Chapter 5: Tier 3 Fuel Sulfur Program
use them for either program. Refiners will not be able to exactly predict their future credit
requirements and should not be penalized by over or under-designating credits for either
program.
In addition, we believe that any party holding Tier 2 credits at the end of 2016 should be allowed
to convert these Tier 2 credits to Tier 3 early credits, which could be used through 2019. Tier 2
credits represent real sulfur reductions against the existing standard which are a benefit to the
environment. The owners of these credits should be able to preserve their value by applying them
in the early years of the Tier 3 program.
Should EPA decide to proceed with the proposed credit designation of Tier 2 or Tier 3 during the
early generation credit period, we recommend that such declaration is made by February 28th of
the following year, consistent with compliance reporting.
The benefits of averaging banking and trading of the Tier 2 rule are not comparable to Tier 2.
The opportunity to generate credits is limited by a refinery's existing equipment and
configuration. Tier 2 sulfur credits are still in circulation, and it would be beneficial, and we
request that EPA publish aggregate data on the remaining Tier 2 sulfur credit balances in order to
determine to what extent an ABT program would provide or could provide a benefit.
With such a short lead time, the opportunity to generate sufficient credits to cover both the Tier 2
requirements and bank credits for Tier 3 between now and January 2017 will be limited.
Commenter: Chevron Products Company
We agree with the format of the ABT program. We generally agree with the structure of the
standard credit generation program, allowing credits to be generated for gasoline that over-
complies with the 10 ppm annual average standard. Credits should be generated annually on a
calendar year compliance period.
We do agree with the inclusion of a credit averaging, banking, and trading (ABT) program,
including a three-year phase-in period from 2014 through 2016 with the ability to generate early
credits. However, we propose some changes to the details of the ABT program which are
described below.
We support the inclusion of a special period of credit generation from 2014 to 2016 to facilitate
the transition from Tier 2 to Tier 3. We agree that the program should be available to refiners and
importers, and agree with the structure of generating credits against the current 30 ppm standard
with no individual refinery baseline. However, the ability to generate credits during this period
does not guarantee that sufficient credits will be available to allow refiners to broadly defer
investments past the proposed January 1, 2017 implementation date. EPA has described a very
optimistic scenario for refineries to generate credits by operating existing pre-treaters and post-
treaters at sub-30 ppm gasoline levels in 2014-16. If refineries fail to meet these optimistic
predictions, credit availability might be limited and will hinder the industry transition to the 10
ppm standard in 2017.
5-180
-------
Tier 3 Summary and Analysis of Comments
We do not agree with the proposed structure of the early credit program during the 2014-2016
period and instead recommend a different program structure to transition from the current Tier 2
program to the start of the Tier 3 program. We believe that any refinery should be able to
generate credits against the 30 ppm standard during the 2014-2016 period, and that these credits
should have a 5 year life. Credit generators should not have to designate these credits as either
Tier 2 or Tier 3 credits. The party who holds a credit generated in 2014-2016 should be able to
use the credit for compliance with Tier 2 standards or bank the credit for future use with Tier 3.
By granting these credits a 5 year life and by allowing a party to bank these credits for future
Tier 3 use, the refiner or importer is incentivized to reduce the sulfur content of their gasoline
pool. This could maximize the removal of sulfur from the pool even in advance of the Tier 3
program. No transition from Tier 2 Standard Credits to Tier 3 Early Credits to Tier 3 Standard
Credits would be required. This single credit program would be easier for EPA to manage and
enforce compared to implementation of an early credit program.
If credits must be designated as Tier 2 or Tier 3, we believe that the program should
acknowledge the transition from Tier 2 to Tier 3 by allowing unused Tier 2 credits to be
converted to Tier 3 credits after 2016. Under the current proposal, all Tier 2 credits will become
stranded if they are not used by the end of the 2016 compliance year. The Tier 2 credits represent
real sulfur reductions in gasoline below the required standard which in turn generated
environmental benefits beyond the requirements of Tier 2. Refiners and importers should not
lose the benefit of these credits that were generated by their over-compliance with the Tier 2
sulfur standard.
Commenter: CHS, Inc.
Currently under the proposed Tier 3 regulation, sulfur credits generated under the Tier 2
regulation for the years 2012 and 2013 can only be used for compliance through the calendar
year 2016. CHS requests that the EPA allow these Tier 2 sulfur credits to be used to demonstrate
compliance for years 2017 and 2018. This will help ensure that refineries meet compliance and
ultimately protect the consumer from any supply disruptions.
Commenter: Irving Oil Terminals Inc.
In addition, EPA has recognized that early credits generated by both refiners and importers will
help industry comply with the new standard and ensure that lower sulfur gasoline is available as
quickly as possible. Early credits also help to advance the associated environmental and health
benefits to an earlier timeframe. With some companies already having the ability to meet the
proposed lower standard, the EPA should begin the early credit program at a minimum of five
years before the effective date.
Irving Oil does not think that the compliance timetable of three years is adequate to meet the Tier
3 sulfur content standard. The use of early credits would certainly provide some flexibility with
refinery project execution in order to meet the requirements of the proposed standard. In
addition, early credits would help bring greater volumes of lower sulfur gasoline to market
earlier and address concerns from the auto manufacturers. Irving Oil strongly supports the
5-181
-------
Chapter 5: Tier 3 Fuel Sulfur Program
generation of early credits by both refiners and importers. EPA should do everything it can to
encourage the production and importation of qualified gasoline eligible for these credits.
It is well known that some refiners and importers need only to make modest adjustments to their
facilities and/or locate foreign gasoline cargoes in order to incrementally lower their sulfur
content downward from 30 ppm. The EPA should initiate the early credit program as early as
possible with a retroactive effective date of January 1, 2013.
It should be noted that, even with this modification, it is not certain that enough early credits
would be generated to allow the regulated community to meet the Tier 3 standard by the current
proposed Effective Date of 2017. However, the more early credits are encouraged, the greater the
chance more will be available to provide the degree of flexibility EPA has anticipated, thereby
facilitating compliance with the 10 ppm annual average standard at the earliest possible date.
Commenter: Marathon Petroleum Company LP (MFC)
We generally agree with the structure of the standard credit generation program, allowing credits
to be generated for gasoline that over-complies with the 10 ppm annual average standard. Credits
should be generated annually on a calendar year compliance period.
We support the inclusion of a special period of credit generation from 2014 to 2016 to facilitate
the transition from Tier 2 to Tier 3. We agree that the program should be available to refiners and
importers, and agree with the structure of generating credits against the current 30 ppm standard
with no individual refinery baseline. However, the ability to generate credits during this period
does not guarantee that sufficient credits will be available to allow refiners to broadly defer
investments past the proposed January 1, 2017 implementation date. EPA has described a very
optimistic scenario for refineries to generate credits by operating existing pre-treaters and post-
treaters at sub-30 ppm gasoline levels in 2014-16. If refineries fail to meet these optimistic
predictions, credit availability might be limited and will hinder the industry transition to the 10
ppm standard in 2017.
During the 2014-2016 period, we do not agree with the proposed structure of the early credit
program and instead recommend a different program structure to transition from the current Tier
2 program to the start of the Tier 3 program. We believe that any refinery should be able to
generate credits against the 30 ppm standard during the 2014-2016 period, and that these credits
should have a 5 year life. Credit generators should not have to designate these credits as either
Tier 2 or Tier 3 credits. The party who holds a credit generated in 2014-2016 should be able to
use the credit for compliance with Tier 2 standards or bank the credit for future use with Tier 3.
By granting these credits a 5 year life and by allowing a party to bank these credits for future
Tier 3 use, the refiner or importer is incentivized to reduce the sulfur content of their gasoline
pool. This could maximize the removal of sulfur from the pool even in advance of the Tier 3
program. No transition from Tier 2 Standard Credits to Tier 3 Early Credits to Tier 3 Standard
Credits would be required. This single credit program would be easier for EPA to manage and
enforce compared to implementation of an early credit program.
5-182
-------
Tier 3 Summary and Analysis of Comments
If EPA finalizes the early credit program as proposed, we do not agree that refiners should have
to designate credits generated from 2014 to 2016 as either Tier 2 credits or Tier 3 early credits.
We believe that these credits, generated against the 30 ppm standard, should be allowed to be
used for either Tier 2 or Tier 3 compliance without prior designation by the generating party.
Generated credits could accrue undesignated in the party's account and should be available to be
freely used for Tier 2 compliance, Tier 3 early compliance, or traded to other parties who could
use them for either program. Refiners will not be able to exactly predict their future credit
requirements and should not be penalized by over or under-designating credits for either
program.
In addition, we believe that any party holding Tier 2 credits at the end of 2016 should be allowed
to convert these Tier 2 credits to Tier 3 early credits, which could be used through 2019. Tier 2
credits represent real sulfur reductions against the existing standard which are a benefit to the
environment. The owners of these credits should be able to preserve their value by applying them
in the early years of the Tier 3 program.
Should EPA decide to proceed with the proposed credit designation of Tier 2 or Tier 3 during the
early generation credit period, we recommend that such declaration is made by February 28th of
the following year, consistent with compliance reporting.
And, we believe that the program should acknowledge the transition from Tier 2 to Tier 3 by
allowing unused Tier 2 credits to be converted to Tier 3 credits after 2016. Under the current
proposal, all Tier 2 credits will become stranded if they are not used by the end of the 2016
compliance year. The Tier 2 credits represent real sulfur reductions in gasoline below the
required standard which in turn generated environmental benefits beyond the requirements of
Tier 2. Refiners and importers should not lose the benefit of these credits that were generated by
their over-compliance with the sulfur standard.
Commenter: Phillips 66 Company
EPA is overly optimistic on early credit generation capability.
Secondly, EPA asserts that the Averaging, Banking and Trading program will provide refineries
with early credits that can be used by refineries that fall outside of the small refinery category to
postpone their project start-up date. Phillips 66 disagrees with this assessment. We believe EPA
is being overly optimistic on the capability of refineries to generate and make available sufficient
credits.
Phillips 66 supports the inclusion of an early credit generation provision in the regulation and we
believe these flexibilities combined with a 5-year implementation lead time will enhance the
chances of a successful implementation of any new standard. As a key implementation
component, ABT provides both 1) incentives for limited early reductions in gasoline sulfur with
existing units and 2) important flexibilities for continuing compliance after the final
implementation date.
5-183
-------
Chapter 5: Tier 3 Fuel Sulfur Program
We offer some suggested modifications to the proposed ABT provisions that we believe will
provide additional utility.
- Allow refiners to designate credits generated during the 2014-2016 time period as generic
sulfur credits and put them into a "bank".
- Allow refiners to use the generic credits for either Tier 2 compliance or leave them in the bank
and use them for Tier 3 compliance.
- Allow any Tier 2 credits in the bank at the end of 2016 (e.g. generated in 2011, 2012, or 2013)
to be converted into Tier 3 credits.
Commenter: PBF Energy Inc.
PBF supports the proposed ABT program and will examine our options to generate early credits.
We ask that the EPA clarify the regulatory language to ensure that the final regulation allows
refiners to decide on credit use as either Tier 2 or Tier 3 early credits at the time of use. It is
extremely difficult to forecast future requirements, given the potential for unplanned equipment
outages. Having to designate credit types prior to actual consumption increases the risk of
"stranding" credits in a less usable type and is a drawback to the rules as currently proposed. The
EPA through the current programs can still track the amount and year credits are generated and
effectively define the credits as to their potential. We see little value in forcing a refiner to
declare the credit type in advance of consumption.
Commenter: Weaver and Tidwell LLP
Additionally, the program allows for early credits to be earned. However, the Tier 2 sulfur credit
inventory is not being addressed. What will happen to the current level of Tier 2 sulfur credits?
To ignore the Tier 2 program sulfur reductions within the context of the Tier 3 program would
leave the Tier 2 sulfur decrease (funded by refiners, both large and small) out of the program.
Thus, the efforts of facilities to comply with Tier 2 are ignored. Weaver recommends that the
agency allow for the conversion of Tier 2 credits to Tier 3, at some reasonable conversion rate.
Early credits are valid through February 29, 2020. We believe this date should be changed to
March 31 (consistent with when the annual compliance reports are due, assuming the deadline is
changed).
Commenter: Wyoming Refining Company (WRC)
Some subtle aspects of the proposed rule, however, have the potential of making pre-2020 Tier 2
compliance more difficult, if not impossible. First, the proposed Tier 3 rule allows refiners to
generate early Tier 3 credits by lowering their gasoline sulfur below 30 ppm during 2014, 2015
and 2016. Considering the proposed compliance time frame, WRC expects there will be
significant demand for Tier 3 early credits that will enable large refiners to delay 10 ppm
production until 2020. The early credits will most likely be generated by increasing operating
severity on existing Tier 2 equipment. Most Tier 3 early credits will, therefore, probably not be
new credits but, instead, will be Tier 2 credits, most of which would have been generated in any
event, converted into Tier 3 early credits.
5-184
-------
Tier 3 Summary and Analysis of Comments
This situation will create a shortage in the Tier 2 credit supply on which WRC must depend to
maintain Tier 2 compliance through 2019.
It does appear that Tier 2 credits generated prior to 2014 (2013 and earlier), however, will be-
come more abundant starting in 2017 as large refiners transition into Tier 3 production and no
longer have a need for Tier 2 credits. These 'dumped' Tier 2 credits, however, will expire, at the
latest, in 2018. WRC, therefore, may very well be exposed to a situation in 2019 where abundant
2013 and earlier Tier 2 credits have all expired and 2014 and later Tier 2 credits have been
mostly converted into Tier 3 early credits. Non-compliance in 2019 is a very real possibility even
though the proposed rule intends to extend flexibility to small refiners and small volume
refineries until 2020.
Our Response:
As discussed in more detail in Section V.D of the preamble to the final rule, in response
to comments, we are finalizing a more flexible approach to the ABT program than proposed. We
believe this will provide for a more seamless transition from the Tier 2 to the Tier 3 credit
programs, and will provide more certainty of credit availability.
Consistent with our proposal, refiners and importers may begin generating credits on an
annual average basis January 1, 2014, and continue through December 31, 2016, for
overcompliance with the current 30 ppm annual average sulfur standard. We believe this simple
approach is possible because U.S. gasoline was averaging around 30 ppm as we started
developing the Tier 3 program, based on compliance data. Since refiners and importers would
need to continue to comply with the existing 30 ppm Tier 2 sulfur standards during this
timeframe absent Tier 3, they would need to maintain this level of performance on an industry
average basis. Accordingly, any additional gasoline sulfur reductions beyond 30 ppm will be
attributed to the proposed Tier 3 program. In response to comments, we are not requiring that
credits be designated as either Tier 2 or Tier 3 and only used for the program for which they
were designated. Refiners and importers may use credits generated prior to January 1, 2017
either for ongoing compliance with the Tier 2 program, or bank them for future compliance with
the Tier 3 program (within the limits of the credit life restrictions). Essentially, the Tier 2 credit
generation provisions simply continue, and any credits generated from January 1, 2014 through
December 31, 2016 that were not used for compliance with the Tier 2 standards would be carried
over for use in complying with Tier 3. We believe that this will allow for more certainty of
credit availability before refiners must make their Tier 3 investment decisions, thus reducing the
cost of the program. It will also avoid any incentive for refiners to use up banked Tier 2 credits
prior to 2017 causing increased in-use sulfur levels and emissions.
Based on our analysis of the Tier 2 credit market for 2012, we believe that there will be a
balance of 2012 banked credits equivalent to approximately two months of compliance, and we
anticipate a similar amount (perhaps more) for 2013. In response to several commenters that
urged EPA to allow these credits to be able to be used for Tier 3 compliance as well to avoid the
risk of them being stranded, we are finalizing a provision for these credits to retain their full five-
year life. We are finalizing this provision for a number of reasons. First, the Tier 2 banked
credits represent real reductions—refiners and importers are currently generating these credits
5-185
-------
Chapter 5: Tier 3 Fuel Sulfur Program
for overcompliance with the Tier 2 gasoline sulfur standard. Second, allowing these banked
credits to receive their full five-year credit life will provide more assurance of the credit
availability for trading for those who need them to comply with Tier 3 without a large capital
investment. As previously explained, this will allow for more certainty of credits available far
before making Tier 3 investment decisions, thus reducing the cost of the program. A lack of
certainty in the credit trading market could lead to refiners banking more credits than usual for
their own use rather than allowing these credits to be available in the market for trading. As
shown in our analysis in Chapter 4.3 of the RIA, refiners tend to hold credits as an insurance
policy until they approach the end of their credit life. If credit-generating refiners continue with
this approach, credits generated in 2012 and 2013 will likely be available for purchase in 2017
and 2018 for those refiners that may want to rely on them for compliance (along with additional
credits generated from 2014 through 2016). Finally, as we anticipate that these credits will be
equal to about four months of compliance, the allowance of 2012 and 2013 banked Tier 2 credits
makes for a more flexible program by effectively allowing for a small amount of additional lead
time without adversely affecting the overall benefits of the Tier 3 program. We believe these
provisions will allow the Tier 3 program to begin on January 1, 2017, with more certainty
regarding the availability of credits for those refiners needing (or choosing) to defer investment
to better align with their existing turnaround/shutdown schedules.
However, one small refiner highlighted an opposite concern—that large refiners may
hold sulfur credits for their own compliance with Tier 3, drying up the market that small refiners
can rely on to continue to comply with the Tier 2 standards (through December 31, 2019). While
we appreciate this concern, we believe it is overstated. Based on our analysis of current sulfur
credit generation and use, we believe that the 5-year credit life will continue to provide an
incentive for credits to be made available in the market through 2019 for Tier 2 compliance. In
fact, since credits generated in 2015 and 2016 cannot be used in 2020 and beyond, there is now
likely to be an increase in credit availability in the years prior to 2020.
Under the final Tier 3 ABT program, the credit generation provisions are nearly the same
as those under the Tier 2 program - in essence, the Tier 2 program simply continues with a lower
standard below which credits are generated. Refiners and importers are allowed to average
within and across companies to meet the standard in the most cost-effective manner possible,
including generating and using credits. Credit generation periods remain 12 months long and
continue to be synchronized with annual compliance demonstration periods. As noted by one
commenter, these annual reports (which will include a refiner or importer's credit generation and
use information) will be due on March 31 of each year, not February 28.
Please see our response to Chapter 5.5.2.4, below, for a detailed discussion of the final
credit life and credit trading provisions.
5.5.2.4 Credit Use - Credit Trading and Credit Life
What Commenters Said:
Commenter: Alliance of Automobile Manufacturers (Alliance) and Association of Global
5-186
-------
Tier 3 Summary and Analysis of Comments
Automakers (Global)
We concur with EPA providing a three-year life for sulfur credits (60). The proposed three-year
early credit life provision would be consistent with the duration of the small volume refinery
provisions. It is important that when Tier 3 is fully phased in for automakers it is likewise fully
phased in for refiners, so that the public can fully benefit from its provisions.
60 - 78 Fed. Reg. 29908 at 29939 (May 21, 2013).
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM)
We agree with the 5 year standard credit life as proposed, but, as explained above, we believe
that the proposed 3 year early credit life should be extended to 5 years. And, we believe that the
program should acknowledge the transition from Tier 2 to Tier 3 by allowing unused Tier 2
credits to be converted to Tier 3 credits after 2016. Under the current proposal, all Tier 2 credits
will become stranded if they are not used by the end of the 2016 compliance year. The Tier 2
credits represent real sulfur reductions in gasoline below the required standard which in turn
generated environmental benefits beyond the requirements of Tier 2. Refiners and importers
should not lose the benefit of these credits that were generated by their over-compliance with the
sulfur standard.
We do not believe the maximum two trade limitation on sulfur credits is required. The credit
program is primarily protected by limiting participation in the credit program to refiners and
importers. The two trade limitation reduces market liquidity without providing a significant
enforcement benefit. Multiple sales transactions of the same credits within this small community
can be adequately tracked with commercial documentation and EPA reporting. To maintain
integrity of the sulfur credit program, the prohibition on outside parties taking ownership of
credits should absolutely be preserved, but the two trade limitation could be eliminated.
Should EPA decide to proceed with two trade limitation, we recommend that EPA clarifies and
clearly states that there are no restrictions to transferring credits within a refiner/refining
company.
Commenter: Chevron Products Company
Credits should be allowed to be traded between companies who are participants in the Tier 3
program. Credits should not be limited to internal transfers between refineries from the same
company.
Regarding credit trading, we do not believe the maximum two trade limitation on sulfur credits is
warranted. The credit program is primarily protected by limiting participation in the credit
program to refiners and importers. The two trade limitation reduces market liquidity without
providing a significant enforcement benefit. Multiple sales transactions of the same credits
within this small community can be adequately tracked with commercial documentation and
5-187
-------
Chapter 5: Tier 3 Fuel Sulfur Program
EPA reporting. To maintain integrity of the sulfur credit program, the prohibition on outside
parties taking ownership of credits should absolutely be preserved, but the two trade limitation
should be eliminated.
Commenter: CHS, Inc.
CHS is providing the following comments regarding the sulfur averaging, banking and trading
(ABT) program. Currently under the proposed Tier 3 regulation, five year sulfur credits
generated under the Tier 2 regulation for the years 2012 and 2013 can only be used for
compliance through the calendar year 2016 thus cutting short the five year credit life. CHS
requests that the EPA allow these Tier 2 sulfur credits to be used to demonstrate compliance for
years 2017 and 2018.
Commenter: Independent Fuel Terminal Operators Association (IFTOA)
Early credits should be valid for more than the three years proposed. The 10 ppm annual average
sulfur standard is an aggressive standard that will take substantial time and money to meet.
Refiners need as much flexibility as possible, and EPA should do all that it can to maximize (a)
the number of early credits available, and (b) the value (credit life) of such credits to facilitate
compliance. Extension of the life of early credits would meet that objective.
Recommendation: The Association recommends that both refiners and importers have the right
to generate early credits to help ensure that an adequate volume of such credits is available to
help refiners stagger their investments to meet the proposed Tier 3 standard. In addition, the life
of such credits should be 5 years from the date Tier 3 is implemented.
Permit both refiners and importers to generate early credits to facilitate delivery of lower sulfur
gasoline into the market at the earliest possible date and to allow such credits to be used for a
period of 5 years from the date that the Tier 3 regulations become effective;
Commenter: Irving Oil Terminals Inc.
Irving Oil endorses EPA's proposal to limit the number of parties who may hold title to Tier 3
credits. Such a limitation would enhance the ability of regulated parties to assess the validity of
the credits generated and maintain the integrity of the credit program.
Moreover, early credits should be valid to demonstrate compliance into 2019, 2020 and 2021.
Commenter: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
EPA has proposed that under the Tier 3 ABT program, sulfur credits must be transferred directly
from the refiner or importer generating them to the party using them for compliance purposes
(61). This would ensure that the parties purchasing them are better able to assess the likelihood
that the credits are valid. EPA has allowed an exception so that if a credit generator transfers
5-188
-------
Tier 3 Summary and Analysis of Comments
credits to a refiner or importer who inadvertently cannot use all the credits, the credits can be
transferred a second time to another refiner or importer. After the second trade, the credits must
be used or be terminated. We agree that allowing a maximum of two trades is consistent with
other recent fuel programs and is sufficiently flexible while still preserving adequate means for
enforcement. One need only look at recent activity in the Renewable Fuel Standard RFS (RIN)
credit mechanism to know that fraud is a concern, and therefore minimizing the trading of these
potentially valuable credits is the correct approach.
61 - 78 Fed. Reg. 29908 at 29940 (May 21, 2013).
Commenter: Marathon Petroleum Company LP (MFC)
We agree with the 5 year standard credit life as proposed, but, as explained above, we believe
that the proposed 3 year early credit life should be extended to 5 years.
We do not believe the maximum two trade limitation on sulfur credits is required. The credit
program is primarily protected by limiting participation in the credit program to refiners and
importers. The two trade limitation reduces market liquidity without providing a significant
enforcement benefit. Multiple sales transactions of the same credits within this small community
can be adequately tracked with commercial documentation and EPA reporting. To maintain
integrity of the sulfur credit program, the prohibition on outside parties taking ownership of
credits should absolutely be preserved, but the two trade limitation could be eliminated.
Should EPA decide to proceed with two trade limitation, we recommend that EPA clarifies and
clearly states that there are no restrictions to transferring credits within a refiner/refining
company.
Commenter: Northeast States for Coordinated Air Use Management (NESCAUM)
Early credits generated under the averaging, banking, and trading (ABT) program should expire
after three years. The three-year early credit life provision offers sufficient flexibility to refiners
and still provides a date certain by when automobile manufacturers can be assured that
introduction of their technology designs correspond with the availability of 10 ppm average
sulfur gasoline.
Commenter: Phillips 66 Company
We strongly support limiting ownership of sulfur credits to refiners and importers. This has been
instrumental in protecting the integrity of the current program.
- Provide a 5 year life to all credits
Our Response:
Credit Life
5-189
-------
Chapter 5: Tier 3 Fuel Sulfur Program
Under the final Tier 3 ABT program, in response to comments, all credits generated
before January 1, 2017 will be valid for five years or until December 31, 2019, whichever is
earlier—no early credits may be used for compliance beginning January 1, 2020. Thus, unlike
our proposal, banked Tier 2 credits generated in 2012 and 2013 will receive their full five-year
life and will not expire at the start of the Tier 3 program. However, credits generated in 2015
and 2016 that are unused as of December 31, 2019 will expire and become invalid. We believe
that structuring the credit program this way will offer considerable flexibility to refiners phasing
in Tier 3 gasoline sulfur controls, while still placing a date at which point the intended sulfur
program will be fully implemented and enforceable—January 1, 2020 (the same date small
refiners and small refineries must begin complying with the 10 ppm sulfur standard). This will
also provide a date certain to give auto manufacturers greater confidence for the design of their
vehicles that all vehicles in-use are running on 10 ppm average fuel. Otherwise, it is possible
that the greater ease of generating early credits relative to 30 ppm sulfur (as opposed to 10 ppm
in 2017 and beyond) would allow higher sulfur levels to continue well beyond 2019.
Consistent with our proposal, all credits generated beginning January 1, 2017 will be for
overcompliance with the Tier 310 ppm annual average sulfur standard, and will have a five-year
credit life. We believe five years will give refiners and importers sufficient time to use credits
generated in previous years, while still placing limitations on credit life to help with
enforcement. Five-year credit life is also consistent with the Tier 2 ABT program, as well as the
current credit life and recordkeeping provisions for other 40 CFR part 80 fuels programs, and
coincides with the applicable five-year statute of limitations for violations by parties who
generate invalid credits. Credits must be used within five years from the year they were
generated (regardless of when/if they are traded), otherwise they will expire and become invalid.
For example, credits generated in 2017 can be applied towards 2018-2022 compliance, as well as
2017 compliance. After March 31, 2023 (when reports for the 2022 annual compliance period
will be due), credits generated in 2017 will expire and become invalid. Similarly, credits
generated in 2018 can be applied towards 2019-2023 compliance, as well as 2018 compliance.
After March 31, 2024, credits generated in 2018 expire, and so on and so forth.
Credit Trading
We are finalizing credit trading provisions in the Tier 3 ABT program that are identical to
those under the current Tier 2 program. As with Tier 2, it is possible that sulfur credits could be
generated by one party, subsequently transferred or used in good faith by another, and later
found to have been calculated or created improperly or otherwise determined to be invalid.
Under the Tier 2 ABT program, as well as other 40 CFR part 80 fuel programs, if this occurs, we
are requiring that both the seller and purchaser will have to adjust their sulfur calculations to
reflect the proper credits and either party (or both) could be determined to be in violation of the
standards and other requirements if the adjusted calculations demonstrate noncompliance with
the 10 ppm standard. Sulfur credits must be transferred directly from the refiner or importer
generating them to the party using them for compliance purposes, to ensure that the parties
purchasing them are better able to assess the likelihood that the credits are valid. However, as
with the Tier 2 ABT program, there are no prohibitions against brokers facilitating the transfer of
credits from one party to another. Further, any person can act as a credit broker, regardless of
whether such person is a refiner or importer, as long as the title to the credits is transferred
5-190
-------
Tier 3 Summary and Analysis of Comments
directly from the generating refiner or importer to the using refiner or importer. We believe that
maintaining a prohibition on outside parties taking ownership of credits will allow for maximum
program enforceability and consistency with all of our other ABT programs for mobile sources
and fuels.
We disagree with comments arguing against the two-trade maximum for inter-company
trading that exists in the current Tier 2 ABT program. Under the final Tier 3 ABT program, after
the second trade, credits must be used or they will expire. While some commenters stated that
they believe the two-trade maximum is not necessary given the fact that credits are only being
traded within a small part of industry, we believe that unlimited trading could result in an
unenforceable program and potentially lead to problems with invalid credit trading. Further,
allowing a maximum of two trades is consistent with other recent fuel programs and we believe
it is sufficiently flexible while still preserving adequate means for enforcement. Also consistent
with the Tier 2 program, the provision for unlimited trading for intra-company credits will
continue. In response to comments, we have clarified this in both the preamble and the
regulatory text.
5.5.2.5 Deficit Carryforward
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM), Chevron Products Company, Marathon Petroleum
Company
The deficit carry-forward is an important regulatory flexibility mechanism that is included in the
Tier 3 regulations but is not discussed in the preamble. We support the inclusion of a deficit
carry-forward provision in Tier 3. However, we suggest that the deficit carry-forward be
extended to three years from the current one year. Compliance with the 10 ppm annual average
will be very challenging for the average refinery. The ability to reduce sulfur significantly below
10 ppm to make up a compliance deficit in a single year may not be possible, depending on the
technology and process configuration of the refinery. Credits may or may not be available in the
market for the refinery to supplement their production of low sulfur gasoline. Extending the
deficit from one to three years allows the refinery to manage their sulfur production within
reasonable limits and still maintain the overall integrity of the Tier 3 program.
In addition, we believe the deficit carry-forward language at 80.1616(a)(6) is in error. It states
that a refiner should use all available credits before recording a compliance deficit. We believe
that this requirement should apply to an individual refinery and not to a refiner. A refiner should
be allowed to carry a deficit at one refinery while having a surplus of credits at another refinery,
if they so choose. We request EPA to make this correction in the regulations.
Commenter: Phillips 66 Company
5-191
-------
Chapter 5: Tier 3 Fuel Sulfur Program
Provide a multi-year deficit carryover provision. Meeting the very stringent 10 ppm average will
be challenging. An operational issue could persist causing a refinery to not be able to meet the
standard. Securing sufficient credits or making up the deficit operationally the following year
may also be very difficult. Allowing a multi-year deficit carryover will be beneficial. We suggest
allowing a three year deficit carryover rather than just a single year.
Section 80.1616(a)(6) states that "a refiner possessing credits must use all credits prior to falling
into a compliance deficit". The use of "refiner" implies an aggregated company basis. We ask
that this be changed to be on a refinery specific basis rather than a refiner basis.
Our Response:
As discussed in the preamble to the final rule, we are finalizing deficit carryforward
provisions similar to the existing Tier 2 program. An individual refinery that does not meet the
10 ppm standard in a given year may carry a credit deficit forward for one year, and will have to
make up the credit deficit and come into compliance with the Tier 3 sulfur standard the next
calendar year.
Regarding comments requesting that the deficit carryforward allowance be extended to
two or three years, we disagree with these comments primarily because of concerns with the
enforceability of a allowing for a deficit beyond one year. Furthermore, an extended deficit
carryforward could further delay Tier 3 sulfur reductions. However, in recognition of
unanticipated circumstances, such as where credits are unavailable or are prohibitively expensive
such that a refiner could not make up the deficit in one year, we are finalizing hardship
provisions that provide EPA with the authority to allow for extended deficit carryforward if a
refiner's hardship petition demonstrates that it meets the required criteria. Thus, as described
more in Section V.E.2 of the preamble to the final rule, a refiner could carry a deficit forward for
up to three years only in cases of hardship situations.
We disagree with comments suggesting that the deficit carryforward provisions should be
applied on a refinery (facility-specific) basis instead of a refiner (company) basis. There is little
need for a company to carry a deficit for one refinery if it has excess credits from its other
refineries. But allowing this could defeat the purpose of the one-year restriction, as a refiner
could simply shift a deficit around from one refinery to another year after year.
5.5.3. Regulatory Flexibility Provisions (Small Refiners and Small Volume Refineries)
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM)
If EPA implements a small refiner/refinery extension, we recommend that any small
refiner/refinery holding Tier 2 credits at the end of 2019 should be able to convert these credits
to Tier 3 credits. Again, these Tier 2 credits represent over-compliance with the Tier 2 standard
5-192
-------
Tier 3 Summary and Analysis of Comments
and would have created a benefit for the environment. The value of these credits should not be
lost due to expiration. Refiners should not be penalized for failing to predict their exact Tier 2
credit requirements prior to the end of 2019.
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM)
Standard credit generation by small refiners and small volume refineries should be allowed
during the 2017-2019 period. The preamble suggests that small refiners would have to
voluntarily opt in to the Tier 3 program. We note that the proposed regulations do not include
any obvious reference to this opt-in mechanism. We are not able to comment on whether this
opt-in would be appropriate without having the regulatory text available to understand how it
would be implemented.
Small refinery exceptions are of limited utility if pipelines set lower specifications, which is what
happened in the Tier 2 rulemaking.
Commenter: Chevron Products Company
If EPA proceeds with the gasoline sulfur reduction, we do support several of the proposed
elements including: small refinery exemption;
We agree with the inclusion of an extension of the compliance deadlines for small refiners and
small refineries as proposed in the Tier 3 NPRM. However, the extension of the compliance
deadline may not be sufficient to ensure the long term viability of these challenged refineries.
The Tier 3 capital investment requirements for certain refineries may be economically
prohibitive regardless of an extension of the start date. EPA should carefully consider the impact
of the Tier 3 regulations on all refineries and whether additional flexibility measures are
warranted.
If the EPA defers the 2017 start date to a later time, the extension for small refiners and small
refinery owners should be maintained, such that the Tier 3 requirements should apply to small
refiners and refineries at least three years after the revised start date.
Commenter: Chevron Products Company
The preamble suggests that small refiners would have to voluntarily opt in to the Tier 3 program
and comply with the 10 ppm standard to be able to generate standard credits. We note that the
proposed regulations do not include any obvious reference to this opt-in mechanism. We are not
able to comment on whether this opt-in would be appropriate without having the regulatory text
available to understand how it would be implemented.
Standard credit generation by small refiners and small volume refineries should be allowed
during the 2017-2019 period.
5-193
-------
Chapter 5: Tier 3 Fuel Sulfur Program
Small refiners/refineries should be able to participate in the credit trading program. We suggest
that the credit generation rules during the transition period, 2014 through 2019 for small
refiners/refineries, should allow flexibility for parties to manage their credits for compliance and
to generate value for their over-compliance. The requirement for small refiners/refineries to opt-
in to Tier 3 to generate standard credits is not clear in the preamble and does not appear in the
regulations. We do not understand what it would mean for a small refiner/refinery to opt-in to
Tier 3. Could an opt-in party also opt back out? We suggest that small refiners/refineries be
allowed to comply with the 30 ppm standard through the 2019 compliance year. Any credits
generated during this period should be able to be used for compliance with the 30 ppm standard,
and unused credits at the end of 2019 should be allowed to be used for Tier 3 compliance.
Conversion of Tier 2 credits to Tier 3 credits preserves the value generated by the refinery by
over-complying with the standard and creating a benefit for the environment.
Commenter: Countrymark, Small Business Refiners (SBR)
We recommend that the early credit generation period be available for the three years prior to
compliance (2014-2020) for small refiners and small volume refineries just as the early credit
generation is available for large refineries in the three years prior to compliance. We also
recommend that early credits are available for use three years after the compliance deadline
(through 2023), just as they are available for large refiners.
A small refiner that is unable to comply with the rule until 2020 will not be penalized for
complying with the rule by the due date. Not having early credits available for use in the three
years after equipment start-up would significantly impact a small refiner and the communities
that they supply if issues occur (such as FCC post-treater shut down) and credits are not available
to continue operations.
As part of the Tier 2 regulation, EPA permitted small refiners, as well as refiners in the
Geographic Phase-in Area, to fully participate in the ABT program at the same time the program
started for the industry at large. This decision provided early compliance incentives for all
refineries and maximized the credits available system-wide. EPA has not provided any
justification on why they have significantly scaled back small volume refinery participation in
the ABT program compared to Tier 2.
Commenter: Countrymark
If an exemption cannot be granted, then we recommend that the ABT program be re-structured
per our comments above to generate early credits in 2014 through 2020 and utilize early credits
for compliance through 2023. Implementing these changes will provide appropriate relief to the
SBRs as they implement their compliance projects.
We prefer that the Small Business Refiners and Small Volume Refineries be exempt from
complying with Tier 3 fuels requirements. As stated above, DOE's estimate that small refiners
only provide 11.9% of the gasoline supply. With this small contribution to overall emissions, we
recommend that an exemption from the rule due to the high cost of implementation compared to
the low impact of the required investment. EPA estimates that the 2030 sulfur reduction will be
5-194
-------
Tier 3 Summary and Analysis of Comments
14,264 tons per year; of which the estimated total contribution from fuels generated by SBR s
will be 2,000 tons per year.
Commenter: Small Business Refiners (SBR)
Clearly, SBRs have important financial differences from large refiners. It is a well-settled fact
that our size limits the options we have to comply economically with new regulations.
Small Business Refiners are important to the economy
- Small refiners foster competition in the petroleum industry.
- Small refiners are critical to easing the tight supply of petroleum products and often are the
only sources of supply in their areas.
- Most small refiners serve as the major economic resource in the small, often rural, communities
in which they operate.
- It is generally agreed that the economic 'multiplier effect' (jobs and other local and regional
investment and businesses) resulting from refinery operations is eight-to-ten times the refinery's
actual budget.
- Many small refiners provide a reliable and competitive supply of military jet fuel to our
country's military bases and thus are important to national security.
Small Business Refiners Do Not Enjoy Economies of Scale
- Large refiners are able to spread compliance and operating costs over much greater product
sales and over a much greater asset base.
- SBRs are not fully integrated like the large refiners. Many do not have upstream crude oil and
gas production, midstream pipelines and terminals, or downstream retail marketing.
- SBRs already have been at a disadvantage with major refiners relative to higher production
costs of ultra-low sulfur and reformulated fuels on a per barrel basis.
- SBRs as a group are most vulnerable to decreasing domestic demand for refined products and
increased competition from renewable fuels.
Small Business Refiners Have Limited Resources and Compliance Flexibility
- Access to capital present much greater obstacles for SBRs than for larger refiners.
- Small refiners do not have large staffs to negotiate and implement permitting, regulatory, and
compliance requirements.
- By contrast, large refiners have access to a larger qualified labor pool and can maintain large
corporate staffs with a diverse range of specialties and in-house expertise.
- Qualified outside engineering consulting is limited even where financial resources to procure
such help are available.
- Due to the smaller size of projects, SBRs are disadvantaged when competing with large refiners
to garner outside engineering resources.
- The majority of SBRs do not have port access like the majority of large refiners and are
therefore more reliant upon local domestic crude supplies; they therefore have little or limited
ability to change crude slate when regulations and specifications change.
- The majority of SBRs are less complex than large refiners and thus have less operational
flexibility and fewer outlets for intermediate products.
- Small refiners owning just one or two facilities have limited internal compliance flexibility
relative to the industry at large with respect to Average, Banking, and Trading (ABT) programs.
5-195
-------
Chapter 5: Tier 3 Fuel Sulfur Program
ABT programs, which are a fundamental aspect of many EPA fuel regulations, inherently
provide more flexibility to companies owning multiple refineries than SBRs owing a single or
few facilities.
The Small Business Refiners would like to thank EPA for adding the definition of a Small
Volume Refinery. This additional definition will assist many members of our Ad Hoc group that
would not have qualified for certain benefits such as delayed implementation status under the
small refinery definition. This change will benefit small refiners and the communities that we
supply with transportation fuels.
While SBRs are able to achieve compliance, we will be disproportionately disadvantaged
compared to larger facilities due to the economy of scale, scope, technical staff availability, and
funding capabilities. Among these challenges, economy of scale to spread the capital cost over
the barrels of produced product will be the hardest to overcome.
We prefer that the Small Business Refiners and Small Volume Refineries be exempt from
complying with Tier 3 fuels requirements. As stated above, DOE's estimate that SBRs only
provide 11.9% of the gasoline supply. With this small contribution to overall emissions, we
recommend that an exemption from the rule due to the high cost of implementation compared to
the low impact of the required investment. EPA estimates that the 2030 sulfur reduction will be
17,267 tons per year; of which the estimated total contribution from fuels generated by SBRs
will be 2,000 tons per year.
If an exemption cannot be granted, then we recommend that the ABT program be re-structured
per our comments above to generate early credits in 2017 through 2020 and utilize early credits
for compliance through 2023. Implementing these changes will provide appropriate relief to the
SBRs as they implement their compliance projects.
The SBRs will be balancing cash flow, personnel availability, and contractor availability to
complete implementation by 2020. Some SRBs may complete their projects in 2020 because this
is the fastest that they are able to complete all of the project work; independent of their desire for
early compliance and credit generation.
Commenter: Marathon Petroleum Company LP (MFC)
Finally, since MFC believes that the average refinery size has increased since the Tier 2
rulemaking, EPA should increase the small refinery exemption size threshold from 75 MBPD to
90 MBPD.
EPA should change the size limit for the small refinery exemption from 75,000 MBPD to 90,000
MBPD
EPA should increase the "average aggregate daily crude oil throughput" amount in the "small
volume refinery" definition from 75,000 MBPD to 90,000 MBPD to reflect changes that have
occurred in the industry in the past several years. According to EIA, in 2000 there were 151
refineries in the US with an average crude capacity of 111.2 MPBD. In contrast, the figures for
5-196
-------
Tier 3 Summary and Analysis of Comments
2012 were 128 refineries with an average crude capacity of 133.5 MBPD or an increase of 20%.
In response to economic pressures, the US refining industry has shifted to a smaller pool of
larger facilities. The concerns that the "small volume refinery" definition is intended to address
still occur—ability to raise capital and access to engineering services—but to larger facilities. In
recognition of that fact, EPA should increase the "small volume refinery" throughput amount
accordingly.
Standard credit generation by small refiners and small volume refineries should be allowed
during the 2017-2019 period. The preamble suggests that small refiners would have to
voluntarily opt in to the Tier 3 program. We note that the proposed regulations do not include
any obvious reference to this opt-in mechanism. We are not able to comment on whether this
opt-in would be appropriate without having the regulatory text available to understand how it
would be implemented.
If EPA implements a small refiner/refinery extension, we recommend that any small
refiner/refinery holding Tier 2 credits at the end of 2019 should be able to convert these credits
to Tier 3 credits. Again, these Tier 2 credits represent over-compliance with the Tier 2 standard
and would have created a benefit for the environment. The value of these credits should not be
lost due to expiration. Refiners should not be penalized for failing to predict their exact Tier 2
credit requirements prior to the end of 2019.
Commenter: Natural Resources Defense Council (NRDC)
Further, small refiners and small volume refineries have a longer period to raise needed capital
and make investments.
Commenter: Shell Oil Products for Shell and Motiva
EPA Should Not Exempt Small Refiners/Refineries EPA proposes to exempt both small
refiners and small refineries until 2020. Shell disagrees and urges EPA to not pick winners and
losers in the market place and allow all refiners and refineries to compete on a level playing
field.
In any event, if EPA does provide a small refinery/small refiner exemption, it should not be
limited solely to refiners that process crude oil. As explained above regarding credit generation,
EPA should treat the obligated parties fairly and extend this same treatment to gasoline blenders,
which are classified as refiners under EPA's rules, even though they do not process crude oil.
Commenter: Private Citizen
The one issue I have is with the EPA giving a 3 year extension to refineries using less than
75,000 barrels per day of oil to comply with the change. Many refineries of that size are usually
owned by large oil companies that have the financial resources to implement the changes
immediately. Giving them a 3 year extension only extends the poor health due to poor air quality
that many residents of the USA have to contend with.
5-197
-------
Chapter 5: Tier 3 Fuel Sulfur Program
Commenter: United Steelworkers Union (USW)
USW supports the inclusion of methods to allow smaller refineries to seek compliance relief, and
proposals to provide credit to refiners already producing lOppm, which will allow for market
trading or banking of the credits to manage compliance until 2019. Methods to ease the transition
and allow for capital investment planning will give refiners and their workers time to adequately
prepare facilities for the upgrades.
Commenter: Weaver and Tidwell LLP
§80.1622(a) - Applications for small refiner or small volume refinery status must be submitted
by March 31, 2014. Weaver recommends that this date be extended to June 30, 2014. Given the
Agency's aggressive time line for approval of this section, EPA will most likely not issue the
final regulations until November 2013. Any slippage of this date would put unnecessary burden
on the registrants.
Commenter: Wyoming Refining Company (WRC)
These comments are prompted by the fact that WRC has not yet installed its Tier 2 compliance
equipment, but, instead, has been demonstrating compliance with Tier 2 sulfur credits generated
under the small refiner provisions of the Tier 2 rule. Under the proposed Tier 3 rule, WRC would
be allowed to purchase and use Tier 2 credits through 2019 rather than construct either a Tier 2
or Tier 3 project now and, therefore, forfeit the intended small entity benefits of the proposed
Tier 3 rule.
WRC, therefore, respectfully suggests that the final rule contain two additional provisions. First,
any Tier 2 or early Tier 3 sulfur credit may be used by a small refiner or small volume refinery
for compliance with Tier 2 requirements through 2019. Second, the life of any 2013 and earlier
Tier 2 credits shall be extended through 2019 so long as they are owned by a small refinery or
small volume refinery. These provisions will broaden the supply of credits available for small
refiner and small volume refinery Tier 2 compliance through 2019. There may be concerns about
small volume refineries owned by multi-refinery companies and the ability to shift credits from
one facility to another. WRC, however, feels these concerns are not valid since, beginning in
2017, the only demand for Tier 2 credits will be from small refiners and small volume refineries.
Tier 2 credits will have no further value for larger refineries and will necessarily be obtained
only for small facility compliance.
Our Response:
Small Refiner/Small Refinery Delay
As discussed in Section XII.C of the preamble to the final rule, refiners who are classified
as small businesses ("small refiners") as a category generally lack the resources that are available
to larger companies to raise capital for investing in a new regulatory program, such as shifting of
internal funds, securing of financing, or selling of assets. As noted by comments above, and
based on information available from small refiners and others, we believe that the category of
5-198
-------
Tier 3 Summary and Analysis of Comments
entities classified as small generally face unique circumstances with regard to compliance with
environmental programs, compared to larger entities. As such, the final Tier 3 gasoline sulfur
program includes provisions that will provide assistance for small businesses in meeting the 10
ppm average sulfur standard. This approach will allow the overall program to begin as early as
possible; achieving the air quality benefits of the program as soon as possible, while helping to
ensure that small businesses have adequate time to raise capital for new fuel desulfurization
equipment or to make any other needed changes. We also believe that small business regulatory
flexibilities can provide these entities with additional help and/or time to accumulate capital
internally or to secure capital financing from lenders, and could spread out the availability of any
needed engineering and construction resources in a manner that they are available by the time
they are needed.
As discussed in Sections V and XII.C of the preamble to the final rule, we are also aware
that there are some refineries that may experience higher compliance costs on a per-gallon basis
than other refineries, and in some cases considerably higher. In many cases, these are refineries
owned by a refiner/company that would not meet the Small Business Administration definition
of a small business (as those entities classified as "small refiners" do). In many cases, these
small refineries compete directly in markets also supplied by small business refineries. In an
oversupplied gasoline market, these refineries may have difficulty justifying capital investments
to comply with new standards. Overall, we believe that these refineries are disproportionally
impacted when it comes to their cost of compliance and ability to rationalize the investment costs
in today's gasoline market. These small refineries also find it more difficult to obtain vendor,
contractor, and engineering firm support, as these firms find it more profitable to first focus on
the larger refinery projects. Giving these refineries additional lead time will allow more time to
invest in desulfurization technology, take advantage of advancements in technology, develop
confidence in a Tier 3 credit market as a means of compliance, and avoid competition for capital,
engineering, and construction resources with the larger refineries.
While the small business refiners as a category, and the small volume refineries as a
category are expected to experience a greater challenge in complying with Tier 3 than larger
refiners/refineries, within these categories are also refineries that are already complying with the
Tier 3 sulfur standards or need take little action to comply with Tier 3. Consequently, providing
a delay in compliance to these categories of refineries as a whole must be balanced against
providing an unnecessary delay in compliance and/or a windfall in credit generation. To avoid
windfall credits for those refineries that are already complying with the Tier 3 standard, we have
also carefully constructed the program such that they generate credits relative to 30 ppm prior to
2017 like all other refineries and relative to 10 ppm beginning in 2017 just like all other
refineries for use in complying with Tier 3. Furthermore, we do not believe that refineries that
do not process crude oil (e.g., produce gasoline from blendstocks) experience the same challenge
in complying with Tier 3, and thus do not warrant such relief. They have the ability to comply
through low or even no capital cost changes relatively quickly and easily in comparison to the
FCC naptha hydrotreating modifications required by many refineries that produce gasoline by
processing crude oil.
We do not believe a complete exemption from the program is appropriate, as this could
create situations where vehicles could receive fuel that is above a 10 ppm average sulfur
5-199
-------
Chapter 5: Tier 3 Fuel Sulfur Program
standard. However, we do believe a short amount of additional lead time for these refiners and
refineries will provide them needed assistance with compliance while still allowing the Tier 3
program to begin on January 1, 2017. Thus, the final Tier 3 program includes a delayed
compliance schedule for small refiners and small volume refineries of January 1, 2020 for small
refiners, which will allow these entities to postpone compliance with the Tier 3 program for up to
three years. Small refiners and small volume refineries will have from January 1, 2017 through
December 31, 2019 to continue production of gasoline with an average sulfur level of 30 ppm
(per the Tier 2 gasoline sulfur program). This delayed compliance schedule is not intended as an
opportunity for these refiners and refineries to increase their production of gasoline with sulfur
levels greater than 10 ppm, but rather will help small refiners/small volume refineries with
compliance with the Tier 3 program. Compliance with the 10 ppm annual average sulfur
standard will begin on January 1, 2020, thus small refiners and small volume refineries may
continue using Tier 2 gasoline sulfur credits through December 31, 2019 to meet their refinery
average 30 ppm sulfur standard. Refiners must apply, and be approved, for small refiner or
small volume refinery status. Applications are due on January 1, 2015.
Small Volume Refinery Definition
We disagree with comments that suggested we increase the size threshold for small
volume refineries. As noted above, we are aware that some refineries that may experience higher
compliance costs than other refineries and that these refineries are owned by a refiner/company
that would not meet the definition of a small business. In recognition of this concern under the
RFS program, Congress granted all refineries with a crude oil throughput of less than or equal to
75,000 barrels per calendar day (bpcd) additional time to comply. The Tier 3 final rule includes a
small volume refinery net crude throughput of less than or equal to 75,000 bpcd, based on the
highest crude throughput for the 2012 calendar year. In analyzing various crude throughput
maximums between 75,000 and 90,000 bpcd, as suggested by a commenter, we do not believe it
is appropriate or necessary to increase the threshold beyond what was previously set by
Congress. The 75,000 bpcd limit set by Congress was to recognize those refineries that would
have difficulty with compliance with a rulemaking (from both a cost and feasibility standpoint),
raising this limit would go beyond Congress' intent.
Nor do we believe it is necessary or appropriate to extend this definition to include
refineries that produce gasoline only from purchased blendstocks. The purpose of the relief is to
allow lead time for capital investments for refineries that are producing gasoline from crude oil.
Refineries using blendstocks can control their sulfur levels through their blendstock purchases.
Furthermore, allowing this could open a significant compliance loophole that would be difficult
if not impossible to enforce.
AST for Small Refiners and Small Refineries
We are also finalizing provisions that would allow approved small refiners and small
volume refineries to generate credits for overcompliance with the 30 ppm Tier 2 standard prior to
January 1, 2020. Such credits generated by small refiners and small volume refineries can be
traded/sold to non-small refiners for use by December 31, 2019, and the credit revenues could be
used to help offset their Tier 3 investments. When the Tier 3 program begins on January 1,
5-200
-------
Tier 3 Summary and Analysis of Comments
2017, small refiners and small volume refineries may continue to generate credits for
overcompliance with the 30 ppm sulfur standard (as they will still be subject to the Tier 2
standards through December 1, 2019), or they may generate credits for overcompliance with the
Tier 310 ppm sulfur standard. Any credits generated for sulfur levels below 10 ppm will be
eligible for compliance with either the Tier 2 or Tier 3 programs. Any credits generated for
sulfur levels above 10 ppm (and below 30 ppm) will only be eligible for use in complying with
the Tier 2 standards from 2017-2019. For example, in 2017, a small refiner with an annual
gasoline sulfur average of 8 ppm could generate 20 ppm-volume Tier 2 credits (30 ppm-10 ppm)
that could be used by other small refiners and small volume refineries, or banked by the refinery
for future Tier 2 compliance. This small refiner would also generate 2 ppm-volume Tier 3
credits (10 ppm-8 ppm) that could be sold to refiners and importers subject to Tier 3, or banked
by the refiner for future Tier 3 compliance. In response to comments, under the final program
there is no longer any need for designation of credits for Tier 2 or Tier 3. The designation occurs
automatically based on the sulfur level. Further, small refiners and small volume refineries will
not be required to "opt-in" to the Tier 3 ABT program.
As discussed above, all credits generated prior to January 1, 2017 will be valid for use for
five years or through December 31, 2019, whichever is earlier. There is a five year statute of
limitations on recordkeeping, and credit life is tied to this limit. Allowing credits to be valid for
longer than five years would violate this statute of limitations. While a previous EPA fuels
program has allowed for a small business flexibility that would allow for credits to be sold to and
used by small refiners only, in no case do these provisions allow for credit life beyond five years.
Further, the January 1, 2020 compliance date for small refiners and small volume refineries is
also the date on which no credits generated relative to 30 ppm may be used for compliance. The
result is that all refiners and importers will be in compliance with the 10 ppm annual average
sulfur standard, whether by producing gasoline with an average sulfur level of 10 ppm or using
credits that were generated against a 10 ppm sulfur standard.
5.5.4 Other
What Commenters Said:
Commenter: Private Citizens
And finally, I was wondering -1 guess the ABT speaks to that - will the refineries get credits of
some sort for alternatives, or will there be some sort of a bonus or subsidy provided so that if an
alternative, say, from an algae or greenfield comes along, it could be dropped in, and there could
be an exchange of some sort.
Our Response:
With regard to the comment regarding credits or bonuses/subsidies for alternatives, this is
not part of the Tier 3 ABT program. The flexibility afforded by the ABT program allows for the
generation of credits relative to gasoline sulfur reduction, and these credits can then be used, or
traded to another refiner, for compliance with the average gasoline sulfur standard. Allowing for
5-201
-------
Chapter 5: Tier 3 Fuel Sulfur Program
credits for other alternatives would not necessarily meet the gasoline sulfur reduction goals of the
Tier 3 program.
5.6. Compliance Provisions (Registration, Reporting, Recordkeeping, Exemptions)
5.6.1 Downstream Added Oxygenate
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM)
API and AFPM do not support EPA's proposal within new §80.1603(d)(3) that anyone adding
oxygenate downstream from a refinery or import facility to assume the sulfur content to be 10.00
ppm. API and AFPM recommend modifying the proposed rule to allow the downstream blender
of DFE into conventional gasoline to either use the actual commercial ethanol sulfur value or
allow laboratory hand blends, as similarly provided for RBOB/RFG in §80.69(a).
Commenter: Marathon Petroleum Company LP (MFC)
MFC does not support EPA's proposal within new §80.1603(d)(3) that anyone adding oxygenate
downstream from a refinery or import facility to assume the sulfur content to be 10.00 ppm. This
requirement would preclude a conventional gasoline producer adding downstream DFE from
using the actual low sulfur level but an artificial (high) value. MFC recommends modifying the
proposed rule to allow the downstream blender of DFE into conventional gasoline to either use
the actual commercial ethanol sulfur value or allow laboratory hand blends, as similarly provided
for RBOB/RFG in §80.69(a).
Commenter: Countrymark, Small Business Refiners (SBRs)
EPA states that oxygenates added downstream of the refinery may be included in the sulfur
calculation. Where oxygenate added downstream from the refinery or import facility, an assumed
10.00 ppm sulfur content must be used.
We believe that this language should be modified to increase blending flexibility and encourage
additional ethanol blending, up to a 10% limit. We recommend that parties blending gasoline and
oxygenates without a Certificate of Analysis (COA) for oxygenates should assume that the sulfur
content is 10.00 ppm. Any oxygenate that is blended with gasoline with a COA demonstrating
that the sulfur content is less than 10.00 ppm may apply that sulfur-volume to the overall blend
calculations.
This additional flexibility in blending will provide gasoline blending options and/or increase the
possible sulfur credits that are available to meet compliance requirements. Credit generation
should be permitted only for the obligated party as defined in the rule. This is important to the
5-202
-------
Tier 3 Summary and Analysis of Comments
CountryMark because we own and operate the refining and terminal assets. This flexibility
provides a tangible relief and real options to u s as we own and operate facilities and are the
obligated party.
Increasing our flexibility in blending low sulfur oxygenate blend components into our fuels
enables us to deliver lower cost fuels to our communities. Including the oxygenate sulfur content
in our final blend calculations will enable u s to reduce our operating costs in the refinery to
offset some of the expected $250,000+ per year increase that we are anticipating as a result of
this new rule.
Our Response:
In demonstrating compliance with the gasoline sulfur standards finalized under the Tier 3
program, gasoline refiners and importers may adjust the sulfur levels in the gasoline and
blendstocks for oxygenate blending (BOBs) that they produce/import to account for the
downstream addition of ethanol. We proposed that the sulfur content of denatured fuel ethanol
(DFE) used for downstream blending would be assumed to be 10 ppm in making such
demonstrations of compliance. We agree with the comments that refiners and importers should
be allowed to use the actual sulfur content of DFE when a sulfur test result is available and when
the refiner can demonstrate that the test result was derived from a representative sample of the
DFE that was blended with the gasoline or BOB. We also agree that the sulfur content of in-use
DFE will typically be lower than the Tier 3 program's 10 ppm sulfur cap for DFE. We assumed
that DFE would have an average sulfur content of 5 ppm in conducting the refinery analysis to
support the Tier 3 final rule. Therefore, the Tier 3 program requires that in determining their
compliance with today's sulfur standards, refiners and importers must either use the actual sulfur
content of the DFE established through testing of the DFE actually blended or assume a 5 ppm
sulfur content for the DFE added downstream. To prevent potential bias, a refiner or importer
must choose to use only one method during each annual compliance period.
5.6.2 Registration, Recordkeeping, and Reporting Requirements
5.6.2.1 Registration
What Commenters Said:
Commenter: Shell Oil Products for Shell and Motiva
There is an inconsistency between the preamble, 78 Fed. Reg. 29942, and the proposed
regulations in 40 CFR 80.1650 pertaining to registration. In the proposed registration regulations
in 40 CFR 80.1650 (a)(l), we recommend that the phrase "unless already registered under 80.76
or 80.103" be added so that new registrations are not required. Similarly, for 40 CFR 80.1650 (a)
(3), we recommend that the phrase "unless already registered under 80.76" be added so that new
registrations are not required. These changes would be consistent with the preamble on page 311
- "Most refiners, importers, and ethanol producers are currently registered with EPA under other
5-203
-------
Chapter 5: Tier 3 Fuel Sulfur Program
40 CFR part 80 fuels programs. We are proposing that those who are already registered would
not have to register again.".
Our Response:
Regarding the comments that suggested an inconsistency between the preamble and
regulations for reporting, we agree with the comments and have changed the regulatory language
accordingly.
5.6.2.2 Reporting
What Commenters Said:
Commenter: Countrymark; Small Business Refiners (SBRs)
Sections 80.1652 and 80.1653 establish reporting and recordkeeping requirements. We believe
that the small refiner's exemption should apply to this section. Small refiners and small volume
refineries should begin reporting and recordkeeping when credit generation begins or by the
compliance deadline of 2020, whichever is earlier. Requiring small refiners to meet this section
of the rule prior to project implementation adds value for neither EPA nor small refiners.
Our Response:
Since the Tier 2 program essentially transitions into the Tier 3 program, all regulated
entities under the Tier 3 program will continue to be subject to the reporting provisions and all
other compliance requirements (e.g., recordkeeping) of the Tier 2 program until they begin
participating in the Tier 3 program. As stated in the regulations at §80.1602, the requirements of
the Tier 3 program (40 CFR part 80 subpart O) apply beginning January 1, 2017 except in the
case of credit generation prior to 2017 and the small refiner/small volume refinery delay; the
requirements of the Tier 2 program (40 CFR part 80 subpart H) continue to apply until that point.
5.6.2.3 Recordkeeping Requirements
What Commenters Said:
Commenter: Consumer Specialty Products Association (CSPA) & Automotive Specialty
Products Alliance (ASPA)
The maximum treatment rate is currently stated on the product transfer document or the
packaging for aftermarket additives. Further, it will not be overly burdensome for aftermarket
additive manufacturers to maintain records of their additive production quality control activities,
demonstrating that the sulfur content of additive production batches complies with the proposed
5-204
-------
Tier 3 Summary and Analysis of Comments
sulfur requirement. Our members are committed to making five years of these reports available
to EPA upon request.
Commenter: Countrymark
Sections 80.1652 and 80.1653 establish reporting and recordkeeping requirements. We believe
that the small refiner's exemption should apply to this section. Small refiners and small volume
refineries should begin reporting and recordkeeping when credit generation begins or by the
compliance deadline of 2020, whichever is earlier. Requiring small refiners to meet this section
of the rule prior to project implementation adds value for neither EPA nor small refiners.
Section 80.1653(g) requires all of the records collected to be made available to EPA upon
request
We strongly oppose collecting and maintaining the names and titles of employees collecting
samples and the names of employees performing laboratory analysis. We believe in protecting
the privacy of our employees and enabling them to perform their job without fear that their
personal information will be released directly to the public or indirectly to the press through a
Freedom of Information request. In May and June 2013, several news outlets covered the release
of farmers' personal information by EPA.
We appreciate EPA's desire for individual accountability at our facilities. If we cannot eliminate
the need to collect employee names and positions, we request that EPA not take custody of this
information upon data inspections. We request that this information remain at our facilities
where we will maintain control of the personal information. Finally, if EPA does require
possession of this data, we request that our employees information be protected under the
Privacy Act and not be released to the public.
While examining recordkeeping requirements in section 80.1653, we notice that the
recordkeeping requirements for denatured ethanol and other oxygenates and the recordkeeping
requirements for fuel producers are significantly different. The recordkeeping requirements
imposed on importers and parties that produce, import, sell, offer for sale, dispense, distribute,
transport gasoline, refine crude oil into finished products are more than double those which
oxygenate producers are required to produce.
We believe that the records requirements for both parties should be equivalent. We recommend
that the recordkeeping burden be reduced as follows for Fuel Producers: eliminate name and title
of person collecting sample and eliminate name of the tester. We further recommend that
oxygenate producers collect the following information, as it is valuable information in
maintaining fuel quality and tractability: Location, Time, Storage tank/truck identification, other
testing results, test methodology used, oxygenate batch number, and ethanol production date.
The table below [of EPA-HQ-OAR-2011-0135-4781-A2] illustrates the differences between the
two parties.
5-205
-------
Chapter 5: Tier 3 Fuel Sulfur Program
Commenter: Small Business Refiners (SBR)
Section 80.1652 and 80.1653 establishes reporting and recordkeeping requirements. We believe
that the exemption for small refiners and small volume refineries should apply to this section.
Small refiners and small volume refineries should begin reporting and recordkeeping when credit
generation begins or by the compliance deadline of 2020, whichever is earlier. Requiring SBRs
to meet this section of the rule prior to project implementation adds value for neither EPA nor
SBRs.
Section 80.1653(g) requires all of the records collected to be made available to EPA upon
request.
We strongly oppose collecting and maintaining the names and titles of employees collecting
samples and the names of employees performing laboratory analysis. We believe in protecting
the privacy of our employees and enabling them to perform their job without fear that their
personal information will be released directly to the public or indirectly to the press through a
Freedom of Information request. In May and June 2013, several news outlets covered the release
of farmers' personal information by EPA.
We appreciate EPA's desire for individual accountability at our facilities. If we cannot eliminate
the need to collect employee names and positions, we request that EPA not take custody of this
information upon data inspections. We request that this information remain at our facilities
where we will maintain control of the personal information. Finally, if EPA does require
possession of this data, we request that our employees information be protected under the
Privacy Act and not be released to the public.
While examining recordkeeping requirements in section 80.1653, we notice that the
recordkeeping requirements for denatured ethanol and other oxygenates and the recordkeeping
requirements for fuel producers are significantly different. The recordkeeping requirements
imposed on importers and parties that produce, import, sell, offer for sale, dispense, distribute,
transport gasoline, refine crude oil into finished products are more than double those which
oxygenate producers are required to produce.
We believe that the records requirements for both parties should be equivalent. We recommend
that the recordkeeping burden be reduced as follows for Fuel Producers: eliminate name and title
of person collecting sample and eliminate name of the tester. We further recommend that
oxygenate producers collect the following information as it is valuable information in
maintaining fuel quality and tractability: Location, Time, Storage tank/truck identification, other
testing results, test methodology used, oxygenate batch number, and ethanol production date.
The table [of EPA-HQ-OAR-2011-0135-4804-A2] below illustrates the differences between the
two parties.
Our Response:
5-206
-------
Tier 3 Summary and Analysis of Comments
Regarding product transfer documents and recordkeeping for aftermarket additives, the
Tier 3 program finalized the requirement that manufacturers of gasoline additives used
downstream of the refinery at less than 1 volume percent must limit the sulfur contribution to the
finished gasoline from the use of their additive to no more than 3 ppm when the additive is used
at the maximum recommended treatment rate. The additive manufacturer will be required to
maintain records of its additive production quality control activities which demonstrate that the
sulfur content of additive production batches is such that when the additive is used at its
maximum recommended treatment rate it will add no more than 3 ppm to sulfur content of the
finished gasoline. The additive manufacturer will also be required to list the maximum
recommended treatment rate on the product transfer document. Records will be required to be
maintained for 5 years by additive manufacturers and made available to EPA upon request. We
agree with the comments that these requirements will not impose a substantial burden on
gasoline additive manufacturers.
As discussed above in Chapter 5.6.2.2, all regulated entities under the Tier 3 program will
continue to be subject to the compliance provisions (such as reporting and recordkeeping) of the
Tier 2 program until they begin participating in the Tier 3 program. As stated in the regulations
at §80.1602, the requirements of the Tier 3 program (40 CFR part 80 subpart O) apply beginning
January 1, 2017 except in the case of credit generation prior to 2017 and the small refiner/small
volume refinery delay; the requirements of the Tier 2 program (40 CFR part 80 subpart H)
continue to apply until that point.
Regarding the comments about concerns about employee privacy in recordkeeping, we
do understand the commenters' concerns. However, in the case of an inspection or enforcement
action, it is important for EPA's enforcement personnel to know who did the actual sampling and
testing in the event that questions arise. We will work to ensure the privacy of individuals whose
names appear in those records, and that they will not be released publicly.
We agree with the comments that enforcement and compliance assurance would be
substantially improved by requiring oxygenate producers/importers and denaturant
producers/importers to maintain records regarding the location, time, storage vessel
identification, other test results, test methodology used, oxygenate batch number, and oxygenate
production date. Hence, the Tier 3 FRM finalizes requirements to this effect.
5.6.2.4 Sampling and Testing Requirements
What Commenters Said:
Commenter: Flint Hills Resources, LP (FHR)
Individual Batch Reporting of Gasoline and Other Fuels Should Not Be Required
FHR believes that reporting of each batch of gasoline under §80.1652 has little value, and EPA
should not consider individual batch reporting for DFE producers. FHR agrees that batch records
should continue to be part of the recordkeeping requirements within §80.1653, but only
5-207
-------
Chapter 5: Tier 3 Fuel Sulfur Program
aggregated reports be submitted to EPA. Similar to how the diesel sulfur program is currently
structured, fuel producers should be required to keep batch records that demonstrate compliance
with standards and support annual reporting. Individual batch records would be made readily
available to EPA upon request or during inspections.
Our Response:
Chapter 5.3 of this Response to Comments document contains comments regarding the
reporting requirements for oxygenate producers. As discussed in Chapter 5.3, we continue to
believe that annual reports from oxygenate producers are an important enforcement and
compliance assurance tool. Therefore, the Tier 3 FRM finalized the requirement that producers
and importers of DFE and other oxygenates must submit annual reports to EPA that include the
total volume of DFE/oxygenate produced and an attestation that all batches met fuel quality
requirements.
It is also important for EPA's enforcement personnel to know who did the actual
sampling and testing in the event that questions arise in the case of an inspection or enforcement
action concerning gasoline, oxygenates, and certified ethanol denaturant. Therefore, the Tier 3
final rule requires that records be maintained by gasoline producers/importers, oxygenate
producers/importers, and certified denaturant producers/importers regarding the name and title of
the person who performs the required sampling and testing.
5.6.3 Sampling and Testing
What Commenters Said:
Commenter: Countrymark; Small Business Refiners (SBRs)
Sections 80.1630 and 80.1631 state that refiners and importers will collect, test, and retain
samples from each batch of gasoline that is produced or imported; using the sample and testing
methodology provided in the rule. We believe that the small refiner's exemption should apply to
this section. Small refiners and small volume refineries should begin collecting, testing, and
retaining samples when credit generation begins or by the compliance deadline of 2020,
whichever is earlier. Requiring small refiners to meet this section of the rule prior to project
implementation adds value for neither EPA nor small refiners and will produce data not
reflecting an operator's post-implementation product.
Our Response:
As discussed in Chapter 5.6.2, above, Tier 2 compliance requirements for small refiners
and small volume refineries, such as sampling and testing, will apply until participation in the
Tier 3 program begins.
5.6.4 National Security Exemptions
5-208
-------
Tier 3 Summary and Analysis of Comments
What Commenters Said:
Commenter: Department of Defense Clean Air Act Services Steering Committee
Our enclosed comment supports EPA's proposed inclusion of a national security exemption for
gasoline used in tactical equipment and other military vehicles and equipment covered by a
National Security Exemption.
The Department of Defense (DoD) supports the inclusion by the Environmental Protection
Agency (EPA) of a national security exemption (NSE) for gasoline used in tactical and other
NSE-covered military equipment. NSEs are necessary for DoD until such time that low sulfur
fuels are available worldwide.
Due to national security considerations, EPA's existing regulations provide either explicit NSEs
to the military or allow the military or manufacturers to request and receive NSEs from
emissions regulations for DoD motor vehicles, engines and equipment if the operational
requirements for such vehicles, engines, or equipment warrant such an exemption.
In order to prevent adverse interaction with after-treatment devices required by EPA regulations
in compliant engines, EPA also established fuel sulfur standards that apply to all fuels used in
regulated vehicles and equipment. The availability of these low or ultra-low sulfur fuels in all
locations and scenarios where DoD operates for national security purposes cannot be guaranteed.
Therefore, EPA regulations for fuels also include NSE language, but only if the fuels are used in
tactical military vehicles, engines, or equipment that are covered by a national security
exemption.
EPA stated in the final 2007 Non-Road, Locomotive and Marine (NRLM) diesel fuel regulation
(69 Fed. Reg. 38958, June 29, 2004), that providing an exemption for military diesel fuel used in
tactical, non-road engines and equipment will not have any significant environmental impact.
This conclusion was based on data provided by DoD for implementing a similar exemption
provision in the highway diesel program. Also, EPA and DoD have developed, and effectively
used for many years, a process to address and obtain NSE for the tactical vehicles, engines, and
equipment covered by fuel exemptions on a case-by-case basis, when required. We believe, as
EPA indicated in final NRLM diesel fuel regulation, that the NSE provides DoD with the needed
flexibility to meet its goals of keeping vehicles, engines, and equipment ready for quick
deployment overseas. We further believe that the exemptions are still needed by DoD to fulfill its
mission on a worldwide basis, to respond adequately and quickly even if emerging economic or
geopolitical situations will not allow the availability of either low or ultra-low sulfur fuels
required for the proper operation of compliant engines.
Though gasoline use is not as common in military tactical vehicles and equipment as diesel, the
same requirements for worldwide deployability exist. Currently, only Europe and a few isolated
countries have regulations in place which meet the proposed 10 part per million (ppm) gasoline
standard. Many parts of the world will not have the new 10 ppm sulfur gasoline available that
will be required for Tier 3 gasoline engines, with a majority of countries producing 100-2,500
5-209
-------
Chapter 5: Tier 3 Fuel Sulfur Program
ppm sulfur gasoline. DoD expects that it may require NSE from the Tier 3 standards for some
gasoline engines and equipment that are used overseas. At the same time, the fuel NSE proposed
in this rule will cover any gasoline purchased overseas and remaining in the fuel tanks of the
equipment when returned to the United States. In addition, DoD has previously and continues to
procure fuel for long term storage in Maritime Prepositioning Ships. The fuel procured for this
purpose may also return to the United States in equipment in which it was used. The sulfur
content of the fuel currently in storage exceeds the proposed limits. Also, in DoD's role in
support of disaster relief activities, often times higher sulfur fuel is procured to ensure a quick
and efficient supply of fuel which meets the demand of these activities.
For example, DoD purchases modified commercial vehicles for our special operations forces
which may not meet EPA Tier 3 emissions standards or may be designed to operate on gasoline
that does not meet the Tier 3 standards. These vehicles are required so that our special operations
forces have the capability to conduct missions where there is no military logistical presence and
where vehicles must be maintained on the local economy.
In the case of our tactical vehicles, even if compliant fuels are available world-wide, there are
cases where in-engine and after-treatment controls are incompatible with military operational
requirements. For example, engine structural improvements allowing for higher pressure
combustion, electronic engine and after-treatment controls, and the after-treatment systems
themselves often do not meet military shock, vibration, electro-magnetic interference, reliability,
and maintainability requirements necessary for operation in a combat environment.
Failure or performance degradation of an engine in combat can have unacceptable dire
consequences such as personnel casualties or military mission failure. Therefore, DoD will
continue to need access to NSE for both engines and fuels included in this rule and other mobile
source regulations for the foreseeable future.
EPA should include the provision that exempts gasoline produced, imported, sold, offered for
sale, supplied, offered for supply, stored, dispensed or transported for use in tactical military
vehicles, engines, or equipment as proposed.
Our Response:
In both our diesel fuel and Tier 2 gasoline sulfur programs, we provided an exemption for
fuel used in tactical military vehicles and nonroad engines and equipment with a national
security exemption (NSE) from the vehicle and engine emissions standards. Due to national
security considerations, some of our existing regulations allow the military to request and receive
NSEs for vehicles, engines, and equipment from emissions regulations if the operational
requirements for such vehicles, engines, or equipment warrant such an exemption. Fuel used in
these applications is also exempt if it is used in tactical military vehicles, engines, or equipment
that are not covered by an NSE but, for national security reasons (such as the need to be ready
for immediate deployment overseas), need to be fueled on the same fuel as those with an NSE.
This exemption is being continued in the Tier 3 gasoline program as well.
5-210
-------
Tier 3 Summary and Analysis of Comments
5.7. Other, including Alternatives
5.7.1. Butane Standard
What Commenters Said:
Commenter: Mid-Continental Energy (MCE)
EPA is proposing to reduce the sulfur content to lOppm in butane blended into gasoline by
terminal blenders. MCE supports this change with the request that butane blenders be allowed to
use averaging on a monthly basis to meet the lOppm. There are several streams of butane that
have an average sulfur content of 15ppm to 20ppm. For example, a butane blender should be
allowed to purchase some streams with 15ppm during the month if the average butane blended
during the month calculates at lOppm. This extra flexibility would be a benefit to potential
butane blenders whose supply of butane with lOppm sulfur could be limited.
Commenter: Independent Fuel Terminal Operators Association (IFTOA)
In the Preamble, EPA described the current butane blending rules and proposed to lower the
sulfur standards that are now applicable to butane to a 10 ppm cap - consistent with the Tier 3
proposed standard for the annual average gasoline. Members of the Association are concerned
that this reduced sulfur level would inhibit butane blending due to the reduced availability of
butane with very low sulfur. The regulated community needs more flexibility because it is
usually difficult and costly to find butane with a sulfur content of 10 ppm or below. Therefore,
instead, EPA should consider focusing on the sulfur content of the final blend - not the butane
component. Blenders downstream of the refiner/import gate should be permitted to add butane
that meets the downstream per-gallon cap so long as the butane blending operation does not
cause the blender's finished gasoline to have an annual average sulfur content above 10 ppm.
Recommendation: The Association recommends that EPA continue to provide flexibility to
butane blending and consider a requirement that butane only need meet the downstream per-
gallon sulfur cap so long as the butane blending operation does not cause the annual average
sulfur content of the gasoline to exceed the proposed Tier 3 standard of 10 ppm. Moreover, the
Association supports broadening the butane blending provisions to include pentane.
Provide greater flexibility for butane blending and expand butane blending provisions to include
pentane.
Our Response:
As discussed in Section V.I. of the preamble to today's final rule, under the Tier 2
gasoline program, "purity" butane blended into gasoline downstream of the refinery is subject to
a 30 ppm sulfur cap and other specifications regarding its composition.15 This is consistent with
the 30 ppm refinery average sulfur standard under the Tier 2 program. The Tier 3 FRM finalized
' 40 CFR 80.82.
5-211
-------
Chapter 5: Tier 3 Fuel Sulfur Program
the proposed 10 ppm sulfur cap for purity butane blended into gasoline effective January 1,
2017. This is consistent with the Tier 310 ppm refinery average sulfur specification.
Butane and has an inherently low sulfur content that can be made to meet a 10 ppm sulfur
cap with relatively mild desulfurization techniques. We anticipate that butane suppliers will
desulfurize these blendstocks to well below 10 ppm sulfur as part of their response to the Tier 3
gasoline sulfur requirements. There is no need to allow higher sulfur levels for butane blended
voluntarily downstream that would result in higher gasoline sulfur levels. Furthermore, allowing
butane used for RVP trimming to exceed a 10 ppm sulfur cap would needlessly complicate
compliance assurance.
5.7.2. CNG/LPG In-Use Fuel Standards
What Commenters Said:
Commenter: Advanced Biofuels USA (ABFUSA)
Consideration of CNG and LPG Emissions Test Fuel: In the past sulfur content regulation of
"natural gas" sources fuels was not seen as necessary for two reasons, 1) the low volume of use,
and 2) the historically low sulfur content of US "conventional" natural gas sources. However,
with "unconventional" (fracked) natural gas sources now providing over 60% of the US NG
supply (US DOE/EIA 2012 data) sulfur content becomes an issue because of the higher content
in the unconventional gas. Therefore, EPA should enact standards for sulfur content of NG and
LNG transport fuels that are equal (in terms of pollutant/mile) of these proposed Tier 3
regulations.
Commenter: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
...but emphasize that EPA should ensure that all market fuels, including LPG and CNG meet the
same "total" sulfur standards as Tier 3 gasoline.
If market CNG or LPG do not meet the proposed Tier 3 emissions certification fuel standards,
then there must be relief in the emissions standards to accommodate potential deterioration in
vehicle emissions systems capability resulting from the use of higher sulfur fuels. The
Coordinating Research Council Performance Committee is in the process of conducting a 10
U.S. city fuel survey on retail CNG. Two samples will be collected from each city and results are
expected during the 4th quarter of 2013. CNG and LPG fuels are not subject to the same degree
of regional refinery differences as conventional liquid fuels. However, there can be large
differences in marketplace gaseous fuel quality as a result of: (l)the amount of processing they
receive as a result of their source, i.e., well-head gas versus refinery gas; (2)the type of
dehydration and filtering they receive just prior to or directly at retail dispensers.
5-212
-------
Tier 3 Summary and Analysis of Comments
U.S. CNG and LPG industries are at the front end of what appears to be the start of a long term
growth trend, so it is an appropriate time to establish market and test fuel standards, including for
sulfur.
Commenter: American Gas Association (AGA)
The American Gas Association appreciates that as natural gas increasingly moves into the
mainstream as a motor vehicle fuel, the industry will need to work closely with the EPA to
ensure a smooth transition. Natural gas is an emerging motor vehicle fuel that we believe offers
tremendous societal benefits, including aiding in the transition to hydrogen fuel cell vehicles. If
the agency elects to develop sulfur standards for CNG, we urge the agency to work with the
natural gas industry to adopt a phased-in approach. Such an approach would allow the industry
time to assess the current state of natural gas fuel quality with respect to sulfur, and to adopt any
remediation practices that may prove to be required. Conversely, we fear an overly aggressive
timeline toward regulation would stymie the adoption of an alternative fuel that offers the
promise of significant benefits to our nation.
As noted by EPA in the proposed rule, there are currently no sulfur standards for the fuel used in
compressed natural gas (CNG). EPA has requested comment on whether it is necessary for EPA
to establish sulfur standards for CNG, and whether a 15 ppm sulfur cap similar to that
established for highway Diesel fuel would be appropriate. EPA further requested comment on
whether and how to address the sulfur contribution from odorants and other additives used in
CNG.
The American Gas Association believes that, to the extent possible, standards for motor vehicle
fuel should be fuel neutral. AGA appreciates the importance of regulating sulfur content in
motor vehicle fuels, due to the potential damage that high levels of sulfur content in motor
vehicle fuels can impose on the exhaust after-treatment systems on-board vehicles. Natural gas
is an inherently cleaner burning fuel than gasoline or diesel, and dedicated natural gas vehicles
do not require the same exhaust after-treatment systems required for gasoline and diesel vehicles
to meet EPA standards for emissions of criteria pollutants. Natural gas sulfur content, when used
for dedicated-use NGVs, is unlikely to be a significant issue.
However, the American Gas Association believes that dual-fuel vehicles - capable of running on
either gasoline or CNG, or potentially diesel and CNG - will be an important market component
as natural gas vehicles gain acceptance as a mainstream motor fuel. AGA recognizes that the
sulfur content of natural gas could impact the efficacy of exhaust after-treatment systems
designed to alternate between CNG and petroleum-based fuels. In other words, the sulfur
content of CNG, when used for dual-fuel vehicles, could have negative impacts on the emissions
of these vehicles while operating in a conventional fuel mode.
NGVs provide a number of social benefits, including reductions in some cases in emissions of
nitrogen oxides and paniculate matter, as compared to gasoline or diesel vehicles. The greater
use of natural gas as a transportation fuel also offers additional societal benefits, including
strengthening our national energy security, and improving our balance of trade, by displacing
imports of foreign petroleum and insulating our economy from the comparative volatility of oil
5-213
-------
Chapter 5: Tier 3 Fuel Sulfur Program
prices vis-a-vis natural gas prices. Projections by the Energy Information Administration (EIA)
consistently indicate that domestic natural gas prices are likely to remain low and stable, while
petroleum prices are likely to climb and exhibit greater volatility in the coming years and
decades.
In the light-duty greenhouse gas/fuel economy standards finalized by EPA and NHTSA in
August 2012, the agencies acknowledged the importance of encouraging NGV adoption into the
marketplace because of the key role NGVs will play as a technology bridge to hydrogen-
powered fuel cell electric vehicles (FCEVs).
The connection between NGVs and FCEVs is two-fold. First, technological developments
related to on-board storage and fueling delivery, of compressed gases are likely to translate to
some degree between both applications. Secondly, establishing a backbone of NGV
infrastructure across the nation could lessen barriers to entry for FCEVs, since on-site reforming
of hydrogen could provide fuel for FCEVs at stations initially established to serve an NGV
market.
Natural gas is an emerging market competitor to gasoline and diesel fuel. Automobile
manufacturers and the petroleum industry have spent decades studying and adjusting the
composition of petroleum-based fuels, balancing engine and exhaust system requirements with
refining specifications.
To date, end uses of natural gas have not had the same level of sensitivity to sulfur that is evident
in vehicle applications. Because of this, little data is readily available concerning the sulfur
content of CNG at the point of use. The American Gas Association is cognizant of the
implications of sulfur content in CNG, and is currently working with the Coordinating Research
Council (CRC) to support a national survey of natural gas fuel quality from the perspective of a
motor fuel. This work was initiated in the fall of 2012, and is expected to yield results in the fall
of 2013. Sulfur is among a number of the constituent components of CNG that is being tested as
part of this work.
We would welcome the opportunity to share the results of this initial survey of CNG fuel quality
at the point of use for motor fuel - the CNG fueling station - when the results become available.
While a valuable first step in obtaining real-world data, we anticipate that additional work will
likely be needed beyond the current CRC effort to provide a more complete national picture of
CNG and sulfur content. AGA believes that a joint effort with EPA on data collection could be a
valuable extension of this effort.
While data on sulfur content of CNG at the point of use is lacking, some information on sulfur
content in the natural gas distribution and transmission system is available. Operators of
interstate pipelines impose limitations on sulfur content for natural gas injected into their
systems, as part of their commitment to ensuring pipeline quality. In the presence of moisture,
sulfur can form highly acidic compounds that corrode the interior of the pipe. Termed "tariff
restrictions," these industry standards for sulfur content vary nationally from 0.5 grains per 100
standard cubic feet (grains/SCF) to approximately 20 grains/SCF. These standards correspond to
levels between 8 ppm and about 300 ppm sulfur content.
5-214
-------
Tier 3 Summary and Analysis of Comments
It is extremely important to note that the 300 ppm standard is a maximum intended for unusual
cases when natural gas producers do not have access to processing prior to injection into the
pipeline. In real-world practice, natural gas with these higher levels of sulfur content are heavily
diluted by the lower sulfur content gas that is in the norm in interstate pipelines.
Since the EPA request for comment was published, AGA has made initial inquiries to its
members seeking data on sulfur content in their systems obtained at the city gate. As noted
previously, traditional end-uses of natural gas in the residential and commercial sectors have not
exhibited high sensitive to sulfur content, and so further data is not readily available. The AGA
effort with the CRC is aimed at reducing uncertainty about CNG sulfur content by acquiring
actual data at the consumer's point of use.
Additionally, it is critical to note that AGA member companies, as natural gas local distribution
companies (LDCs), place the highest priority on the safety of their customers. For this reason, a
sulfur-based odorant called mercaptin is introduced into the natural gas system by LDCs when
they take possession of the natural gas at the city gate. This odorant is a critical safety measure
that allows consumers to detect the presence of any natural gas leakage. The amount of sulfur
introduced into the natural gas by mercaptin is approximately 5 ppm. We ask that, in
considering any future sulfur standard for natural gas used as a motor fuel, that EPA weigh the
importance of mercaptin as a safety measure.
Commenter: Countrymark; Small Business Refiners (SBR)
We do not believe that sulfur standards need to be established for CNG or LNG. According to
DOE statistics, natural gas is less than 5 ppm of all sulfur compounds. This is sulfur associated
with natural gas production and an odorant that is required for leak detection. Additional
standards would be unnecessary testing and documentation required by parties involved in the
CNG and LNG business. This unnecessary cost of compliance will be passed on to end users,
increasing energy costs with no benefits attained.
Adding a sulfur standard for CNG and LNG increases the switching cost for consumers and may
decrease corporation's willingness to participate in the market. Additional testing and
documentation without tangible benefits to suppliers, customers, or environmental benefits
without positive results does not deliver value for any party involved in the process.
Commenter: Private Citizen
We must be careful not to encourage CNG vehicles because the process used to extract natural
gas in deep shale causes climate instability through methane releases, contaminates water, air,
and food sources, and destroys communities.
Commenter: Private Citizen
Please do not make the mistake of supporting a switch from oil to fracked gas. The extraction
process for fracked gas pollutes air, water, and cropland, destroys wildlife habitat, depletes water
5-215
-------
Chapter 5: Tier 3 Fuel Sulfur Program
supplies, and buries millions of gallons of chemically compromised water thousand of feet
underground where it can migrate and foul streams, lakes and rivers.
As if this weren't enough, thousands of trees are clearcut to make roads for the trucks that
transport fracked waste water,, wildlife is impacted, communities are disrupted and the health of
millions is in jeopardy.
Support renewables. A nation cannot support itself without its natural resources upon which
everything depends.
Commenter: Manufacturers of Emission Controls Association (MECA)
MECA also supports defining fuel sulfur average limits or caps on any alternative transportation
fuels (e.g., blends of alcohols with gasoline, natural gas) that are consistent with the proposed 10
ppm national average for gasoline or the existing 15 ppm national sulfur cap on diesel fuel.
Commenter: National Propane Gas Association (NPGA)
With respect to fuel specifications, EPA is proposing to reduce the maximum sulfur content in
gasoline from the current 30 parts per million (ppm) to 10 ppm on an annual average basis
beginning January 1, 2017. EPA is also proposing a sulfur maximum on a per-gallon basis with
caps set at 80 ppm at the refinery gate and 95 ppm at any downstream locations. The Agency is
also seeking input on changes to emissions test fuel specifications.
Although EPA's Tier 3 proposal is described as being fuel-neutral, there are elements of the
proposed rule that could impact propane autogas, and as such, NPGA submits the following
comments on the Tier 3 proposal.
As previously stated, EPA is seeking to lower the maximum sulfur content for gasoline, on an
annual average basis, from the current 30 ppm to 10 ppm. In doing so, EPA states that this will
"enable vehicles designed to the proposed Tier 3 tailpipe exhaust standards to meet these
standards in-use for the duration of their useful life and facilitate immediate emission reductions
from all the vehicles on the road at the time the sulfur controls are implemented."
The proposed changes to fuel specifications specifically apply to gasoline. However, EPA also
requests input on whether it is necessary to establish sulfur standards for alternative fuels such as
liquefied petroleum gas (LPG), i.e. propane autogas, and compressed natural gas (CNG) given
that no such standards exist for these vehicles. In addition, EPA asks whether a 15 ppm sulfur
cap similar to that established for highway diesel fuel would be appropriate. EPA also solicits
input on the contribution of sulfur from odorants used in LPG.
With respect to sulfur standards for alternative fuels, NPGA understands that EPA is simply
seeking input on this subject and not actually specifying a proposed sulfur maximum at this time.
We further recognize that a specific proposal to this effect would have to be addressed under a
separate rulemaking activity to allow for proper review by the public, which it cannot perform
5-216
-------
Tier 3 Summary and Analysis of Comments
within the context of this NPRM since there is no maximum number proposed upon which to
evaluate.
Nevertheless, the establishment of a sulfur maximum for LPG, i.e. propane autogas, presents a
unique challenge that does not exist for gasoline or diesel fuel. Unlike gasoline or diesel fuel,
propane autogas is odorless. Therefore, for safety reasons, federal hazardous materials
regulations require that it be odorized to a level that indicates the presence of gas down to a
concentration in air of not over 1/5 the lower flammability limit of LPG.
Further to this, NFPA 58, LP-Gas Code, requires that the fuel be odorized prior to delivery to a
bulk plant (4), where the fuel is stored for subsequent distribution for vehicle fuel purposes or for
other applications such as space or water heating. To meet these odorization requirements, it has
been a long-standing industry practice to inject an odorant named ethyl mercaptan to the fuel.
Ethyl mercaptan is a sulfur-based odorant and is, without question, the most commonly used and
available odorant in the United States. It is injected at a recommended rate of 1 pound per 10,000
gallons of LPG. For conservatism, it is not uncommon to see a slightly higher injection rate at
1.5 pounds per 10,000 gallons. This equates to approximately 15-20 ppm of sulfur that would be
present in the fuel after it leaves the refinery or gas processing plant and prior to arrival at a retail
marketer's bulk plant.
NPGA does not believe that EPA should place maximum sulfur content requirements on propane
autogas at this point in time for several reasons. First, consider that the establishment of sulfur
maximums for gasoline and diesel require actions on the part of refiners or gas processors that
would remove sulfur from the fuel. Likewise, with propane autogas, similar actions would be
required for the removal of sulfur that may exist in the form of constituents such as carbonyl
sulfide or hydrogen sulfide.
However, where things differ between the fuels is the required addition of a substance that
contains sulfur as a constituent, namely ethyl mercaptan, which is an important and necessary
safety measure. In doing so, this adds a layer of complexity to the issue of sulfur removal that
would need to be fully vetted outside the construct of this proposed rulemaking to ascertain, if
possible, where the proper balance exists between appropriate levels of sulfur removal and the
need to maintain safety for a fuel that is used in applications beyond just vehicles.
Second, EPA would have to account for a variety of economic impacts including, among other
things, the cost to produce low-sulfur propane autogas. In addition, EPA would need to take into
account the cost impact on new or potentially separate fuel storage needs and capabilities for
retail marketers of propane autogas to avoid intermixing the low-sulfur fuel with the fuel for
non-vehicle applications.
Lastly, EPA solicits input on whether a 15 ppm sulfur cap similar to that established for highway
diesel fuel would be appropriate. NPGA does not believe it would be appropriate to follow a
similar approach to diesel. The current 15 ppm diesel requirement applies to all downstream
locations, including distributors and retailers, which could be very problematic for retail autogas
marketers.
5-217
-------
Chapter 5: Tier 3 Fuel Sulfur Program
In addition to all of the aforementioned factors that EPA must take into account when
considering this subject, NPGA has a practical concern related to the establishment of sulfur
maximums. In particular, by establishing a cap for gasoline, it begs the question as to what
should be an appropriate maximum level for alternative fuels such as propane autogas and CNG,
and it suggests there could be unintended consequences without a sulfur maximum. However, it
would be impossible to realistically answer these questions given the extreme time constraints
for responding to this complex NPRM imposed by EPA as it would require broad, collaborative
input from all interested stakeholders such as fuel suppliers, fuel providers and those who
develop the fuel specifications to determine the answers.
4 - National Fire Protection Association (NFPA) 58, LP-Gas Code, 2011 Edition, Section 4.
Commenter: New York State Department of Environmental Conservation
Gaseous fueled vehicles need to be held to the same standards as liquid fuels.
EPA has requested comment regarding the sulfur content of natural gas and propane when used
to fuel light duty vehicles. This issue is complicated by the fact that sulfur containing chemicals
(mercaptans) are added to these gases to facilitate leak detection.
We believe that the competing demands of maintaining catalyst efficiency and ensuring adequate
odorant levels can be met. We do not support relaxing the Tier 3 emission standards for gaseous
fueled vehicles as an accommodation to mercaptan sulfur levels in these fuels.
Commenter: NGVAmerica
The Tier 3 rulemaking proposes to establish new, lower limits on the sulfur levels in gasoline
and diesel fuel. The expected advantages of lower sulfur are the ability to provide not only lower
emissions in new vehicles but also continued lower emissions over the life of such vehicles. The
rulemaking also requests comment on whether EPA should impose similar sulfur requirements
on natural gas when used as a motor vehicle fuel.
The proposal requests comment on whether natural gas should have to meet an in-use standard as
well as a test fuel standard. EPA regulations currently include test fuel standards for natural gas
but do not include any controls on sulfur levels.
EPA's notice specifically asks whether it should impose a 15 ppm sulfur limit for natural gas.
While we can understand the desire to impose similar performance standards on all motor
vehicle fuels, we believe that it is not appropriate to impose any limits on natural gas for the
following reasons:
1. The rulemaking record does not demonstrate that current natural gas sulfur levels need to be
controlled in order for NGVs to meet the proposed emission standards;
2. The rulemaking record has not identified how sulfur levels would be controlled while
maintaining appropriate odorant levels;
5-218
-------
Tier 3 Summary and Analysis of Comments
3. The rulemaking record has not identified who - producers, pipelines, gas utilities, fueling
station operators - would be responsible for controlling sulfur limits in natural gas; and
4. The rulemaking record has not identified appropriate technologies for lowering sulfur in
natural gas or the cost of these technologies and, therefore, no cost-benefit analysis has been
conducted.
Based on the foregoing and as explained here, it would be premature to establish test-fuel
specifications or in-use specifications for natural gas sulfur levels.
In preparing these comments, NGVAmerica consulted with industry experts in order to
determine whether the industry currently has data on current sulfur levels. The most
comprehensive assessment of natural gas fuel quality of which we are aware was conducted in
1992 by the Gas Research Institute (now Gas Technology Institute) (1). According to GTI
officials, total sulfur was typically in the range of 0.34-0.39 grains/100 scf or about 9-13 ppm,
mass basis. Maximum levels of 0.6-0.7 grains (18-22 ppm) were noted, but this was relatively
rare. Transmission pipeline (non-odorized) gas was typically around 0.12 grains/100 scf (about 4
ppm). Based on these figures it appears that the levels contemplated by EPA would not present a
problem for natural gas providers. Conversations with a major natural gas producer and also with
one of the nation's largest natural gas utilities indicates that the natural gas they produce and
deliver to customers also would be in line with the levels proposed by EPA.
We believe that these factors suggest that, rather than supporting the setting of a new national
standard for natural gas, there likely is no need for such a standard. Moreover, we are extremely
hesitant to support a 15 ppm level because the GRI data is now more than 20 years old, and we
do not have enough comprehensive, recent data to support a 15 ppm level. Further, it would be
unwise to support such a level without having an understanding of how sulfur would be
controlled if, in fact, it needs to be controlled, and by whom would it be controlled.
Unlike the regulatory record compiled with respect to the petroleum industry and sulfur controls,
there has not been an engagement with the natural gas industry - producers, pipelines, fuel
providers, equipment suppliers, etc. - with respect to the technology that would be used to
control sulfur levels. And unlike the example of the petroleum industry, it is not at all certain that
the best means or the most efficient means of controlling sulfur limits in natural gas would be to
control levels upstream as is the case with petroleum where sulfur is controlled at the refinery.
Given that most natural gas is currently consumed for non-transportation purposes (electric
generation, heating, industrial processes), it seems unlikely that EPA would or could legally
impose upstream sulfur controls on the natural gas industry without a strong showing that such
levels are necessary for other regulatory purposes. Therefore, it would appear that, if controls are
necessary, they would be imposed at the fueling station. This presents more questions about what
technologies would be used and how this would impact odorant levels since the odorant itself
contains sulfur. The odorant (mercaptan) is added as a safety feature so that leaks of natural gas
will be recognized. If the sulfur is pulled out presumably some odorant will have to be put back
in at added costs and difficulty at the fueling station. All of these issues must be fully explored
before a national standard can be set.
5-219
-------
Chapter 5: Tier 3 Fuel Sulfur Program
The questions raised by this rulemaking and any proposed sulfur limits on natural gas must be
fully explored and considered before moving forward. Therefore, we would urge the EPA to
forgo establishing new controls on natural gas sulfur levels. EPA should instead work with the
natural gas industry to assess whether there is a need for a national sulfur standard, i.e. whether
sulfur levels in natural gas present an emission issue for certifying new natural gas vehicles.
These discussions should also begin to assess what types of technologies are available to control
sulfur levels if in fact sulfur levels need to be controlled.
1 - Gas Research Institute, American Gas Association Laboratories, GF Steinmetz, Institute for Gas
Technology, Variability of Natural Gas Composition in Select Major Metropolitan Areas of the United
States 1990 - 1992 (March 1992).
Commenter: Pennsylvania Department of Environmental Protection (DEP)
The EPA should decide whether to require the same sulfur content in natural gas in order to
attain the same emissions standards in natural gas-powered vehicles as those proposed for
gasoline-powered vehicles.
As specified in the Preamble for the proposed Tier 3 rule, there are currently no sulfur
specifications for the test fuel used for certifying natural gas vehicles nor for the test fuel used
for certifying liquefied petroleum gas (LPG) vehicles (78 FR 29,914). In addition to seeking
comments on the appropriateness of amending 4OCFR section 86.113 to reference 40 CFR Part
1065 for both natural gas and LPG test fuels, EPA also requested comments on amending the
specifications to better reflect in-use fuel characteristics. Additionally, EPA requested comment
on the appropriateness of aligning the sulfur specifications with those that apply for gasoline test
fuel.
Sulfur levels in natural gas are typically lower than sulfur levels in gasoline by two orders of
magnitude. Sulfur levels in gasoline average 0.034 percent currently and will be reduced to 0.012
percent. Sulfur levels in natural gas average 0.0005 percent sulfur from the wellhead. Natural gas
providers add odorants to the natural gas, however, that can have high levels of sulfur. Natural
gas-powered vehicles will have the same control equipment installed as is installed on gasoline-
powered vehicles. EPA should determine if it is appropriate to establish a standard for sulfur in
natural gas that will be used in natural gas-powered vehicles, including sulfur in the odorants that
would meet the same vehicle standards that Tier 3 gasoline-powered vehicles would be required
to meet under EPA's proposed rule.
Commenter: Robert Bosch GmbH, Gasoline Systems, Germany
The best approach would be a standardization on a basis of the actual CEN standardization for
CNG fuel in Europe.
Commenter: Private Citizen
The solution to the pollution problem is right under your noses! Natural gas! All internal
combustion and turbine engines run quite well on it. It's clean, non-polluting, and cheap! Why
won't you acknowledge it? Has big oil tied your hands? With the finding of huge pockets of it
5-220
-------
Tier 3 Summary and Analysis of Comments
popping up everyday, one would think of it was a God send. What's holding up start? City and
state, and federal government would use it and benefit hugely. Saving monies, cleaning the
atmosphere, etc.. Don't send it overseas to benefit countries that are not our friends. Benefit us!
Equipment using it lasts longer, goes longer between oil changes, requires less tune ups, burns
cleaner. It doesn't require a rocket scientist to figure this out. Jump on the band wagon, do
something right for a change. Now, everything has been built to run on up to 10% ethanol. I hear
whispers about you people wanting to raise that limit to 15% ethanol. Won't work! As of today,
some small engine manufacturers are telling their users to use mid-grade gasoline to correct
starting, running, and performance problems. Think of the impact on people if they have to
purchase new equipment to do lawn work. What of the people who are forced to purchase new
cars?
Commenter: VNG.CO
VNG.CO (VNG) submits these Comments in the above-referenced docket. Based in
Pennsylvania, VNG is developing a nationwide, retail compressed natural gas (CNG) fueling
network specifically for light duty vehicles. VNG will be collocating its CNG fueling facilities at
existing gasoline stations in order to provide a convenient and familiar fueling experience. While
initially serving corporate fleets, VNG's fueling facilities will also support mass market adoption
of natural gas vehicles (NGVs) and can be adapted to dispense advanced gaseous fuels like
renewable natural gas (RNG) and hydrogen. VNG offers its recommendations below because the
EPA's consideration of motor vehicle emissions and fuel standards may impact the development
of the NGV market in ways the EPA may not have considered.
It is widely recognized that the shale gas revolution has transformed the U.S. energy landscape
with its promise of abundant, low-cost natural gas supplies for decades to come. This has led to a
growing interest in natural gas vehicles (NGVs), which offer major environmental, economic,
and security benefits compared to petroleum-based gasoline and diesel fuels. While the NGV
industry has historically focused on the heavy-duty segment, this sea change in US energy
production has also led to increasing attention on the potential of light-duty NGVs by
automakers, compressed natural gas (CNG) refueling service providers like VNG, and
policymakers - particularly at the federal level:
• President Obama has repeatedly hailed the importance of both light-duty and heavy-duty
natural gas vehicles, including in his 2012 Blueprint for an America Built to Lastl and his 2013
Blueprint for a Clean and Secure Energy Future. (2)
• The National Petroleum Council of the U.S. Department of Energy and the National
Academy of Sciences have both recently released comprehensive studies of alternative fuels
which have found NGVs to have the greatest potential for mass-market adoption and reductions
of oil use of any light-duty vehicle technology over of the next two decades. (3,4)
• EPA acknowledged the potential of light-duty NGVs to accelerate the commercialization
of zero-emission hydrogen fuel cell electric vehicles (FCEVs) in the 2017-2025 light-duty GHG
rules, awarding these vehicles special multiplier incentives to stimulate their development. (5)
However, despite this growing recognition of the potential of light-duty NGVs to reduce air
pollution, reduce petroleum imports and reduce consumers' fuel costs, the market for these
vehicles is still at a nascent stage. While refueling service providers such as VNG are making
5-221
-------
Chapter 5: Tier 3 Fuel Sulfur Program
progress in expanding natural gas fuel availability, there are still just over 1,200 CNG fueling
stations nationwide - fully half of which are not accessible to the public6 - compared to about
100,000 gasoline stations.? Light duty NGV product offerings are limited to the Honda Civic
Natural Gas, dual-fuel NGV pickups from GM and Chrysler, and a variety of offerings from
conversion companies (notably Westport, IMPCO, and others) - all of which are currently being
produced and sold in the U.S. at very low volumes and high incremental costs compared to
similar models powered by gasoline or diesel fuel.
Given the tenuous stage of the development of this market, it is critical that new regulations such
as the Tier 3 regulations do not place unnecessary burdens on the industry. Rather, wherever
possible the Tier 3 regulations should encourage the development of the market for light duty
NGVs. This can be assured in two ways:
• Avoiding the implementation of sulfur content regulations for CNG. There is no evidence
that such rules are needed, and the implementation of such regulations for natural gas
transportation fuel raises unique issues compared to regulations applicable to gasoline.
By avoiding unnecessary regulatory burdens on CNG fuel as well as reducing existing
and future inequalities in the vehicle certification process, EPA could play an important role in
accelerating the development of the NGV industry by reducing incremental costs to consumers,
with no reduction in the overall efficacy of the Tier 3 program. Achieving this would be
consistent with EPA's mandates under the Clean Air Act as well as Executive Order 13563:
Improving Regulation and Regulatory Review - which, among other things, calls for regulations
that "promote economic growth, innovation, competitiveness, and job creation," through use of
"the least burdensome tools for achieving regulatory ends" as well as "flexible approaches." (8)
In the Notice of Proposed Rulemaking (NPRM), EPA requests comment on whether it is
necessary to "establish sulfur standards for CNG and LPG fuels, and whether a 15 ppm sulfur
cap similar to that established for highway diesel fuel would be appropriate," potentially on the
grounds of EPA's intent to make the regulations "fuel neutral."9 However, because of the
inherent differences between natural gas and gasoline, it would be inappropriate for EPA to
regulate the sulfur content in CNG.
There is no evidence that the sulfur content of CNG fuel is sufficient to impact the catalyst
efficiency of NGVs;
There is no data on the average sulfur content of natural gas, and gas composition will in any
case vary significantly between different regions;
There is no evidence or information about the means or costs for regulating the sulfur content
of natural gas motor vehicle fuel; and
Unlike gasoline, which is only used as vehicle fuel, the natural gas compressed for CNG is
used for a wide range of purposes (power plants, home heating, etc.) which do not and will not
have sulfur regulated.
Thus, imposing sulfur regulations on CNG vehicle fuel has not been shown to be necessary,
would have uncertain costs, and risks discouraging the use of natural gas as a vehicle fuel. If
sulfur regulations were imposed on upstream natural gas producers or utilities, these entities (and
their many customers) would object to the additional cost of processing 100 percent of their
natural gas supplies to comply with a regulations that would currently apply to the less than 0.2%
5-222
-------
Tier 3 Summary and Analysis of Comments
of total US gas demand used for transportation.10 Alternately, if the burden of compliance were
to fall on downstream CNG retailers, it would impose unknown additional costs at every CNG
dispensing site that would have to be recovered from relatively low volumes of fuel, making it
even more difficult for natural gas to replace petroleum-based motor vehicle fuels, contrary to
EPA'sgoals.
Regardless, much more data on the levels of sulfur currently present in natural gas, the impacts
of this sulfur on NGV emissions, and the means and costs for reducing this sulfur content should
be obtained before any potentially costly decisions to regulate the sulfur content of natural gas
motor vehicle fuel are made.
1 Slack, Megan. "Everything You Need to Know: President Obama's Blueprint for American-Made
Energy." White House Blog. 26 Jan 2012. http://www.whitehouse.gov/blog/2012/01/26/everything-you-
need-know-president-obamas-blueprint-american-made-energy?twp=twt
2 White House. "FACT SHEET: President Obama's Blueprint for a Clean and Secure Energy Future." 5
Mar 2013. http://www.whitehouse.gov/the-press-office/2013/03/15/fact-sheet-president-obama-s-
blueprint-clean-and-secure-energy-future
3 National Petroleum Council. "Future Transportation Fuels: Natural Gas Analysis." 28 May 2013.
http://www.npc.org/FTF-report-080112/Chapterl4-NaturalGas-052813.pdf
4 National Academy of Sciences. "Transitions to Alternative Vehicles and Fuels." March 2013.
http: //www .nap .edu/catalog .php ?record
5 Environmental Protection Agency. "2017 and Later Model Year Light-Duty Vehicle Greenhouse Gas
Emissions and Corporate Average Fuel Economy Standards; Final Rule." Federal Register, 15 Oct 2012,
p. 62815.http://www.gpo.gov/fdsys/pkg/FR-2012-10-15/pdf/2012-21972.pdf
6 Alternative Fuels Data Center. "Alternative Fueling Station Locator." Department of Energy, Energy
Efficiency and Renewable Energy Program. Accessed 28 June 2013.
http: //www .afdc. energy .gov/locator/stations/
7 U.S. Census Bureau. "Economic Census, Industry Snapshot: Gasoline Stations with Convenience
Stores." 2007. http://www.census.gov/econ/census/pdf/44711.pdf
8 The White House. "Executive Order 13563 - Improving Regulation and Regulatory Review." 18 Jan
2011.
http://www.whitehouse.gov/the-press-office/2011/01/18/improving-regulation-and-regulatory-review-
executive-order
9 Environmental Protection Agency. "Control of Air Pollution From Motor Vehicles: Tier 3 Motor
Vehicle Emission and Fuel Standards; Proposed Rule." Federal Register, 21 May 2013, p. 29826.
10 Energy Information Administration. "Annual Energy Outlook
2013." http ://www.eia.gov/forecasts/aeo/
Our Response:
As discussed in Section VJ. of the preamble to the final rule, EPA is deferring finalizing
in-use sulfur requirements for CNG/LPG in this final rule to provide additional time to work
with stakeholders to collect data on current CNG/LPG sulfur content, to determine whether
additional control of in-use CNG/LPG sulfur content is needed, and to evaluate the feasibility
and costs associated with potential additional sulfur controls. Given that the information in the
public comments suggests already low sulfur levels in CNG/LPG, the Tier 3 program vehicle
emissions standards will apply to CNG/LPG vehicles in addition to vehicles fueled on gasoline,
diesel fuel, or any other fuel.
5-223
-------
Chapter 5: Tier 3 Fuel Sulfur Program
For those comments that suggested that EPA should either promote the use of natural gas
or constrain its use due to concerns over fracking, such comments are beyond the scope of the
Tier 3 rulemaking. We sought comment only on the sulfur standards for CNG and LPG if they
are used in motor vehicles.
5.7.3. Other
What Commenters Said:
Commenter: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
Market fuel gasoline RVP for the summer season should be capped at 9 psi RVP, including
within that the statutory one pound waiver for El 0.
Our Response:
The regulation of gasoline RVP is beyond the scope of the Tier 3 final rule.
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM)
Combined with other regulations, such as the RFS2, this rule would impact domestic fuel
supplies, result in increased consumer costs, and affect energy security.
Our Response:
Given the estimated costs and impacts of the Tier 3 program, as discussed in both the
preamble to the final rule and in the RIA, we do not expect this rule to negatively effect U.S.
energy security - or the supply or distribution of gasoline or other fuels.
What Commenters Said:
Commenter: Private Citizen
As an intermediate transition from the continued dependence on poisonous polluting carbon-
based fuels, I support the new Vehicle and Gasoline Standards. However, I also support rapid
replacement of such fuels by one that is totally clean (no pollution) and requires no new
technological development: hydrogen, derived from water by electrolysis, will make us totally
energy-independent. Air, land, and oceanic vehicles can burn liquid hydrogen. The mass-
produced cost of transforming existing motor vehicles' engines to run on hydrogen instead of
5-224
-------
Tier 3 Summary and Analysis of Comments
gasoline will be one-tenth the cost of replacing them all with electric vehicles, and no mining or
drilling is required as is the case with needed battery-replacement raw materials. The electricity
needed to produce hydrogen needs to come from natural solar, wind, and tidal-current sources. A
solar-hydrogen economy will result in 99% less pollution! Study descriptions by International
Society of Hydrogen Engineers scientist Harry Braun at http://PhoenixProjectFoundation.US ...
also consider the arguments for Lunar Solar Power (eight times more efficient, without
occupying any human habitat areas, and allowing direct energy transfer to all urban centers on
Earth), by scientist David Criswell - a noble plan using proved technology to bring cheap
electricity to every person on Earth... http://lunarsolarpowersystem.blogspot.com (developing
now, a new website, http://lunarsolarpower.org will promote LSP).
Our Response:
The Tier 3 program vehicle emissions standards are fuel neutral (i.e., they apply
regardless of the type of fuel that vehicle uses, including hydrogen). The promotion of
alternative energy sources (including solar energy) is beyond the scope of the Tier 3 final rule.
What Commenters Said:
Commenter: Private Citizen
Let's transform our cars so they can run on alcohol fuel. Please make it legal for retailers to buy
wholesale from local alcohol fuel plants E-100 Ethanol. Make it legal for local communities to
operate local alcohol fuel plants in zones recognized as industrial for sustainable fuel and energy
rather than in zones recognized as heavy industrial. Please make it the law that automobile
manufactures must install dedicated ethanol engines in a high percentage of all new cars. If all of
the above is done, then local economies will be revived; our air, land, and water would become
clean again; and perpetual war in Afghanistan and the Mid East Region will end. We must
become a nation dedicated to sustainable fuel. Support ethanol fuel and support the Tier 3 clean
vehicle and gasoline standards.
Our Response:
The Tier 3 program vehicle emissions standards are fuel neutral (i.e., apply regardless of
the type of fuel that vehicle uses, including El 00.) The consideration of a potential requirement
to mandate the production of dedicated El 00 engines by automobile manufacturers is beyond the
scope of the Tier 3 final rule.
What Commenters Said:
Commenter: Private Citizens
The proposed fuel and emissions law changes are good, but the increases in automobile prices
and gasoline are not. These so called higher prices for cars and the fuel used, is not the answer.
5-225
-------
Chapter 5: Tier 3 Fuel Sulfur Program
The use of more alcohol in motor fuel is. Cleaner burning engines of today can use 10-15
percent gasoline to 80-85percent alcohol, that is the combination that can reduce emissions. Fuel
additives can come from all types of renewable sources, and reduce the oil imports plus lower
fuel cost. The keystone pipeline should be built, and have a second source of refining, the east
coast of NY and NJ—there are huge refineries there and can handle the oil to be processed.
Our Response:
The Tier 3 program vehicle emissions standards are fuel neutral (i.e. apply regardless of
the type of fuel that vehicle uses, including higher level ethanol blends.) The construction of the
Keystone pipeline is beyond the scope of the Tier 3 final rule.
What Commenters Said:
Commenter: Mercedes-Benz USA, LLC on behalf of Daimler AG
To provide Mercedes-Benz and other OEMs a degree of certainty and incentive that costs
associated with the design, development, certification and production of vehicles dedicated to
operation on Tier 3 Fuel are wise investments in the future of the US and global environments,
Mercedes-Benz requests the following:
EPA grant a reduction in the Mercedes-Benz (or OEM) total annual corporate CO2 compliance
burden, based on the following equation per vehicle model certification family and a CO2
Burden Reduction Multiplier:
CO2 Burden Reduction =
CO2 output on Tier 3 certification fuel/equivalent CO2 output on 91 AKI Tier 2 EO certification
fuel x projected annual sales percentage of corporate gasoline-powered vehicles optimized for
operation on Tier 3 fuel by MY / total corporate gasoline-powered vehicle sales projection by
MY x annual OEM corporate CO2 projected output
As an incentive for early adoption of Tier 3 Fuel capability, the CO2 Burden Reduction
Multiplier would be applied according to this schedule:
MY 2016 and prior: 100% of CO2 Burden Reduction
MY 2017-2019: 50% of CO2 Burden Reduction
MY 2020-2022: 25% of CO2 Burden Reduction
MY 2023 and later: no additional CO2 Burden Reduction
In the event of limited Tier 3 fuel availability nationwide, but where the necessary hardware to
support Tier 3 fuel usage has already been incorporated in vehicle models MY2016-2022 as well
as prior, Mercedes-Benz requests the following:
EPA grant a reduction in the Mercedes-Benz (or OEM) total annual corporate CO2 compliance
burden, based on the CO2 Burden Reduction equation per vehicle model line:
5-226
-------
Tier 3 Summary and Analysis of Comments
MY 2016 and prior: 50% of the CO2 Burden Reduction
MY 2017-2019: 25% of CO2 Burden Reduction
MY 2020-2022: 15% of CO2 Burden Reduction
MY 2023 and later: no additional CO2 Burden Reduction
These corporate CO2 burden reductions would end MY 2023 or 90% retail station penetration of
a high-octane mid-blend ethanol retail fuel, whichever comes first.
In order to provide incentive for fuel producers to achieve an expedient and substantial
marketplace penetration of a Tier 3 fuel similar to the proposed certification fuel, it is further
proposed that a Federal Fuels Tax discount or equivalent RIN credit be applied to each gallon of
Tier 3 fuel produced for US consumption. This discount or credit would be equal to:
Producer Credit = 1.00 - (net heating value of the Tier 3 certification fuel divided by the NHV of
91 AKI Tier 2 EO certification fuel).
This Producer Credit would also expire MY2023 or after 90% retail penetration of a Tier 3 is
achieved, whichever comes first.
13 - Resources for the Future, Reassessing the Oil Security Premium (February 2010), available at
http://www.rff.org/RFF/Documents/RFF-DP-10-05.pdf.
14 - Renewable Fuels Association, Battling for the Barrel: 2013 Ethanol Industry Outlook, p. 1, supra.
15 - See Stein et al., An Overview of the Effects of Ethanol-Gasoline Blends on SI Engine Performance,
Fuel Efficiency, and Emissions, supra.
16 - MathPro Inc., Analysis of the Refining Costs and Associated Economic Effects of Producing 92 AKI
Gasoline in the U.S. Refining Sector (October 30, 2012).
17 - Stillwater Associates, The Cost of Introducing an Intermediate Blend Ethanol Fuel for 2017- and-
Later Vehicles, (October 17, 2012).
Our Response:
Consideration of greenhouse gas requirements for vehicle manufactures and potential
federal fuel tax credits are beyond the scope of the Tier 3 rulemaking.
What Commenters Said:
Commenter: Private Citizens
I am disappointed in the fact that neither the EPA nor UCS can't come up with a better method
of cleaning up automobile exhausts. It is time to stop using fossil fuels to produce our energy and
switch to a hydrogen economy in which hydrogen is used in combination with fuel cells. The
only exhaust is pure drinkable water.
As a weak secondary alternative, I would support the Tier 3 standards for tailpipe emissions and
low-sulfur fuel, and urge the EPA to finalize these standards by the end of 2013.
5-227
-------
Chapter 5: Tier 3 Fuel Sulfur Program
Every major automobile manufacturer is designing new fuel cell automobiles. Why can't the
EPA and federal government learn more about what's going on in many countries where the
number of fuel cell autos and buses are on the increase. It's time to get with it!
While we've made progress cleaning up our cars and trucks in recent years, they remain a
leading source of air pollution. More than one in three Americans still live in areas where air
pollution levels exceed at least one federal health standard. Hydrogen fuel cell autos would
correct this.
The Tier 3 standards would only reduce smog-forming pollutants and soot, which pose a
particular danger to those suffering from asthma and other lung ailments. A hydrogen economy
would not only REDUCE the pollution but would STOP it.
Meeting the Tier 3 standards is achievable and necessary for our health. That's why public health
and environmental justice advocates, automakers, and state air quality regulators, environmental
and science-based advocacy groups all support Tier 3. The only one standing in the way is the oil
industry and their allies in Congress, but we can't wait for their green light to clean up our air.
More important, we can't wait for the EPA and UCS to stop supporting the fossil fuel industry.
Our Response:
The Tier 3 program vehicle emissions standards are fuel neutral (i.e., apply regardless of
the type of fuel that vehicle uses, including hydrogen.) The promotion of the use of hydrogen as
a motor fuel is beyond the scope of the Tier 3 final rule.
What Commenters Said:
Commenter: Monroe Energy, LLC
Monroe operates a petroleum refinery in Trainer, Pennsylvania. As a merchant refiner, Monroe
can only comply with its RFS2 obligations by purchasing renewable identification numbers
(RINs). The recent and ongoing spike in the price of RINs will require Monroe to spend, on an
annual basis, a very significant amount of money to comply with RFS2. Requiring Monroe to
spend such additional capital investments to comply with the Tier 3 gasoline sulfur standards,
and its nominal environmental benefit, will compound Monroe's (and other merchant refiners')
economic injury. Therefore, EPA should not promulgate a Tier 3 sulfur standard until it has
corrected the errors in the RFS2 rule.
The costs of compliance should not be evaluated in a vacuum; rather, they need to be balanced
against other federal and state regulations, including the renewable fuel standard (RFS2).
Our Response:
Issues pertaining to a refiner's annual RFS2 obligations are being addressed in separate
EPA actions and are beyond the scope of the Tier 3 rulemaking. We do not believe there is
compelling reason to link the completion of the annual RFS2 rulemakings on the RFS volume
5-228
-------
Tier 3 Summary and Analysis of Comments
obligations to the promulgation of the Tier 3 final rule. Issues associated with the costs to
refiners in complying with the Tier 3 sulfur requirements are addressed in Chapter 7.2 and the
cost effectiveness of the Tier 3 sulfur requirements is addressed in Chapter 7.3 of this Summary
and Analysis of Comments document.
5-229
-------
Tier 3 Summary and Analysis of Comments
6 Regulatory Streamlining and Technical Amendments
What We Proposed:
EPA proposed a range of technical amendments and regulatory streamlining actions as
part of the Regulatory Review initiative. Some of these may have some bearing on
implementation of the Tier 3 vehicle and fuel standards, while others deal with other aspects of
EPA's existing fuel and vehicle regulations. The comments in this chapter correspond to Section
VI of the preamble to the proposed rule. A summary of the comments received and our response
to those comments are located below.
6.1. Fuels Program
6.1.1. Regulatory Streamlining and Tech Amendments
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM)
We appreciate EPA effort to make regulatory streamlining a priority. We support the elimination
of unnecessary and outdated provisions in order to improve administrative efficiency and reduce
regulatory compliance burdens. These proposed streamlining provisions are independent of Tier
3 and should be finalized earlier than the Tier 3 final rule. We agree with the Agency that these
are straightforward and should be implemented quickly.
EPA has made regulatory streamlining a priority and we appreciate the Agency's efforts. We
agree that regulatory streamlining will result in more efficient and less costly compliance. We
support the elimination of unnecessary and outdated provisions. These provisions are
independent of Tier 3 and should be promulgated in a final rule earlier than the Tier 3 final rule.
We agree with the Agency that these are straightforward and should be implemented quickly.
Commenter: Irving Oil Terminals Inc.
EPA has proposed a number of regulatory changes to the Gasoline Program such as (1) reducing
testing and reporting requirements of certain RFG parameters during the winter months; (2)
updating of test methods for various fuel parameters; (3) designation of an alternative,
independent laboratory; (4) elimination of unnecessary diesel fuel pump labeling; (5) extension
of Performance Based Measurement System to other fuel parameters and test methods; (6)
provision for de minimis changes in batch volume reports, and (7) attest engagements. Irving Oil
supports these proposals. They would streamline registration, recordkeeping and reporting,
reduce costs and free up personnel to focus on more significant areas of compliance.
Commenter: Marathon Petroleum Company LP (MFC)
6-1
-------
Chapter 6: Regulatory Streamlining and Technical Amendments
MFC also supports the fuel streamlining steps that EPA has identified and is willing to work with
EPA to identify further streamlining actions that can be taken. EPA has made regulatory
streamlining a priority and we appreciate the Agency's efforts. We agree that regulatory
streamlining will result in more efficient and less costly compliance. We support the elimination
of unnecessary and outdated provisions. These provisions are independent of Tier 3 and should
be promulgated in a final rule earlier than the Tier 3 final rule. We agree with the Agency that
these are straightforward and should be implemented quickly.
Commenter: PBF Energy Inc.
PBF supports the regulatory streamlining efforts included in the proposed rule.
Commenter: Phillips 66 Company
We are appreciative of the effort to streamline various portions of existing regulations. With
changes over time, there are several areas that need "clean-up" and this effort will reduce
confusion and burden on the regulatory parties. We offer the following comments on the
proposed revisions as well as suggestions for other provisions that we feel would add value and
should be considered.
Our Response:
EPA appreciates the industry support for the current effort to streamline applicable
regulatory provisions in a timely and environmentally sound manner.
6.1.1.1. Complex Model Testing & Reporting
EPA proposed several amendments related to the reduction of testing and reporting of
complex model gasoline parameters.
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM)
We support the initiative to streamline testing and reporting for EPA's Complex Model,
including the eliminating API gravity and oxygenates. We also recommend that the requirement
for aromatics, distillation, and olefins be eliminated completely for winter Reformulated and
Conventional gasoline batches. API and AFPM recommend additional regulatory changes to
eliminate testing that provides no value, is redundant or otherwise unnecessary. In addition, we
recommend reporting frequency and deadlines be changed to meet the needs of EPA and the
regulated parties, and EPA update regulatory references to the most up to date standards.
6-2
-------
Tier 3 Summary and Analysis of Comments
Testing and reporting for Complex Model fuel parameters: EPA proposes to reduce the testing
and reporting burden for individual batches of RFG and conventional gasoline (CG). Many of
the tests for individual parameters are no longer needed because of the phase out of complex
model standards. The complex model standard for NOx was replaced by the Tier 2 gasoline
sulfur standard. The toxics complex model standard was replaced by MSAT2 (43). The only
remaining complex model standard for most refineries is summer RFG VOCs.
We support the Agency's proposal to eliminate API gravity. This is proposed to be effective on
January 1, 2013 (see proposed 80.65(e)(3)). Given that the proposal was published in the Federal
Register on May 21, 2013, this proposed effective date may be confusing because of its
retroactive nature. Refiners recommend that the effective date remain January 1, 2013 in the final
rule.
We support EPA's proposal to eliminate testing and reporting for oxygenates (unless necessary
because oxygenates added downstream are used in calculations). The effective date of this
elimination should be no later than the effective date of the Tier 3 final rule.
Testing for RFG, CG, RBOB and CBOB batches: EPA proposes to allow monthly composites
for winter RFG for aromatics, olefms, and distillation which is already allowed for Conventional
batches. This is an example of a regulatory burden that is not needed to support out-of-date
complex model standards. We propose that the requirement for aromatics, distillation, and
olefms be eliminated completely for winter RFG and all Conventional batches. The effective
date of this elimination should be no later than the effective date of the Tier 3 final rule.
EPA states that the "values for aromatics, distillations and olefms may continue to be determined
from monthly composites" for CG. We interpret "may" to mean a refiner has the discretion to
test and report or not. We see no value to test and report aromatics, distillations and olefms for
monthly composites of CG or CBOB for parties that are subject to MSAT2, not the anti-dumping
toxics complex model. A clear elimination would result in a reduction in paperwork and
reporting burden. The effective date of this elimination should be no later than the effective date
of the Tier 3 final rule.
We support the continuation of testing and reporting of all summer complex model parameters
for summer RFG/RBOB batches because of the summer RFG VOC standard.
Commenter: Independent Fuel Terminal Operators Association (IFTOA)
EPA is proposing to reduce the testing and reporting requirements for certain fuel parameters
associated with the complex model. The Association supports the Agency's efforts to eliminate
regulatory obligations that are no longer needed. Such streamlining would reduce compliance
costs and allow companies to focus more efficiently on more important aspects of compliance.
IFTOA recommends that EPA adopt the proposed amendments to streamline testing and
reporting obligations.
6-3
-------
Chapter 6: Regulatory Streamlining and Technical Amendments
Commenter: Sutherland Asbill & Brennan LLP (Sutherland)
Testing and Reporting of Fuel Parameters. EPA proposes to streamline the requirements for
testing and reporting. Sutherland encourages all attempts at streamlining and simplifying the
reporting requirements. Additionally, EPA should clarify whether it would like daily batch level
production reports.
Commenter: Marathon Petroleum Company LP (MFC)
EPA proposes to reduce the testing and reporting burden for individual batches of RFG and
conventional gasoline (CG). Many of the tests for individual parameters are no longer needed
because of the phase out of complex model standards. The complex model standard for NOX was
replaced by the Tier 2 gasoline sulfur standard. The toxics complex model standard was replaced
by MSAT2.30 The only remaining complex model standard for most refineries is summer RFG
VOCs.
We support the Agency's proposal to eliminate API gravity. This is proposed to be effective on
January 1, 2013 (see proposed 80.65(e)(3)). Given that the proposal was published in the Federal
Register on May 21, 2013, this proposed effective date may be confusing because of its
retroactive nature. Refiners recommend that the effective date remain January 1, 2013 in the final
rule as it will be providing relief from a requirement. EPA should specify in the Preamble of the
final rule that this will be retroactively applied.
We support EPA's proposal to eliminate testing and reporting for oxygenates (unless necessary
because oxygenates added downstream are used in calculations). The effective date of this
elimination should be no later than the effective date of the Tier 3 final rule.
Testing for RFG, CG, RBOB and CBOB batches: EPA proposes to allow monthly composites
for winter RFG for aromatics, olefms, and distillation which is already allowed for Conventional
batches. This is an example of a regulatory burden that is not needed to support out-of-date
complex model standards. We propose that the requirement for aromatics, distillation, and
olefms be eliminated completely for winter RFG and all Conventional batches. The effective
date of this elimination should be no later than the effective date of the Tier 3 final rule.
EPA states that the "values for aromatics, distillations and olefms may continue to be determined
from monthly composites" for CG. We interpret "may" to mean a refiner has the discretion to
test and report or not. We see no value to test and report aromatics, distillations and olefms for
monthly composites of CG or CBOB for parties that are subject to MSAT2, not the anti-dumping
toxics complex model. A clear elimination would result in a reduction in paperwork and
reporting burden. The effective date of this elimination should be no later than the effective date
of the Tier 3 final rule.
We support the continuation of testing and reporting of all summer complex model parameters
for summer RFG/RBOB batches because of the summer RFG VOC standard.
6-4
-------
Tier 3 Summary and Analysis of Comments
Commenter: Phillips 66 Company
- Sampling and testing of fuel properties - We agree with the proposed elimination to test and
report API gravity. Although EPA has proposed other modifications to potentially reduce the
sampling and testing burden, we think the proposal does not go far enough. Currently, the
complex model is only used to certify VOC controlled reformulated gasoline. Therefore, the full
slate of properties that must be analyzed and input into the model for calculation of VOC
emission reductions should only be required for reformulated gasoline or RBOB designated as
VOC controlled. There should be no requirement to test conventional gasoline or non-VOC
controlled reformulated gasoline or RBOB for olefins, aromatics, and distillation. This change
would significantly reduce the laboratory testing burden and should be implemented effective
upon promulgation of the final rule.
Commenter: Shell Oil Products for Shell and Motiva
The preamble, under the Reporting Requirements section, 78 Fed. Reg. 29942, proposes
denatured fuel ethanol batch reports to include properties and volume effective January 1, 2017.
It is suggested that 40 CFR 80.1451 be amended to include the requirements of this additional
reporting and 40CFR 80.1464 be amended to include attestation engagement procedures for this
additional reporting. In this way, there is consistency with both ethanol and gasoline batch
reports on how they are reported and attested.
Our Response:
We are streamlining and reducing the reformulated gasoline (RFG) and conventional
gasoline (CG) testing and reporting burden of gasoline refiners and importers by reducing the
testing and reporting requirements of certain fuel parameters associated with the complex model.
With the phasing out of complex model standards1, reduced testing and reporting is appropriate,
particularly for RFG. In cases where a refiner is subject to only benzene, RVP, and sulfur
standards, certain parameters no longer need to be tested and reported on an every-batch basis.
However, refiners producing RFG during the summer volatile organic compound (VOC) control
season will still need to use the complex model to determine VOC performance, and thus must
still measure and report the relevant complex model fuel parameters. Sulfur and benzene will
continue to be tested and reported on an every batch basis as these values are necessary for the
Tier 2 and Tier 3 gasoline sulfur and MSAT2 benzene programs. In addition, small refiners that
are subject to the delayed compliance option for the 0.62 volume percent benzene standard will
have to use the complex model (and thus measure all complex model parameters) until 2015 for
CGMSAT1 compliance.2
Currently, there are 17 complex model parameters on the RFG/anti-dumping batch
report. We are reducing testing and reporting requirements for some of these parameters for
RFG, CG or both in this final rule. All other reporting requirements not changed by the final rule
1 Per §§80.41(e) and (f), and §80.101(c), applicable NOx and toxics emissions requirements are superseded by the
Tier 2 gasoline sulfur standards and MSAT2 benzene standards, respectively.
2 61 FR 17230 (March 29, 2001).
6-5
-------
Chapter 6: Regulatory Streamlining and Technical Amendments
are still in effect. For both RFG and CG, we are eliminating testing and reporting of American
Petroleum Institute (API) gravity. Regarding the effective date for this change, we have applied
it retroactively in keeping with the proposed date. In addition, we are finalizing reporting
requirements for refiners that either have gasoline containing oxygenates or that include
oxygenates added downstream in compliance calculations.
For winter RFG, we are eliminating the requirement to test and report aromatics,
distillations and olefins on an every batch basis and instead are allowing testing and reporting of
monthly composites. Commenters from the refining industry strongly suggested that we
eliminate testing of these parameters altogether, since they are not needed, and their elimination
would further reduce the burden on regulated parties. While we agree that a reduction in burden
would occur if refiners were not required to test, even on a monthly composite basis, for these
parameters, the many interconnected aspects of the RFG program make any seemingly
innocuous change potentially fraught with unintended consequences. Thus, we are will evaluate
completely eliminating the testing of these parameters in any future fuel program restructuring.
In the proposal, we stated ".. .values for aromatics, olefins, and distillation terms may
continue to be determined from monthly composites." Some commenters viewed the term
"may" in that sentence as meaning that reporting for those parameters is optional. That is an
incorrect interpretation. In this final rule, we are finalizing requirements for use of either
monthly composites or batch testing. The parameters must still be measured and reported.
Commenters in the refining industry do not think these parameters need to be measured at all for
parties subject to the MSAT2 benzene standard, as compliance with that standard is not
dependent on aromatic, olefm, or distillation values. As mentioned earlier, there are several areas
of the RFG and/or anti-dumping programs where testing and reporting burden could likely be
reduced; however, we have not fully evaluated the implications of changing the current
requirements, and thus we are leaving consideration of such changes to a future broader program
restructuring.
6.1.1.2. De Minimis Threshold for Batch Volumes
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM)
We agree with EPA's concept of a de minimis value, below which a party would not be required
to correct and resubmit their batch reports. However, the proposed level is of little practical
value.
We suggest that a de minimis threshold value of 0.5 percent of any batch is a practical level that
will provide the intended relief for regulated parties but will still sufficiently protect the integrity
of EPA reporting and compliance programs.
We agree with the concept of including a de minimis value for reporting of batch volumes,
below which a party would not be required to correct and resubmit their batch reports. However,
6-6
-------
Tier 3 Summary and Analysis of Comments
the proposed de minimis level of the lesser of 500 gallons or 1 percent of the true batch volume
is so small that it is of little practical value. A de minimis volume of 500 gallons is equivalent to
approximately 12 barrels and will almost always be less that 1 percent of the true batch volume.
For example, on a typical pipeline batch volume of 25,000 barrels, the 500 gallon de minimis
volume represents approximately 0.05 percent of the total volume. On a large refinery
production batch of 250,000 barrels, the de minimis volume of 500 gallons represents
approximately 0.005 percent of the total volume. Such a small threshold value would fail to
provide the intended relief and would not prevent a party from having to make inconsequential
volume corrections.
We suggest that a de minimis threshold value of 0.5 percent be applied regardless of batch size.
Individual batch volumes that fall within plus or minus 0.5 percent (+/- 0.005 expressed as a
decimal) of the true volume would not need to be corrected. The 0.5 percent de minimis value is
a practical level that will provide the intended relief for regulated parties but will still sufficiently
protect the integrity of EPA reporting and compliance programs.
Regarding the impact of de minimis batch volume corrections on compliance with the benzene,
sulfur, RFG, RFS and other Clean Air Act fuel standards, we recommend that EPA should delete
the regulatory text at 80.10 (c) and 80.10(d). The application of a de minimis threshold implies
that the small volume errors in batch reporting are truly inconsequential and do not have an
impact on compliance with fuel standards. Therefore, no separate demonstration of material
impact should be required. The normal, unintentional variation in batch volumes will be
distributed both greater than and less than the true volume. These variations will cancel each
other over time and do not represent any degradation of the standard.
Commenter: Chevron Products Company; Marathon Petroleum Company LP (MFC)
We agree with the concept of including a de minimis value for reporting of batch volumes,
below which a party would not be required to correct and resubmit their batch reports. However,
the proposed de minimis level of the lesser of 500 gallons or 1 percent of the true batch volume
is so small that it is of little practical value. A de minimis volume of 500 gallons is equivalent to
approximately 12 barrels and will almost always be less that 1 percent of the true batch volume.
For example, on a typical pipeline batch volume of 25,000 barrels, the 500 gallon de minimis
volume represents approximately 0.05 percent of the total volume. On a large refinery
production batch of 250,000 barrels, the de minimis volume of 500 gallons represents
approximately 0.005 percent of the total volume. Such a small threshold value would fail to
provide the intended relief and would not prevent a party from having to make inconsequential
volume corrections.
We suggest that a de minimis threshold value of 0.5 percent be applied regardless of batch size.
Individual batch volumes that fall within plus or minus 0.5 percent (+/- 0.005 expressed as a
decimal) of the true volume would not need to be corrected. The 0.5 percent de minimis value is
a practical level that will provide the intended relief for regulated parties but will still sufficiently
protect the integrity of the EPA reporting and compliance programs.
6-7
-------
Chapter 6: Regulatory Streamlining and Technical Amendments
Regarding the impact of de minimis batch volume corrections on compliance with the benzene,
sulfur, RFG, RFS and other Clean Air Act fuel standards, we recommend that EPA should delete
the regulatory text at 80.10(c) and 80.10(d). The application of a de minimis threshold implies
that the small volume errors in batch reporting are truly inconsequential and do not have an
impact on compliance with fuel standards. Therefore, no separate demonstration of material
impact should be required. The normal, unintentional variation in batch volumes will be
distributed both greater than and less than the true volume. These variations will cancel each
other over time and do not represent any degradation of the standard.
Commenter: National Biodiesel Board (NBB)
NBB Opposes the Proposed Application of a "De Minimis" Correction to Obligated Parties
Renewable Volume Obligations Under the Renewable Fuel Standard Program.
In the Tier 3 Proposal, EPA proposes new § 80.10, which would provide that obligated parties
would not be required to "correct unintentional errors in reporting batch volume on previously
submitted batch reports." 78 Fed. Reg. at 30,003. With respect to the RFS2 program, this
provision would apply "only to the volume of fuels produced or exported that result in a
renewable volume obligation under subpart M." Id. The Tier 3 Proposal's entire explanation for
this provision is one paragraph and is as follows: We are proposing for the correction of batch
volume reports to allow for de minimis changes in reporting compliance that would not require a
complete resubmission of compliance reports when a minor discrepancy of a few barrels is
uncovered. This allowance for the correction of batch volume reports would apply to reporting
for: RFG, anti-dumping, gasoline and diesel sulfur, MSAT2 and the RFS renewable volume
obligation (RVO) (for RFS, this would only apply to the volume of fuels produced or exported
that result in a RVO for obligated parties). We are proposing a new section 80.10 to define de
minimis, for the purpose of this allowance, as no more than 500 gallons or by no more than 1
percent of the true batch volume in gallons, whichever value is less. We request comment on
whether or not a different amount would be more appropriate.
Id. at 29,952. This one paragraph buried in a 377-page proposal is wholly insufficient for the
public to meaningfully comment. NBB opposes the application of any so-called "de minimis"
corrections to reporting renewable volume obligations.
In the RFS2 program, Congress "directed" EPA "to ensure that transportation fuel sold or
introduced into commerce in the United States . . ., on an average annual basis, contains at least
the applicable volume of renewable fuel, advanced biofuel, cellulosic biofuel, biomass-based
diesel" under the statute. NPRA v. EPA, 630 F.3d 145, 147 (D.C. Cir. 2010), reh 'g denied, 643
F.3d 958 (D.C. Cir. 2011), cert, denied, 132 S. Ct. 571 (2011); see also 42 U.S.C. §
7545(o)(2)(A)(i), (iii), (o)(3)(B). Thus, EPA's obligation is to enforce the statutory volumes. The
means of enforcing the statutory volumes that EPA has established is through the annual
renewable volume obligations for each obligated party.
Unlike the other programs where the "de minimis" reporting correction may apply,1 the RFS2
program is a volume mandate and compliance is contingent on the actual volume of gasoline or
diesel fuel produced or imported. While the Tier 3 Proposal does not provide adequate
6-8
-------
Tier 3 Summary and Analysis of Comments
information as to how the correction would apply for the RFS2 program, on its face it would
appear as though an obligated party can under-report its gasoline production and, thereby, reduce
its renewable volume obligations. While EPA claims the volume is de minimis (500 gallons or
no more than 1 percent of the true batch volume, whichever is less), EPA provides no
explanation as to whether that means, for example, each shipment overseas or each batch of fuel
produced. Even if the correction applies to the one time volume reported at the end of the year,
each obligated party can do so, potentially resulting in significantly less RINs actually being
required. This is in clear violation of EPA's obligation to "ensure" the minimum volumes are
met. While an agency may have certain inherent authority to provide for de minimis exemptions
in certain statutory schemes, such authority is not available where there is "an 'extraordinarily
rigid' statutory mandate," as is the case here. See Sierra Club v. EPA, 705 F.3d 458, 468 (D.C.
Cir. 2013) (citation omitted); see also Kentucky Waterways Alliance v. Johnson, 540 F.3d 466,
491 (6th Cir. 2008) (recognizing such authority "is not an ability to depart from the statute . . .")
(quoting Ala. Power Co. v. Costle, 636 F.3d 323, 360 (D.C. Cir. 1979)).
This de minimis "correction" would also appear to be yet another means to reduce the statutory
volumes in a manner that is beyond EPA's authority. The statute provides EPA with limited
waiver authority, imposing stringent procedural and substantive criteria on when EPA can reduce
the statutory mandates. See 42 U.S.C. § 7545(o)(7). In addition to the clear requirements that
EPA ensure "at least" the applicable volumes of renewable fuel are sold, this limited authority
evidences that Congress sought to restrict EPA's authority to reduce the statutory volumes, and a
"de minimis" error in reporting of gasoline or diesel fuel production or imports is not among the
reasons EPA can use its authority. The Tier 3 Proposal provides no explanation for EPA's
authority to allow for such correction. Such an omission is likely because there is no such
authority for EPA to do so. See Sierra Club v. Tennessee Valley Auth., 430 F.3d 1337, 1346-49
(11th Cir. 2005) (finding Alabama's 2% de minimis rule impermissibly modified requirements in
State Implementation Plan without undergoing proper procedures).
Further, the general rationale for providing de minimis exemptions is administrative burdens. See
Sierra Club, 705 F.3d at 462 (citation omitted). While such administrative burdens may justify
providing renewable fuel producers with a de minimis exemption for certain under-or over-
reporting of RIN generation, which does not appear to even be contemplated by EPA in the
proposal, there is no such administrative burden with respect to obligated parties annual
reporting of their renewable volume obligation. Nonetheless, EPA's proposal is limited only to
obligated parties with no discussion whatsoever as to the administrative burdens to support such
an exemption. Thus, EPA has not (nor can it) meet its burden to show why this de minimis
"correction" is necessary here. Kentucky Waterways Alliance, 540 F.3d at 491 ("Determination
of when matters are truly de minimis naturally will turn on the assessment of particular
circumstances, and the agency will bear the burden of making the required showing.") (quoting
Greenbaum v. EPA, 370 F.3d 527, 534 (6th Cir. 2004)).
In fact, EPA cannot identify any administrative burden or need for such a "correction"
allowance. Unlike the numerous calculations necessary for the RIN generation and transactions
that a producer must undertake on a per batch basis, obligated parties and exporters are only
required to submit an annual compliance report. Such report is required two months after the end
of the year, and EPA does not explain why an obligated party or exporter cannot obtain accurate
6-9
-------
Chapter 6: Regulatory Streamlining and Technical Amendments
r\
numbers for the volume of gasoline or diesel fuel it produced or imported within that time.
Indeed, in this same proposal, EPA is proposing to extend the time for obligated parties to submit
their compliance reports by another full month. EPA's rationale for this extension is to give
obligated parties more time to address volume discrepancies identified through "volumetric
auditing." 78 Fed. Reg. at 29,952.
Because EPA failed to provide any legal or factual support for its proposal, it also should not be
allowed to manufacture such support in any final rule. Even if responding to comments, EPA
cannot circumvent the notice and comment requirements of the Clean Air Act and
Administrative Procedure Act. The Clean Air Act requires EPA to provide notice of its proposed
rule through a statement of its basis and purpose, and to give the public a meaningful opportunity
to comment. 42 U.S.C. § 7607(d)(3), (4), (5); see also 5 U.S.C. § 553(b), incorporated by
reference in 42 U.S.C. § 7607(d)(3) (requiring "reference to the legal authority under which the
rule is proposed" and notice of "either the terms or substance of the proposed rule or a
description of the subjects and issues involved"). "The significance of rulemaking cannot be
underemphasized. It gives parties affected by a decision an opportunity to participate in the
decision-making process and forces EPA to articulate the bases for its decisions." Dormer Hanna
Coke Corp. v. Costle, 464 F. Supp. 1295, 1305 (W.D.N.Y. 1979) (citation omitted); see also
Envtl. Integrity Project v. EPA, 425 F.3d 992, 996 (D.C. Cir. 2005) ("[N]otice requirements are
designed (1) to ensure that agency regulations are tested via exposure to diverse public comment,
(2) to ensure fairness to affected parties, and (3) to give affected parties an opportunity to
develop evidence in the record to support their objections to the rule and thereby enhance the
quality of judicial review.") (citation omitted); Weyerhaeuser Co. v. Costle, 590 F.2d 1011, 1028
(D.C. Cir. 1978) (Court will defer to agency "so long as we are assured that its promulgation
process as a whole and in each of its major aspects provides a degree of public awareness,
understanding, and participation commensurate with the complexity and intrusiveness of the
resulting regulations.") (citations omitted). Given EPA's obligations under the RFS2 program
and the apparent potential for the proposal to reduce the overall volumes actually required, it was
incumbent on EPA to ensure that it provided adequate notice and a meaningful opportunity for
the public to comment.
Commenter: Phillips 66 Company
De Minimis Reporting Changes - We are very appreciative that EPA has recognized this issue
and proposed an allowance for de minimis volume errors and corrections. We strongly support
the concept of a de minimis volume allowance. However, the de minimis level proposed is far
too small to be of any utility. Fortunately, the systems in place to measure and report batch
volumes are very efficient and accurate. Unfortunately, there are some instances where slight
errors do occur. Currently when an error is discovered, no matter how small the volume
difference, it results in having to change the batch reports and resubmit with the corrected
volumes. EPA has proposed a de minimis level of 500 gallons or 1% of the batch volume,
whichever is less. Typical pipeline batch volumes are 25,000 barrels. A 500 gallon difference in
a 25,000 barrel batch constitutes a 0.05% difference.
6-10
-------
Tier 3 Summary and Analysis of Comments
We suggest the de minimis level be set at 0.5% for each batch. No further action should be
necessary such as recalculation of property averages with the uncorrected volumes versus the
corrected volumes. These should be considered de minimis from all aspects.
Our Response:
We proposed that parties who submit batch reports would not be required to correct
inconsequential errors in reporting batch volumes under certain conditions, the primary condition
being that the discrepancy met the definition of "de minimis", which as proposed, was an amount
no greater than the smaller of 500 gallons or one (1) percent of the true batch volume. Under the
proposal, regulated parties would no longer be required to provide a complete resubmission of a
compliance report when a minor discrepancy of a few barrels was uncovered. We proposed that
this new provision would apply to reporting for RFG, anti-dumping, gasoline and diesel sulfur,
and MSAT2. We proposed that it would also apply to the RFS renewable volume obligation
(RVO), but would only apply to the volume of fuels produced or exported that result in a RVO
for obligated parties.
We have decided not to finalize the proposed de minimis provisions at this time. One of
the primary motivations for proposing the de minimis provision was to avoid the need for
inconsequential corrections to production volumes that would have no impact on compliance
with standards that rely on production volumes (e.g., average standards and RFS). Late changes
to production volumes for whatever reason can necessitate simultaneous changes to compliance
calculations that, if de minimis, would have no meaningful impact on compliance. However, as
comments highlight, it is has proven difficult to set an acceptable de minimis threshold that can
apply across all potential situations. Furthermore, the proposed provision focused only on
volume corrections, but in reality our experience suggests that corrections often take place for
other purposes such as data entry, coding, formating or other typographical errors - not only
minor corrections to reported volume. We believe that adjusting reports for inconsistencies will
become less burdensome, as EPA intends to transition all reporting to electronic reporting in the
future. In addition, EPA received input from the regulated community about a de minimis
provision specifically as it related to the RFS program. We are finalizing a one-month delay in
the RFS reporting deadline, which we believe will provide obligated parties with more time to
review and correct their records and reports, and help to minimize the need for late corrections.
We will revisit the need for a de minimis threshold in the future if these changes prove
insufficient.
6.1.1.3. Reporting Deadlines
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM)
EPA should eliminate quarterly reporting for the RFG/RBOB batch reports. Data for all summer
RFG/RBOB batches should be submitted once year with the annual RFG VOC compliance
6-11
-------
Chapter 6: Regulatory Streamlining and Technical Amendments
report. The Agency has proposed, and we support, the elimination of aromatics, distillations, and
olefins for winter RFG/RBOB. Therefore, there is no purpose served for the first and fourth
quarters. In addition, there is no point in splitting up summer RFG/RBOB between the second
and third quarters.
We support EPA's proposal to set the due date for all fuel annual compliance reports at March
31. This extension will provide some flexibility for refining company personnel while having no
impact on emissions, air quality or compliance with the standards.
The Agency proposes a new due date for additional reporting for refiners blending butane with
RFG or RBOB at 80.75(o), March 1. This should be revised to March 31 to conform with the
uniform due date above.
Commenter: Chevron Products Company
Chevron is very much in favor of EPA's proposal to align reporting dates between the various
Part 80 programs and appreciates the extension of fourth quarter and annual reporting deadlines
to March 31. We recommend that EPA should extend the attest engagement deadlines by an
additional month from May 31 to June 30 to allow sufficient time for the significant data-
gathering and back-and-forth communications required to complete those engagements. While it
may be possible to begin some of this work ahead of the new reporting deadlines, it is much
more effective to focus on the reporting itself before shifting to the attest activities. We believe
that extending these annual deadlines will significantly reduce the risk of error and rework in
annual compliance reporting.
Commenter: Marathon Petroleum Company LP (MFC)
EPA should eliminate quarterly reporting for the RFG/RBOB batch reports. Data for all summer
RFG/RBOB batches should be submitted once a year with the annual RFG VOC compliance
report. The Agency has proposed, and we support, the elimination of aromatics, distillations, and
olefins for winter RFG/RBOB. Therefore, there is no purpose served for the first and fourth
quarters. In addition, there is no point in splitting up summer RFG/RBOB between the second
and third quarters.
We support EPA's proposal to set the due date for all fuel annual compliance reports at March
31. This extension will provide some flexibility for refining company personnel while having no
impact on emissions, air quality or compliance with the standards.
The Agency proposes a new due date for additional reporting for refiners blending butane with
RFG or RBOB at 80.75(o), March 1. This should be revised to March 31 to conform with the
uniform due date above.
Commenter: Phillips 66 Company
6-12
-------
Tier 3 Summary and Analysis of Comments
Change in reporting dates - Overall, the concept of aligning the various reporting dates and
being able to develop a unified and simplified reporting form is a good one. Providing additional
time is beneficial. We appreciate the Agency providing this change.
Commenter: Weaver and Tidwell LLP
Assuming the EPA implements the reporting deadline changes contemplated in the NPRM,
Weaver suggests that the deadline for submission of the annual attest reports be extended by one
month (June 30 following the end of the compliance year). The proposed change of the annual
reporting date for gasoline and RFS reports from the last day of February to March 31 following
the end of the compliance year, without a corresponding change in the due date for annual attest
reports, shortens the period of time for conducting the required attest work by one month.
Regulated parties are often subject to attest engagements at multiple facilities and for various
activities that are all inter-related (i.e., gasoline refiner being subject to gasoline attest
engagements [each applicable subpart], as well as an RFS attest engagement). Therefore, the
attest engagements can take a substantial amount of time to complete.
Weaver would like to emphasize the importance of the attest function in relation to compliance
with the EPA's fuels regulations as a whole. The EPA receives completed annual attest reports
which, in many cases, reflect compliance with the regulations. However, in our experience,
during the performance of the attest procedures, a myriad of reporting and compliance issues are
frequently discovered that can be and are remediated by the regulated party prior to the
completion of the attest engagement (hence, a "clean/cleaner" attest report is prepared and filed
with the EPA). If not for the independent attest engagement, it is our belief that many of the
issues unearthed and remedied as part of the attest would otherwise go unchecked and
undiscovered, thus shifting an increased level of review, investigation, and enforcement burden
to the EPA. Without sufficient time to find, research and resolve these errors, omissions, and
other instances of non-compliance, the efficacy of the attest function is diluted.
Commenter: Turner, Mason & Company
The change in the reporting date for the fourth quarter and annual reports could significantly
impact those who report to the EPA under 40 CFR 79 & 80. The reporting deadline for those
reports is being pushed back by a month to March 31. The reporting deadline for the first quarter
of the year is being pushed back by one day to June 1. In contrast, the reporting deadline for
attestation reports remains the same at May 31.
Because entities tend to delay submission of their reports to the EPA until all activities have been
completed, the time available to conduct an attestation under the current regulations is three
months. Under the current reporting schedule, it is already difficult to obtain data, conduct the
attestation, review the attestation with the affected party, revise the attestation based on
additional information provided by the affected party, obtain management representation letter(s)
signed by the responsible corporate officer, and produce the final attestation report within the
three month timetable currently existing.
6-13
-------
Chapter 6: Regulatory Streamlining and Technical Amendments
It is our experience at Turner, Mason & company that those involved in generating and
submitting reports to the EPA are the same people that provide data for the attestations. Based on
these new reporting deadlines, the months of April and May will become even busier for them.
Under the new reporting schedule, they will be preparing the first quarter reports simultaneously
to providing data to the attester(s) because of the increased workload in the shorter timeframe for
these workers, we expect an increase in reporting errors. Thus, we find that extending the
deadline for attestation reports by a month to June 30 would provide a month of relatively lighter
workload for those involved in the attestations and reduce the probability of reporting error.
Commenter: Shell Oil Products for Shell and Motiva
In regards to Amendments Related to Reporting 78 Fed. Reg. 29949, we concur with the
proposal to adjust and be consistent with reporting dates for various EPA programs. However,
with the annual reporting date being moved from February 28 to March 31, the attestation
engagement reporting date needs to be adjusted also. Currently, the attestation engagement report
is due May 31 and we recommend the due date be changed to June 30 in order to continue to
provide 3 months between the due dates of the annual report and the attestation engagement
report. As proposed, a timeframe of two months between the annual report due date and
attestation engagement report due date is too short of a time period.
Our Response:
In the final rule, EPA is amending various provisions to reduce the number of different
reporting deadlines that regulated parties must meet and to enable the future use of a unified and
simplified reporting form. Currently under 40 CFR parts 79 and 80, there are ten separate
cyclical reporting dates each year (eleven in a leap year). Streamlining reporting deadlines will
allow EPA to develop a single, user-friendly, electronic form that will collect all required data,
maximizing the capability of electronic reporting to provide reuse of data and avoid duplicate
data submission. EPA's goal is to simplify reporting and reduce the number of hours parties
spend preparing and submitting reports while simultaneously improving data received from
stakeholders. This overall effort responds to Executive Orders 13563 and 13610, which direct
government agencies to simplify rules and to achieve reductions in paperwork and reporting
burdens, and is part of EPA's agency-wide effort to streamline regulatory reporting
requirements.
We are amending these deadlines so that all affected programs use the same four
reporting deadlines. Programs that will be affected by this change include:
• the fuels and fuel additives registration program (40 CFR part 79, subpart A);
• the Reformulated Gasoline and Anti-Dumping program (40 CFR part 80, subparts D and
E);
• the Gasoline Sulfur program (40 CFR part 80 subpart H);
• the Motor Vehicle, Nonroad, Locomotive, and Marine Diesel program (40 CFR part 80
subpart I);
• the Gasoline Benzene program (40 CFR part 80 subpart L);
• the Renewable Fuel Standard program (40 CFR part 80 subparts K and M); and
• the Tier 3 program being finalized today (40 CFR part 80 subpart O).
6-14
-------
Tier 3 Summary and Analysis of Comments
Consistent with the proposal, we are finalizing that reporting deadlines will be
standardized as follows: June 1, for all reports covering quarter 1 of the compliance year
(January through March); September 1, for quarter 2 (April through June); December 1, for
quarter 3 (July through September); and March 31 for quarter 4 (October through December).
End-of-year compliance reports will also be due on March 31. These changes will either delay or
maintain current deadlines for nearly all required reports. Deadlines for all other annual reports
will either be maintained if they matched the new quarterly deadline, or extended to match the
new quarterly deadline. It should be noted that even with the changes finalized today,
respondents will still have the option to report earlier than any given deadline. Some
commenters in the refining industry suggested eliminating all quarterly reporting for RFG if
winter aromatics, olefms and distillation are no longer required to be reported as they suggested.
These values are normally reported in the 1st and 4* quarters. As one commenter stated, "In
addition, there is no point in splitting up summer RFG/RBOB reporting between the 2n and 3r
quarters." We disagree. As noted above, winter aromatics, olefms and distillation are still
required to be reported, albeit on a monthly composite basis. Thus, the need for quarterly
reporting remains. However, we expect that the need for quarterly reports will be evaluated as
part of a future broader program restructuring.
EPA proposed not to include "Attest Engagements" (currently due May 31 of the
following year) or reporting related to specific events under the Fuels Program, such as trading
Renewable Identification Numbers (RINs) in EPA's Moderated Transaction System (EMTS), in
the reporting revisions described above. Rather, all reporting deadlines for Attest Engagements
and reporting specific events will remain the same. Some commenters suggested that if the
annual and fourth quarter reporting deadlines were to be extended to March 31 of each year, then
the attest engagement date should also be shifted one month from May 31 to June 30 to allow
sufficient time for the significant data-gathering and communications required to complete those
engagements. EPA is not changing the compliance period associated with the new extended
reporting date, but rather, is simply allowing additional time for data review, preparation and
reporting. EPA does not believe that extending the attest engagement date is necessary because
companies have the flexibility to use the extra reporting time to begin preparation for the attest
engagement if they prefer. As stated previously, the extension to reporting deadlines does not
preclude stakeholders from reporting before the deadline. EPA is not extending the date for
attest engagements from May 31 to June 30. That said, EPA is streamlining reporting dates to
aid in our development of a single electronic reporting format and is updating the due date for
attest engagements found in part 80 from May 31 to June 1. This change will not affect
compliance or increase burden of reporting entities. Rather, the purpose of streamlining various
reporting deadlines is to ease reporting burden and help aid EPA in the development and
implementation of a single electronic reporting format.
6.1.1.4. Previously Certified Gasoline (PCG)
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
6-15
-------
Chapter 6: Regulatory Streamlining and Technical Amendments
Manufacturers (AFPM); Marathon Petroleum Company LP (MFC)
The Tier 3 proposal includes amendments for Previously Certified Gasoline. We support the
proposed changes at 80.1235(a)(6) and 80.1347(a)(6).
Commenter: Independent Fuel Terminal Operators Association (IFTOA)
Under current regulations, when refiners and importers are blending gasoline with Previously
Certified Gasoline ("PCG"), they must test (1) the PCG for benzene before addition of the
blendstock, (2) test the combined blend of PCG and blendstock for benzene, (3) calculate the
properties of the blendstock by treating the PCG as a negative volume batch, and (4) treat the
blended product as a positive batch. As EPA properly recognizes, due to variability in the
benzene test method, this approach can, at times, result in calculated blendstock benzene
concentrations that are negative, which is physically impossible. Therefore, EPA is proposing an
alternative that would allow greater flexibility. EPA would amend the regulations to permit
refiners and importers to sample and test each batch of the blendstock directly, and treat the
blendstock as the produced batch for blendstock that is blended into PCG. This approach is very
sensible and would eliminate illogical test results that have occurred in the field.
IFTOA recommends that EPA adopt the alternative test method that has been proposed. It would
provide a far more accurate test result and allow for more practical compliance.
Our Response:
Consistent with the proposal, for blendstock that is blended into previously certified
gasoline (PCG), we are finalizing an alternative requirement that allows refiners to directly
sample and test each batch of blendstock, and treat the blendstock as a produced batch. We note
that this practice is already allowed under the Tier 2 sulfur program.
6.1.1.5. Imports by Truck
What Commenters Said:
Commenter: Independent Fuel Terminal Operators Association (IFTOA)
Under current regulations, when conventional gasoline is imported into the United States by
truck, compliance with fuel parameters may be demonstrated by testing fuel in the storage tank
from which the trucks are loaded. See 40 CFR Part 80, Subpart E, Section 80.101(i)(l) and (3).
However, pursuant to 40 CFR Part 80, Subpart L, Section 80.1347, each batch of gasoline
produced or imported (RFG or conventional gasoline) must be sampled and tested for
compliance with the benzene standard. The inconsistency of these two regulations has resulted
in importers spending thousands of dollars testing each truck that enters the United States and
this additional testing - testing each truck versus testing the tank from which the trucks are
loaded - is unnecessary. It provides no greater testing accuracy and is so costly that it has a
chilling effect on truck imports. Therefore, the Association welcomes the EPA's proposal to
6-16
-------
Tier 3 Summary and Analysis of Comments
make the regulations consistent and allow gasoline imported into the U.S. by truck to
demonstrate compliance by testing the gasoline in the storage tank from which the trucks are
loaded.
IFTOA endorses EPA's proposal to allow gasoline imports made by truck to demonstrate
compliance with applicable fuel parameters by testing the gasoline in the tank from which the
trucks are loaded in accordance with the requirements set forth in 40 CFR Part 80, Subpart E.
Commenter: Irving Oil Terminals Inc.
EPA has proposed to allow entities that import conventional gasoline into the United States by
truck to demonstrate compliance by following the sampling and testing requirements in Subpart
E — sampling and testing of the storage tank that feeds the truck imports — as an alternative to
the sampling and testing requirements of Subpart L that some have interpreted to require the
testing of every truckload. Irving Oil strongly supports this amendment and commends the EPA
for allowing importers to use a practical and more economic means of testing that provides the
same degree of accuracy and makes the provisions in Subpart E and Subpart L consistent.
Our Response:
We are allowing importers who import gasoline into the United States by truck to use the
sampling and testing requirements in subpart E for truck importers as an alternative to the
sampling and testing requirements in subpart L. EPA provided these alternative requirements in
subpart E to eliminate the need to test every truckload of imported conventional gasoline for all
complex model parameters, including benzene. Since Subpart L also requires importers to test
every truckload of imported gasoline for benzene, EPA believes it is appropriate to allow truck
importers of gasoline to use the sampling and testing requirements in subpart E as an alternative.
6.1.1.6. Diesel Tech Amendments
What Commenters Said:
Commenter: Chevron Products Company
Diesel Pump Labels: Chevron supports the removal of the requirement for diesel fuel pump
labels for 15 ppm highway diesel fuel. We do not expect that this will cause confusion for
consumers and agree with EPA's assertion that voluntary labeling is always an option for retail
and wholesale facilities should that become an issue. It should be noted that §80.572(a) already
contains an expiration date for these labeling requirements. The current regulations show an
expiration date of September 31 [sic], 2014 and the amendment proposed in this rulemaking is
actually extending the expiration date to November 30, 2014. There appears to be no reason for
any extension and we assume this is an error. As EPA states in its preamble, these labels have
been unnecessary since December 1, 2010. Therefore, an immediate removal of the requirement
would be entirely appropriate.
6-17
-------
Chapter 6: Regulatory Streamlining and Technical Amendments
Should the labeling requirements remain in place, we oppose the regulatory changes proposed
that do nothing but underline the words 'prohibits' and 'required' in the current labeling
requirements. If this requires that existing labels, which are no longer required after 2014, be
updated, it's an expensive and burdensome proposition with little tangible benefit.
Commenter: Magellan Midstream Partners, L.P.
Transmix exemption — EPA has proposed per-gallon and marker requirements for NRLM diesel
fuel, EGA marine fuel, and heating oil downstream of the refiner or importer. EPA has also
clarified that this provision does not apply to LM diesel fuel produced from transmix or interface
fuel that is sold or intended for sale in areas other than those listed in §80.510(g)(l) or (g)(2), as
provided by §80.513(f).
Magellan Comment — we agree with EPA on the transmix exemption and strongly encourage
the agency to adopt this provision. The high levels of sulfur in jet fuel create challenges for
transmix processor because access to certain markets has been restricted.
Our Response:
We are removing the requirement for diesel fuel pump labels for 15 ppm highway diesel
fuel. Beginning December 1, 2010, all highway diesel fuel was required to be 15 ppm or less;
thus, highway diesel fuel labels are no longer needed to distinguish it from 500 ppm highway
diesel fuel. However, we do recognize that it may confuse consumers who are accustomed to
seeing the highway diesel fuel pump labels if those labels were to disappear, thus, retail and
wholesale purchaser-consumer facilities will be free to continue labeling to eliminate confusion
if they so choose. The elimination of this requirement from the regulations does not preclude
retail and wholesale purchaser-consumer facilities from keeping 15 ppm highway diesel fuel
pump labels, it only eliminates the EPA requirement that such labels must be present.
We are finalizing the proposed change to §80.511.
6.1.1.7. Test Method Updates
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM); Marathon Petroleum Company LP (MFC)
Amendments to Update Test Methods: We note that, in some instances, the test method which
EPA is proposing to update in this rulemaking does not necessarily represent the most recent
version adopted by ASTM International. For instance, the proposal references ASTM D4057-06
(2011) the manual sampling standard practice for petroleum and petroleum products. However,
this standard practice was significantly revised and was recently re-issued as ASTM D4057-12.
We strongly urge EPA to update its test method references to the most current versions available
when it publishes the final rule, or, at the very least, provide a rationale for not doing so.
6-18
-------
Tier 3 Summary and Analysis of Comments
Commenter: Chevron Products Company
Test Method Reporting: Numerous updates are proposed to add the requirement that test
methods be reported whenever fuel properties are reported under existing programs.
Understanding that this is intended to enable the Performance Based Measurement System, we
urge the EPA to consider how best to simplify the adoption of this requirement in its reporting
procedures. Adding a test method field corresponding to every property field on existing report
forms would greatly increase the size and complexity of those forms and, where the PBMS is not
employed, would add no value. We recommend allowing obligated parties to report standard test
methods in one separate notification and indicate exceptions, if any, separately.
Commenter: Independent Fuel Terminal Operators Association (IFTOA)
In the Preamble that accompanied the proposed rules, EPA explained that refiners, importers and
oxygenate blenders producing gasoline and diesel are required to test various fuel parameters. In
a number of the regulations, there are references to test methods that have been revised or
updated since the regulation was adopted. Therefore, EPA is proposing to amend the regulations
and to reflect the new or updated ASTM test methods.
The Association supports the basic concept of the proposal - updating test methods. However,
the Association believes that the regulations should be amended to authorize refiners, importers
and oxygenate blenders to use the current version of the test method at the time of the testing -
not one current at the time the final rule is adopted. In this way, industry would be employing
methods that ASTM deemed appropriate and would facilitate better operation for the user of the
test method. In addition, EPA should retain those alternative test methods it has already
approved for use.
Recommendation: The Association supports EPA's efforts to reference updated/revised test
methods in the regulations used to test gasoline and diesel fuel parameters. However, the
Association recommends that EPA go further with its amendment and authorize testing using the
current version of the applicable test method as of the date of testing or an alternative test
method approved by EPA.
Organization:Irving Oil Terminals Inc.
Irving Oil supports the updating of test methods used for various fuel parameters and also
recommends that the regulation require compliance to 'the most current version of each method'
instead of specifying a revision status date. This would allow the EPA and industry to perform
testing by responding with the most current technology available.
Commenter: Phillips 66 Company
Update Test Method references - We support the proposal to update the referenced test methods
to more current versions but ask that EPA check for the most current versions.
Our Response:
6-19
-------
Chapter 6: Regulatory Streamlining and Technical Amendments
Refiners, importers and oxygenate blenders producing gasoline and diesel motor vehicle
fuel are required to test RFG, CG and diesel fuel for various fuel parameters including aromatics,
benzene, distillation, olefms, oxygenate content, RVP, and sulfur. As stated in the proposal, a
number of relevant regulatory provisions had references to test methods that have been revised
and updated. Several comments were received indicating that since the time of the proposal,
some of the proposed updates have been further revised and updated and the desire was
expressed for the regulated community to have the opportunity to use the most current version of
each of these test methods. Consistent with the proposal, in this final rule we are updating those
test methods to reflect current test methods in order to ensure that all test methods are readily
available to the regulated community. (See Section VI.A.l.c of the preamble to the final rule and
more specifically Table VI-2 for the list of test methods.)
6.1.1.8. Broader Regulatory Streamlining
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM); Marathon Petroleum Company LP (MFC)
There are additional topics that should be included in regulatory streamlining. Some are very
simple and straightforward and should be implemented quickly and easily. These were suggested
to EPA by API and AFPM in a memo sent to EPA on May 11, 2011 (Attachment No. 2).
Although EPA addressed a few suggestions, the following issues remain.
MSAT2:
§80.1347(a)(3) requires that each batch of gasoline be sampled and tested for benzene content.
Refiners are required to test each batch at a significant expense. The cost of a single benzene test
ranges between $130 and $350, a cost which becomes disproportionate for small volume
blenders. For all gasoline this program imposes averaging standards rather than any per-gallon
limits. Allowing the individual batch samples to be composited prior to analysis would reduce
costs and simplify reporting and recordkeeping.
Batch reports and the addition of oxygenates to CBOB:
In section 80.69, EPA describes an alternate QA program for RBOB to confirm that the proper
amount of ethanol is being added to the RBOB. That confirmation essentially enables the
refiners to account for the 10% dilution of sulfur and benzene in their batch reporting. The
alternate QA program addresses the situation where the RBOB is distributed through a fungible
pipeline system like the one that serves the Mid-Atlantic and northeast states. If a refiner wants
to account for the ethanol dilution for conventional gasoline via fungible shipments of CBOB,
there is no parallel QA program allowed and EPA has declined to clearly indicate the RBOB
approach for CBOB is acceptable. EPA should revisit and clarify to allow the same survey
approach used for RBOB to apply to CBOB distribution systems, as well. EPA should allow
refiners to account for the ethanol dilution for conventional gasoline via shipments of CBOB.
6-20
-------
Tier 3 Summary and Analysis of Comments
Under 40 CFR 80.101(d)(4), EPA Anti-Dumping regulations require a refiner or importer to
include in its compliance calculations any conventional gasoline blendstock (CBOB) that is
produced or imported which becomes conventional gasoline solely upon the addition of
oxygenate. Refiners and importers must conduct a program of quality assurance testing at the
downstream oxygenate blending facility in order to include the oxygenate in their compliance
calculations. Under the current regulations, refiners must conduct a program of sampling and
testing (quality assurance) at the downstream oxygenate blending facility in order to include the
oxygenate in their compliance calculations. This rule provides an alternative QA requirement for
these refiners and importers.
When oxygenate is to be added to produce a finished conventional gasoline at a downstream
oxygenate blending facility, refiners produce a product called conventional gasoline blendstock
for oxygenate blending, or CBOB. CBOB is certified by the refiner, or by an importer who
imports CBOB, as complying with all of the conventional gasoline requirements. The oxygenate
blender is responsible for complying with the oxygen requirement when the oxygenate is added
to the CBOB to produce a finished conventional gasoline at the oxygenate blending facility.
Oxygenates such as ethanol, have a propensity to attract water, and, as a result, cannot be added
at the refinery, particularly where the finished gasoline will be shipped through a fungible
pipeline on its way to terminals and retail gasoline stations. As a result, CBOB is typically
produced for blending with ethanol at a blending facility downstream from the refinery that
produced the CBOB.
Where a specific type and amount of oxygenate is designated for CBOB, the regulations require
the refiner or importer to conduct downstream oversight sampling and testing quality assurance
(QA) of the downstream oxygenate blending facility (40 CFR 101(d)(2)). This is to ensure that
the specific type and amount of oxygenate that is designated and claimed by the refinery or
importer for compliance, is in fact added to the CBOB by the oxygenate blender. In addition, the
refiner or importer must have a contract with the oxygenate blender which requires the blender to
comply with the blending procedures specified by the CBOB refiner or importer and allows the
refiner or importer to conduct the required QA sampling and testing (40 CFR 80.101(d)(l)). If
the refiner or importer does not meet the contractual and quality assurance requirements for
CBOB, the refiner or importer may not include the oxygenate for compliance purposes.
Due to the complexities of the gasoline distribution system, it would be extremely difficult, if not
impossible, to track CBOB from the refinery where it was produced to the terminal where it was
blended with ethanol in order to comply with the downstream QA sampling and testing
requirements specified in the regulations. In order to facilitate ethanol blending, effective August
1, 2006, EPA amended the RFG regulations to permit refiners and importers seeking to add
ethanol to RBOB an alternative quality assurance and downstream sampling and testing
program. This program needs to be extended to refiners and importers supplying CBOB for
downstream oxygenate blending in conventional gasoline areas as well.
RFG Sections 80.68 and 80.79:
Section 80.79 addresses liability for violations and addresses the defense elements which must be
present. One of those defense elements is that the refiner must have a QA program at each of the
points in a distribution system, excepting the truck carrier. Section 80.68 covers the need for
6-21
-------
Chapter 6: Regulatory Streamlining and Technical Amendments
retail surveys, surveys which are conducted by the RFG Survey Association for their members.
A Q&A that dates back to the mid-nineties is as follows: 16. Question: May survey samples be
used as a substitute for a refiner's quality assurance program for enforcement purposes? Answer:
Surveys may not be used as a substitute for a regulated party's own quality assurance program.
Industry has been conducting independent surveys of their retail sites whereas they should be
allowed to rely on the RFG Survey Association survey samples of individual brands.
Commenter: Chevron Products Company
Regulatory Streamlining General Requirements and Deadlines: Chevron supports the reduced
reporting requirements for reformulated gasoline and EPA's focus on simplifying reporting to
those elements critical to compliance. We strongly encourage EPA's notion of performing a
comprehensive restructuring of the Part 79 and 80 regulations to improve comprehension, reduce
compliance costs and simplify reporting. Identifying all of the various opportunities for
consolidation and simplification would go well beyond the timing and scope of this rulemaking
but we encourage EPA to immediately form a working group with industry representatives to
pursue a follow-up rulemaking with the goal of proposing regulatory updates by the first quarter
of2014.
Commenter: Irving Oil Terminals Inc.
Certification of Fuels (RFG and CG) and Commingling: EPA has asked for comments on the
need to revise the regulations surrounding RFG and conventional gasoline relating to such issues
as areas of sale and commingling of product. Irving Oil supports efforts to streamline the
regulations and had anticipated proposals to address these issues. There are at least two reasons
to move forward with such revisions.
First, there have been several times when Irving Oil has had gasoline certified as conventional
gasoline that meets the specifications/requirements for RFG. When an emergency occurred in a
nearby RFG-market and there were shortages of product, Irving OH could have moved its
conventional gasoline to the RFG area to alleviate the problem. However, it was prevented from
doing so because it could not recertify the gasoline, even though it met the RFG standards and
would not have adversely affected air quality.
Second, at present, a company may not commingle RFG and conventional gasoline in the same
tank. However, in those instances when gasoline meets both the RFG and conventional gasoline
standards, use of commingled storage, coupled with robust recordkeeping, would substantially
reduce costs and facilitate more effective operations.
Commenter: Magellan Midstream Partners, L.P.
80.82 Sulfur Sampling Results at Time of Transfer:
Magellan Comment — The revised 40 CFR 80.82(i)(2)(iii) specifies that butane blenders must
obtain a copy of the butane supplier's test results at the time of each transfer of butane. This
requirement is an unnecessary burden because the same test results often apply to multiple
shipments of butane over the course of days or weeks. Rather than requiring redundant
6-22
-------
Tier 3 Summary and Analysis of Comments
conveyance of the same information, it should be adequate assurance to require that the blender
obtain test results that represent each batch. Magellan proposes that 80.82(i)(2)(iii) be revised to
include the following language: 'The butane blender must obtain a copy of the supplier's test
results to represent each transfer of butane to the butane blender.'
80.1347(6) Benzene sampling upon receipt at refinery vs. refiner:
Magellan Comment — The revised 40 CFR 80.1347(a)(6) specifies that when sampling blend
stock for benzene representation the refiner must take the sample when the blend stock is
received at the refinery. We propose that the refiner should be allowed to take this representative
sample upon receipt by the refiner, but that the sample need not be taken at the specific refinery
where the blendstock is blended into previously certified gasoline. We propose changing the
words 'at the refinery' to 'by the refiner.'
As discussed in the recent call and in our comments to the proposed rule, Magellan strongly
supports a revision to 80.1347 (a) (6) from 'when it is received at the refinery' to 'when it is
received by the refiner'. The reason for the modification is to allow for a gasoline blender to
receive, sample, and test blend stock at a central company facility and then distribute that same
blend stock to various other company owned refineries. This change would avoid the need to
build blend stock storage at each refinery facility or to sample every truck load of blend stock
blended at a location that did not have blend stock storage. Construction of dedicated blend stock
storage at all facilities would be extremely expensive or in some cases is physically impossible
due to space constraints and sampling of individual truck loads is logistically challenging and
unduly expensive. An oversight program similar to that found in 80.82 could be implemented to
ensure that blend stock quality is maintained during transport.
Commenter: National Biodiesel Board (NBB)
EPA Should Review and Conform the Various Product Transfer Document Requirements that
May be Applicable to Biodiesel: In the Tier 3 Proposal, EPA notes that the fuel industry has
requested EPA to review its Part 79 and Part 80 regulations to look for further ways to streamline
the regulations. "EPA requests comments on potential areas in the fuel regulations that may
benefit from such a more comprehensive streamlining effort." 78 Fed. Reg. at 29,953. As NBB
noted in its comments on the Quality Assurance Program proposal, see EPA-HQ-OAR-2012-
0621-0069 at 57-58, there are numerous disclosure requirements that must be included on
product transfer documents for biodiesel blends, and NBB requests that EPA review and
conform these requirements.
Biodiesel is subject to product transfer document requirements under the RFS2 program and the
ultra-low sulfur diesel fuel program. This may be in addition to any state requirements as well.
EPA should consider whether these statements can be consolidated to address the various
regulatory programs without conflicting with potential state requirements. While the definitions
for other diesel fuels may be different under the various programs, it is not for biodiesel.
Biodiesel is fuel that meets ASTM D 6751. This is the standard that is applicable to biodiesel
regardless of its use, although the finished fuel may be subject to another ASTM standard such
as heating oil (ASTM D 396). There simply is little need for EPA to require long and
overlapping statements on biodiesel product transfer documents. Thus, NBB recommends that
6-23
-------
Chapter 6: Regulatory Streamlining and Technical Amendments
EPA require one statement that can cover these various requirements. For example, it should be
sufficient to state that: "This volume of fuel is or contains biodiesel for which RINs have been
generated under § 80.1426, is designated or intended for use as or in transportation fuel, heating
oil or jet fuel." For blends, EPA may also require disclosure of the sulfur content to ensure
compliance with 40 C.F.R. § 80.590.
Commenter: Phillips 66 Company
Other Streamlining Issues - EPA has asked for comment on other provisions that we feel should
be considered that would increase efficiency and reduce costs. There are several issues that the
industry has discussed with EPA that we still feel should be pursued. In addition, Phillips 66 has
internally identified an issue that we ask the Agency to consider. These are:
CBOB - simplification of process to allow inclusion of ethanol blended downstream - The
current process for RBOB certification and downstream quality assurance should be replicated
for CBOB. Although there is a process in the regulations for capturing ethanol volumes blended
into conventional gasoline or CBOB downstream, it is virtually impossible to utilize in situations
where a refinery is shipping fuel on large fungible pipeline systems (such as Colonial Pipeline).
Back in 2006, EPA finalized regulations that provide refiners an alternative quality assurance
and downstream sampling and testing program for RBOB. This program needs to be replicated
and extended to CBOB. The industry has previously done extensive work in looking at sections
of the regulation and potential language changes necessary to provide this option. In addition, the
RFG Survey Association has outlined a program to provide the quality assurance sampling and
testing that would be required, similar to the current program for RBOB. Our industry could
provide updated information to the Agency in an expedited manner for inclusion in this
rulemaking. Currently, refineries have to test properties on CBOB or conventional gasoline on a
clear basis for batch reporting purposes to EPA. However, they are also having to hand blend
ethanol into the samples and test to assure the fuel will meet performance standards on a blended
basis. Providing the requested regulatory change would simplify the laboratory sample handling
and testing.
Our Response:
Today's rule also clarifies the list of products that are not to be included in a refinery's or
importer's compliance determination under §80.1240. Refiners and importers are currently
required under §80.1235(b)(2) to exclude oxygenate added to finished gasoline, RBOB or CBOB
downstream of either the refinery that produced the gasoline or the import facility where the
gasoline was imported. This conflicts with EPA's intended approach in §80.1238(b), which
allows refiners and importers to include oxygenate blended downstream of a refinery or import
facility in their annual average benzene calculation, provided the refiner or importer meets
certain requirements. We are finalizing changes that will allow refiners and importers to include
oxygenate blended downstream of their facility and that will make these related sections
consistent. EPA received significant support for this action from commenters.
The current set of fuel regulations is the result of programs that have been established
over the years to reduce emissions from mobile sources. These programs include gasoline
6-24
-------
Tier 3 Summary and Analysis of Comments
volatility (RVP), reformulated gasoline and anti-dumping, sulfur control (which today's Tier 3
program will revise), mobile source air toxics (MSATI), benzene control (MSAT2), and the
renewable fuel standards (RFS). Most of these regulations have been amended numerous times.
The RFG and anti-dumping regulations in particular contain some of the more extensive
requirements on sampling, testing, and reporting. They also have some of the more stringent
restrictions on gasoline use (e.g., restricting where fuel produced can be sold, what it may be
commingled with, etc.). EPA used the RFG and anti-dumping rules as the foundation for many
aspects of subsequently developed fuel regulatory programs. However, the subsequent rules,
considered as a whole, have supplanted most of the RFG and anti-dumping standards. For this
reason, we proposed to streamline the regulations in several places as described above. Initial
discussions with fuel industry representatives have indicated that a comprehensive review of the
complete set of fuel regulations contained in 40 CFR parts 79 and 80 ("Registration of Fuels and
Fuel Additives" and "Regulation of Fuels and Fuel Additives," respectively) could lead to further
streamlining of the regulations beyond the streamlining provisions being finalized today. EPA
expects that further streamlining would result in more efficient and less costly compliance
determinations for affected parties while maintaining the environmental benefits of the programs.
In many cases such changes could require not just the removal or streamlining of existing
provisions but also the replacement of several provisions with new, less onerous ones that
require further development, notice and comment. We intend to continue to seek comment in
future actions on potential areas in the fuel regulations that may benefit from such a more
comprehensive streamlining effort. For example, it may be possible for the RFG VOC standard
to be met if a sufficiently stringent RVP level is attained. Under this scenario, sampling and
testing requirements at the refinery would be reduced. Another potential scenario could involve
consolidation of some RFG and anti-dumping rules; for example, a single set of rules governing
the treatment of downstream ethanol blending and in-use surveys could provide greater
efficiency and flexibility regarding fuel distribution.
We received a number of comments supporting the concept of further streamlining in 40
CFR parts 79 and 80, including suggestions for additional areas of the regulations to consider in
the future.
6.1.1.9. Other
Alternative Laboratory
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM); Marathon Petroleum Company LP (MFC)
We support the proposed changes at 80.65(f)(5) that would allow a facility to use an alternate
independent laboratory. This will provide needed flexibility when there is an unexpected
problem.
6-25
-------
Chapter 6: Regulatory Streamlining and Technical Amendments
Commenter: Phillips 66 Company
Alternative designated lab - We support the proposed provision to allow designation of an
alternative lab. There are occasions when the primary designated lab has an instrument outage or
other problem preventing them from being able to complete the necessary sample analysis.
Providing another option will alleviate these issues when they arise.
Our Response:
Refiners have indicated to EPA that significant problems are created when a facility's
designated lab is nonoperational and testing cannot be performed at the lab during that time
period. We are thus finalizing (at §80.65(f)(5)) that a facility will have the ability to designate a
back-up or alternative lab for testing during such times. In no case could this alternative lab be
used to select the best test result, rather it may only be used on those occasions where operational
necessity causes a need for it (e.g., the normal lab is closed, the apparatus for certain test
methods are down, or independent lab personnel are not available).
Responsible Corporate Officer
What Commenters Said:
Commenter: Phillips 66 Company
Changes needed in registration update procedures - Currently, any update to a company or
facility registration requires signature from a responsible corporate officer (RCO) no matter the
nature of the change. This process, which requires senior level approvals for relatively minor
administrative updating, is very burdensome for executives and can be streamlined using
sufficient company delegations. An example of this is a change in the facility contact
information: a simple revision to a phone number requires RCO signature. We strongly
encourage the EPA to provide flexibility in this process by granting authority to the RCO's
designated company editor to make these changes without requiring additional RCO signatures.
Also, there should be a mechanism for an RCO to designate which notifications are directed to
him/her directly or to specifically choose to have notifications go to his chosen editor instead. An
example is the notifications that go out when reports are submitted or resubmitted. In the
situation where the company has a significant number of facilities, these notifications add greatly
to the RCOs e-mail load. Our company editor would be happy to work directly with the EPA to
help outline some specific changes that would alleviate burden and frustration in this area.
Our Response:
Regarding the suggestion that EPA revise the instances when a "responsible corporate
officer" signature is required, we understand the concern expressed, but do not believe a
regulatory action is necessary to address this concern. Rather, this issue is best addressed via
administrative procedures. EPA plans to further evaluate this specific concern and similar
6-26
-------
Tier 3 Summary and Analysis of Comments
"responsible corporate officer" issues and address the concerns in the future, outside of the
regulatory framework.
Diesel Cetane Number
What Commenters Said:
Commenter: Ford Motor Company (Ford)
Diesel - Need for Increased Market Fuel Cetane Number: Ford would like to take this
opportunity to recommend the need to increase the minimum cetane number in typical diesel
market fuel to the EPA's attention for future regulatory action.
A higher cetane number is preferred to support a shorter ignition delay and better ignition quality
as these are important to achieving the stringent Tier 3 emission standards. An increase in cetane
number generally results in decreased emissions of HC, CO, and NOx. Increased cetane number
results in a slower rise in combustion pressure, which leads to lower gas temperatures and in turn
reduces the formation of NOx. Low levels of in-use cetane rating can increase white smoke
emissions during cold start and lower cetane ratings result in higher combustion noise - the low
ignition quality of the fuel causes rapid burning of the larger amount of fuel injected before
ignition occurs, which results in higher cylinder peak pressure and creates the characteristic
diesel noise. Additionally, running an engine on diesel fuel with a lower cetane number than
required will result in cold starting difficulties, reduced performance, and increased emissions.
As a result of refineries optimized for maximum gasoline production, specified cetane levels are
typically lower in North America than in Europe. In order to maintain the balance between the
highest fuel economy and lowest emissions, diesel engines are designed based on fuels that are
most representative of the market fuels where the vehicle will be sold. This makes it imperative
for market fuel specifications to have tolerances that allow little variation in the fuel properties
so that the engines will perform as close to the design standard as possible. The current ASTM
D975 minimum cetane rating specified is too low to achieve maximum fuel economy and
emissions improvements that are required in the near future.
Recommendation: Ford supports raising the minimum cetane level to 51 in order to align with
European Union (EU) specifications, which will foster diesel engine technology advancements in
global product development.
Our Response:
As noted by the commenter, this comment is outside the scope of the Tier 3 rule. If a
future rule is promulgated regarding diesel cetane number, the commenter's suggestion could be
considered at such time.
6.1.1.10. RFS
6-27
-------
Chapter 6: Regulatory Streamlining and Technical Amendments
General
What Commenters Said:
Commenter: National Biodiesel Board (NBB)
While NBB appreciates the opportunity to comment on the proposed amendments to the RFS2
program, it is concerned with EPA's decision to make these proposals as part of an unrelated
rulemaking and with insufficient explanation as to the legal authority or rationale supporting
such changes. This is even more troubling given the extensive proposals on the RFS2 program
that were recently made in the notices regarding the Quality Assurance Program and technical
amendments. Thus, NBB requests EPA provide more explanation for the amendments, as
required by the Clean Air Act, and a meaningful opportunity to comment on them.
Our Response:
We disagree with the comment. The Agency utilized the Tier 3 rulemaking to also
propose and finalize a number of relatively minor regulatory streamlining actions cutting across
our vehicle and fuel programs, including several that related to the renewable fuels program.
These actions were fully noticed in the proposal.
At proposal, we explained the need to revise and streamline certain provisions in order to
reduce the burden on industry or in other words to provide additional flexibility. We explained that
these revisions would have no expected environmental impact. Specifically, we proposed changes
to the RFS2 annual report date from the last day of February to March 31. EPA intended the
change would 1) alleviate staffing problems for regulated entities because the regulatory
requirements are often handled by the same personnel and 2) provide relief from the urgent need
to obtain RINs when small discrepancies in gasoline production or import volumes are realized.
We also explained that streamlining reporting dates would allow EPA to develop a single, user-
friendly, electronic form that would collect all required data, maximizing the capability of
electronic reporting to provide reuse of data and avoid duplicate data submission. Additionally,
we sought comment on whether the same date extension from the last day of February to March
31 should apply to those transferring RINs in EMTS for satisfying RVOs under the previous
compliance year. We are not finalizing this change in this final rule, in response to comments.
Further, EPA proposed not to include "Attest Engagements" (currently due May 31 of the following
year) or reporting related to specific events under the Fuels Program, such as trading Renewable
Identification Numbers (RINs) in EPA's Moderated Transaction System (EMTS), in the reporting
revisions described above. Rather, all reporting deadlines for Attest Engagements and reporting
specific events will remain the same. Finally, we proposed that parties who submit batch reports
would not be required to correct inconsequential errors in reporting batch volumes under certain
conditions, the primary condition being that the discrepancy met the definition of "de minimis",
which as proposed, was an amount no greater than the smaller of 500 gallons or one (1) percent of
the true batch volume. (See 78 FR 29950-29953). As discussed in Chapter 6.1.1.2 above, we are not
finalizing any de minimis provisions at this time.
6-28
-------
Tier 3 Summary and Analysis of Comments
Reporting Dates
What Commenters Said:
Commenter: Chevron Products Company
In conjunction with the RFS compliance deadline, the ability to trade RINs in EMTS for
satisfying previous year RVOs should be extended to March 31 as well to enable maximum
compliance flexibility. An end-of-February cutoff date would mean a de facto end-of-February
deadline to finalize compliance reporting given the potential need to purchase additional RINs
for compliance.
Commenter: Independent Fuel Terminal Operators Association (IFTOA)
March 31 Annual Reporting Date and Trading: EPA is proposing to amend the annual reporting
date for the Renewable Fuel Standard Program. Currently, compliance must be demonstrated by
February 28 for the prior year. EPA is proposing to change that date to March 31. In addition,
the Agency has asked for comments on whether trading of RINs used to demonstrate compliance
should also continue through March 31.
The Association supports the change to the annual reporting date, but believes that all trading of
RINs to meet the prior year's obligations should cease on February 28. In this way, members of
the regulated community would have sufficient time to (a) calculate their precise RVOs for the
prior year, (b) determine the volume of RINs held in inventory,(c) verify those RINs, and (d)
prepare the reports in a timely manner. Separating the close of RIN trading from RIN reporting
by a month should improve compliance and reporting accuracy.
Recommendation: The Association supports the proposed change in the annual reporting date
from February 28 to March 31, but opposes any change to the date when RIN trading ceases; the
later should remain at February 28.
Commenter: Irving Oil Terminals Inc.
Change in Annual Reporting Date — March 31: EPA has proposed to change the RFS2 annual
reporting date from February 28 to March 31. EPA is also seeking comment on whether the
ability to transfer RINs into the EMTS to meet an obligated party's previous compliance year's
Renewable Volume Obligation ('RVO') should be extended to March 31 as well. Irving Oil
supports the annual reporting change to March 31. However, it believes that the February 28
deadline for RIN trading and meeting a company's RVO should be retained. By allowing a
month to elapse between the two activities, obligated parties have sufficient time to complete
their accounting, verify their RVO and RIN inventory and work through their approval process
with the Responsible Corporate Officer. This separation would facilitate more accurate reporting
to EPA.
Commenter: National Biodiesel Board (NBB)
6-29
-------
Chapter 6: Regulatory Streamlining and Technical Amendments
NBB Requests Clarification as to the Proposal to Amend the RFS2 Annual Report Date and
Potential Implications for the REST Market. In a purported attempt to address "a staffing problem"
with respect to the RFS2 annual compliance reporting and reporting under other fuels programs,
EPA is proposing to revise the annual compliance reporting date in 40 C.F.R. § 80.1451 for
obligated parties from February 28 to March 31. 78 Fed. Reg. at 29,952. NBB does not
necessarily oppose such an extension, but it is wholly unclear whether the proposal is truly
addressing such a staffing issue and, thus, is necessary. In particular, EPA cites the RFG and
anti-dumping compliance deadlines, but EPA is also proposing to move several of those
February 28 due dates to March 31. See id. at 30,012, 30,015. NBB requests clarification as to
EPA's rationale for the need to move the dates.
EPA also contends that the extension will allow obligated parties to conduct a "volumetric
auditing . . . and would have no impact on ... compliance with the standard." 78 Fed. Reg. at
29,952. However, EPA provides no explanation as to what this "volumetric auditing" is, if it
applies to all gasoline/diesel (or, e.g., just RFG audits), or when it occurs. It simply is unclear
why obligated parties need 90 days to identify their volumes and obligations under the RFS2
program. That being said, NBB does not necessarily object to a March 31 deadline, so long as
the extension is only intended to allow additional time to ensure accurate volumes are reported,
and not to affect the RIN market. The Tier 3 Proposal provides no explanation as to whether the
additional month will have an effect on the RIN market. Instead, EPA simply concludes that it
will have no impact on compliance with the standard. There is substantial RIN activity in
January and February to allow parties to true-up their RIN inventory to ensure they have
sufficient RINs. As EPA is aware, there is much speculation in the RIN market, and it is unclear
how this extension may affect that market. Moreover, many negotiations for fuel contracts for
the coming year are done at the beginning of the year, when obligated parties apparently will still
be addressing the prior year's compliance.
Further, NBB requests more information before EPA considers allowing ongoing RIN transfers
in the EMTS through March 31 for compliance with the prior year. The RFS2 program is
intended to spur production, and such an extension may only lead to more speculation and ease
of purchasing prior year RINs rather than actual purchase of wet gallons. In either case, EPA
does not provide sufficient explanation as to its proposed change, and, thus, NBB requests that
EPA carefully consider such impacts and allow the public time to comment on its assessment.
In addition, in its comments on the Quality Assurance Program, NBB requested more
transparency and consistency as far as reporting of available RINs and compliance by obligated
parties. See EPA-HQ-OAR-2012-0621-0069 at 60-61 (Apr. 18, 2013). We are also concerned
that the additional month will further delay EPA's release of such information. Again, NBB
requests that EPA carefully consider the potential impacts of the proposed change.
NBB notes that EPA also has proposed to revise the due dates for the quarterly reports under 40
C.F.R. § 80.145 l(f). In particular, for the last quarter, EPA proposes to extend the deadline by an
additional month. Again, NBB does not necessarily oppose the extension, but if the February 28
deadline remains for obligated parties, their RIN activity and transaction reports should continue
to be due the same date.
6-30
-------
Tier 3 Summary and Analysis of Comments
Commenter: Shell Oil Products for Shell and Motiva
EPA requested comment on Amendments Related to RFS2 Annual Report Date,78 Fed. Reg.
29952, specifically the date to be able transfer RINs in EMTS for satisfying RVOs for the
previous compliance year. We request that the ability to transfer RINs in EMTS for satisfying
RVOs under the previous compliance year be extended to March 31 to coincide with the
proposed annual report date for other EPA programs.
Commenter: Sutherland Asbill & Brennan LLP (Sutherland)
Reporting Deadlines. The Proposed Rule would change the reporting deadlines to enable the use
of a unified and simplified reporting form and reduce the number of deadlines facing regulated
parties. EPA solicited comments regarding whether the ability to transfer Renewable
Identification Numbers ('RINs') in EMTS for the purpose of satisfying Renewable Volume
Obligations ('RVOs') under the RFS from the previous compliance year should be extended to
March 31. Sutherland believes this is a logical extension, and encourages EPA to extend the
deadline to transfer such RINs to March 31.
Our Response:
EPA is finalizing the proposed changes to the RFS2 annual report date from the last day
of February to March 31. EPA intended the change would 1) alleviate staffing problems for
regulated entities because the regulatory requirements are often handled by the same personnel
and 2) provide relief from the urgent need to obtain RINs when small discrepancies in gasoline
production or import volumes are realized. Streamlining reporting dates will allow EPA to
develop a single, user-friendly, electronic form that will collect all required data, maximizing the
capability of electronic reporting to provide reuse of data and avoid duplicate data submission.
Commenters generally agreed with the proposed date change. One commenter pointed out that
the staffing issue described would not be addressed by the change date because EPA is changing
both RFG and Anti-dumping compliance deadlines to March 31 - same as the RFS2 annual
reporting. EPA understands the point made but nonetheless, believes that delaying the RFS
reporting date will give reporting entities additional time and flexibility to review data and will
have no impact on emissions, air quality, or compliance with the standard. EPA also believes that
the overarching goal of streamlining reporting dates to develop a single electronic reporting
format, in and of itself, will provide general relief to regulated parties. Comments received
generally support this view.
EPA requested comment on whether or not the same date extension from the last day of
February to March 31 should apply to those transferring RINs in EMTS for satisfying RVOs
under the previous compliance year. Some commenters wanted to have RIN trading through the
new deadline (March 31), while others wanted trading to end on the last day of February to allow
time for data cleanup which could improve compliance and reporting accuracy. One commenter
also pointed out that changing the RIN trading deadline might have an unforeseen impact on the
RIN market and suggested EPA carefully consider this possibility before making a decision on
extending the RIN trading date. EPA agrees that providing enough time for data cleanup is
6-31
-------
Chapter 6: Regulatory Streamlining and Technical Amendments
important to ensure reporting accuracy and meet compliance goals. In addition, EPA understands
stakeholders' concerns about how such a change in REST trading could conceivably impact the
RIN trading market dynamics. As such, RIN transfers in EMTS for satisfying RVOs under the
previous compliance year will continue to end on the last day of February.
6.1.2. Performance-Based Measurement Systems (PBMS)
6.1.2.1. Comments in Support of our Proposed PBMS Requirements
What Commenters Said:
All the comments we received were in support of PBMS.
Commenter: Chevron Products Company
Chevron supports to adoption of the Performance Based Measurement Standards concept.
Chevron actively participated in the development of the API/AFPM comments on this proposal.
We refer you to those comments as our detailed feedback on this section.
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM)
We commend EPA for embracing a performance-based approach to specifying analytical testing
requirements for fuel property measurements, as doing so allows for flexibility and encourages
innovation. We broadly support the adoption of a PBMS for fuels, concur with EPA's proposed
categorization of fuel parameters as absolute- or method-defined, and urge the Agency to
expedite its effort in extending PBMS to on-line analytics and automated sampling.
We support the proposal of requiring Agency Approval for only non-VCSB methods. The dates
and timing stated in the Preamble discussion section entitled "Agency Approval of Only Non-
VCSB Methods" and the dates contained in the proposed regulatory text on this topic, for the
two sections are confusing and inconsistent. In addition, the exemption language in the two
sections is not clear. Specifically, the Preamble states "We are also proposing to exempt existing
(i.e., in use for six months prior to publication of this proposal) installations of designated test
methods that are method-defined parameters from the qualification requirement." We support the
exemption clause as stated in the proposed regulatory language in §80.47 and suggest that the
Preamble be written more clearly in order to properly reflect this text.
Comment on whether the Agency should require qualification of all analytical test methods for
the fuel parameters at 40 CFR 80. (78 Federal Register p. 29965)
We do not support qualification requirement for all analytical test methods as suggested above.
Commenter: Marathon Petroleum Company LP (MFC)
6-32
-------
Tier 3 Summary and Analysis of Comments
MFC supports adopting the Performance Based Measurement Systems proposal although we
have provided some clarifications. We commend EPA for embracing a performance-based
approach to specifying analytical testing requirements for fuel property measurements, as doing
so allows for flexibility and encourages innovation.
Qualification criteria for designated test method installations that are "Method-Defined"
parameters instruments and not used to qualify other "Method-Defined" methods (p. 29961): We
agree with EPA's proposal for only requiring implementation of a Statistical Quality Control
(SQC) program as the sole qualification criterion for Designated Test Method installations that
are not used to qualify alternate method-defined parameter instruments.
Qualification criteria for method defined parameter instruments other than designated test
methods (p. 29961): We agree with the proposed qualification criteria for Method Defined
Parameter Instruments for VCSB Method-Defined Parameter Test Methods.
Statistical Quality Control for non-VCSB methods used to measure method-defined parameters:
We support the proposal of requiring Agency Approval for only non-VCSB methods.
Commenter: PBF Energy Inc.
PBF supports the inclusion of the proposed Performance Based Measurement Standard (PBMS)
in the new rule for sulfur measurement.
Commenter: Shell Oil Products for Shell and Motiva
In general, we support the PBMS concept.
Commenter: Phillips 66 Company
Performance Based Measurement Systems - We are supportive of this effort - some
modifications are suggested.
We are appreciative of the effort EPA has put forth to include a PBMS proposal in this
rulemaking. The industry has been working collaboratively with EPA for many years to develop
this concept - one that was first put into practice with the ultra-low sulfur diesel rulemaking. We
support the proposal to exempt existing installations of designated test methods that have been in
use for six months. As EPA states, requiring their qualification could be disruptive and
burdensome to those operators and are already being used to certify fuels.
Our Response:
We acknowledge the support. For those companies that participated in the issuance of
the API/AFPM comments on PBMS compliance deadlines, applicability of PBMS to in-line
blending and specific PBMS requirements, please see below for our response to these
AFI/AFPM comments.
6-33
-------
Chapter 6: Regulatory Streamlining and Technical Amendments
6.1.2.2. Comments on the Proposed Compliance Period for PBMS
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM)
A one year grace period for ASTM D6708 Assessments on Voluntary Consensus-based
Standards Body (VCSB) alternate method candidates for method-defined fuel parameters is
inadequate. We suggest a minimum grace period of 18 months, given the time required to plan
and complete such an endeavor.
Commenter: Marathon Petroleum Company (MFC)
The dates and timing stated in the Preamble discussion section entitled "Agency Approval of
Only Non-VCSB Methods" and the dates contained in the proposed regulatory text on this topic,
for the two sections are confusing and inconsistent. In addition, the exemption language in the
two sections is not clear. Specifically, the Preamble states "We are also proposing to exempt
existing (i.e., in use for six months prior to publication of this proposal) installations of
designated test methods that are method-defined parameters from the qualification requirement."
We support the exemption clause as stated in the proposed regulatory language in §80.47 and
suggest that the Preamble be written more clearly in order to properly reflect this text.
Commenter: Phillips 66 Company
The proposed timing of November 30, 2014 for labs to have completed their qualification of test
methods is too aggressive. A lab could be currently using multiple alternative methods that will
need to be qualified under these new provisions. We would anticipate some modifications to the
proposed rule based on comments received. Labs would wait for final rulemaking and additional
needed detailed guidance before beginning the qualification process. There is a need for further
clarification and implementation guidance beyond what is in the proposed rule in order to
develop uniform qualification processes. Also the qualification process for some methods
requires reference instruments to meet certain criteria for at least five months prior to
application. A minimum of 18 months from publication of the final rule is a much more
reasonable time requirement for labs to complete the test qualification process.
Commenter: Shell Oil Products for Shell and Motiva
We do believe, however, that in some ways the proposal is unnecessarily complicated and should
be simplified. We agree with the proposal to exempt existing (in use for six months prior to
publication of the proposal) installations of designated test methods that are method-defined
parameters from the qualification requirement. We also agree with the proposed 40CFR
80.47(b)(3) to exempt existing (in use for six months prior to publication of the proposal)
installations of D2622 from the accuracy and precision qualification requirements.
6-34
-------
Tier 3 Summary and Analysis of Comments
Our Response:
PBMS will be effective January 1, 2016, see Section VI.A.2.h of the Tier 3 final rule as
well as 40 CFR 80.47. The Agency has chosen the approval option of only requiring Non-VCSB
test methods to receive Agency approval starting January 1, 2016. VCSB test methods will be
self qualified either through their respective VCSB or on a site specific basis. Please see Section
VI.A.2.h of the Tier 3 final rule preamble as well as the final regulations at 40 CFR 80.47.
6.1.2.3. Applicability of PBMS to in-line Blending Methods
What Commenters Said:
We received several comments requesting our extension of PBMS to in-line blending
methods.
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM); Marathon Petroleum Company LP (MFC)
EPA should expand PBMS to include sampling or in-line blending methods: Extension of the
PBMS to sampling and in-line blending is a logical extension of this performance-based
approach to the sampling and analysis of physical and chemical properties. We note that most if
not all in-line analytical instrumentation is equal or superior to laboratory-based test methods in
terms of precision as well as provision of a more representative analysis for the complete batch,
as opposed to a laboratory analysis on an aliquot of a batch taken from a tank. Superiority in
precision as well as overall representativeness of the batch is achieved by virtue of taking the
average of many in-line analytical results during the manufacturing process.
We note that there is a series of ASTM Standard Practices and Guides that prescribes industry-
consensus best practices associated with the automated sampling and in-line analysis using
process analyzers. See ASTM D7825 Standard Guide for Generating a Process Stream Property
Value through the Application of a Process Stream Analyzer. We urge EPA to expedite its effort
in extending PBMS to on-line analytics and automated sampling.
Our Response:
The final requirements apply to the qualification of analytical test instrumentation and
methods used to measure various characteristics of individual fuel samples. Consistent with our
proposal, it does not apply to sampling methods or in-line blending methods. This is because, in-line
blending already has a certification process that sets forth qualification criteria that take into account
the unique combinations of sampling, control, and analysis that are involved with in-line blending.
See 40 CFR 80.65(f).
6.1.2.4. Categorization of Fuel Parameters as Absolute or Method Defined
What Commenters Said:
6-35
-------
Chapter 6: Regulatory Streamlining and Technical Amendments
We received several comments in support of our categorization of fuel parameters as
either absolute or method defined.
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM)
We commend EPA for embracing a performance-based approach to specifying analytical testing
requirements for fuel property measurements, as doing so allows for flexibility and encourages
innovation. We broadly support the adoption of a PBMS for fuels, concur with EPA's proposed
categorization of fuel parameters as absolute- or method-defined qualification criteria for method
defined parameter instruments other than designated test methods (78 Federal Register p. 29961).
We agree with the proposed qualification criteria for Method Defined Parameter Instruments for
VCSB Method-Defined Parameter Test Methods.
Commenter: Marathon Petroleum Company LP (MFC)
Categorization of fuel parameters as absolute or method defined: We support the proposed
categorization of fuel parameters. We further support designation of Sulfur as the only Absolute
parameter.
Our Response:
The Agency is finalizing the categorization of fuel parameters as absolute or method
defined as proposed. The absolute fuel parameters are sulfur in gasoline and sulfur in butane.
All other fuel parameters besides sulfur have been finalized as method defined fuel parameters.
The following fuel parameters are method defined fuel parameters: Olefms, Reid Vapor Pressure
(RVP), Distillation, Benzene, Aromatic content of gasoline and diesel fuel, and Oxygen and
oxygen content analysis.
6.1.2.5. Comments on the PBMS Requirements
EPA Should Not Extrapolate Precision Qualification Criteria for Absolute Parameters to
Those Used for Method-Defined Parameters.
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM); Marathon Ashland Petroleum LP (MFC)
We commend EPA for embracing a performance-based approach to specifying analytical testing
requirements for fuel property measurements, as doing so allows for flexibility and encourages
innovation. While we broadly support the adoption of a PBMS for fuels as proposed by EPA, we
do have a number of questions and concerns regarding the specific details as spelled out in the
Notice of Proposed Rulemaking. Briefly, several of our main concerns include:
6-36
-------
Tier 3 Summary and Analysis of Comments
EPA should not extrapolate Precision Qualification Criteria for Absolute Parameters to those
used for Method-Defined Parameters. We recommend that the precision standard deviation
qualification criterion for method-defined parameters be based on a Test Performance Index
(TPI) approach as per ASTM D 6792 Standard Practice for Quality System in Petroleum
Products and Lubricants Testing Laboratories. We disagree with extrapolating the same
precision qualification criteria for absolute parameters based on published method repeatability
(r) to method-defined parameters. The basis for our disagreement is similar to EPA basis for
recognizing the need for method-defined parameters. Specifically, the precision criteria of
method-defined parameters are sensitive to the matrix of the material. This degree of sensitivity
is different for different test methods / techniques / instrumentation that claim to measure the
same property. We recommend that the precision standard deviation qualification criterion for
method-defined parameters be based on a Test Performance Index (TPI) approach as per ASTM
D 6792 Standard Practice for Quality System in Petroleum Products and Lubricants Testing
Laboratories. Using the TPI as outlined in ASTM D6792 also is consistent with OMB Circular
119 which directs agencies to use voluntary consensus standards in lieu of government-unique
standards except where inconsistent with law or otherwise impractical. The TPI approach in
ASTM D6792 sets minimum site precision performance criteria based on test method
reproducibility (R) and the Precision Ratio (PR) of the published test method.
Our Response:
EPA has adopted the TPI approach for setting precision criterion for method defined
parameters, see Section VI.A.2.c.iii of the preamble to the Tier 3 final rule as well as 40 CFR
80.47.
EPA's Proposed Accuracy Qualification Requirements for Reference Installations are Overly
Restrictive
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM); Marathon Ashland Petroleum LP (MFC)
EPA has proposed Accuracy Qualification Requirements for Reference Installations and for
Designated Method Installations used to qualify method-defined parameter instruments which
are overly restrictive. We are concerned that the proposed requirement to stay within the middle
50% of the distribution of measurements of the industry monthly inter-laboratory crosscheck
program for at least 5 months will severely restrict the number of participants. We suggest that a
requirement of 3 out of 5 successive exchanges is more realistically achievable. The proposed
qualification criteria for designated method installations used to qualify method-defined
parameter instruments should be relaxed to be more realistically achievable.
Our Response:
6-37
-------
Chapter 6: Regulatory Streamlining and Technical Amendments
EPA has changed the requirement that reference installations of the designated test
method should be shown to be in the middle 50% of the distribution measurements of an industry
monthly inter-laboratory crosscheck program for 3 out of 5 successive exchanges. See Section
VI.A.2.d of the preamble to the Tier 3 final rule as well as 40 CFR 80.47.
Sites Should Be Granted Greater Flexibility in the Choice of Procedures to Comply with the
Proposed Statistical Quality Control (SQC) Requirements
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM); Marathon Ashland Petroleum LP (MFC)
Sites should be granted greater flexibility in choosing procedures to comply with the proposed
Statistical Quality Control (SQC) Requirements. We suggest that a site should be given the
option of using either one of the two SQC procedures outlined in ASTM D6299, and not
mandated to use both.
Our Response:
The Agency agrees the Q-procedure is functionally equivalent to the I-procedure and that
laboratories should be given the flexibility to use either of these two procedures instead of both,
as proposed.
Recommend that EPA Specify ASTMD86-07
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM); Marathon Ashland Petroleum LP (MFC)
Table VI-6 (on p. 29958) lists EPA's proposed precision criteria for Method-Defined Fuel
Parameters with no Alternatives to the Designated Test Method. For the gasoline distillation fuel
parameter, we recommend that EPA specify ASTM D86-07 for the following reasons:
The precisions as published in later versions are not consistently supportable by actual ASTM
ILCP program data; this is clearly stated in Note 31 of the current ASTM D86 test method:
'NOTE 31—A new inter laboratory study is being planned to address concerns that laboratories
are not able to meet the precision for percent evaporated temperature at fifty percent.'
A plausible explanation is that the ILS study used to derive the current precision required several
runs on the ILS material to select the heating profile, and hence does not reflect how the method
is actually conducted in routine production environment. In addition, the ILS study sample set
may not adequately cover the range of real world production gasoline available.
6-38
-------
Tier 3 Summary and Analysis of Comments
There is general consensus that the precision of this test method is sensitive to the slope of the
boiling curve; hence, a constant precision as articulated in the cited ASTM D86-12 method may
not be appropriate as a 'one-size-fit-aH' measure for gasolines with different matrices and
volatility properties (winter versus summer). Directionally, based on on-going discussions in
ASTM, it appears that the ASTM test method precision may revert to a boiling curve slope based
approach, similar to version ASTM D86-07.
For gasoline containing 10% ethanol by volume (herein referred to as E-10), the boiling curve
slope and hence precision is impacted by the location of the azeotrope point relative to the
distillation points of interest (T10, T50, and T90). The azeotrope point is a function of the base
stock (neat gasoline) composition, and therefore can vary with different matrices. We believe
that precision function as stated in D86-07 is a more realistic representation for E-10 gasoline
precision with different neat matrices.
To support our concern that use of D86-12 precision and 1.5r is too restrictive, see the example
of control chart data in Figure 15 [of EPA-HQ-OAR-2011-0135-4276-A2] supplied by a
producer.
Our Response:
EPA has based the distillation precision criterion on ASTM D86-07, see preamble
Section VI.A.2.c.iii and Table VI-9, as well as 40 CFR 80.47. In addition, the accuracy
demonstration must be based on the reference installation of the ASTM D86-07 test method
along with use of ASTM D6708 in determining if a correction equation is warranted.
EPA's Proposed Accuracy Qualification Requirements for Reference Installations Are Overly
Restrictive
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM); Marathon Ashland Petroleum LP (MFC)
For a single entity that wishes to qualify alternate test methods per ASTM D6708 by using a
single reference installation, we believe that the 'middle 50 percent' requirement as proposed by
EPA is a reasonable requirement to impose on the single reference installation. However, we are
concerned that the proposed requirement to stay within the middle 50% of the distribution of
measurements of the industry monthly inter-laboratory crosscheck program for at least 5 months
is overly restrictive. Based on a work up (provided as a separate attachment to these comments)
of the proposed requirements (as we understand them) using ASTM D5599 Total Oxygen results
on eleven RFG distributions, RFG1205 through RFG1303, we note that less than 15% of the
participants met EPA requirement of staying within the central 50 percent for 5 successive
exchanges. We specifically note that the EPA lab (lab 47) failed to meet this requirement.
Therefore, we suggest a requirement of 3 out of 5 successive exchanges is more realistically
6-39
-------
Chapter 6: Regulatory Streamlining and Technical Amendments
achievable. We have attached the spreadsheet and description of the work up with these
comments. See Attachments No. 4 and 5.
The proposed qualification criteria for designated method installations used to qualify method-
defined parameter instruments should be relaxed to be more realistically achievable (78 Federal
Register p. 29960).
We support in principle the qualification criteria described in the Preamble (at Section VI.3.e) for
sites that intend to qualify other Method-Defined parameters using only a single designated
method installation. However, as per our comment above in section IV.2 on Precision
Qualification criteria, we believe that a more appropriate and realistically achievable standard
deviation under site precision conditions for the method-defined parameter for both the
designated method and alternate method installation, even by a 'good' lab, should be set based
on the TPI approach in ASTM D 6792.
We agree in principle with the additional accuracy requirement for a single Designated Method
installation used to qualify alternate method-defined parameter instruments. However, as noted
in an earlier comment above, we disagree with the requirement of staying in the middle 50
percent for a minimum of 5 successive exchanges, and we suggest that 3 out of 5 successive
exchange requirements is more realistically achievable.
Our Response:
EPA has changed the requirement that reference installations of the designated test
method should be shown to be in the middle 50% of the distribution measurements of an industry
monthly inter-laboratory crosscheck program for 3 out of 5 successive exchanges. See section
VI.A.2.d of the Tier 3 final rule as well as 40 CFR 80.47.
ILCPData
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM); Marathon Ashland Petroleum LP (MFC)
For the situations where ILCP (such as those conducted by ASTM CS92) data and summary
statistics for VCSB designated and alternate methods exist for the same materials, we believe
that the summary statistics (mean and standard error = standard deviation / square root [ no. of
results]) from these ILCP data can be used as is, i.e., without imposing the reference installation
criteria, to conduct an ASTM D6708 assessment on VCSB alternate test methods, provided that
the number of non-outlying results is >16 for both designated and alternate methods, since this is
the current de facto methodology for determination of ARV of check standards as specified in
ASTM D6299, clause 6.2.2.1 and Note 7. Therefore, per OMB Circular 119, we suggest that the
ASTM D6299 protocol for establishing ARV be followed. We note also that in actual fact,
6-40
-------
Tier 3 Summary and Analysis of Comments
ASTMILCP data for the method-defined parameters of interest exceeds this number (16)
significantly.
We note that it is neither necessary nor is it statistically justified to apply the reference
installation precision and 'middle 50 percent' criteria to the ILCP data for designated test method
because the relevant ILCP statistics are calculated using outlier-free data, and, the number of
data points is large, hence providing a better statistical 'sample' of the laboratory population. The
mean calculated using the full ILCP, outlier-free data set is a 'truer' representation of the
population parameter (u) than the mean calculated using only the middle 50 percent. We note the
standard error for the arithmetic mean calculated using the full ILCP data set is significantly
reduced due to the square root [number of non-outlying results] term in the denominator for
calculation of standard error.
We urge that EPA clearly state that the use of ILCP data as described above is suitable for an
ASTM D6708 assessment of VCSB alternate test methods.
Our Response:
The Agency has made the change in the final regulations to allow VCSBs to make use of
ILCP data for conducting ASTM D6708 assessments for method defined fuel parameters.
EPA Should Provide an Example of an Acceptable ASTMD6708 Assessment
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM); Marathon Ashland Petroleum LP (MFC)
We request that EPA provide a worked example of what the Agency deems to be an acceptable
ASTM D6708 assessment. Doing so will provide valuable guidance to the regulated community
with respect to understanding and implementing the provisions of the PBMS as outlined in the
proposed rule.
Accuracy - the average of 10 consecutive results on a 1-10 ppm gravimetric sulfur standard
cannot be more than 0.71 ppm and the average of 10 consecutive results on a 10-20 ppm
gravimetric sulfur standard cannot more than 1.00 ppm."
To be consistent with the language in the NPRM, this statement should be revised as follows:
"Accuracy - the average of 10 consecutive results on a 1-10 ppm gravimetric sulfur standard
cannot be differ by more than 0.71 ppm from the ARV of the standard and the average of 10
consecutive results on a 10-20 ppm gravimetric sulfur standard cannot differ by more than 1.00
ppm from the ARV of the standard." [Note: Strikeout = deleted text. Underline = added text.]
6-41
-------
Chapter 6: Regulatory Streamlining and Technical Amendments
We appreciate the effort undertaken by EPA to develop flowchart examples of qualification
requirements for Absolute Fuel Parameters and Method Defined Fuel Parameters and to place
this material in the Docket for public comment. (See EPA-HQ-OAR-2011-0135-17841.) While
these flowcharts are helpful, we urge EPA to develop more detailed schematics for inclusion in
the final rule, as such information will be extremely helpful in assisting our members to
implement the PBMS provisions. We also note that there is an error in the flowchart labeled
"Flow Chart Example: Absolute Fuel Parameter (Absolute) - Sulfur in Gasoline" in Docket item
EPA-HQ-OAR-2011-013 5-17841.
Our Response:
The flowchart examples have been corrected and placed in the docket to reflect the
editorial comments for absolute fuel parameters and also to reflect specific comments the
Agency has accepted in regards to method defined fuel parameters. For the absolute fuel
parameter of sulfur, PMBS has been successfully demonstrated by over 800 test facilities for
sulfur in diesel fuel. Therefore we believe the regulated industry is familiar with how to meet the
accuracy and precision criterion for the absolute fuel parameters of sulfur in gasoline and sulfur
in butane. For the method defined fuel parameter, such an example already exists in ASTM
D6708 that is available to the industry and with which they are already familiar. If further
explanation is needed for either the absolute or method defined fuel parameter approaches, the
Agency can do so through future discussions and guidance.
EPA Should Remove the Term 'Robust' from the Final Rule Preamble and Regulatory
Language
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM); Marathon Ashland Petroleum LP (MFC)
We note that the current 'robust' outlier treatment methodology for the ASTM CS92 ILCP
program will be replaced with a statistically more rigorous approach using the Generalized
Extreme Studentized Deviation (GESD) technique (37). We suggest that EPA remove the term
'robust' from the Regulation (and Preamble) wording.
Our Response:
EPA agrees with this comment and has thus removed the term "robust" from the
applicable final regulations.
Sites Should be Granted Greater Flexibility in the Choice of Procedures for Statistical Quality
Control (SOC) Requirements (78 FR 29962)
6-42
-------
Tier 3 Summary and Analysis of Comments
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM); Marathon Ashland Petroleum LP (MFC)
We agree in principle with the SQC requirements, and that each instrument should be under its
own SQC oversight. We note that in ASTM D6299, for the handling of QC material batch
transition, the Q-procedure is intended to be an alternate approach to the concurrent testing
(overlap) protocol. We suggest that the site should be given the option of using either one of the
two procedures, and not mandated to use both. The Q-procedure is technically equivalent to the
I-procedure. We suggest that for sites opting to use the Q-procedure, the very first run on the
new QC batch should be validated by either an overlap in-control result of the old batch, or, by a
single execution of an accompanying SRM. The new result is considered validated if the single
result of the SRM is within the established site precision (R') of the ARV.
We suggest that because the standard error of the ARV in consensus-named fuels may not in all
cases be negligible when compared to 0.75R, the expanded uncertainty of the ARV should be
incorporated into the accuracy qualification criterion as follows:
Accuracy qualification criterion = sqrt [(0.75R)A2 + (0.75R)A2/L], where L = the number of
single results obtained from different labs used to calculate the consensus ARV.
Our Response:
The Agency agrees the Q-procedure is functionally equivalent to the I-procedure and that
laboratories should be given the flexibility to use either of these two procedures instead of both
of these two procedures as proposed in the Statistical Quality Control requirements for PBMS.
The Agency also agrees for sites opting to use the Q-procedure, the very first run on the new QC
batch should be calibrated by either an overlap in-control result of the old batch, or by a single
execution of an accompanying standard reference material (SRM). Thus the new result would be
considered validated if the single result of the SRM is within the established site precision of the
ARV of the SRM. We have revised the regulations to reflect these changes. See preamble
Section VI. A.2.g as well as 40 CFR 80.47.
Qualification Criteria and Test Method Approval Date
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM); Marathon Ashland Petroleum LP (MFC)
We support in principle the notion of requiring laboratories to meet prescribed qualification
requirements for specifically cited measurement data quality assurance. We suggest that this can
be further streamlined by selecting a subset of the regulated parameters for the purpose of
demonstrating to the agency measurement capability, with the expectation that the quality
6-43
-------
Chapter 6: Regulatory Streamlining and Technical Amendments
assurance work process and oversight for this subset of parameters are extended to all methods
used to take measurement for regulatory purposes.
The Preamble and the proposed Regulatory language are inconsistent with respect to the
exemptions from the precision criteria for approval of the method defined fuel parameters
granted for test methods in use prior to May 30, 2014. We support the exemption language stated
in the proposed regulatory text at §80.47(b)(3), §80.47(c)(3), §80.47(d)(2), §80.47(e)(2),
§80.47(f)(2), §80.47(g)(2). §80.47(h)(2), §80.47(h)(2), §80.47(i)(2), and §80.47(j)(2) and we
suggest that the Preamble be revised in the final rule to more clearly reflect this wording.
The proposed regulatory language exempting VCSB method-defined parameter test methods in
use prior to May 30, 2013 (§80.47(1)(4)) is different than the May 30, 2014 date specified in all
of the other precision criteria exemptions contained in the proposed regulatory text at §80.47 and
noted above. Is this different date intentional?
Our Response:
The Agency is going forward with the exemption date for designated test methods as
explained in the proposed preamble and has made this correction in the final regulations. The
Agency has chosen the approval option of only requiring Non-VCSB test methods to receive
Agency approval starting January 1, 2016. VCSB test methods will have the option to either self
qualify either through their respective VCSB or on a site specific basis. Please see Section
VI.A.2.h of the Tier 3 final rule preamble as well as the final regulations at 40 CFR 80.47.
Requirement for Test Method Qualification Applications to Include a Complete Operational
Description of the Test Method in Question
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM); Marathon Ashland Petroleum LP (MFC)
If a VCSB method is used to measure and qualify either an absolute or a method-defined fuel
parameter, we believe that a simple citing of the VCSB Test Method Number and Title should
provide EPA with sufficient documentation with respect to its operational description We note
that there is an inconsistency between the language of the preamble and that of the proposed
regulation: the proposed regulation requires full documentation (see §80.47(1)(1)), while the
preamble suggests otherwise. The final rule should specify that Test Method Qualification
Applications need only to cite the VCSB by test method number and title.
Our Response:
The Agency has chosen the approval option that only Non-VCSB test methods will be
submitted for approval to that Agency. VCSB test methods will either qualify through their
respective VCSB or on a site specific basis. The Agency agrees that during a potential audit,
6-44
-------
Tier 3 Summary and Analysis of Comments
reference to a particular VCSB test method designation number should suffice for the method
utilized through self-qualification by that particular facility as long as the VCSB follows the
requirements specified at 40 CFR 80.47(1). The Agency agrees that a simple citing of the ASTM
test method designation number would suffice for requirements of the operational description of
an ASTM test method. Please see 40 CFR 80.47.
Temporal Distribution of Precision Tests
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM); Marathon Ashland Petroleum LP (MFC)
We support option 2 (arranging tests into no fewer than five batches of five or fewer tests each,
with only one such batch allowed per day) since it provides the most flexibility and is easier to
implement.
Our Response:
The Agency has chosen option 2 (arranging tests into no fewer than five batches of five
or fewer tests each, with only one such batch allowed per day) for temporal distribution of
precision tests. Please see 40 CFR 80.47.
Statistical Control Requirements Governing the Operation of Reference Installations
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM); Marathon Ashland Petroleum LP (MFC)
We support EPA's proposal that the reference installation "...must be shown to be in statistical
control, as provided for in ASTM D6299-10el ... and that the applicant must submit control
charts showing a record of in-control operation for at least five months" but only with the
proviso that:
Regular maintenance and/or re-calibration conducted during the 5 month in-control qualification
period is considered as part of in-control normal operation, and
If an assignable cause for 'out of control' is found, mitigated, and the system is brought back in-
control during the period that the reference installation is attempting to meet the 5 month in-
statistical-control requirement, the 'clock' for the 5 month period does not restart. In other
words, the system is still considered as being 'in control'.
Our Response:
6-45
-------
Chapter 6: Regulatory Streamlining and Technical Amendments
The Agency agrees and has made this change to the regulations. Please see applicable
Statistical Quality Control requirements at 40 CFR 80.47.
Use of Reference Materials in Qualifying and Maintaining Alternative Analytical Techniques
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM); Marathon Ashland Petroleum LP (MFC)
We support the philosophy and principle behind the use of the three types of Standard Reference
Materials (SRMs) as discussed in the proposed rule. We suggest that for the Consensus-named
fuels (SRM), EPA confirms that the relevant clauses described in ASTM D6299 (6.2) are
acceptable in the determination of the Accepted Reference Value (ARV).
Use of reference materials in qualifying and maintaining alternative analytical techniques (p.
29960): We support the philosophy and principle behind the use of the three types of Standard
Reference Materials (SRMs) as discussed in the proposed rule. We suggest that for the
Consensus-named fuels (SRM), EPA confirms that the relevant clauses described in ASTM
D6299 (6.2) are acceptable in the determination of the Accepted Reference Value (ARV).
Our Response:
The regulations state that a facility conduct tests on every instrument with a
commercially available gravimetric reference material, or check standard as defined in the
ASTM D6299 at least 3 times a year using good laboratory practices.
5O) Qualification criteria for designated test method installations that are "Method-Defined"
parameters instruments and not used to qualify other "Method-Defined" methods (78 Federal
Register p. 29961)
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM); Marathon Ashland Petroleum LP (MFC)
We agree with EPA's proposal for only requiring implementation of a Statistical Quality Control
(SQC) program as the sole qualification criterion for Designated Test Method installations that
are not used to qualify alternate method-defined parameter instruments.
Our Response:
6-46
-------
Tier 3 Summary and Analysis of Comments
We acknowledge the support of this comment.
Test Performance Index
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM); Marathon Ashland Petroleum LP (MFC)
Test Performance Index (TPI) is defined in ASTM D6792 as follows: "an approximate measure
of a laboratory's testing capability, defined as the ratio of test method reproducibility to site
precision." 37 - Rosner, Technometrics, Vol. 25, May 1983
Our Response:
The Agency has incorporated by reference ASTM D6792 into the regulations.
Reference Installations
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM); Marathon Ashland Petroleum LP (MFC)
We support EPA's proposal that the reference installation ".. .must be shown to be in statistical
control, as provided for in ASTM D6299-10el ... and that the applicant must submit control
charts showing a record of in-control operation for at least five months" but only with the
proviso that:
-Regular maintenance and/or re-calibration conducted during the 5 month in-control qualification
period is considered as part of in-control normal operation
-If an assignable cause for 'out of control' is found, mitigated, and the system is brought back in-
control during the period that the reference installation is attempting to meet the 5 month in-
statistical-control requirement, the 'clock' for the 5 month period does not re-start. In other
words, the system is still considered as being 'in control'.
Our Response:
The Agency agrees and has made this change to the applicable accuracy and precision
statistical control section of the regulations at 40 CFR 80.47(n)(2)(ii), 40 CFR 80.47(o)(2)(ii) and
40 CFR 80.47(p)(2)(ii).
Repeatability Criterion Alternatives
6-47
-------
Chapter 6: Regulatory Streamlining and Technical Amendments
What Commenters Said:
Commenter: PBF Energy Inc.
PBF will evaluate the option provided in the proposal to voluntarily convert other RFG test
methods to PBMS. We are aware that some of the methods may have higher variability and it
may be difficult to meet the precision qualification measurement requirements as proposed. EPA
should examine alternatives to the 1.5 times repeatability criterion discussed in the proposal if it
is desired to have more refiner participation in the voluntary portion of the program.
Our Response:
EPA has adopted the TPI approach for setting precision criterion for method defined
parameters, see Section VI.A.2.c.iii of the preamble to the Tier 3 final rule as well as 40 CFR
80.47.
6.1.2.6. Comments on the Proposed Regulatory Text
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM); Marathon Petroleum Company LP (MFC)
We also agree with the proposed qualification criteria for Method Defined Parameter Instruments
for non-VCSB Method-Defined Parameters. However, we believe that in addition to having the
degrees in Chemistry or Statistics, qualifications for third party oversight service providers
should also have a good working knowledge of ASTM D6708 and ASTM D6299. In addition,
believe that limiting the third party oversight qualification to only US degree holders will
exclude non-US degreed subject matter experts with equivalent knowledge and qualification. We
suggest that the wording be expanded to include non-US equivalent degreed or industry
recognized subject matter experts.
Commenter: Turner, Mason & Company
Under the proposed regulations, after November 29, 2014, the determination of sulfur in gasoline
must be performance based, and other gasoline tests may be performance based. The
performance based approval procedures are similar to those for the diesel sulfur determination.
However, the EPA has added the additional step that the performance based procedures require
an independent third party scientific review and written report, and verification of the
information provided. The report and verification requires a site visit and review by an
independent chemist and statistician who has received at least a bachelor's degree in science
from an accredited college in the United States, with professional work experience in the
petroleum or oxygenate field.
6-48
-------
Tier 3 Summary and Analysis of Comments
We propose that a chemical engineer with demonstrated experience in analytical techniques can
perform the same function. Chemical Engineers have a strong background in chemistry, and,
based on trends within industry, they will also have experience in statistical process control.
Thus, we propose that the EPA should allow chemical engineers who can demonstrate their
experience to be acceptable for providing an independent third party review and verification.
Our Response:
EPA agrees and has amended the final PBMS regulations to include chemical engineers
and non-US degreed subject matter experts with equivalent knowledge and qualifications as third
party oversight reviewers. See Section VI.A.2.f of the Tier 3 final rule preamble and 40 CFR
80.47.
What Commenters Said:
Commenter: Shell Oil Products for Shell and Motiva
We do believe, however, that in some ways the proposal is uncessarily complicated and should
be simplified. The Agency requested comments on several items. For Agency Approval of Only
Non-VCSB Methods, 78 Fed. Reg. 29964, we concur that only Non-VCSB methods need agency
approval for method-defined parameters and absolute fuel parameters. Documentation should be
submitted to the agency only for non-VCSB methods for qualifications for method-defined
parameters and absolute fuel parameters.
In regards to qualification of test methods, id., we believe the Agency should not require
qualification of all analytical test methods for the fuel parameters in 40 CFR 80.
In the proposed 40 CFR 80.47(k)(2), pertaining to criteria for reference installations, it mentions
the use of an industry monthly inter-laboratory crosscheck program. In order to have the option
to use other crosscheck programs besides ASTM, we recommend that the criteria for the
reference installations refer to the use of a commercially available monthly inter-laboratory
crosscheck program.
Our Response:
The Agency is finalizing the exemption date for designated test methods as explained in
the proposed preamble and has made this correction in the final regulations. The Agency has
amended the regulations to allow commercially available inter-laboratory crosscheck program
data for meeting PBMS qualification requirements described in final regulations at 40 CFR
80.47(k).
What Commenters Said:
6-49
-------
Chapter 6: Regulatory Streamlining and Technical Amendments
Commenter: International Federation of Inspection Agencies (IFIA), Americas Petroleum
Regional Technical Committee (APTRC)
It appears that EPA is proposing to add precision and accuracy requirements for tests that are put
into use after May 30, 2014. These requirements are summarized in Table 1 [of EPA-HQ-OAR-
2011-0135-4292-A1].
There does appear to be any exclusions for Alternate Test Methods in Section 80.47 ((i)(4) and
also in Sections ((b)(3), (d)(2), (f)(2), (g)(2), (h)(2), (i)(2). IFIA recommends that these sections
be excluded from this requirement as well.
Section 80.47 (k)(2): IFIA interprets this section that reference installation of method defined
fuel parameters (i.e., excluding Sulfur Test Methods) must be shown to be within an acceptable
tolerance range of consensus value averages in Proficiency Testing Programs. They also must be
monitored by SQC in accordance with D6299.
Based upon this interpretation is appears that correlations of Alternate Test Methods to Primary
Test Methods may only be performed using a Reference Installation. IFIA recommends that
there be a provision for correlating to Primary Test Method ARV's of ILCP samples as well.
Section 80.47 (l)(l-4): This section states that for test methods that are not listed as being a
Primary Test Method (including Alternate Test Methods, VCSB Test Methods that are not listed
as being an Alternate Test Method and Non-VCSB Test Methods), their precision should be
evaluated according to sections (b) through (i) and accuracy should be evaluated according to
D6708. Also, regarding the accuracy check, if the difference is "null" then a correlation need not
be applied but if "not null" then a correlation needs to be applied to the Primary Test Method.
Based on the above section, it seems evaluation by D6708 is mandatory. However IFIA
recommends that it should be up to the laboratory to do a correlation in lieu of this, even if by
D6708 it would not be required. The reasoning for this is that as time elapses a lab may find
itself in and out of meeting D6708 tolerance requirements.
Sections 80.47 (1) and (2): IFIA requests confirmation of the following interpretations:
For Sulfur determinations, Gravimetric Reference Standards or Check Standards will need to be
analyzed at least quarterly and once 15 or more results are gathered both an I and MR Chart
should be constructed with control limits as defined in this section of the CFR.
That SQC be in place according to D6299 and that the Q procedure be used when replacing QC
Material.
Sections 80.47 (o)(l) and (o)(2): This section is applicable to the Non-Sulfur Test Methods and
basically the same as 80.47(n)(l), the accuracy check standard must be on an "ordinary fuel".
IFIA has two recommendations on this element of the section. Regarding the proposal that the
accuracy check standard be an "ordinary fuel", this may cause issues over time as the integrity of
a gasoline sample will likely be compromised. This is of particular concern for RVP and Olefins
where there will be a propensity for drift and also Distillation where there will be a sizeable
6-50
-------
Tier 3 Summary and Analysis of Comments
sample consumption. First, IFIA recommends that labs be allowed to plot the delta from the
ARV [Accepted Reference Value] (obtained value minus ARV) instead of the obtained value.
Second, instead of being mandated that this be done quarterly, IFIA would recommend three (3)
times per year so that this requirement is in-line with ASTM ILCP sample cycle frequency.
This section states that the ARV must be determined according to D6299. IFIA believes this is
an odd statement as the ARV is usually representative of the consensus value or mean (after
outliers have been rejected).
This section also includes a mandate for monitoring the average of the most recent result and the
one preceding it to determine if that average is greater than 0.75R. IFIA views this as overly
complex. IFIA proposes that each value be compared against R instead.
Below 80.46 (h)(l)(vii), it seems as though the next section should list the test method for RVP
(D5191-10bEPAEq).
IFIA recommends adding a Section 80.47 (a)(14) to further explain the term reference
installation as referred to in 80.47 (k). This is defined (at least somewhat) in the EPA Tier III on
page 29958, column 1 midway down, as a method used for qualifying an alternate test method.
Our Response:
Effective January 1, 2016, for the reasons discussed in the Tier 3 proposed and final rule,
all alternative test methods must meet the PBMS criterion described at 40 CFR 80.47. The
Agency has amended the regulations to allow the use of both commercially available and
voluntary consensus standards board inter-laboratory cross check data in meeting the PBMS
criterion described at 40 CFR 80.47. If a laboratory chooses to use either a VCSB or non-VCSB
method defined fuel parameter test method, in meeting the PBMS criterion for accuracy, the lab
must use ASTM D6708 to determine the need for a correction equation. Please see our final
regulations at 40 CFR 80.47. The Agency has made rule language changes to reflect that test
facilities can conduct verification of new QC material three times a year in order to align this
requirement with ASTM ILCP program sample cycle frequency.
What Commenters Said:
Commenter: Weaver and Tidwell LLP
i. As currently proposed, the primary regulatory methods and alternative regulatory methods
listed in 80.46 are exempted from the performance based testing requirements in 80.47. Is the
exemption for the methods listed in 80.46 dependent on 80.46 complying exactly with all the
requirements listed in those methods (i.e., apparatus, reagents, calibration, quality control,
procedures, calculations, etc.)? If the requirements of these methods are not followed exactly or
are modified in any manner, do the modifications constitute a new method that must be qualified
as required in 80.47?
6-51
-------
Chapter 6: Regulatory Streamlining and Technical Amendments
ii. Change the wording in 80.46(a)(l) - The wording in this section uses the word "must"
concerning the use of 80.47 after the appropriate date in 2014. Other paragraphs in this section
use the word "may". We request that EPA change the word in 80.46(a)(l) from "must" to "may"
for consistency.
Some specific modifications to the 80.46 method are listed below. Would any of the listed
modifications to the method individually or in combination be sufficient to void the exemption
from the performance based requirements of 80.47? As a secondary associated question, would
the EPA cite any of these modifications in enforcement actions?
(a) For olefm content (80.46(b)), ASTM D1319, Section 6.5 requires the use of a hypodermic
syringe with a needle having a length of 102 mm. Would the use of a hypodermic syringe with a
needle having a length of- 50mm be sufficient to void the exemption from the requirements of
80.47?
(b) For olefm content (80.46(b)), ASTM D1319, Section 7.2 requires that the fluorescent
indicator dyed gel be stored in a dark place under an atmosphere of nitrogen. Would the failure
to store the fluorescent indicator dyed gel in a dark place under an atmosphere of nitrogen be
sufficient to void the exemption from the requirements of 80.47?
(c) For olefm content (80.46(b)), ASTM D1319, Section 10.1 requires that the appropriate
amount of fluorescent indicator dyed gel (3-5 mm) be added to the analytical column. Would
the addition of an inappropriate amount of fluorescent indictor dyed gel (less than 3 mm or more
than 5mm) be sufficient to void the exemption from the requirements of 80.47?
(d) For olefm content (80.46(b)), ASTM D1319, Section 10.3 requires that the hypodermic
syringe be chilled to less than 4 °C. Would failure to chill the hypodermic syringe to less than 4
°C be sufficient to void the exemption from the requirements of 80.47?
(e) For olefm content (80.46(b)), ASTM D1319, Section 10.5 requires that the second set of
readings of the zone measurement boundaries be made after the sample has advanced at least
another 50 mm down the column. Would making the second set of readings before the sample
had advanced another 50 mm down the column be sufficient to void the exemption from the
requirements of 80.47?
(f) For olefm content (80.46(b)), ASTM D1319, Section 11.1 requires that percentages of
hydrocarbon types be calculated for each set of readings and then that the percentages of
hydrocarbon types from each set of readings be averaged to obtain the reported values. Would
calculating the reported values in a different way (averaging the zone lengths from each of set of
readings and then calculating the percentages of hydrocarbon types using the averaged zone
lengths) be sufficient to void the exemption from the requirements of 80.47?
(g) For Reid vapor pressure (80.46(c)), ASTM D5191, Section 8.4.1 requires that, for transparent
containers, the sample container be opened to momentarily, even if the appropriate liquid level
can be verified externally. Would failure to open the transparent container prior to beginning the
6-52
-------
Tier 3 Summary and Analysis of Comments
aeration cycles required in Section 8.4.2 volume can be verify the be sufficient to void the
exemption from the requirements of 80.47?
(h) For Reid vapor pressure (80.46(c)), ASTM D5191, Section 11.1 requires a verification
standard be tested every day the analyzer is used to perform testing. Recommended verification
standards are listed in Section 7 and those suggested verification standards having calculated
acceptable testing range intervals listed in Table 1. Section 11.2 requires that, if the verification
standard results are outside the acceptable testing range intervals listed in Table 1, the
verification standard must be rechecked or the calibration of the analyzer must be checked.
Would the use of a verification standard not listed in Section 7 or Table 1 that does not have an
established acceptable testing range interval (i.e. hexane) be sufficient to void the exemption
from the requirements of 80.47?
(i) For distillation (80.46(d)), ASTM D86, Section 10.19 requires that the residue be measured in
a 5 mL graduated cylinder. Would measurement of the residue using a graduated cylinder having
a different volume (i.e. 10 mL marked with 0.1 mL graduations) be sufficient to void the
exemption from the requirements of 80.47?
(j) For aromatic content (80.46(f)), ASTM D5769, Section 11.1 requires that ~ 10 grams of
sample be mixed with the appropriate volume of internal standard during the sample preparation
procedure. Would the use of a significantly different amount of sample (i.e. ~ 1 gram) during the
sample preparation procedure be sufficient to void the exemption from the requirements of
80.47?
(k) For oxygenates (80.46(g)(2), ASTM D4815, Section 13.1 requires that the sample and
internal standard be prepared in a capped 10 mL volumetric flask. Given the sample is being
prepared gravimetrically, would preparation of the sample in a different type of container (i.e. a
capped 10 mL scintillation vial) be sufficient to void the exemption from the requirements of
80.47?
iii. Questions relating to 80.46 - Can the procedures in ASTM D6708 be used to avoid the
requirement to correlate the results from the alternate methods listed in 80.46 to the primary
regulatory methods listed in that section? In our reading of 80.46, sections 80.46(a)(3) for sulfur
content, 80.46(b)(2)(ii) for olefm content, 80.46(f)(3)(ii) for aromatic content, and 80.46(g)(2)(ii)
for oxygen content refiners and importers must report results from the alternate methods that
have individually been correlated to the primary regulatory methods. Our interpretation appears
to be supported by the attached excerpt from the EPA Q&A documents (see Appendix 1).
Demonstrating acceptable correlation between the alternate methods and the primary methods
using the procedures in ASTM D6708 would not ensure that an analytical result from an
alternate method that was determined by an individual measurement system would be
mathematically identical to the result determined by the primary reference method (consistent
with the specified reporting significant figures). If the EPA contemplates allowing the use of the
procedures in ASTM D6708 to eliminate the requirement to correlate the alternate methods to
the primary regulatory methods in 80.46, then the text in sections 80.46(a)(3), 80.46(b)(2)(ii),
80.46(f)(3)(ii), and 80.46(g)(2)(ii) will have to be revised.
6-53
-------
Chapter 6: Regulatory Streamlining and Technical Amendments
iv. 80.47(1)(3) - This section requires that a VCSB Method-Defined Parameter test method that
is a candidate for qualification provide information about whether or not a correlation equation
must be applied that predicts the designated test method results from the applicable method-
defined alternative test method.
In our experience, it is likely that some laboratories may wish to qualify ASTM D6378 as a
method-defined alternative test method for measuring Reid vapor pressure. ASTM D6378
includes a correlation equation to convert test results from ASTM D6378 to an equivalent value
determined by ASTM D5191. However, that correlation equation relates the result from ASTM
D6378 to a Reid vapor pressure determined by ASTM D5191 using the correlation equation in
ASTM D5191, Section 14.2 (DVPE, psi = (0.965*X) - 0.548) rather than the equation required
in 80.46(c) (RVP, psi = (0.956*X) - 0.347) where X = the total vapor pressure. For results
generated according to ASTM D6378, the correlation equation specified in the method is
insufficient to produce a Reid vapor pressure result equivalent to the original method and must
be modified further to yield a result equivalent to the results required by 80.46(c).
v. Remove wording from 80.47 - EPA should consider removing the requirement that "All
sample handling, testing, procedures and tests must be conducted using good laboratory
practices" from 80.47. We agree that use of good laboratory practices is an excellent idea,
however, the definition of "good laboratory practices" is somewhat subjective and the
regulations provide no referenced list of "good laboratory practices" - providing a legally
enforceable set of criteria.
Our Response:
i) Only those designated test method in use 6 months prior to the publication of the Tier 3
final rule are exempt from the PBMS accuracy and preicsion criterion described at 40 CFR
80.47. Those designated test methods will still be required to meet the applicable SQC
requirements described at 40 CFR 80.47. In addition, if the user of a particular VCSB test
method decides to deviate from the required apparatus, reagents, calibration, quality control,
procedures or calculations, etc., that are specified in a particular VCSB test method, then by
definition, the user would not be meeting the operational description of the VCSB test
method, and thus have the option to qualify such a test method as a non-VCSB test method.
ii) Once PBMS is effective, January 1, 2016, all test methods must meet the requirements
of 40 CFR 80.47, except those designated test methods in use 6 months prior to the publication
of the Tier 3 final rulemaking. In regards to questions 6(D)(ii)(a) through 6(D)(ii)(k), as stated
above, if a user of a particular VCSB test method decides to deviate from the required apparatus,
reagents, calibration, quality control, procedures or calculations, etc., that are specified in a
particular VCSB test method, then by definition, the user would not be meeting the operational
description of the VCSB test method, but that user does have the option to qualify such a test
method as a non-VCSB test method. The Agency notes one exception, for question B(f) on
olefin content, where Weaver and Tidwell LLP asks about a particular modification to D1319,
Section 11.1 requires that percentages of hydrocarbons types from each set of readings be
averaged to obtain reporting values. Would calculating the reported values in a different way
(averaging the zone lenghts from each set of readings and then calculating the percentages of
6-54
-------
Tier 3 Summary and Analysis of Comments
hydrocarbon types using the averaged zone lenghts) be sufficient to void the exemption from the
requirements of 90.47? The answer is no, unless the mathematics, including the oxygenate
correction, yields different results.
iii) Once PBMS is effective, on January 1, 2016. the user of a VCSB method defined test
method or Non-VCSB test method must utilize ASTM D6708 to determine the need for a
correction equation for the candidate alternative test method. Please see the final regulations at
40 CFR 80.47.
iv) For RVP, once the user of an alternative test method such as ASTM D6378 has
demonstrated thourgh the use of ASTM D6708 whether a correction equation is needed or not,
they will still need to use the equation described in 40 CFR 80.46(c) for reporting purposes for
the RVP fuel parameter.
v) EPA believes that all sample handing, testing, procedures and tests must be conducted
using good laboratory practices and thus has retained the term good laboratory practices in the
regulations.
6.1.3. Downstream Pentane Blending
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel &
Petrochemical Manufacturers (AFPM); Marathon Petroleum Company LP
(MFC)
EPA has asked for comment regarding allowing downstream pentane blending into gasoline
using a similar construct as currently exists for butane blending. The Agency has not proposed
any specific regulatory language but rather is looking for input on the concept of downstream
pentane blending. Pentanes exist in the gasoline mixtures being sold today as gasoline
encompasses a broad range of hydrocarbons that result from the refining and gasoline blending
processes. Theoretically, it would be possible to establish pentane specifications and processes
akin to the existing butane blending provisions. EPA poses additional questions regarding
whether further downstream blending of pentane into PCG would have any vehicle emissions
and operability effects. API does not have any specific data to respond to this specific question.
However, it would seem that if the volumes of pentane expected to be blended result in
concentration levels that are already found in existing gasoline blends that would provide some
assurance. We would encourage EPA to look at the range of C5s contained in the gasolines being
marketed today and then examine some pentane blending scenarios to determine possible C5
content of the blended gasoline. There should be a theoretical limit on the volume of C5 that
could be blended based on vapor pressure and volatility limitations.
Commenter: Independent Fuel Terminal Operators Association (IFTOA)
6-55
-------
Chapter 6: Regulatory Streamlining and Technical Amendments
EPA also explains in the Preamble that there is a substantial increase in the production of natural
gasoline liquids (such as pentane) occurring as a by-product from the increased production of
domestic natural gas and crude oil in the United States. In addition, butane blenders have asked
EPA to broaden the downstream butane blending provisions to include pentane to help increase
the domestic supply of gasoline and to take advantage of this enhanced natural gasoline liquids
market. Finally, EPA states that pentane blending may have a marginally beneficial effect on
vehicle emissions. For these reasons, the Association supports expanding the downstream
butane blending provisions to include pentane.
Recommendation: The Association recommends that EPA continue to provide flexibility to
butane blending and consider a requirement that butane only need meet the downstream per-
gallon sulfur cap so long as the butane blending operation does not cause the annual average
sulfur content of the gasoline to exceed the proposed Tier 3 standard of 10 ppm. Moreover, the
Association supports broadening the butane blending provisions to include pentane. Provide
greater flexibility for butane blending and expand butane blending provisions to include pentane.
Commenter: Magellan Midstream Partners, L.P.
EPA currently provides flexibility for butane a blender which reduces the compliance burden.
Should EPA provide the same flexibility for pentane blenders?
Magellan Comment — Yes, absolutely. We agree that it would be appropriate to allow pentane
meeting certain specifications (specifications which include RVP and volatility limits) to be used
in the production of gasoline using a method which would reduce the compliance burdens
normally associated with refiner sampling and reporting. However, we do not believe this should
be limited to 'pentane' but should as also include gasoline range hydrocarbons. For instance,
heavy naphtha can be an appropriate gasoline blendstock, but would not fit in the pentane or
natural gasoline definition.
Also as discussed, we encourage EPA to include a provision in the final rule which will provide
compliance requirements for refiners that blend pentane or other gasoline range hydrocarbons
downstream similar to those that apply to butane blenders. We feel it is appropriate to extend this
opportunity to all gasoline range hydrocarbons rather than just for pentane blending. As greater
levels of domestic oil and gas are produced, our nation will have increased natural gas liquids
(NGLs) available for domestic utilization. We believe it is appropriate to allow gasoline range
hydrocarbons meeting certain specifications to be used in the production of gasoline using a
method which would reduce the compliance burdens normally associated with refiner sampling
and reporting.
Commenter: Sunoco Logistics Partners L.P.
Sunoco Logistics also supports the broadening the downstream butane blending provisions to
include pentane, provided the finished gasoline meets all applicable federal and state motor fuel
quality standards. There has been rapid and unprecedented growth in the production of natural
gas liquids (ethane, propane, butane and natural gasoline) - by products of the increased domestic
6-56
-------
Tier 3 Summary and Analysis of Comments
natural gas and crude oil development in multiple shale regions in the U.S. We see a substantial
opportunity to increase domestic gasoline supply with the downstream blending of the pentane
fraction of natural gasoline streams. Pentanes (C5) are clean burning alkanes like butane (C4).
Like butane the pentane fraction has always been present in gasoline. With a significantly higher
boiling point-
Pentanes 82-97 °F
Butanes 11-31°F
downstream blending of pentane at a constant volatility will result in reduced vehicle evaporative
emissions than downstream butane blending.
Pentane blending and the specifications required to assure proper finished fuel performance and
vehicle operability should generally mirror existing butane specifications:
Commercial Grade Pentane-
95% purity minimum
Sulfur <= 30ppm, then <=10ppm consistent with Tier 3 implementation date
Benzene <= 0.03 volume percent
Olefms <= 1.0 volume percent
Aromatics <= 2.0 volume percent
C6+ content <= 5.0 volume percent
Non- Commercial Grade Pentane-
Sulfur <= 30ppm, then <=10ppm consistent with Tier 3 implementation date
Benzene <= 0.03 volume percent
Olefms <= 10.0 volume percent
Aromatics <= 2.0 volume percent
C6+ content <= 5.0 volume percent
Downstream testing and oversight provisions for the commercial and non-commercial pentane
streams could similarly parallel existing regulations for butane blending.
With these specifications assuring fuel performance, vehicle operability and the aforementioned
reasons of increased domestic gasoline supply and lower evaporative emissions, the effective
date of new pentane blending regulations should be set upon Final Rule publication.
Downstream Butane and Pentane Blending Sulfur Credit or Averaging Provisions: The proposed
reduction in gasoline sulfur levels requires significant lead time and flexible compliance options
to ensure a stable, adequate and affordable supply of fuel for the consuming public. Historically,
EPA has devised various credit and averaging options for some transition period to ease supply
concerns due to non-compliance. Like refiners, downstream blenders face the same challenges
with the Tier 3 sulfur rules. A major supply source of the downstream blend components are the
impacted domestic refiners and importers. Supply of lOppm butane and pentane five years into
the future is difficult to forecast. During the most recent transition from 140 to 30ppm in butane
blending gasoline in 2003-5, a significant portion of the butane supply for blenders was non-
compliant. Projecting forward to the implementation of the new Tier 3 Sulfur rules, it is realistic
to expect a similar and perhaps more significant portion of supply from refiners and importers
would be similarly noncompliant with the proposed lOppm level. Accordingly, annual averaging
for butane and pentane blenders to the lOppm with the present 30ppm cap seems reasonable,
6-57
-------
Chapter 6: Regulatory Streamlining and Technical Amendments
uncomplicated and sufficient to maintain adequate, stable supply during the transition. At a
minimum, blenders should have the same type of compliance flexibility as refiners during the
transition period.
Our Response:
As discussed in Sections V.I and VI.A.S.of the preamble to the final rule, and consistent
with our proposal, we are finalizing pentane blending provisions that are similar to those in place
for butane under the Tier 2 program.3 We are finalizing a 30 ppm sulfur cap for pentane blended
into PCG (blender-grade) that will be effective until December 31, 2016, after which a 10 ppm
sulfur cap will become effective. This is consistent with the Tier 310 ppm refinery average
gasoline sulfur specification and Tier 3 specifications for butane blended into PCG.
As is the case with butane used by downstream butane blenders, we believe that blender-
grade pentane that can be manufactured to meet the Tier 3 program sulfur caps with relatively
mild desulfurization techniques and/or the choice of low-sulfur feedstocks. We further believe
that allowing pentane used for RVP trimming to exceed a 10 ppm sulfur cap would needlessly
complicate compliance assurance, defer some of benefits of the Tier 3 sulfur requirements, and
would be inconsistent with the premise of implementing additional programmatic flexibilities
that do not detract from the environmental goals of the Tier 3 program.
During our discussions with stakeholders following the proposal and from the review of
public comments, we became aware of additional potential issues associated with assuring the
quality of pentane for gasoline blending beyond those that exist for butane. In response to
comments and to further limit variability in pentane quality, C6 and higher hydrocarbons in
pentane blended into gasoline must be limited to 5 volume percent or less. In phone
conversations with stakeholders during our review of the written comments, we were also made
aware of the possibility that parties that handle natural gasoline liquids (NGL) might misinterpret
the pentane blending provisions finalized today to apply to natural gas liquids. 4 A pentane
stream for gasoline blending does not currently exist, and there are currently varying definitions
of NGL, which is sometimes referred to pentanes-plus. Concerns also arose regarding potential
contamination if the same equipment is used to transport blender grade and NGL. Consequently,
in today's rule, we are finalizing the following additional requirements, that will preclude
potential confusion of NGL with blender grade pentane, help ensure that the quality of blender
grade pentane is maintained throughout the distribution system, and facilitate EPA enforcement
and compliance assurance of the quality requirements for blender grade pentane.5
We believe that the butane blending provisions have reduced the burden of compliance
with EPA gasoline quality requirements. We anticipate that expanding the provisions to allow
pentane to be blended into PCG that we are finalizing in the Tier 3 FRM will further reduce the
burden of compliance. The requirement that final blends must comply with maximum gasoline
volatility requirements will ensure that the flexibility to conduct downstream RVP trimming will
3 The requirements for butane blenders are found in 40 CFR 80.82.
4 Memorandum to the docket entitled "Tier 3 Phone and E-Mail Log".
5 The requirements for producers of blender-grade pentane and for pentane blenders are discussed in Section VI.A.3.
of the Tier 3 FRM preamble.
6-58
-------
Tier 3 Summary and Analysis of Comments
not reduce the environmental benefits of EPA's gasoline quality requirements. Due to its lower
volatility compared to butane, larger volumes of pentane than butane can be blended into
gasoline while still meeting the gasoline RVP standards. Thus, allowing pentane to be blended
into gasoline downstream of the refinery may displace butane blending. Since pentanes have a
lower boiling point than butane, this could result in some environmental benefit from reduced
vehicle evaporative emissions.
We are not expanding the downstream blending provisions beyond pentane in today's
rulemaking. Pentane is a clean burning alkane like butane. The only concern with respect to the
effect of downstream pentane and butane blending on vehicle emissions can be addressed
through compliance with gasoline maximum volatility requirements. Less is known about the
potential impacts on vehicle emissions of the downstream blending of blendstocks other than
butane and pentane. We note that such blendstocks can still be utilized by refiners.
We therefore disagree with the comment that we should allow blending of heavier boiling
range hydrocarbons, such as heavy naptha, would also likely raise additional compliance
assurance issues. Similar to butane, blender grade pentane requires special pressure vessels for
transport, storage, and blending into gasoline due to its relatively high vapor pressure / boiling
point.6 These special equipment and handling needs present significant barriers to entry into the
pentane blending market, thereby limiting the potential number of parties engaged in the market.
The substantial investments needed for such special equipment also provides assurance that
parties engaged in the pentane blending market will be motivated to comply with EPA
requirements. These factors make us confident that the compliance assurance requirements
finalized in the Tier 3 FRM are sufficient to support provisions for pentane blending. On the
other hand, in the case of heavier hydrocarbons that are liquid under ambient conditions, gasoline
handling equipment could be used. This would greatly multiply the number of potential parties
that could supply product to downstream blenders, thereby substantially increasing compliance
assurance concerns. Therefore, we are not finalizing provisions for downstream blending of
hydrocarbons other than blender-grade pentane.
6.1.4. Top Tier Detergent Equivalence
What Commenters Said
Commenter: Chevron Products Company
Regulatory Streamlining General Requirements and Deadlines - Gasoline Deposit Control
Program
Chevron's experience with respect to compliance with EPA's and the top tier deposit control
requirements confirms the following statement in this section: 'It is widely accepted that
conformance with the Top Tier intake valve deposit (IVD) and fuel injector deposit (FID) control
testing is more challenging than complying with the EPA IVD and FID testing requirements.' In
6 The boiling point of pentane is ~ 97° F and butane is -30° F.
6-59
-------
Chapter 6: Regulatory Streamlining and Technical Amendments
fact, our years of extensive experience with respect to deposit control and related vehicle/engine
performance testing, combined with the vast body of technical literature on the subject, shows
that compliance with the Top Tier IVD/FID requirements results in enhanced improvements in
emissions and performance compared to fuels with deposit control additive levels that merely
meet EPA's deposit control requirements.
Chevron supports EPA's proposal to accept test data demonstrating conformance with the Top
Tier program as alternative data under EPA's Deposit Control Program.
However, in proposed regulation §80.177(b)(l)(iv), EPA proposes that test fuel used for intake
valve deposit testing using the procedure specified in §80.176(b) (the Top Tier IVD test) contain
'no less than 240 ppm sulfur.' The Top Tier Deposit Control Standard section 1.3.1.2 clearly
states that the fuel shall 'Contain no less than 24 mg/kg sulfur.' Given EPA's consideration of
permitting Top Tier IVD/FID data as an alternative, we believe the proposed 240 ppm minimum
sulfur requirement in this proposed section is the result of a typographical error and should be
corrected to 24 ppm (mg/kg) in the final rule.
Our Response:
As discussed in Section VI.A.4. of the preamble to the final rule, we are adopting the
proposed amendments to EPA's gasoline deposit control regulations to accept test data collected
for the industry-based "Top-Tier" deposit control program as demonstration of compliance with
EPA's intake valve deposit (IVD) and fuel injector deposit (FID) control requirements. We note
that we received comments that supported the "Top Tier" deposit control gasoline standards
developed by four major automakers because they provide a more robust level of the control of
vehicle engine and fuel systems than that provided by the EPA deposit control requirements.7
We also note that several major gasoline marketers have adopted Top Tier for their gasoline.
Accepting IVD/FID test data that complies with the Top Tier requirements in place of the
standard EPA IVD/FID testing requirements will provide significant savings to industry from
reduced deposit control testing while maintaining the emissions benefits of EPA's gasoline
deposit control program.
In response to the comments that identified a typographical error in the proposed
regulations to codify the Top Tier testing requirements at §80.177(b)(l)(iv), i.e., test fuel used in
IVD testing must contain no less than 240 ppm sulfur, we have corrected the error in the
regulations finalized in the Tier 3 final rule. The revised §80.177(b)(l)(iv), now states that test
fuel used in IVD testing must contain no less than 24 ppm sulfur, consistent with the Top Tier
deposit control standard.
6.2. Engine, Vehicle, and Equipment Program Technical Amendments
6.2.1. Engine and Vehicle Test Procedures (including Reference Methods)
7 The industry-based Top Tier deposit control program is discussed at http://www.toptiergas.com/
6-60
-------
Tier 3 Summary and Analysis of Comments
6.2.1.1. Heavy-Duty Vehicle Testing
What Commenters Said:
Commenter: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
HDV Testing: Discusses HDV evap test procedures. To simplify and streamline the CFR
regulations industry proposes that EPA reference the LD evap test procedures and remove the
HD evap test procedures from the HD subpart. LD evap test procedures would be utilized for all
chassis certified evaporative emission testing.
Our Response:
We agree with the comment and have drafted the final rule accordingly.
6.2.1.2. Durability
What Commenters Said:
Commenter: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
Component Durability: We note that proposed §§ 86.1823-08(g) and 86.1824-08(h) direct
manufacturers to "use good engineering judgment to determine that all emission-related
components are designed to operate properly for the full useful life of the vehicles in actual use."
The Alliance and Global Automakers support this language.
Automobile manufacturers have strong incentives to produce vehicles whose emission-related
components (as well as other components) are durable. The failure of an emission-related
component in substantial numbers, within a class or category of vehicles, during a vehicle's
useful life can give rise to the need for an emission-related recall. Manufacturers seek to avoid
such recalls, which can be very costly and can also result in reduced customer satisfaction. As a
result, manufacturers have developed a variety of techniques to evaluate the durability of
emission-related components, with the goal of ensuring that they remain functional throughout
the full useful life of the vehicle. Such techniques may include vehicle testing, bench testing of
selected components, computer modeling, and other methods. In some cases, manufacturers
provide suppliers with durability-related component specifications and require the supplier to
attest that the specification is met using whatever data the supplier deems necessary to support
the attestation. Collectively, these techniques come under the rubric of "good engineering
judgment." The manufacturer, rather than the government, decides what information it needs to
determine that its emission-related components will be durable. Manufacturers have honed these
techniques over the years, and in the overwhelming majority of cases they work very well.
6-61
-------
Chapter 6: Regulatory Streamlining and Technical Amendments
In a Supplemental Notice of Proposed Rulemaking issued on January 17, 2006, EPA discussed
the issue of component durability and discussed possible options for further rulemaking in this
area. 71 Fed. Reg. 2843. Option A would allow manufacturers to continue making "good
engineering judgment" determinations. EPA outlined two other options, B and C, that would also
require manufacturers to conduct "whole-vehicle testing" of certain vehicles. The problem with
Options B and C is that they would seek to impose a brand new set of "one-size-fits-all" tests
across the industry. Clearly, Options B and C would impose significant additional costs on
vehicle manufacturers, who would be forced to deploy additional human and facility resources to
conduct time-consuming vehicle tests. In contrast to the certain and substantial costs associated
with these tests, the benefits of such tests would be negligible. Mandated, cookie-cutter tests are
unlikely to be particularly effective in assessing component durability; at best, they may produce
data that is duplicative of information that the manufacturer has already obtained at less cost by
other methods.
The CAA gives EPA discretion to determine what types and sources of information
manufacturers and the Agency may rely on in the course of the vehicle certification process.
Nothing in the Act requires EPA to mandate uniform tests for every conceivable aspect of
vehicle certification. Here, manufacturers have a strong track record of producing vehicles with
durable emission-related components through their use of good engineering judgment. Moreover,
they have strong incentives to continue doing so.
Recommendation: In light of the above, we support EPA's proposal to maintain the "good
engineering judgment" standard for component durability and refrain from imposing costly and
unnecessary new tests on the automobile industry.
Commenter: Afton Chemical Corporation
EPA's Tier 3 proposal includes "a variety of technical amendments to certification-related
requirements for engines and vehicle emission standards." As explained more fully below, Afton
Chemical's comments relate solely to those technical amendments, and in particular, the test
methods and procedures that EPA has proposed for use by vehicle manufacturers to demonstrate
compliance with the new Tier 3 emission standards. As the Agency should be aware, it has
worked diligently in recent years to improve the transparency of the vehicle certification process
under § 206 of the Clean Air Act ("CAA" or "Act"). Afton Chemical has very much appreciated
those efforts. EPA's Tier 3 proposal provides a further opportunity for EPA to enhance the
transparency of the vehicle certification process in accordance with the rulemaking requirements
ofCAA§206(d).
Afton Chemical markets specialty chemicals for use in motor vehicle fuels and lubricants. One
of Afton Chemical's objectives is to market products that improve the performance of fuels and
lubricants in motor vehicles. For this reason, Afton Chemical has a direct and substantial interest
in the nature and scope of the test methods and procedures used by EPA to evaluate motor
vehicle performance. How EPA decides to test motor vehicles under the requirements of § 206 of
the Act (42 U.S.C. §7525) may impact the design or viability of Afton Chemical's products. This
impact can be direct, to the extent that Afton Chemical's products may potentially influence the
outcome of the tests, or indirect, to the extent the test methods and procedures impact the way in
6-62
-------
Tier 3 Summary and Analysis of Comments
which motor vehicle manufacturers design their vehicles to operate. In either case, access to the
Agency's test methods and procedures provides Afton Chemical an important tool for evaluating
its own products.
When EPA adopted the existing "Tier 2" vehicle emission standards in 1999, Afton Chemical
(then Ethyl Corporation) sought judicial review of the Tier 2 regulations seeking to ensure that
EPA developed all test methods and procedures used for motor vehicle "certification" under
CAA § 206 by rulemaking as specified in CAA § 206(d). Afton Chemical's legal challenge
culminated in a decision vacating the Tier 2 regulations and remanding the regulations to EPA
for further rulemaking. See Ethyl Corporation v. EPA, 306 F.3d 1144 (D.C. Cir. 2002).
In 2006, following additional public notice and comment, EPA issued new certification
regulations intended to comply with the court's mandate in Ethyl Corp. v. EPA. At the same
time, EPA issued a supplemental notice of proposed rulemaking intended to address one of the
comments made by Afton during the rulemaking process related to test methods and procedures
used to establish emission control "component durability" as part of EPA's certification
program. As EPA noted at the time, "Afton has raised an important issue." To address that issue,
EPA proposed three options for establishing component durability and vehicle manufacturers
and Afton Chemical thereafter submitted extensive comments on those three options (6).
In March 2006, Afton Chemical sought judicial review of EPA's new certification regulations.
Those petitions for review are being held in abeyance, however, pending EPA action on an
administrative petition for reconsideration of the then-new certification regulations Afton
Chemical submitted to EPA a few weeks later. Afton Chemical's petition addressed three general
issues:
1. EPA's implementation of a procedure using "equivalency factors" that relates various
proprietary test methods used by automakers to test their vehicles to a "standard road cycle"
established by EPA as part of the then-new certification regulations. Among other things, Afton
Chemical expressed concern that (a) the automakers might assert that the equivalency factors are
confidential business information thereby foreclosing release of equivalency factors to the public
at large, and (b) the equivalency factors might be of limited value if automakers opted to report
only a single "worst-case" equivalency factor covering their entire product line (8).
2. EPA's treatment of vehicles certified after issuance of the mandate in Ethyl Corp. v. EPA and
prior to the effective date of the new certification regulations; and
3. EPA's failure to adopt by regulation test methods and procedures for assessing the durability
of vehicle emission control system components, either individually or together as an integrated
system.
In April 2007, EPA developed a new durability webpage that can be found at
www.epa.gov/otaq/regs/ld-hwy/durability. Guidance included on the webpage indicated then
(and now) that it would provide "more information specifically related to the implementation of
the new emissions durability rule and the future component durability rule, once it is finalized."
Consistent with that statement, in March 2009, EPA posted an initial report containing
equivalency factors for all certified 2008 model year vehicles. Thereafter (presumably sometime
during 2010), EPA posted a second report containing equivalency factors for all certified 2008
model year vehicles.
6-63
-------
Chapter 6: Regulatory Streamlining and Technical Amendments
Finally, the court order holding in abeyance Afton Chemical's legal challenge to the 2006
certification regulations also required the parties to submit status reports to the court on a rolling
90 day basis. Since the court's order, twenty-seven status reports have been submitted to the
court, the most recent on April 1, 2013. In all of those reports, EPA represented that it was
continuing its examination of "the administrative record and comments regarding Afton's
administrative Petition for Reconsideration" to determine whether it "has merit" and that the
Agency would "make a final determination whether or not to grant that administrative petition as
soon as is practicable" (14).
EPA's Tier 3 Proposal as It Relates to Vehicle Certification under CAA § 206: EPA's Tier 3
proposal includes a number of proposed revisions to the pre-existing CAP 2000 vehicle
certification regulations. One such proposed regulation is 40 C.F.R. § 86.1823-08(g), pertaining
to emission component durability testing. EPA has proposed adoption of the following
regulatory text:
(g) Emission control component durability. The manufacturer shall use good engineering
judgment to determine that all emission-related components are designed to operate properly for
the full useful life of the vehicles in actual use.
Despite the fact that the foregoing language exactly matches the first of the three options EPA
proposed in 2006 in its supplement notice to address Afton Chemical's concerns about the test
methods and procedures used to demonstrate component durability, EPA nowhere addresses (a)
why the Agency has apparently opted to adopt the first of the three options for component
durability proposed in 2006 (and presumably rejected the other two options (or any others that
might have been adopted)); (b) Afton Chemical's administrative petition for reconsideration
(which addresses the same issue), or (c) whether the proposed regulatory text is intended to
provide EPA's response to Afton Chemical's reconsideration petition (as it unavoidably does)
(16).
As previously acknowledged, EPA has substantially improved the transparency of the vehicle
certification program and Afton Chemical fully supports the Agency's efforts in that regard. The
Tier 3 proposal provides EPA another opportunity to enhance transparency by adopting
appropriate test methods and procedures for establishing the durability of emission control
system components by rulemaking as required by § 206(d) of the Act (17). Afton Chemical
therefore requests that the Agency address this important issue (including the long-pending
administrative petition for reconsideration) as part of its continuing assessment of the Tier 3
proposal (18).
6 - See id. at 2847-2848. The comments submitted by Afton Chemical and automaker associations can be
found at www.regulations.gov in docket: EPA-HQ-OAR-2002-0079. Afton Chemical incorporates by
reference (as if included directly in these comments) all of the documents contained in the foregoing
docket.
8 - See www.regulations.gov: docket EPA-HQ-OAR-2002-0079-0033, pp. 3-7.
14 - See, e.g., Afton Chemical Corporation v. Environmental Protection Agency, Case Nos. 06-1095 and
06-1096, Status Report (April 1, 2013).
16 - EPA states only that "[manufacturers and catalyst suppliers perform detailed studies evaluating the
cost and emission control performance of aftertreatment systems." Draft Regulatory Impact Analysis:
Tier 3 Motor Vehicle Emission and Fuel Standards (March 2013) (EPA-420-D-13-002), p. 1-26. As
6-64
-------
Tier 3 Summary and Analysis of Comments
Afton Chemical noted in comments responding to EPA's supplemental notice concerning emission
control system component durability, "[i]f good engineering judgment is to be the test for component
durability, then EPA has to specify in its regulations what the test entails with sufficient specificity that
interested members of the public will know how good engineering judgments are made by each
automaker and can potentially replicate the exercise of engineering judgment in any particular case." See
www.regulations.gov: docket EPA-HQ-OAR-2002-0034, pp. 3-4.
17 - Several other provisions in EPA's Tier 3 proposal also identify "good engineering judgment" as the
metric for establishing the durability of emission-related components, including those that control
evaporative and re-fueling emissions. See, e.g., 40 C.F.R. § § 86.1810.1(c), 86.1824(h) and 86.1825-
08(h). As with EPA's proposed revision to 40 C.F.R. § 86.1823(g), these other provisions fall short of the
rulemaking directive in § 206(d) of the Act (as Afton has explained in its administrative petition for
reconsideration and related comments).
18 - The passage of time has obviated the need for EPA to address the second issue addressed in Afton
Chemical's administrative petition for reconsideration so Afton Chemical formally withdrawals that part
of its petition. See www.regulations.gov: docket EPA-HQ-OAR-2002-0033, pp. 8-15. Regarding the first
issue addressed in Afton Chemical's petition, release by EPA of equivalency factors for all vehicles in the
manner employed by EPA (i.e., reporting equivalency factors for each vehicle and each catalytic
converter on each vehicle) addresses most, but not all of Afton Chemical's concerns about
implementation of the equivalency factor procedure. Afton Chemical notes, for example, that EPA has
not updated its durability webpage to include any equivalency factors other than those for the 2008 model
year even though EPA represents on its webpage that it would update equivalency factors on a semi-
annual basis. See http//www.epa.gov/otaq/regs/ld-hwy/durability ("The equivalency factor must be
provided by manufacturers of light-duty vehicle, light-duty trucks and heavy-duty vehicles each model
year and will be published semi-annually on this site.").
Our Response:
In the Tier 3 proposal, EPA included purely administrative changes to the component
durability regulations for vehicle certification to remove regulatory citations from the text. The
component durability regulations referenced older versions of the component durability
regulations promulgated in 1999. In the Tier 3 proposal, EPA's intent was merely to remove the
1999 citations and replace them with "plain language" describing the same, unchanged
regulatory requirements. That proposed change may have inadvertently led to the mistaken
impression that EPA was proposing a final decision on a separate rulemaking proposal, the
component durability proposal, published in 2006 (71 FR 2843, Jan. 17, 2006). Given the
confusion regarding the purpose and effect of the proposed changes, and our intent not to make
any substantive changes or decisions related to these provisions in this rulemaking, we are not
finalizing the proposed revisions.
As noted in comments, Afton Chemical has also submitted a petition for reconsideration
for the 2006 Durability Final Rule (71 FR 2810, Jan. 17, 2006). The Tier 3 rule does not address
that petition and does not include a response to that petition for reconsideration. EPA will
finalize its response to that petition separately.
6.2.2. Other
6.2.2.1. Fuel Economy Labeling
6-65
-------
Chapter 6: Regulatory Streamlining and Technical Amendments
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM); Marathon Petroleum Company LP (MFC)
Current vehicle fuel economy labels include a smog rating ranging from 1 to 10. Vehicle labels
need to convey complex information in a simple manner enabling an equitable comparison of
vehicle attributes by new vehicle buyers. At no time should two differing smog rating scales be
used, as it will only confuse consumers by demonstrating a false comparison. Current smog
scales should be maintained until every vehicle manufactured is certified to the new Tier 3
standards, at which time, all labels can be transitioned to an adjusted smog scale.
Commenter: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
Under Fuel Economy Labeling: We request comment on applying smog ratings to TZEV
vehicles; in particular, we request comment on any appropriate differentiation of TZEV vehicles,
such as assigning different smog ratings based on the particular SULEV exhaust emission
standards or the allowance value for different all-electric range values. We request comment on
these and other alternative approaches to revising the specifications for the smog rating on the
fuel economy label.
We also request comment on whether one or more transition rating scales would be appropriate
to gradually adjust the smog rating scale as the fleet average standards become more stringent.
With respect to the Smog Ratings for TZEVs, we recommend EPA work with industry to
develop an appropriate method that will account for real-world emissions based on factors such
as exhaust certification level and all-electric range. We believe that the appropriate accounting
for real-world emissions of TZEV should be addressed in BOTH the smog ratings AND the
value used to calculate fleet average NMOG+NOx.
With respect to transition to different smog ratings, the intent of smog ratings are to inform
consumers. Consequently, consistency is very important. It is possible, and in fact likely, that
two vehicles identical in all respects except MY, will have different smog ratings for no other
reason than changes to the scoring. Such discrepancies lead to confusion and lead customers to
ignore the label all together or incorrectly believe that a newer cleaner vehicle has the same
emissions as an older vehicle. While it might be necessary to make changes to the label at some
point, we recommend severely limiting those changes.
Our Response:
In addition to the current rating scale, which will apply through MY 2017, EPA is
finalizing one interim scale between now and MY2025. Though the standards will become more
stringent annually, we believe it is important to strike a balance between maintaining consistency
of the smog rating scales across multiple model years and targeting the midpoint of the smog
rating to be at the current fleet average standard. An interim scale will allow for both. EPA
6-66
-------
Tier 3 Summary and Analysis of Comments
understands that the first model year of a new rating scale may need additional explanation to
ensure consumers understand that the rating scale has changed. We intend to help communicate
these changes via fueleconomy.gov and EPA's Green Vehicle Guide.
EPA plans to develop guidance for smog ratings for TZEV vehicles in its annual fuel
economy guidance letter. Therefore we are not finalizing a smog rating for California TZEV
vehicles at this time.
In addition to the current rating scale, which will apply through MY 2017, EPA is
finalizing one interim scale between now and MY2025. Though the standards will become more
stringent annually, we believe it is important to strike a balance between maintaining consistency
of the smog rating scales across multiple model years and targeting the midpoint of the smog
rating to be at the current fleet average standard. An interim scale will allow for both. EPA
understands that the first model year of a new rating scale may need additional explanation to
ensure consumers understand that the rating scale has changed. We intend to help communicate
these changes via fueleconomy.gov and EPA's Green Vehicle Guide.
6.2.2.2. Vehicle Labeling
What Commenters Said:
Commenter: Ford Motor Company
Ford ... recommends that the requirement for vehicle ULSD labels be removed. Such vehicle
labeling would be unnecessary for the same reasons that the ULSD pump label is no longer
needed.
Our Response:
EPA has already modified §86.1807-0l(h) to discontinue the vehicle labeling
requirement related to ULSD for model year 2014 and later vehicles subject to chassis-based
standards (77 FR 34146, June 8, 2012). We proposed in the Tier 3 NPRM to adopt this same
provision for vehicles equipped with engines certified to the engine-based standards in 40 CFR
part 86, subpart A. We are including this provision without modification in the final rule.
6-67
-------
Tier 3 Summary and Analysis of Comments
7. Costs
What We Proposed:
The comments in this chapter correspond to Section VII of the preamble to the proposed
rule and address the estimated costs of the program. Summaries of the comments received and
our responses to those comments are located below.
7.1. Costs of the Vehicle Standards
What Commenters Said:
Commenter: Consumers Union
In their comments, the Consumers Union states that, in addition to the tremendous public health
benefits, vehicle owners will also see direct financial benefits from the proposed rule. The
Consumers Union argues that lowering the sulfur content in gasoline cleans up exhaust from
older cars, reduces corrosion of emissions control systems for existing vehicles and increases
the lifespan of catalytic converters which can cost hundreds to thousands of dollars to replace.
Commenter: Peoples Republic of China
In their comments, the Peoples Republic of China had a somewhat opposite view stating that,
since the useful life is being increased to 150,000 miles, manufacture costs will increase due to
the need for more durability testing—costs which will result directly in the rise of the sales
price of each vehicle. Their comments went on to recommend that the useful life under Tier 3
be the same as that under Tier 2.
Our Response:
We agree with the comment that reducing the sulfur level of gasoline will clean up
exhaust from the majority of older cars as observed in our Tier 2 sulfur test program and
analysis (see the final RIA). However, current sulfur levels have not demonstrated a
deleterious impact on system durability and, therefore, we have not included any change in
catalytic converter lifespan or included any such savings in our cost analysis.
Regarding the comment from the People's Republic of China, we have carefully
considered the vehicle costs associated with the Tier 3 standards and present those costs in
detail in Chapter 2 of the final RIA.
What Commenters Said:
Commenter: Emissions Control Technology Association (ECTA)
7-1
-------
Chapter 7: Costs
In their comments, the Emissions Control Technology Association (ECTA) states their belief
that EPA's track record in estimating future technology costs has traditionally been shown to be
very conservative (SAE 2002-01-1980). ECTA goes on to state their belief that
The Tier 3 hardware cost estimates are also on the high side. As an example, ECTA notes that
they are aware of advanced substrate developments being tested with customers that will
decrease precious metal usage possibly 10-20% below current levels.
Commenter: International Council on Clean Transportation (ICCT)
In their comments, the International Council on Clean Transportation (ICCT) states that the
costs to comply are modest and likely overstated in the proposal [EPA-HQ-OAR-2011-0135-
4304-Al, p. 1]. ICCT also states that major advancements have occurred in vehicle emission
control technology: catalysts have improved dramatically; fuel injection is more precise;
feedback of actual air/fuel ratio is faster; software algorithms to predict air/fuel ratio have
improved; drive-by-wire systems allow air and fuel to be changed simultaneously; and,
development of initial idle retard for cold starts can bring the catalyst above light-off
temperature before the initial 20-second idle is done. ICCT claims that, as these and other
improvements are primarily due to better software algorithms, meeting the vehicle standards
will be easier and will cost much less than assumed in the proposed rule. ICCT highlights their
own analyses which found that catalyst precious metals will cost only about a third as much as
estimated in the draft RIA and that optimized close-coupled catalyst, optimized thermal
management, secondary air injection, and hydrocarbon adsorbers will not be needed on the vast
majority of vehicles and/or will cost much less than estimated in the draft RIA.
ICCT also points to recent research by Honda and Johnson Matthey showing platinum group
metal (PGM) usage could be reduced by 25% with respect to current Tier 2 Bin 5 levels and
still provide LEVIII SULEV30 compliance (i.e., Tier 3 Bin 30) by using an improved layered
catalyst and improved oxygen storage capacity (OSC). ICCT also notes research by Umicore
in which Umicore stresses the importance of a combined NMOG+NOx standard as being less
demanding than separate NMOG and NOx standards ICCT also provided their assessment that
a 20% increase in PGM loading is the most that would be required for Tier 3.
ICCT also states that the proposal appears to place too much emphasis on early LEVII-SULEV
vehicles, some of which used secondary air injection or HC adsorbers to ensure compliance.
ICCT argues that these were the first vehicles to comply with lower emission standards and
they were relatively low volume making it easier and cheaper for manufacturers to add existing
hardware than to invest the engineering resources to fully optimize precise air/fuel control and
fast light-off strategies, or to develop new hardware. ICCT adds that many manufacturers were
able to meet the LEVII-SULEV standards without such additional hardware, even on their first
attempt.
Commenter: Manufacturers of Emission Controls Association (MECA)
In their comments, the Manufacturers of Emission Controls Association (MECA) states general
agreement with the proposed cost estimates but, given that the Tier 3 vehicle cost estimates
7-2
-------
Tier 3 Summary and Analysis of Comments
were assembled approximately two years before the proposal was publicly released, MECA
encouraged an update to the costs with the latest projections from the automobile industry and
emission control industry. MECA states their belief that a number of factors will allow
projected Tier 3 compliance costs to be lower than the proposed estimates. MECA notes that
these factors include available and continued improvements in three-way catalyst technology,
the additional flexibility of meeting a combined NMOG+NOx Tier 3 standard versus meeting
individual NMOG and NOx emission standards, and the trend of engine/vehicle downsizing
that auto manufacturers will use to comply with the future fuel economy/greenhouse gas
emission standards.
Commenter: Sierra Club
In their comments, the Sierra Club argued that the proposal had demonstrated that the
substantial increase in stringency of the proposed rule does not have correspondingly high
costs. The Sierra Club suggested that this is because federal fleets already are demonstrating
actual emissions performance much cleaner than the level to which they are certified. The
Sierra Club also argues that not every technology will be required on all vehicles to meet the
proposed standards, and only the most difficult powertrain applications will require very
expensive emissions control solutions.
Our Response:
EPA agrees with these commenters that our proposed vehicle cost estimates
used dated information and the estimated vehicle costs presented in the proposal were
inappropriately high. For the final rule, as described fully in Chapter 2 of the final RIA, we
have updated our vehicle cost estimates. In doing so, we have incorporated information from
ICCT's study (SAE 2013-01-0534), ICCT's docket comments on the proposal, and a Umicore
study (SAE 2012-01-1245). As a result, the vehicle cost estimates in the final rule and, in
particular, the catalyst loading costs are considerably lower than in the proposal. Further, we
consider the final estimates to be much more transparent in that they provide much greater
detail behind the estimates, and we consider them far more robust given the greater level of
supporting information
Regarding MECA's statement that engine downsizing in the context of green house gas
standards would help reduce Tier 3 costs, we agree and have accounted for that effect by using
the EPA-projected fleet mix resulting from the 2017-2025 GHG standards. That fleet mix
contains a predominance of gasoline 14 engines which are the least costly engines at meeting
Tier 3 emissions standards. The fleet mix used in the final rule and how it differs from the
proposed rule is discussed in Chapter 2.7 of the final RIA.
Regarding the comments from the Sierra Club, we agree that some vehicles are
demonstrating compliance with the Tier 3 standards today and have incorporated that fact into
the final vehicle cost analysis. We also agree that not every technology will be required on all
vehicles and have incorporated that fact into our analysis via the penetration (or application)
rates presented in Tables 2-16 and 2-17 of the final RIA.
7-3
-------
Chapter 7: Costs
7.2. Costs of the Fuel Program
What Commenters Said:
Commenter: International Council on Clean Transportation (ICCT)
There are two additional very important points. First, reducing gasoline sulfur will result in
large emission reductions not just from future Tier 3 vehicles, but also from the entire in-use
fleet. The impacts of sulfur on older vehicles, in grams/mile, are fully as large as on Tier 3
vehicles, if not larger. As most of the sulfur impacts on catalysts are reversible, reducing
gasoline sulfur will result in immediate and very large reductions of in-use emissions.
Second, analyses of catalyst precious-metal loadings and cost are generally done without
considering changes in fuel sulfur. Reducing gasoline sulfur will enable reductions in catalyst
precious-metal loadings, further reducing the cost of compliance and offsetting much of the
cost of reducing fuel sulfur.
I also want to emphasize that the impact of sulfur on older vehicles in grams per mile are fully
as large as on Tier 3 vehicles, if not larger. And as most of the sulfur impacts are reversible,
reducing the sulfur will result in immediate and very large end use emission reductions.
Commenter: PBF Energy Inc.
PBF has reviewed and evaluated the proposed rule to control Air pollution from Motor
vehicles: Tier 3 Motor Vehicle Emission and Fuel Standards. PBF is concerned that this
regulation will place a large economic burden on refiners who must modify facilities to meet
the proposed new fuel standards.
Commenter: State of Utah
In moving forward, EPA must minimize disruption and costs to refineries, costs which likely
will be passed on to consumers. Manufacturing Tier 3 gasoline will require significant
investments by Utah's petroleum refineries.
Our Response:
As shown in Sections V and VII of the preamble to the final rule, we believe that the
costs imposed is reasonable in light of the environmental benefits the Tier 3 program provides
and it is therefore a reasonable action for EPA to require under our statutory authority. As also
discussed in Chapter 5.5 of this Summary and Analysis of Comments document, we designed
the program with considerable flexibility to minimize the cost and burden of complying with
the standards.
7-4
-------
Tier 3 Summary and Analysis of Comments
7.2.1. Impact of the Sulfur Standard (on Refinery Closures, Price Spikes, Fuel
Shortages)
What Commenters Said:
Commenter: Advanced Engine Systems Institute (AESI)
During the Tier 2 rulemaking process, petroleum industry executives predicted dire outcomes if
that rule were to go into effect, including refinery closures, price spikes, and fuel shortages.
Some stated it melodramatically, 'For some refiners, EPA's proposed regulation will be the
straw that broke the camel's back. Facilities will close and jobs will be lost.'
Since then, refining activity has remained fairly constant, while shortages and price spikes
driven by de-sulfurization requirements failed to materialize. In fact, an analysis done in late
2007 by experts from the Federal Reserve Bank of Dallas found no discernible effect on
gasoline prices of the fully phased in Tier 2 sulfur requirements. And studies show that Tier 2
benefits in ozone reduction alone far outweighed oil industry costs of implementation.
For the proposed Tier 3 rule, the oil industry is again claiming that their costs of making sulfur
reductions do not justify the benefits to the public's health and the economy. Yet the oil
industry's most recent study for the Tier 3 proposal shows that no refineries will be closed or
become uneconomic to operate, and it fails to accurately model the actual rule as proposed.
Commenter: Environmental Defense Fund (EDF)
Furthermore, erroneous claims of high costs by API are all too familiar. API made similar
claims about the Tier 2 program and opposed a national program to address sulfur from
vehicles and instead advocated for a less protective regional approach, suggesting western
states had good air quality and that there would be only modest benefits of a national program
(115). Yet, in recent testimony to EPA, API acknowledged that the Tier 2 program, "yielded
measurable environmental benefits. The air quality benefits of the Tier 2 program are still being
realized' (116).
115 - Environmental Protection Agency, Tier 2 Motor Vehicle Emission Standards and Gasoline Sulfur
Control Requirements: Response to Comments, December 1999: 13-5.
116 - American Petroleum Institute, Testimony by Patrick Kelly at the EPA Philadelphia Public Hearing
(Apr. 24,2013).
Our Response:
We received comments that disputed claims of the refinery industry on the impacts of
the sulfur standards. We agree with the commenters that the oil industry has made very
dramatic claims in the past regarding refinery closures and shortfalls in product availability,
and the most dramatic claims occurred in response to nonroad and highway diesel fuel
7-5
-------
Chapter 7: Costs
standards. After the first Baker and O'Brien cost study which made similar conclusions
associated with low sulfur and low RVP standards prior to our Tier 3 NPRM, we compared
what actually transpired to diesel fuel supply and refinery closures after 2006 when the
highway diesel fuel sulfur standards took effect to the dramatic predictions made by Baker and
O'Brien and provided our comparison in Chapter 5 of the draft RIA for the NPRM. Our
findings were that none of the dire predictions made by Baker and O'Brien with respect to
diesel fuel had occurred. We also note that the revised Baker and O'Brien cost study for Tier 3
gasoline, which solely evaluated the cost of lower sulfur gasoline, neither forecasted any
refinery closures nor reductions in the supply of gasoline like the first Baker and O'Brien study.
7.2.2. Cost (Average Cost and Marginal Cost, Such as 6 - 9 c/gal)
7.2.2.1 Average Cost
What Commenters Said:
Commenter: Natural Resources Defense Council (NRDC)
API has a history of inflated cost claims. A review of several clean fuels regulations finds that
API estimates have been 3.6 to 6.4 times higher than actual observed price changes.
The API high cost estimates have the potential to raise concerns of industry investors, and some
refining companies are backing away from the API estimates to present more realistic figures.
For example, Valero Energy, which is the largest independent refiner in the U.S., recently told
a group of financial analysts that its Tier 3 compliance will cost to $300 to $400 million. A
finance expert colleague of mine at NRDC used the EPA's marginal cost assumptions and
found that the $400 Valero cost claim would equate to a compliance cost of approximately .6
cents per gallon, more than 10 times less than the API high estimate.
For Valero, the .6 cent per gallon translates into annual costs of about $95 million, or just two
percent of the four and a half billion in pretax earnings that the company made on refining in
2012. While Valero and other companies point out the cost of compliance is minimal, API
continues to inflate costs. Regulatory cost inflation is part of their history. By exaggerating
costs, API has attempted to scare consumers into potential large gas price increases at the
pump.
The reality is that these large increases don't appear. According to a peer review report by
EPA, API's estimates of compliance costs for previous clean fuel regulations adopted in the
1990s have been 3.6 to 6.4 times higher than the actual observed price changes. Similarly, a
recent study by Navigant Economics found that the Tier 2 gasoline sulfur regulations, which
API estimated would cost about three cents a gallon, have had no discernible impact on retail
prices.
Commenter: Navigant Economics
7-6
-------
Tier 3 Summary and Analysis of Comments
First, there are serious flaws in the approach employed by Baker and O'Brien on behalf of the
American Petroleum Institute to estimate the compliance cost to refineries. Second, I've
examined EPA's cost benefit analysis. It comports with economic logic and best practices, and
it validates our conclusions. Third, we examined the effects of the Tier 2 sulfur rule on the
price of gasoline and found little to no impact suggesting the Tier 3 effect will be negligible.
So going to my first point on the flaws in the Baker and O'Brien approach, there are several,
but today I just want to address three. Baker and O'Brien failed to account for averaging and
trading of credits, which reduces the compliance costs to refineries. As EPA explained,
refineries with pre and post treaters are able to achieve further sulfur reductions below the 10
parts per million standard at a relatively low incremental cost, and sell credits to refineries who
would otherwise be faced with grass roots investments.
Averaging and trading reduces EPA's estimated compliance cost from .97 cents per gallon to
.79 cents per gallon, a reduction of nearly 20 percent with averaging in trading. Between eight
and 25 high-cost refineries out of 11 would prefer to consume credits than to upgrade their
facilities. Baker and O'Brien's mistaken assumption that those high-cost refineries reduce the
sulfur content of their gasoline they produce to 10 parts per million severely inflates their
compliance cost estimates.
Second, Baker and O'Brien assume without support that the refinery with the highest cost of
compliance will set the price of gasoline. However, if this were true, then the profits of the
refinery industry, defined as the difference between the marginal cost and the average cost per
gallon, would be four to seven cents, at which point the refiners should endorse Tier 3. Given
the refiners' opposition, it seems that Baker and O'Brien's logic is faulty.
Third, Baker and O'Brien's estimated capital cost for a grass roots FCC post treater is not
reasonable according to companies that construct or install desulfurization units. Indeed, Baker
and O'Brien's capital cost estimates are between two and four times higher than industry
benchmarks. Baker and O'Brien states that their estimates were based on actual installation for
the Tier 2 program, but as EPA explains, a grass roots FCC post treater installed for Tier 2
would remove about 10 times more sulfur than one installed for Tier 3, and, thus, would have
significantly higher capital costs. Correcting this one defect in the Baker and O'Brien study
causes their cost estimates to come into alignment with those of MathPro and of EPA.
Thank you for the, I guess, the support for our cost analysis. But if you could just spend just a
moment, because you were cut a little short. So what are the main differences between, you
know, our one cent a gallon and the six to nine cent estimate from Baker and O'Brien?
So I think the points that I rattled off were the main drivers. One is this conflation of the
average and the marginal. I think that what — when you actually saw for the average of the
Baker and O'Brien, you don't get that big of a divergence. You get something like one versus
two.
But even there, the difference can be entirely explained away with just one change, and that's
this capital cost component of these treaters. And the way that we tried to ascertain that it was
7-7
-------
Chapter 7: Costs
exaggerated was by interviewing actual providers of these equipments, these types of
equipment and founding out whether their costs were reasonable. And they told us they
weren't.
Commenter: Natural Resources Defense Council (NRDC)
API's cost estimates for Tier 3 are inconsistent with API's own members, which estimate costs
as either immaterial to their business or less than a penny a gallon.
Commenter: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
We recognize that reducing the sulfur content of fuel will have a cost. However, as EPA details
in the proposed rule, these costs have been exaggerated by the petroleum industry (66). Indeed,
as a general matter, petroleum product prices are due primarily to changing global crude oil
prices. The compliance costs of the sulfur reductions contained in the proposed rule will not
have an appreciable effect on the price of gasoline at the pump.
In the proposed rule EPA found that the cost of compliance with the fuel standards will be an
average of 0.89 cents per gallon (67). The Baker and O'Brien study (API) found a 1.9 cent per
gallon average increase from Tier 2 (6-9 cents marginal cost with no RVP change) (68). In
contrast, the MathPro study for the International Council on Clean Transportation (ICCT)
found a 0.8 to 1.4 cent per gallon increase from Tier 2 (69). In the proposed rule, EPA's
analyses were closer to the MathPro estimates, and exposed the invalid assumptions of the
Baker and O'Brien study (70). It is extremely unlikely that a small (1 cent or less) increase in
refining costs would be passed on to consumers in an increased gasoline price (71).
In short, both EPA's data in the proposed rule and the Navigant study demonstrate that the
overall cost impact of reduced sulfur will be very low on an average per-gallon basis, and
significantly lower than the benefits of the rule. Navigant states that Tier 2 had no statistically
significant impact on gas prices after controlling for cost of crude oil, refinery margins and
other factors, and Tier 3 would impose only half the average cost of Tier 2 (approximately one
cent per gallon versus 2 cents). Therefore, as noted above, it is very likely the retail pump
impact of the reduced sulfur requirements will be close to or at zero. Even the Baker and
O'Brien study, which both EPA and Navigant has discredited for a number of overestimates of
cost, still found that no refinery shutdowns are projected from the reduction in sulfur proposed
under the rule. The benefits of the rule far outweigh the costs, and the reductions in sulfur will
be a crucial component of allowing the OEMs to achieve the required reductions in criteria
pollutants.
65 - The (octane) grade weighted (pool average for US (excluding California) would be closer to 27
ppm. Survey can be purchased from the Alliance website www.autoalliance.org.
66 - 78 Fed. Reg. 29908 at 29972-29979 (May 21, 2013).
67 - 78 Fed. Reg. 29908 at 29972 (May 21, 2013).
68 - See March 2012 addendum to the original July 2011 Baker and O'Brien study. This was analyzed
by the Navigant study.
7-8
-------
Tier 3 Summary and Analysis of Comments
69 - Id.
70 - EPA cited to the Navigant study which found that Baker and O'Brien cost predictions have been
inflated in the past (e.g., cost of reducing sulfur in diesel fuel was significantly less than predicted). 78
Fed. Reg. 29908 at 29978 (May 21, 2013). EPA also points to Navigant's data showing that the Baker
and O'Brien study does not account for reductions in cost from averaging and trading. In addition, the
Baker and O'Brien study also has other logical flaws, such as erroneously large capital cost assumptions
and a lack of quality data about the individual U.S. refiners, both of which were noted by EPA.
Navigant believes that the Baker and O'Brien study exaggerates the costs by a factor of 2 to 1. EPA and
Navigant note that estimated capital cost for new FCC naphtha desulfurization unit used in the Baker &
O'Brien study is not reasonable. 78 Fed. Reg. 29908 at 29978 (May 21, 2013). Navigant notes that
differences in the refinery model methodologies (i.e. individual refinery versus PADD-based) between
the Baker and O'Brien and Math Pro studies is not enough to account for their varying cost estimates.
Instead, Navigant found that different assumptions about capital costs account for most of the difference
in cost estimates.
71 - See also recent MathPro Tier 3 testimony at the EPA public meeting.
Commenter: American Fuel & Petrochemical Manufacturers (AFPM)
A Baker and O'Brien analysis estimates that a gasoline sulfur content reduction will cost
refiners $10 billion in capital costs, and $2.4 billion per year in operating costs, increasing the
cost of producing gasoline by six to nine cents per gallon.
Commenter: American Lung Association
Other Analyses Support EPA's Findings that Standards Have Very Low Cost for Consumers:
Two independent studies support EPA's conclusions that the refining cost associated with Tier
3 sulfur standards will be very low compared to the benefits. In 2011, MathPro, a consulting
firm specializing in economic analysis of petroleum refining and related industries,
commissioned by the International Council for Clean Transportation, found that the per-gallon
refining cost of a Tier 3 program with a 10 parts per million sulfur standard would be 1.4 cents
(47).
In 2012, Navigant Economics, commissioned by the Emissions Control Technology
Association, estimated the cost of low sulfur gasoline standard would be about one cent per
gallon. This estimate is closer to EPA's and MathPro's estimates rather than the higher cost
claims from the oil industry (48).
47 - MathPro. Refining Economics of a National Low Sulfur, Low RVP Gasoline Standard: A study
performed for The International Council for Clean Transportation. MathPro Inc: Maryland, 2011.
Available at
http://www.theicct.org/sites/default/files/publications/ICCT04_Tier3_Report_Final_v4_All.pdf
48 - Schink GR, Singer HJ. Economic Analysis of the Implications of Implementing EPA's Tier 3
Rules: Prepared for the Emissions Control Technology Association. Navigant Economics: Washington
B.C., 2012. Available at http://www.ectausa.com/061212-Economic-Analysis-of4he-Implications-of-
Tier-3-Sulfur-Reduction-Finalembargoed.pdf
Commenter: American Petroleum Institute (API) and the Association of Fuel &
7-9
-------
Chapter 7: Costs
Petrochemical Manufacturers (AFPM)
Reducing gasoline sulfur to an average of 10 parts per million is expensive. Research API
shared with the Agency shows nearly $10 billion in capital costs. The annual compliance cost
is about $2.4 billion, or six to nine cents per gallon in marginal costs. EPA should not consider
these costs in a vacuum. Refiners need to balance these expenses with the cumulative costs of
other Federal and States regulations.
EPA: Did you say that the annual cost from your study was $2.4 billion?
API: The annual compliance costs. That includes the - it is the annual cost of compliance and
includes capital recovery.
EPA: And so what is that in terms of cents per gallon then because that's the average cost
then?
API: That's a total cost, the $2.4 billion. And it equates to a six to nine cent per gallon
marginal cost in most markets. Does that answer your question?
EPA: I guess - well, what is the $2.4 equate to in terms of an annual - in terms of cent per
gallon?
API: In marginal costs, it would be six to nine.
EPA: Do you have an average cost?
API: We didn't do an average cost. While average cost is an important factor, in commodity
markets, when all of the demand is needed for that commodity, the marginal cost is very
relevant in setting the market price. And so, therefore, we looked at the marginal cost because
of its importance to the marketplace.
EPA uses a cost estimate based on average cost that ignores the important role that marginal
cost plays in gauging the market response to the regulation. EPA should not consider these
costs in a vacuum. Refiners need to balance these costs with the cumulative cost of other
Federal and States regulations.
Commenter: Attorney General of Connecticut, et.al; Attorney General of New York, et.al
Indeed, EPA concluded that "[t]he proposed fuel standards are projected to cost on average less
than one cent per gallon of gasoline." 78 Fed.Reg. at 29,827.
Commenter: Chevron Products Company
We believe that EPA has underestimated the cost and complexity of the changes which will be
required in the refining industry to achieve this reduction. EPA's economic impact analysis
relies entirely on the national average cost increase for gasoline production, instead of the
marginal or incremental cost of supply as proposed by Baker and O'Brien (1). The incremental
cost of supply is usually correlated with market price effects when demand must be met with
more expensive sources of supply. Therefore, using the incremental value is more
representative of the true societal cost than using the average. By using the national average
cost increase, EPA is underestimating the true cost of the Tier 3 program.
1 - Baker & O'Brien, 'Potential Supply and Cost Impacts of Lower Sulfur, Lower RVP Gasoline', 2011
7-10
-------
Tier 3 Summary and Analysis of Comments
Commenter: Emissions Control Technology Association (ECTA)
We agree with EPA's methodology for estimating the cost that the refining industry will incur
to comply with the sulfur content requirement of Tier 3. We also agree with EPA's estimate
that the average cost of compliance for the refining industry is only 0.89 cents per gallon (16).
We believe EPA's estimated costs are reasonable for two reasons as described below.
It is also noteworthy that EPA's average cost per gallon estimates comported with those of
Mathpro (0.8 cents), further validating its cost model (20).
Based on a study by Baker O'Brien sponsored by the American Petroleum Institute, the
refining industry cost estimates implies that the average cost of compliance would be 2.12 cents
per gallon or more than double EPA's estimate of 0.89 cents per gallon (24). We believe that
EPA's estimate is far more reliable than the refining industry's estimate for three reasons.
First, Baker & O'Brien assumed an after-tax ROI for capital invested for the hydrotreaters and
hydrogen plants of 10%, which is unreasonably high. EPA assumed before-tax ROI of 7% (25).
This before-tax ROI is equivalent to an after-tax rate of return of approximately 5.2 %, and it is
consistent with the rates used by EPA in related matters (26). To be fair, Baker & O'Brien's
higher rate of return may be closer to what refiners use when evaluating conventional refinery
investment opportunities (27). However, EPA correctly uses a lower 7% ROI when estimating
the social costs of regulations. In other words, the EPA's 7% before-tax rate of return is
effectively a social discount rate; the same value is used by EPA to discount future benefits
from environment regulations. Merely changing Baker & O'Brien's assumed ROI to reflect the
perspective of society (rather than private industry) reduces its estimated average compliance
costs from 2.12 to 1.58 cents per gallon.
20 - EPA Proposal, Fed. Reg. at 29977.
24 - Id. at 29977.
25-Id. at 29978.
26 - See, e.g., EPA, Regulatory Impact Analysis: Heavy Duty Engine and Vehicle Standards and
Highway Diesel Fuel Sulfur Control Requirements, EPA420-R-00-026, Dec. 2000.
27 - Mathpro, Refining Economics of National Low Sulfur, Low RVP Gasoline, Oct. 2011, at p. 13
(hereinafter Mathpro Study).
Commenter: Environmental Defense Fund (EDF)
The fuel standards in the Tier 3 proposal are cost-effective. EPA estimates that the cost to
refiners of reducing the sulfur in gasoline to an average of 10 ppm will be 0.89 cents per gallon
averaged over all gasoline.
Even if there were no ABT program, or no refiners decided to participate in the program, EPA
estimates that the fuel costs would still only be 0.97 cents per gallon on average. Moreover, a
study by Navigant Economics projects that these costs will not be passed on to consumers
(106).
7-11
-------
Chapter 7: Costs
The International Council for Clean Transportation: The International Council for Clean
Transportation (ICCT) commissioned a study by MathPro in 2011, building upon and updating
a 2009 MathPro study commissioned by the Alliance of Automobile Manufacturers. The ICCT
study analyzed the economic impact of a 10 ppm sulfur limit for all gasoline and found that the
cost of lower sulfur standards to oil refiners would be 0.8-1.4 cents per gallon (107). MathPro
estimates that $3.9 billion would need to be invested to meet low-sulfur standards and that the
annual refining costs would be $1.5 billion. These estimated costs are very close to EPA's
estimates.
Emissions Control Technology Association Study: ANavigant Economics study commissioned
by the Emissions Control Technology Association (ECTA) in June 2012 found that the cost to
oil refineries to produce 10 ppm sulfur gas would equate to approximately one cent per gallon
(108). Furthermore, the study found that these small costs would likely not be passed on to
consumers through higher retail gasoline prices.
Navigant also conducted a multiple regression analysis to test whether the increased cost to
U.S. refiners of implementing the Tier 2 gasoline sulfur reductions from 300 ppm to 30 ppm
were passed on to consumers in the form of higher retail gasoline prices. The analysis found
that the increase in cost to U.S. refiners of about two cents per gallon were not passed on to
consumers, suggesting that it is highly unlikely that the Tier 3 costs will be passed on to
consumers.
American Petroleum Institute Study: The American Petroleum Institute (API) commissioned a
study by Baker & O'Brien (B&O) in 2011, which estimated that the sulfur reduction
requirement of Tier 3 would increase the marginal cost of refining by 6 to 9 cents per gallon
(109). The marginal costs reflect the cost to the refineries that would experience the highest
costs, not the average costs as estimated by EPA.
The API commissioned study also indicates that the fuel program would be economically
beneficial to the refining industry. B&O do not anticipate any refinery shutdowns as a
consequence of Tier 3 and the analysis indicates that all U.S. refineries would find it cost-
effective to make the investments necessary to comply with these standards and that the capital
required to make the investments will be available. According to EPA, API's analysis "would
suggest that the oil industry would profit from 10 ppm low sulfur standard by roughly 4 to 7
cents per gallon, or roughly $4 to $8 billion dollars per year as a result of gasoline sulfur
control."
Opponents to Tier 2 claimed the cost to consumers would be 5 cents per gallon (117).
However, the ECTA study and other sources have found that Tier 2 did not have any impact on
retail gasoline prices (118). Additionally, EPA projected that the monetized health benefits of
reductions of PM and ozone under the Tier 2 program would outweigh the costs by about 5:1
(119).
106 - Navigant Economics, Economic Analysis of the Implications of Implementing EPA's Tier 3
Rules, prepared for the Emissions Control Technology Association (June 14, 2012).
107 - MathPro, Inc., Refining Economics of A Natural Low Sulfur, Low RVP Gasoline Standard,
prepared for The International Council on Clean Transportation (Oct. 25, 2011).
7-12
-------
Tier 3 Summary and Analysis of Comments
108 - Navigant Economics, Economic Analysis of the Implications of Implementing EPA's Tier 3
Rules, prepared for the Emissions Control Technology Association (June 14, 2012).
109 - Baker & O'Brien, Incorporated, Potential Supply and Cost Impacts of Lower Sulfur, Lower RVP
Gasoline, prepared for the American Petroleum Institute (July 2011).
117 - Environmental Protection Agency, Tier 2 Motor Vehicle Emission Standards and Gasoline Sulfur
Control Requirements: Response to Comments, 23-20 (Dec. 1999).
118 - Navigant Economics, Economic Analysis of the Implications of Implementing EPA's Tier 3
Rules, prepared for the Emissions Control Technology Association (June 14, 2012).
119 - Environmental Protection Agency, Regulatory Announcement, EPA's Program for Cleaner
Vehicles and Cleaner Gasoline (Dec. 1999), available
at http://www.epa.gov/tier2/documents/f99051 .pdf.
Commenter: Environmental Defense Fund (EDF)
While the B&O study did not report an estimate of the average U.S. refiner cost of reducing the
sulfur content to 10 ppm, EPA calculated it from the information B&O provided as 2.12 cents
per gallon. The Navigant study compared the B&O average cost estimates with the MathPro
estimates and concluded that the approximately one cent per gallon difference in the two
estimates is due almost entirely to the difference between MathPro's and B&O's estimates of
the U.S. refiners' compliance-related investment costs.
Commenter: International Council on Clean Transportation (ICCT)
The cost of reducing gasoline sulfur from 30 ppm to 10 ppm is very modest. The ICCT
contracted with MathPro in 2011 to evaluate the cost of reducing sulfur from 30 to 10 ppm.
MathPro found that the cost would be 0.8 to 1.4 cents per gallon, and these results are likely to
be conservative.
Gasoline Sulfur Reduction Cost
The cost of reducing gasoline sulfur from 30 ppm to 10 ppm is very modest. The ICCT
contracted with MathPro in 2011 to evaluate the cost of reducing sulfur from 30 to 10 ppm.
MathPro found that the cost would be 0.8 to 1.4 cents per gallon. The cost estimates include
revamping the FCC naphtha hydrotreaters, historical rates of return on investment, supplying
the additional hydrogen needed, replacing small losses in both gasoline volume and octane, and
expanding sulfur recovery.
In fact, these cost estimates are likely overstated, as MathPro assumed that all existing FCC
naphtha hydrotreating capacity would require revamping even though many hydrotreaters
installed to meet the Tier 2 sulfur requirements are already capable of meeting the 10 ppm
standard.
A separate study carried out by Baker and O'Brien for the American Petroleum institute (API)
estimated that production of 10 ppm sulfur gasoline would increase the marginal refining cost
by 6-9 cents/gallon. However, there are two main reasons why these marginal cost estimates in
the API study are unrealistic and should not be used.
7-13
-------
Chapter 7: Costs
First, in the Baker and O'Brien methodology, the indicated marginal cost is the highest cost of
sulfur control that would be incurred by the least efficient refinery or refineries in the US. The
petroleum market in the US is regional, i.e., there is not a single market. It is possible that the
least efficient and highest cost refinery may be in, for example, PADD 4 and have a particular
market to itself. That refinery might be able to pass the marginal costs on to producers in that
market but will not have any effects in the rest of the US. However, this is not the typical case,
as most refineries are not so isolated. High cost refineries that are not isolated will not be able
to pass the marginal costs onto consumers, due to competition from other efficient refineries.
As a result, it is the average costs, not the marginal costs that represent the actual increase in
the refining cost.
If the average refining costs are considered instead of marginal costs and Baker and O'Brien's
investment costs for FCC naphtha hydrotreating are properly adjusted, Baker and O-Brien's
estimated refining costs would be similar to those estimated by MathPro.
Commenter: Manufacturers of Emission Controls Association (MECA)
MECA agrees with EPA's assessment that the cost of lowering average gasoline sulfur levels
from today's 30 ppm national average to the proposed 10 ppm national average is on the order
of one penny per gallon of fuel. EPA's gasoline cost assessment is supported by gasoline cost
studies commissioned by the National Association of Clean Air Agencies (NACAA) and the
Emission Control Technology Association (ECTA) that also indicate that costs for reducing
gasoline sulfur levels to a 10 ppm national average are approximately one penny per gallon.
Commenter: MathPro
In the summer of 2011, the International Council on Clean Transportation commissioned us to
estimate the U.S. refining sector's investment requirements and average cost for reducing
average sulfur content of the U.S. gasoline pool from 30 ppm, the current standard, to 10 ppm,
the proposed Tier 3 standard, and also for reducing the summer re-vapor pressure standard of
U.S. conventional gasoline. Our report from that study was submitted to EPA in October of
2011.
My testimony today deals exclusively with sulfur control.
And that corresponds to an average cost of eight-tenths of a percent to 1.4 cent per gallon of
gasoline.
This range of estimated average cost reflects different assumptions regarding two economic
factors: the rate of return on refinery investment and the capital cost of revamping FCC naphtha
hydro treaters, a particular type of unit in U.S. refineries to enable those units to carry out the
more severe de-sulfurization needed to meet the 10 ppm standard.
The higher estimate of 1.4 cents a gallon reflects a 10 percent after tax return on investment and
a capital cost for revamping of about 50 percent, about half, of the capital costs for a grass roots
unit of the same kind and size. The lower estimate of eight-tenths of a cent a gallon reflects a
7-14
-------
Tier 3 Summary and Analysis of Comments
seven percent pre-tax return on investment, and about a 30 percent factor applied to the capital
cost of revamping a grass roots unit.
Commenter: Medical Advocates for Healthy Air
By opponents of the Tier 3 standard, gasoline prices will increase by $.06 to $.09 per gallon.
The EPA's estimate is a mere $.01 per gallon. It is important to remember that when the EPA
introduced Tier 2 standards to reduce the sulfur content of gasoline from 300 to 30 ppm, the
increased cost of gasoline was much lower than the EPA's estimate of $.02 cents per gallon.
There is no reason to believe that the EPA's estimate this time is conservative, and the ultimate
cost may turn out to be less than 1 cent per gallon.
Commenter: National Association of Clean Air Agencies (NACAA)
While the emission benefits of Tier 3 are very high, the costs of the program are very low. In
our 2011 study, NACAA predicted an increase in the cost of gasoline of less than a penny a
gallon and EPA has found the same (EPA's analysis estimated 0.89 cents per gallon).
Commenter: Natural Resources Defense Council (NRDC)
NRDC supports the EPA's assessment that the average costs to reduce gasoline sulfur to 10
parts per million is less than one cent per gallon. Recent estimates by the American Petroleum
Institute (API) of 6 to 9 cents per gallon are not justified as average compliance costs. Further,
the API estimates include per-gallon profits of 4 to 7 cents per gallon, which translates into $4
to $8 billion in windfall profits for the oil industry each year.
NRDC supports the EPA's assessment that the average costs to reduce gasoline sulfur to 10
parts per million is less than one cent per gallon. Recent estimates of 6 to 9 cents per gallon by
the American Petroleum Institute (API), an association of oil companies, are not justified as
average compliance costs and are inconsistent with figures by API's own members, which
estimate costs as either immaterial to their business or less than a penny a gallon.
The 1.23 cents/gallon difference between the 2.12 cents/gallon average cost and EPA's 0.9
cents/gallon estimate can be explained by two factors. First, the API commissioned study
assumes an additional 10 percent rate of return on investment (ROI) but ignores refiners' ability
to depreciate their investments so implicitly assumes a higher ROI than normal. In contrast,
EPA assumes that the industry requires a 10 percent ROI after tax deductions are considered
which is equivalent to a before tax ROI of 7 percent. EPA finds that when the API estimate is
adjusted to a 7 percent before tax ROI, the API cost would be 1.58 cents per gallon.
Second, the remaining 0.69 cents/gallon difference can be attributed to differences in annual
costs to install and operate the desulfurization equipment. The API commissioned study
estimated a total annual cost of $2.4 billion compared to the EPA estimate of about $1.3 billion
due to differences in equipment costs. As EPA notes in the DRIA, API's claimed cost for a key
piece of equipment to desulfurize oil, the fluidized catalytic cracker postreater, is 350 percent
higher than EPA's estimate.
7-15
-------
Chapter 7: Costs
EPA has assessed the pre-regulation cost estimates by API to comply for clean fuel regulations
with the actual impact seen in the marketplace. As shown in the figure below [Figure 1 can be
found on p. 10 of Docket number EPA-HQ-OAR-2011-0135-4286-A1] from EPA's peer-
reviewed report "Comparison of EPA and Other Estimates of Mobile Source Rule Costs to
Actual Price Changes" (21) API's estimates for regulations adopted in the 1990s have been 3.6
to 6.4 times higher than actual observed price changes.
While the EPA report did not include the actual cost of the most recent gasoline sulfur
reductions, a recent study by Navigant Economics (22) found that Tier 2 gasoline sulfur
regulations, which API estimated would cost about 3 cents per gallon, (23) have had no
discernible impact on retail prices.
Still, however, the oil industry claims that the Tier 3 sulfur reductions are too expensive. It
appears that the industry believes that the public health protection should only be pursued if
they can generate huge profit margins for the industry. API recently estimated the cost of
complying with Tier 3 sulfur requirements at six to nine cents per gallon. EPA estimates the
average cost at less than a penny, or .9 cents per gallon using the API assumptions. Most of the
difference between the API and the EPA estimates is API's projected profit margin.
As EPA described in the draft regulatory impact analysis, API's cost estimate assumes a profit
of four to seven cents per gallon, which translates into $4 to $8 billion in windfall profits for
the oil industry each year. NRDC estimates that the oil industry is projecting an astounding
rate of return on investment of 180 to 340 percent.
18 - EPA, "Draft Regulatory Impact Analysis: Tier 3 Motor Vehicle Emission and Fuel Standards",
EPA-420-D-13-002, March 2013.
19 - David C. Tamm and Kevin P. Milburn (Baker and O'Brien Incorporated), "Addendum to Potential
Supply and Cost Impacts of Lower Sulfur, Lower RVP Gasoline," March 2012.
20 - EPA, op. cit. Figure 5-1.
21 - Anderson, John F. and Sherwood, Todd, "Comparison of EPA and Other Estimates of Mobile
Source Rule Costs to Actual Price Changes", SAE Technical Paper 2002-01-1980, May 2002.
22 - Schink, George R., Ph.D. and Hal J. Singer, Hal J., Ph.D., Navigant Economics, "Economic
Analysis of the Implications of Implementing EPA's Tier 3 Rules", Prepared for the Emissions Control
Technology Association, June 14, 2012.
23 - Anderson, op. cit.
Commenter: Pennsylvania Department of Environmental Protection (DEP)
DEP recognizes that the proposed rules may increase costs for some refineries located in this
region, and raise the cost of gasoline.
The EPA should reexamine the estimated increase in production and at-the-pump costs.
According to the results of a study performed by the energy consulting firm Baker and O'Brian,
the American Petroleum Institute (API) estimates a cost increase to consumers of six to nine
cents per gallon with the proposed Tier 3 low sulfur gasoline. The EPA estimates a cost
increase of about one cent per gallon. EPA used the industry's cost models to model costs, and
had the cost results reviewed by industry consultants. The EPA offers several reasons as to why
7-16
-------
Tier 3 Summary and Analysis of Comments
the cost discrepancy exists, but never fully closes the gap in costs. Recently, refining
companies, including the nation's largest, Valero, have disputed the API cost estimate. These
refiners have indicated that the capital costs associated with complying with the Tier 3
standards will be lower than the $10 billion API estimate. EPA should consult directly with
refining companies to close the gap between these cost estimates.
Commenter: Petroleum Marketers Association of America (PMAA)
PMAA represents over 8000 independent petroleum marketing companies who market liquid
fuels at both wholesale and retail. We have examined the two cost related reports prepared by
Mathpro and Baker/O'Brien. We are concerned that the penny per gallon cost identified by
Mathpro is not accurate and consumers will be burdened with unjustified and burdensome
higher gasoline costs.
If in fact EPA eventually concludes the cost burden will exceed a penny per gallon, we believe
EPA should revise the final rule to keep the costs at that threshold.
Commenter: Refinery Automation Institute, LLC
To get an idea of costs, using DOE EIA June 3, 2013 price differential between New York
gasoline and Los Angeles CARB gasoline of 13 cents/gallon, we are talking about a country-
wide price burden increase of approximately 8.7 million bpd X 42 gal/bbl X $0.13/gal=$47.5
million/day, or an additional $17.3 billion a year that consumers have to pay. In the current
economic environment with 23+ million unemployed, this seems to me unconscionable, even
though 2017 seems far in the future.
Commenter: Sierra Club
In addition, EPA provides sufficient support that the 10 ppm sulfur level is economically
feasible, including a refinery-specific cost of compliance study. On average, a 10 ppm sulfur
level would cost refineries $0.97/gallon without the ABT program, and the costs decline with
the inclusion of ABT and intra-company credit transfers. EPA further justifies the 10 ppm as a
practical level because costs of further desulfurization would be costly at the present time.
While the American Petroleum Institute's scare tactics claim that these standards would cause
gas prices to skyrocket, independent studies have shown that these standards can be achieved
for less than a penny a gallon. This is a clear case where we can clean our air, protect our
public health, and create jobs by reducing pollution from our cars and trucks.
Commenter: Sierra Club, Clean Air Watch, Respiratory Health Association
While the American Petroleum Institute's scare tactics claim that these standards would cause
gas prices to skyrocket, independent studies have shown that these standards can be achieved
for less than a penny a gallon. This is a clear case where we can clean our air, protect our public
health and create jobs by reducing pollution from our cars and trucks.
7-17
-------
Chapter 7: Costs
Commenter: Private Citizen
If my math is correct, Americans consume around 64.3 billion gallons of gas annually. A one
cent increase per gallon would equate to a cost of $64,300,000, plus applicable taxes. Over a 17
year period (2013 - 2030) the cost to US car owners would be over $109 trillion, and that's just
at a cost of 1 penny more per gallon and excluding city, county, state and federal taxes.
Our Response:
We received comments in support of and against our costs assessments, which included
average and marginal costs for the sulfur standards. Commenters that agreed with our costs
assessments viewed them as reasonable in light of the benefits of the Tier 3 program. Several
comments disagreed with the refinery industry's costs assessments that characterized our costs
assessments as underestimating the average and marginal costs of sulfur controls.
As explained in Chapter 5 of the Final Regulatory Impact Analysis, we made a number
of changes to the cost model since our proposal. Some of these changes are in response to
comments we received on the proposed rulemaking, others are in response to the first round of
peer reviewer comments and others were due to our own initiative to further improve the
refinery model. Because of the number and scope of changes to the cost model, we conducted
a second round of three independent peer reviews. We then addressed or incorporated the
comments of the second round of peer reviews as appropriate. We also reviewed how credits
are traded under the Tier 2 sulfur control program and found that sulfur credits are frequently
traded between refining companies. Discovering that sulfur credits are widely traded, we
changed the basis for the cost analysis to assume nationwide credit trading. The refinery cost
model estimates that Tier 3 will cause refiners to make about $2 billion in capital investments
and the program will cost 0.65 cents per gallon when the program costs are averaged over the
entire U.S. gasoline pool. We believe that these costs are reasonable for achieving much lower
sulfur gasoline and are far outweighed by the benefits due to lower emissions of environmental
pollutants.
We also reviewed cost estimates derived by several commenters who conducted their
own cost studies. The cost studies conducted to estimate sulfur control costs for Tier 3 gasoline
sulfur control to a 10 ppm standard derived very similar costs. The principal differences in
costs can be attributed in large part to the impact of the averaging banking and trading (ABT)
program. In total, three cost studies were conducted which estimated the cost of a 10 ppm
average sulfur standard. In addition to our cost study, the American Petroleum Institute (API)
and the International Council for Clean Transportation (ICCT) conducted cost studies to
estimate the cost of complying with the proposed 10 ppm average cost standard. The
following table summarizes the cost estimates conducted by the various cost studies.
7-18
-------
Tier 3 Summary and Analysis of Comments
Summary of Average Per-Gallon Costs (7% before-tax ROI)
With Credit
Trading
EPA
0.65
No Credit Trading
EPA
0.87
API
0.97
ICCT
0.8
The above table shows that our estimated costs for a 10 ppm average gasoline sulfur
standard which includes an ABT program is 0.65 cents per gallon. For the three cost studies
which assessed the costs without allowing for credit generation and trading, estimate costs fall
in the range of 0.8 to 1.0 cents per gallon.1
To derive the per-gallon costs by API shown above, we had to combine the results of
two separate cost studies conducted by API. One API cost study conducted by Baker and
O'Brien estimated the cost of a 10 ppm average sulfur standard with a 20 ppm cap standard,
which is 1.58 cents per gallon (before-tax 7% ROI). The second API cost study conducted by
Turner Mason and Co. estimated the cost of a 20 ppm cap standard relative to the current 80
ppm cap sulfur standard. The 1.58 cents per-gallon cost estimated by Baker and O'Brien study,
which is based on an estimated capital cost of 9.6 billion dollars, is offset by removing API's
estimated 6 billion capital cost for a 20 ppm cap standard, thus reducing API's per-gallon cost
estimate from 1.58 cents per gallon to 0.97 cents per gallon. While we used the two API cost
studies to estimate a per-gallon cost without a 20 ppm cap standard, we in no way endorse these
cost studies as both studies have overestimated the capital costs. The result is nevertheless
instructional. The costs of all three organizations looking at the same standards, but without
assuming the ability for refineries to generate and trade sulfur credits result in very similar
costs. This serves to support EPA's belief that had these studies also evaluated the impact of
trading sulfur credits between refineries they would have also resulted in cost estimates similar
to our 0.65 cent per gallon estimate.
One of the commenters suggested that EPA should consult directly with refining
companies to estimate the costs. We did so extensively during the development of the rule,
meeting individually with companies that represented more than 60% of the refining capacity
in the US. In these discussions the input we received ranged broadly from general support for
our estimates to general support for API's estimates or comparably high cost estimates. During
the development of the rule we also consulted heavily with the contractors, technology firms,
engineering firms, and vendors that service the refining industry, including submitting our
modeling to two separate rounds of independent peer review by industry experts. This
expertise was all utilized in our modeling and our assessments. The Natural Resources Defense
Council (NRDC), in their comments on the proposal provided an overview of the public
1 As summarized and discussed below in Chapter 7.2.5 of this Summary and Analysis of Comments document,
the oil industry states that the LP refinery modeling cost analysis conducted by Mathpro for ICCT reflect some
averaging, like a credit trading program, which is inherent in LP refinery models. In its cost analysis for ICCT,
Mathpro assumed that each refinery would either add a grassroots unit or revamp an existing FCCpostreater and
chose a representative unit size for the refineries in each PADD that it modeled, Thus, we believe that Mathpro
avoided most all the potential overoptimization, and related underestimation in costs, which can occur using such
refinery models.
7-19
-------
Chapter 7: Costs
statements made by some of the publicly traded refiner (see Chapter 7.2.3 below). While this
could not be used in our modeling or analysis, it may provide some general perspective on
expected Tier 3 compliance costs.
7.2.2.2. Marginal Cost
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM)
Economic theory and discussion of average vs. marginal costs: Throughout EPA's analysis, and
especially in Chapter 5.2 of the DRIA, EPA places extraordinary emphasis on their average
cost calculations, which for reasons explained above, are biased downward. At the same time,
EPA mistakenly ignores the marginal cost impacts. EPA implicitly acknowledges the relevance
of marginal costs as evidenced by EPA's analysis of cost impacts refinery-by-refinery that
highlighted the marginal cost.
Adjusting for shortcomings mentioned above in EPA's modeling efforts, their results could
align very well with the Baker & O'Brien study. The marginal cost graph below [See Figure 9
on p. 31 of Docket number EPA-HQ-OAR-2011-0135-4276-A2] (from the Baker & June 28,
2013 31 O'Brien report) illustrates the point that a larger volume of gasoline production will
incur higher than average costs of production. By definition, half the volume falls on either side
of the average. Figure 9 below includes a point estimate of approximately 1 cent per gallon
(i.e., EPA and MathPro) and illustrates approximately 50% of gasoline production is covered
by the average cost, the other half is not. Even EPA's graphs demonstrate this fact.
The Baker & O'Brien study concluded that allocating compliance and capital cost on an
annualized basis resulted in a marginal cost of 6 to 9 cents per gallon in most markets. EPA's
own analysis in the DRIA (Figure 5-4) illustrates a marginal cost curve that appears to exceed
6.5 cents per gallon. Does EPA believe that gasoline will be delivered to market at the average
cost of production?
EPA needs to recognize that basic economics teaches that the market will have to bear the
marginal cost of bringing the last increment of demanded supply into the market. Thus the
marginal cost represents the true cost of this proposal to the consumer.
In section 9.3 of the DRIA, EPA reports that they do not estimate consumer price impacts of
the proposed Tier 3 program. However, EPA goes on to note that the increase "should be
positive and up to the increase of manufacturers' cost of gasoline production."
In this contradiction, not only does EPA point out their analytical shortcomings, but EPA has
also ignored economic theory. Increasing marginal costs of production, (illustrated in both
Baker & O'Brien and EPA) results in a positive relationship between total quantities supplied
and price of that quantity in the market place (i.e. supply curve).
7-20
-------
Tier 3 Summary and Analysis of Comments
EPA focuses on the average cost, which is misleading because refineries producing up to /^ of
gasoline production will by definition incur costs higher than average, some significantly
higher, in order to comply with the Tier 3 proposed regulation. In order to ensure these
marginal supplies of gasoline are brought to the market place, market signals will be required to
incentivize these incremental supplies.
Gasoline costs. Navigant's portrayal of market mechanisms is misleading - the fact is that
consumers ultimately bear the cost of regulations. This has been true with numerous
regulations, including Tier 2, and will likely be repeated with Tier 3 costs. Navigant's use of
European Brent spot prices as an example to compare against U.S. gasoline retail prices makes
little sense, since the Brent price has been very unrepresentative of actual average U.S. crude
costs for the past 5 years.
15 - MathPro Inc. Refining Economics of a National Low Sulfur, Low RVP Gasoline Standard.
Performed for The International Council For Clean Transportation. October 2011.
EPA's modeling shortcomings: It is interesting how EPA spends a large portion of Chapter 5 of
the DRIA discussing their refinery by refinery model. However, EPA fails to mention their
refinery by refinery results of 4.5 cents per gallon marginal costs even with an Averaging,
Banking, and Trading (ABT) program in place. Given the shortcomings mentioned above of
EPA modeling efforts, this result if adjusted properly would align very well with the Baker &
O'Brien results which did not include an ABT program.
Commenter: American Petroleum Institute (API) and the Association of Fuel &
Petrochemical Manufacturers (AFPM)
EPA's calculation errors: The costs that EPA reports on Figure 5-6 [of the DRIA] and in the
proposal text are a nationwide average cost of 0.79 cents per gallon. However, if one simply
uses the data on the graph, it appears that the average cost is 1.08 cents per gallon. If one takes
the same steps with Figure 5-1, which is very similar to the Figure VII-2 used in the proposal,
the average cost calculates to be 1.21 cents per gallon instead of the 0.97 cents per gallon that
EPA uses in the text of the proposal. One would suspect that these differences are due to
volume weighting but for EPA to simply slap an average cost on a graph which does not
demonstrate that average is very misleading and confusing. At the very least, EPA should add
an explanatory footnote.
Commenter: Marathon Petroleum Company LP (MFC)
EPA makes the assumption that mandated 10 ppm average sulfur level will result in refineries
targeting exactly 10 ppm. Refinery operations do not always run exactly as planned and as a
result refiners must aim below the 10 ppm target, since ending the year above 10 ppm is
unacceptable. Refineries will aim for somewhere around 8 ppm, so that an upset at the end of
the year does not result in non-compliance for the entire year. Even with Averaging, Banking
and Trading (ABT) program, the refineries generating credits and those using the credits will
target around 2 ppm below the average that they must have for compliance. The ability to
create Tier 2 sulfur credits is in part caused by this under targeting of sulfur levels. EPA
assumes that refineries operate with perfection and in that world everyone will on average
7-21
-------
Chapter 7: Costs
achieve exactly 10 ppm sulfur in gasoline. The fact is that the refining industry has almost
always overachieved when meeting mandated product specifications. EPA needs to redo their
analysis based on an 8 ppm sulfur average.
EPA's modeling shortcomings: It is interesting how EPA spends a large portion of Chapter 5 of
the DRIA discussing their refinery by refinery model. However, EPA fails to mention their
refinery by refinery results of 4.5 cents per gallon marginal costs even with an Averaging,
Banking, and Trading (ABT) program in place. Given the shortcomings mentioned above of
EPA modeling efforts, this result if adjusted properly would align very well with the Baker &
O'Brien results which did not include an ABT program.
Economic theory and discussion of average vs. marginal costs: Throughout EPA's analysis, and
especially in section 5.2 of the DRIA, EPA places extraordinary emphasis on their average cost
calculations, which for reasons explained above, are biased downward. At the same time, EPA
mistakenly ignores the marginal cost impacts. EPA implicitly acknowledges the relevance of
marginal costs as evidenced by EPA's analysis of cost impacts refinery-by-refinery that
highlighted the marginal cost.
Adjusting for shortcomings mentioned above in EPA's modeling efforts, their results could
align very well with the Baker & O'Brien study. The marginal cost graph below (from the
Baker & O'Brien report) illustrates the point that a larger volume of gasoline production will
incur higher than average costs of production. By definition, half the volume falls on either side
of the average. Figure 7 includes a point estimate of approximately 1 cent per gallon (i.e. EPA
and MathPro) and illustrates approximately 50% of gasoline production is covered by the
average cost, the other half is not. Even EPA's graphs demonstrate this fact.
The Baker & O'Brien study concluded that allocating compliance and capital cost on an
annualized basis resulted in a marginal cost of 6 to 9 cents per gallon in most markets. EPA's
own analysis in the DRIA (Figure 5-4) illustrates a marginal cost curve that appears to exceed
6.5 cents per gallon. Does EPA believe that gasoline will be delivered to market at the average
cost of production?
EPA needs to recognize that basic economics teaches that the market will have to bear the
marginal cost of bringing the last increment of demanded supply into the market. Thus the
marginal cost represents the true cost of this proposal to the consumer.
In section 9.3 of the DRIA, EPA reports that they do not estimate consumer price impacts of
the proposed Tier 3 program. However, EPA goes on to note that the increase "should be
positive and up to the increase of manufacturers' cost of gasoline production."
In this contradiction, not only does EPA point out their analytical shortcomings, but EPA has
also ignored economic theory. Increasing marginal costs of production, (illustrated in both
Baker & O'Brien and EPA) results in a positive relationship between total quantities supplied
and price of that quantity in the market place (i.e. supply curve).
7-22
-------
Tier 3 Summary and Analysis of Comments
EPA focuses on the average cost, which is misleading because refineries producing up to /^ of
gasoline production will by definition incur costs higher than average, some significantly
higher, in order to comply with the Tier 3 proposed regulation. In order to ensure these
marginal supplies of gasoline are brought to the market place, market signals will be required to
incentivize these incremental supplies.
Gasoline costs: To believe Navigant, refiners did not pass any Tier 2 costs, average or
marginal, to the consumer and are therefore highly unlikely to pass the Tier 3 costs on to the
consumer. One has to wonder what kind of economist would truly believe that businesses don't
attempt to pass their costs on to consumers. If businesses can't pass any of their costs on to the
consumer; how do they stay in business? Playing sleight of hand with multiple regression
analysis does not change the economic tenant that businesses must recover their costs to stay in
business. Using a European Brent spot price to compare against U.S. gasoline retail prices
makes little sense, since the Brent price has been very unrepresentative of actual average U.S.
crude costs for the past 5 years.
Commenter: United States Congress, House of Representatives, Pennsylvania, 7th District
Some estimates show that the proposed new rules will cost the American motorist an additional
6 to 9 cents per gallon. Additionally, the EPA should consider the example of California, where
it already has imposed regulations similar to Tier 111 and its motorists pay 40 cents more per
gallon than the national average.
Commenter: MathPro
Another difference is that our focus was on estimating average refining costs. This is
sometimes called a social cost or a national cost. And we did not - well, we did not return or
publish a marginal cost, whereas the headline number in the Baker and O'Brien study was a
marginal cost, you know, the economics of a high cost refiner. And we did not do that because
we didn't think it was appropriate for this analysis.
Commenter: Emissions Control Technology Association (ECTA)
It should be noted that Baker and O'Brien's estimate implies that Tier 3 will generate profit for
the refining industry. Baker & O'Brien estimated that the marginal cost of compliance—that is,
the cost of compliance to the refinery that would incur the highest costs—would be 6 to 9 cents
per gallon. Baker & O'Brien also assumed (without justification) that the highest cost refineries
would set the price of gasoline. The difference between the average cost and marginal cost is
the oil industry's profit. Thus, Baker & O'Brien's analysis suggests that the oil industry will
profit from 10 ppm low sulfur standard by roughly 4 to 7 cents per gallon. If this were the case,
then the oil industry would support Tier 3.
20 - EPA Proposal, Fed. Reg. at 29977.
24 - Id. at 29977.
25-Id. at 29978.
7-23
-------
Chapter 7: Costs
26 - See, e.g., EPA, Regulatory Impact Analysis: Heavy Duty Engine and Vehicle Standards and
Highway Diesel Fuel Sulfur Control Requirements, EPA420-R-00-026, Dec. 2000.
27 - Mathpro, Refining Economics of National Low Sulfur, Low RVP Gasoline, Oct. 2011, at p. 13
(hereinafter Mathpro Study).
Commenter: Natural Resources Defense Council (NRDC)
The main difference between the API 6 to 9 cents/gallon estimates and EPA's 0.9 cents/gallon
estimate is expected profit for the oil industry. As EPA describes in the DRIA, API's cost
estimate assumes a profit of 4 to 7 cents per gallon, which translates into $4 to $8 billion in
windfall profits for the oil industry each year (18).
The API estimates are based on a study by consulting firm Baker and O'Brien (19) that finds 6-
9 cents per gallon is the marginal cost of upgrades to refineries that need the highest cost
upgrades. The study assumes that the market price for gasoline would be set by the small
number of high upgrade-cost refineries. A more appropriate measure of gasoline price impact is
the average industry cost of compliance because the market prices will be set by the majority of
gasoline volume that is produced by refineries that do not need to make expensive upgrades.
Based on EPA's assessment of refinery-by-refinery costs without a sulfur credit averaging,
banking and trading (ABT) program, fewer than 10 refineries of the 111 U.S. refineries would
incur costs above 3 cents per gallon and average refinery costs would be about 0.97
cents/gallon (20). With the proposed ABT program, the average cost is further lowered to 0.89
(or 0.9) cents/gallon.
Using data from the Baker and O'Brien study, EPA was able to assess average industry
compliance costs for 10 ppm gasoline. Sensitivity Case 4 from the Baker and O'Brien study
shows annual costs of $2.39 billion with an expected gasoline production of 7,343 thousand
barrels per day, not including California refineries that already produce an average of 10 ppm
gasoline. These industry assumptions result in an average cost of 2.12 cents/gallon. Compared
to the average cost of 2.12 cents/gallon, API's gasoline price impacts of 6 to 9 cents/per gallon
would result in windfall profits of 3.88 to 6.88 cents/gallon, or $4.3 billion to $7.7 billion in
annual profits at their projected production rates. With these windfall profits, the oil industry is
projecting an astounding rate of return on investments of 180 to over 330 percent.
Our Response:
We received comments in support of and against our costs assessments, which included
average and marginal costs for the sulfur standards. Commenters that agreed with our costs
assessments viewed them as reasonable in light of the benefits of the Tier 3 program. Several
comments disagreed with the refinery industry's costs assessments that characterized our costs
assessments as underestimating the average and marginal costs of sulfur controls.
API reports a cost estimate of 6 - 9 cents per gallon based on the refineries with the
highest Tier 3 gasoline sulfur compliance costs in its cost analysis. API and commenters citing
their study are trying to make the case that the refineries with the highest Tier 3 gasoline sulfur
7-24
-------
Tier 3 Summary and Analysis of Comments
compliance costs will translate into the price increase for gasoline at retail and that this
represents the true cost of the Tier 3 sulfur control program. This argument fails on both
counts.
First, while economic theory would support that the marginal cost supplier of gasoline
may in fact set the price of gasoline in its various submarkets, there are far more and larger
factors that impact the marginal cost supplier of gasoline than the cost of gasoline sulfur
control. These include crude oil costs, refining costs, refinery complexity, operating costs,
distribution costs and on and on. There is no evidence to suggest, and no evidence provided by
the commenters, that refineries with the highest marginal costs for Tier 3 gasoline
desulfurization bear any relationship to the marginal cost suppliers of gasoline in a market. A
simple example highlights this. According to the Pace refining margin report for the fourth
quarter of 2013, the US refineries with the poorest margins are those refineries refining a light
sweet crude oil (Bonny Light). Also, the refinery likely setting the marginal price is probably a
non-conversion refinery (does not have an FCC unit) because the refinery is not realizing the
cost benefit of upgrading the poorest quality, lowest value part of the crude oil barrel.
However, because this refinery does not have an FCC unit, its gasoline sulfur level is likely
already at 10 ppm or below, and thus, it would not be incurring any cost under Tier 3 (and it
could actually earn credits if its gasoline sulfur is less than 10 ppm). The price setter refinery
in this case will not incur a gasoline production cost increase under Tier 3, which could result
in no impact on gasoline prices in the gasoline submarket.
'Second, even if the refineries with the highest marginal cost of gasoline production and
the highest marginal Tier 3 gasoline desulfurization costs in each submarket were one and the
same refineries, it may help inform potential market price impacts resulting from Tier 3, but it
would not be appropriate to use this in assessing the cost of the Tier 3 program. The Office of
Management and Budget (OMB) lays out the methodology to be used for estimating the costs
and benefits of new regulatory programs.2 OMB states that the estimated costs not include any
transfer payments. API's cost analysis (not adjusting for the 20 ppm cap standard that the API
cost analysis assumed) estimates that the social cost of the Tier 3 sulfur control program is 1.58
c/gal (7% before-tax ROI), however the reported marginal costs is 6 - 9 cents per gallon. Thus,
the amount of the transfer payment is 4.42 - 7.42 cents per gallon, which corresponds to 4 to 8
billion dollars per year. Thus, the marginal cost estimated by the API cost assessment would
overestimate the social cost by a factor of 300 to 500 percent. It assumes that consumers would
pay this additional cost, resulting in the oil industry realizing a substantial profit from Tier 3.
Furthermore, the marginal cost estimates assumed by the commenters are inflated. As
discussed in Section VII.B of the preamble to the final rule and in Chapter 5 of the RIA, the
cost analysis conducted by Baker and O'Brien modeled a more stringent sulfur control program
than the one finalized by EPA. The Baker and O'Brien cost analysis assumed a 20 ppm cap,
which precluded refineries from utilizing the trading flexibility inherent in the Tier 3 program.
This caused every refinery to have to invest capital to comply rather than allowing them the
flexibility to comply through the use of credits. This in turn drove up the marginal costs. In
2 Economic Analysis of Federal Regulations under Executive Order 12866, Office of Management and Budget,
January 11, 1996.
7-25
-------
Chapter 7: Costs
addition, as discussed in Chapter 7.2.3 below, the Baker and O'Brien study assumed capital
costs well above what is reasonable, further driving up the marginal cost estimates. In contrast
to the Baker and O'Brien marginal cost range of 6-9 cents per gallon, our final rule estimates
marginal costs only as high as 2.8 cents per gallon, and that is only for a single refinery that
very well may have other compliance options available to it that we were unable to model. The
next several highest marginal cost refineries are in the 1.5-2.1 cent per gallon range, and only
15 refineries are projected to have costs higher than 1 cent per gallon.
What Commenters said:
Commenter: Marathon Petroleum Company LP (MFC)
EPA's calculation errors:
The costs that EPA reports on Figure 5-6 and in the proposal text are a nationwide average cost
of 0.79 cents per gallon. However, if one simply uses the data on the graph, it appears that the
average cost is 1.08 cents per gallon. If one takes the same steps with Figure 5-1, which is very
similar to the Figure VII-2 used in the proposal, the average cost calculates to be 1.21 cents per
gallon instead of the 0.97 cents per gallon that EPA uses in the text of the proposal. One would
suspect that these differences are due to volume weighting but for EPA to simply slap an
average cost on a graph which does not demonstrate that average is very misleading and
confusing. At the very least, if EPA has adequate justification for this average cost, it must add
a detailed explanatory footnote.
Our Response:
We changed the format of these figures in our final rule document to indicate the full
range of costs across all the refineries. The relationship between the average costs and the
range of refinery-specific costs is clearer to the reader.
7.2.3. Capital Cost
What Commenters Said:
Commenter: American Fuel & Petrochemical Manufacturers (AFPM)
A Baker and O'Brien analysis estimates that a gasoline sulfur content reduction will cost
refiners $10 billion in capital costs, and $2.4 billion per year in operating costs, increasing the
cost of producing gasoline by six to nine cents per gallon. [This comment can also be found in
section 7.2.2 of this comment summary.]
Commenter: American Petroleum Institute (API) and the Association of Fuel &
Petrochemical Manufacturers (AFPM)
EPA has chosen to underestimate the costs of a proposed program in an effort to improve the
appearance of their cost/benefit analysis. EPA significantly underestimated refinery capital
7-26
-------
Tier 3 Summary and Analysis of Comments
costs, e.g., FCC post-treater 1,500 $/B vs. 6,800 $/B in Baker & O'Brien. Even though EPA
criticized the Baker & O'Brien capital costs, they admitted that EPA received two vendor
estimates for grassroots FCCU post-treating and elected to ignore the higher cost estimate.
Significant differences between preliminary vendor estimates and actual refinery project costs
EPA presumes a high degree of accuracy from vendor estimates and makes no adjustments to
reflect that vendors are typically incentivized to provide the lowest possible estimates. After
more thorough negotiations between refiners and vendors, engineering complexities are
addressed and costs for specific technologies at specific refineries typically increase. Expertise
in capital improvements at refineries is necessary to evaluate vendor estimates; EPA is not
appropriately challenging estimates to the point that they provide realistic cost estimates or
design factors.
Refiners and refinery construction contractors typically have an idea of the variability in capital
costs estimates, and they typically adjust vendor estimates accordingly based on their past
experiences with particular vendors. EPA lacks this expert knowledge and thus makes a low
estimate of capital costs. In addition, adapting theoretical design to a real refinery platform
always results in extra costs to handle these interface adjustments. Refiners and refinery
construction contractors typically add some percentage above the estimate to cover these
contingencies, but EPA has not factored in these costs.
Critique of Navigant study (16) - The Navigant study makes incorrect claims on:
Refinery capital costs. Navigant starts their analysis with the assumption that MathPro and EPA
are the "experts" in the area of sulfur reduction costs. In this particular case this is not an
accurate assessment; as both MathPro and EPA simply relied on rough vendor cost estimates.
On the contrary, as mentioned earlier, the Baker & O'Brien study was based on discussions
with many refiners to determine actual on the ground costs at refineries. In reality refineries
cannot add on small increments of new units and the supporting utilities as the MathPro and
EPA approach would suggest. As a result real world compliance costs are higher than rough
estimates using a notional or theoretical refinery.
16 - Schink, G. and Singer, H. (2012). Economic Analysis of the Implications of Implementing EPA's
Tier 3 Rules
Commenter: Emissions Control Technology Association (ECTA)
Second, Baker and O'Brien's capital cost assumptions are inflated compared to those used by
EPA. In particular, the grassroots FCC postreater costs used by Baker & O'Brien are much
higher than what EPA used (28). EPA's capital costs are consistent with those estimated by
both the National Petroleum Council and Jacobs. Baker & O'Brien based their capital cost
estimates on the actual installation for the Tier 2 program. Because a grassroots FCC postreater
installed for Tier 2 would typically remove ten times more sulfur than one installed for Tier 3, a
grassroots FCC postreater installed for Tier 3 would be significantly less capital-intensive than
a comparable Tier 2 FCC postreater. Yet Baker & O'Brien ignored this consideration entirely
when relying on the Tier 2 benchmark. Moreover, interviews of experts at companies who sell
and install desulfurization equipment to refiners also suggest that Baker & O'Brien's capital
cost estimates are not aligned with industry norms. For example, Baker & O'Brien's cost
7-27
-------
Chapter 7: Costs
assumption for naptha desulfurization is 257 percent higher than the experts' estimates.
Correcting Baker & O'Brien's capital cost estimates (along with its ROI assumption) brings its
average compliance cost estimates into alignment with those of EPA (29).
28 - EPA Proposal, Fed. Reg. at 29978.
29 - Id.
Commenter: Environmental Defense Fund (EDF)
Navigant interviews with companies engaged in implementing the refinery upgrades required
to reduce the sulfur content of gasoline confirm that the MathPro estimates are in the
reasonable range and that the Baker & O'Brien estimates are too high. The B&O study also
does not take into account the well-established flexibility provided in the program through
averaging and trading.
In fact, several individual refiners have countered API's claims about Tier 3. For example, the
Northern Tier Energy LP, a refining and retail gasoline company, states that "preliminary
engineering assessments predict that no capital spending will be required to comply with the
new standards' (111). CVR Refining, LP announced that "estimated costs to meet Tier 3
standards are less than $20 million' (112). Alon US Partners estimated Tier 3 will cost less than
$30 million (113) and Calumet Specialty Products Partners also made a public statement that
Tier 3 will not materially impact their business (114).
111 - Northern Tier Energy, Northern Tier Energy LP comments on EPA's proposed Tier III standards
(2013) [Press release], retrieved from http://finance.yahoo.com/news/northern-tier-energy-lp-comments-
183900329.html.
112 - CVR Energy, CVR Energy subsidiary CVR refining reports minimal investment required to meet
EPA proposed Tier III standards (2013) [Press release], retrieved from
http://mvestors.cvrenergy.com/phoenix.zhtml?c=203637&p=irol-
newsArticle&ID=1803055&highlight=.
113 - Alon USA Energy,. Alon USA Energy comments on EPA's proposed Tier III standards (2013)
[Press release], retrieved from http://www.reuters.com/article/2013/04/08/tx-alon-epa-upgrades-
idUSnPnDA90077+160+PRN2013 0408.
114 - Calumet Specialty Products Partners, Calumet Specialty Products Partners, L.P. comments on
EPA's proposed Tier III standards (2013) [Press release], retrieved from
http://markets.on.nytimes.com/research/stocks/news/press release.asp?docTag=201304100830PR
NEWS USPRX DE91948&feedID=600&press symbol=4043615.
Commenter: International Council on Clean Transportation (ICCT)
Second, the API marginal costs are upwardly biased. Baker and O'Brien overestimated the
investment costs for FCC naphtha hydrotreating, as indicated by MathPro's informal contacts
with companies involved in refinery upgrading. Their survey showed that investment costs
used by MathPro are in a reasonable range.
Commenter: Marathon Petroleum Company LP (MFC)
7-28
-------
Tier 3 Summary and Analysis of Comments
The Agency has significantly underestimated the costs for the required gasoline sulfur
reductions, overestimated the benefits and failed to take into consideration the accuracy of the
ozone and PM measurements used in their justification for this rule. Compliance with this Tier
3 proposal will cost the refining industry about $10 billion, which is approximately what
compliance with the Tier 2 gasoline regulations cost the industry.
DRIA Cost/Benefit Analysis - Underestimated Costs: Well in advance of the release of the
EPA Tier 3 proposed rule, API published two studies conducted by Baker & O'Brien: a 2011
study titled "Potential Supply and Cost Impacts of Lower Sulfur, Lower RVP Gasoline" and a
2012 study "Addendum to Potential Supply and Cost Impacts of Lower Sulfur, Lower RVP
Gasoline."
EPA has chosen to underestimate the costs of a proposed program in an effort to improve the
appearance of their cost/benefit analysis. EPA significantly underestimated refinery capital
costs, e.g., FCC post-treater 1,500 $/B vs. 6,800 $/B in Baker & O'Brien.
Significant differences between preliminary vendor estimates and actual refinery project costs
EPA presumes a high degree of accuracy from vendor estimates and makes no adjustments to
reflect that vendors are typically incentivized to provide the lowest possible estimates. After a
more thorough negotiations between refiners and vendors will engineering complexities that
additional costs for specific technologies at specific refineries become fully acknowledged.
Expertise in capital improvements at refineries is necessary to evaluate vendor estimates, and
EPA is not appropriately challenging estimates to the point that they provide realistic cost
estimates or design factors.
Refiners and refinery construction contractors typically have an idea of the variability in capital
costs estimates, and they typically adjust vendor estimates accordingly based on their past
experiences with particular vendors. EPA lacks this expert knowledge and thus makes a low
estimate of capital costs. In addition, adapting theoretical design to a real refinery platform
always results in extra costs to handle these interface adjustments. Refiners and refinery
construction contractors typically add some percentage above the estimate to cover these
contingencies, but EPA has not factored in these costs.
Critique of Navigant analysis (6): The Navigant study makes incorrect claims on: Refinery
capital costs. Navigant starts their analysis with the assumption that MathPro and EPA are the
"experts" in the area of sulfur reduction costs. In this particular case this is not an accurate
assessment; as both MathPro and EPA simply relied on rough vendor cost estimates. On the
contrary, as mentioned earlier, the Baker & O'Brien study was based on discussions with many
refiners to determine actual on the ground costs at refineries. In reality refineries cannot add on
small increments of new units and the supporting utilities as the MathPro and EPA approach
would suggest. As a result real world compliance costs are higher than rough estimates using a
notional or theoretical refinery.
6. Schink, G. and Singer, H. (2012). Economic Analysis of the Implications of Implementing EPA's
Tier 3 Rules
7-29
-------
Chapter 7: Costs
Commenter: MathPro
Our analysis indicated that the refinery investment for meeting the 10 ppm gasoline sulfur
standard would be in the range of $3 to $4 billion per year, and that the corresponding total
annual refining cost would be in the range of $.9 to $1.5 billion a year.
In turn, we estimated that the capital cost of a grass roots FCC naphtha hydro treater would be
about $1,800 per barrel per day, which is a standard unit of capital cost measurement and
refining, and that includes both on-site and offsite investments.
Meeting the 30 ppm standard, we posited that the method of choice for refiners would be to
revamp their existing units rather than build grass units of another type. Refineries that now
meet the 30 ppm standard with these units would add facilities, go through a revamp, and
through that achieve the 10 ppm standard.
We understand that many of these units that are already in place already have the capability to
meet a 10 ppm standard, but in our analysis, we assume that all units in the U.S. refining sector
would require revamping.
The U.S. refining industry has already installed about 75 of these advanced units in order to
comply with Tier 2, and many similar units have been installed in other countries as well.
Hence, the range of capital costs and operating costs for these units is now reasonably well
established.
EPA: Earlier today we heard some testimony about a study that was commissioned by the
American Petroleum Institute. Are you familiar with that?
Mathpro: I am, Baker and O'Brien.
EPA: Would you care to comment on the primary differences that account for the cost range in
your work versus the six to nine cents that was cited this morning?
Mathpro: Sure. There are several. One has to do with the estimated capital cost for building
grass roots FCC post treaters and, hence, for revamping them. The estimate that we use, which
I mentioned in my testimony, is about $1,800 of daily barrel. I believe the Baker and O'Brien
study used a number that was about double that approximately. So that was one difference.
Commenter: National Association of Clean Air Agencies (NACAA)
In terms of impacts on refiners, EPA estimates that 111 refineries could potentially be affected
by Tier 3. Of this total, 16 would need to install new equipment to comply with Tier 3. Of the
remaining 95 refineries, 66 could meet the requirements by modifying their existing equipment
and 29 already comply with Tier 3 or could do so by making operational changes.
Commenter: Natural Resources Defense Council (NRDC)
API's Estimates are Not Supported by Member Companies. Public statements by several
refineries indicate that the expected cost of compliance with Tier 3 regulations will be lower
than API's 6 to 9 cents/gallon estimate, if material at all. Valero Energy, which is the largest
7-30
-------
Tier 3 Summary and Analysis of Comments
independent refiner in the U.S., recently told a group of financial analysts that its Tier 3
compliance will cost $300 to $400 million (24). Using Valero's high cost estimate of $400
million, with a 7 percent pre-tax return on capital and a 5 year depreciation, NRDC estimate
that Valero's s marginal cost is approximately 0.6 cents per gallon, which is more than 10 times
less than the API high estimate. For Valero, the 0.6 cents per gallon translates into annual costs
of approximately $97 million or just 2 percent of the $4.45 billion in pre-tax earnings that the
company made on refining in 2012.
- Northern Tier, which operates a refinery in Minnesota, stated in a press release that its
"current assessment is that the proposed updated standard will not have a material financial
impact to the Company. Preliminary engineering assessments predict that no capital spending
will be required to comply with the new standards" (25).
- CVR Energy, operating refineries in both Oklahoma and Kansas, announced that "the
proposed U.S. Environmental Protection Agency Tier III sulfur regulations will have no impact
on the company's previously announced full year 2013 distribution guidance... The full amount
of this upgrade was reserved as cash on the company's balance sheet at the close of its IPO on
Jan. 23, 2013... Operating cost increases as a result of the higher desulfurization levels of
gasoline are not expected to be material to the company's performance" (26).
- Calumet Specialty Products Partners, L.P., which is based in Indiana and operates eleven
facilities located in Louisiana, Wisconsin Montana, Pennsylvania, Texas and Missouri,
announced that its "current assessment is that the proposed updated standards are not expected
to have a material financial impact to Calumet. Calumet's preliminary assessments of its
operations indicate that capital spending requirements, if any, will be immaterial to comply
with the proposed updated standards" (27).
- Alon USA Partners, which owns and operates a refinery in Texas, issued a press release
stating that "[operating cost increases as a result of the higher desulfurization levels of
gasoline are not expected to be material to the partnership's performance" (28).
24 - Siders, Alison, "Valero: EPA Gasoline Plan Would Cost Hundreds of Millions of Dollars," Wall
Street Journal, April 2, 2013.
25 -http://www.reuters.com/article/2013/04/03/ct-northern-tier-epa-
idUSnPnNY88352+160+PRN20130403.
26 -http://www.prnewswire.com/news-releases/cvr-energy-subsidiary-cvr-refining-reports-minimal-
investment-required-to-meet-epa-proposed-tier-iii-standards-201264271 .html.
27 - http://online.wsj.com/article/PR-CO-20130410-907218.html.
28 - http://www.alonpartners.com/phoenix.zhtml?c=251560&p=irol-
newsArticle&ID=1803979&highlight.
Commenter: Phillips 66 Company
Project Costs in Excess of EPA Estimate:
Given the significant project scopes required, we believe EPA has significantly underestimated
the cost to comply with the lOppm proposal. In an independent study commissioned by API,
consultants Baker and O'Brien determined that the capital cost required by industry will be
about $10 billion in contrast to EPA's estimate of $2.1 billion.
7-31
-------
Chapter 7: Costs
Based on preliminary analysis of our potential refinery projects, Phillips 66 believes the
investment costs required to comply with the proposed standard are well in excess of EPAs
estimates. The high investment required for the rule continue to place competitive pressures on
an already capital intensive industry.
Commenter: Union of Concerned Scientists (UCS)
Meeting the 10 ppm standard is estimated to cost the refiners less than a penny per gallon. An
independent study by Navigant Economics confirms EPA's estimates, finding that the cost of
complying with Tier 3 for U.S. refiners is "in the vicinity of 1 cent per gallon" (13).
Additionally, Navigant states that "this expression of the compliance costs on a cents per gallon
basis does not imply that these compliance costs will be passed through fully to consumers."
According to analysis by the Center for American Progress, the oil industry is one of the
world's most profitable industries. The five largest companies took home a combined $118
billion in profits in 2012 alone (15). It's time for the oil industry to stop blocking progress -
and given their record profits, less than a penny a gallon is a small price to pay to protect the air
we breathe.
13 - Schink, George R. and Hal Singer. 2012. Economic Analysis of the Implications of Implementing
EPA's Tier 3 Rules. Online at
http://naviganteconomics.com/docs/061212%20Economic%20Analysis%20of%20the%20Implications
%20of%20Tier %203%20Sulfur%20Reduction%20Final_embargoed%20copy.pdf, p. 14, accessed on
June 21, 2013.
15 - Center for American Progress. 2013. Speed Trap: Big Oil Profits from High Gasoline Prices.
Online at http://www.americanprogress.org/issues/green/news/2013/02/06/51967/big-oil-profits-from-
high-gasoline-prices/ accessed on June 28, 2013.
Commenter: United States Congress, House of Representatives, Pennsylvania, 7th District
Refiners, including in the 7th congressional district and the greater Philadelphia area, are
currently meeting the Tier II vehicle standards that reduced sulfur in gasoline from 300 parts
per million to 30 parts per million — a 90-percent reduction.
American refiners have already spent billions of dollars to achieve a 90 percent reduction in
sulfur levels since 2004, and Tier III will require refiners to spend an estimated $10 billion in
new infrastructure and another $2.4 billion per year in operating costs. These Washington
requirements will further jeopardize our refining industry and workers, and increase the cost of
end product fuels for consumers.
Commenter: Navigant Economic
First, there are serious flaws in the approach employed by Baker and O'Brien on behalf of the
American Petroleum Institute to estimate the compliance cost to refineries. Second, I've
examined EPA's cost benefit analysis. It comports with economic logic and best practices, and
it validates our conclusions. Third, we examined the effects of the Tier 2 sulfur rule on the
price of gasoline and found little to no impact suggesting the Tier 3 effect will be negligible.
7-32
-------
Tier 3 Summary and Analysis of Comments
So going to my first point on the flaws in the Baker and O'Brien approach, there are several,
but today I just want to address three. Baker and O'Brien failed to account for averaging and
trading of credits, which reduces the compliance costs to refineries. As EPA explained,
refineries with pre and post treaters are able to achieve further sulfur reductions below the 10
parts per million standard at a relatively low incremental cost, and sell credits to refineries who
would otherwise be faced with grass roots investments.
Averaging and trading reduces EPA's estimated compliance cost from .97 cents per gallon to
.79 cents per gallon, a reduction of nearly 20 percent with averaging in trading. Between eight
and 25 high-cost refineries out of 11 would prefer to consume credits than to upgrade their
facilities. Baker and O'Brien's mistaken assumption that those high-cost refineries reduce the
sulfur content of their gasoline they produce to 10 parts per million severely inflates their
compliance cost estimates.
Second, Baker and O'Brien assume without support that the refinery with the highest cost of
compliance will set the price of gasoline. However, if this were true, then the profits of the
refinery industry, defined as the difference between the marginal cost and the average cost per
gallon, would be four to seven cents, at which point the refiners should endorse Tier 3. Given
the refiners' opposition, it seems that Baker and O'Brien's logic is faulty.
Third, Baker and O'Brien's estimated capital cost for a grass roots FCC post treater is not
reasonable according to companies that construct or install desulfurization units. Indeed, Baker
and O'Brien's capital cost estimates are between two and four times higher than industry
benchmarks. Baker and O'Brien states that their estimates were based on actual installation for
the Tier 2 program, but as EPA explains, a grass roots FCC post treater installed for Tier 2
would remove about 10 times more sulfur than one installed for Tier 3, and, thus, would have
significantly higher capital costs. Correcting this one defect in the Baker and O'Brien study
causes their cost estimates to come into alignment with those of MathPro and of EPA.
EPA: Thank you for the, I guess, the support for our cost analysis. But if you could just spend
just a moment, because you were cut a little short. So what are the main differences between,
you know, our one cent a gallon and the six to nine cent estimate from Baker and O'Brien?
Navagant: So I think the points that I rattled off were the main drivers. One is this conflation
of the average and the marginal. I think that what - when you actually saw for the average of
the Baker and O'Brien, you don't get that big of a divergence. You get something like one
versus two.
But even there, the difference can be entirely explained away with just one change, and that's
this capital cost component of these treaters. And the way that we tried to ascertain that it was
exaggerated was by interviewing actual providers of these equipments, these types of
equipment and founding out whether their costs were reasonable. And they told us they
weren't.
Our Response:
7-33
-------
Chapter 7: Costs
We received many comments in support of and against our capital costs estimates.
Commenters in support of our estimates viewed them as reasonable while other commenters
from the oil industry argued that we had underestimated those costs. Specifically, API
commented that EPA significantly underestimated Tier 3 refinery capital costs and compared
the capital cost it used in its cost study for a grassroots postreater, which is $6800 per barrel per
day, to EPA's capital cost which is $1,500 per barrel per day. Several commenters viewed the
oil industry's cost as inflated.
'We disagree with the cost comparison made by API.3 The total capital cost that we
used for a grassroots postreater for the NPRM is higher than what API reported in its
comments. The $1,500 per barrel per day value solely represents the inside battery limits costs
(the direct costs associated with the hydrotreater), but when the offsite factor (representing the
indirect costs of the hydrotreater) our capital costs total $1875 per barrel per day. When a
separate overdesign factor that we applied is included (which was not included in the cost table
in Chapter 5 of our Regulatory Impact Analysis document which summarized our capital cost,
but was described in our description of our refinery-by-refinery model), our capital costs for a
grassroots postreater increases to $2,015 per barrel per day.
Before discussing the methodology used in deriving two cost estimates and the final
cost estimates, we will review the steps that EPA and API took to estimate the capital costs for
installing a grassroots postreater.
EPA's Postreater Capital Costs
The cost estimates that we are making for refineries are made in two steps. The first
step is to estimate the inside battery limits (ISBL) costs, which are the costs for the primary
units (i.e., reactor, heater compressors) subunits (pumps, heat exchangers strippers) and the
piping, electrical and control hardware. The vendors understand their technology very well and
understand how to make such cost estimates because they know how their technologies are
currently configured and are being operated in refineries today (are the hydrotreaters being
operated severely or more mildly), and therefore understand how the grassroots and revamps
would have to be configured to achieve the targeted lower sulfur standard.
The second step is estimating other costs called outside battery limit (OSBL) costs
which include piping the new unit to other units, control building for control equipment and the
operations personal, cooling towers, electrical switchgear laboratory facilities, etc. Vendor
companies are generally not very good at estimating offsite costs because it requires experience
installing such units in refineries, and the vendor companies rarely do that sort of work. For
estimating the offsite costs for Tier 3 desulfurizing units we spoke to two engineering
3 In Baker and O'Brien's report for API, the capital cost that they used is $6540 per barrel per day ($228.8 MM
for a 35,000 barrel per day unit). However, when API submitted its comments to EPA, it reported a capital cost of
$6800 per barrel per day. We believe that API adjusted its capital cost to be on the same basis as our costs which
is 30,000 barrels per day instead of 35,000 barrels per day and this is why they report the value of 6800 instead of
6540.
7-34
-------
Tier 3 Summary and Analysis of Comments
companies (Foster Wheeler and Bechtel) which installed desulfurization units for Tier 2. In
both cases, the engineering companies estimated that the offsite costs would be about 35% of
the onsite costs and we used that offsite cost factor for our cost estimate (our offsite factor is
within the range of 30% to 50% estimated by one vendor). Thus, we believe we are making a
very good estimate for the ISBL and OSBL capital investments expected to be made for Tier 3.
This has been corroborated by two rounds of independent peer review by knowledgeable
experts in refining.
API's (Baker & O'Brien) Postreater Capital Costs
API's costs are based on Tier 2 compliance costs. API obtained either estimated
installation costs or actual installation costs (API did not specify) for FCC postreaters for
installation in 5 different refineries for complying with the Tier 2 30 ppm gasoline sulfur
standard. The postreater capital cost information which reflected cost information from the
years 2003 to 2005 was adjusted upward to reflect mid-year 2009 capital costs using the
Nelson-Farrar index and normalized to reflect a 35 thousand barrel per day unit. This resulted
in an average ISBL cost of $144.5 million for installing a Tier 2 compliant FCC postreater.
After API discussed this capital cost estimate with several refiners who built several of the
units in recent years, those refiners felt that the estimated capital costs that API had calculated
were too low, and one refiner thought that the estimated capital costs should be doubled. Based
on the opinion expressed by that one refiner, API doubled its estimated capital costs for a 35K
bbl/day FCC postreater to $228.8 million. For estimating the capital cost for revamps of
existing postreaters, API applied a generic 30% cost estimation factor as their means for
estimating the revamp costs from the grassroots cost.
Discussion of EPA and API Estimated Postreater Capital Costs
EPA and API used two totally different cost estimation methodologies for estimating
the capital costs. We estimated the FCC postreater costs using a bottom-up approach, while
API used a top-down approach. Either approach is viable, although we favor our method for
several reasons.
1) For complying with the Tier 2 gasoline sulfur standard, a typical refinery would have
installed an FCC postreater to desulfurize the FCC naphtha from about 800 ppm down
to about 75 ppm, a 725 ppm, or a 91 percent sulfur reduction. In the case of a
grassroots postreater that would be installed for Tier 3, the postreater would treat FCC
naphtha already low in sulfur due to the pretreater installed before the FCC unit (these
refineries are currently complying with Tier 2 using an FCC pretreater). Thus, the new
grassroots FCC postreater installed for Tier 3 would only have to reduce the FCC
naphtha from 100 ppm to 25 ppm, a much smaller 75 ppm or 75 percent sulfur
reduction. A grassroots FCC postreater installed for Tier 2 would typically remove 10
times more sulfur than one installed for Tier 3. This is important because a significant
portion of the FCC postreater capital cost is devoted to avoiding the recombination
reactions which occur when hydrogen sulfide concentrations are high and react with the
olefins contained in the FCC naphtha. Thus, a grassroots FCC postreater installed for
Tier 3 would be expected to be significantly lower in capital cost compared to a Tier 2
7-35
-------
Chapter 7: Costs
FCC postreater. API's costs are based solely on Tier 2 compliance costs, which is one
reason why their costs are high (we believe about 30% higher).
2) Refiner project cost estimates for adding an FCC postreater for Tier 2 may have
included additional costs for other subprojects that the refiner may have included with
the Tier 2 project, even though they may be unrelated to the FCC postreater project or
for reducing the sulfur level of gasoline. API would not be able to distinguish these
other costs from the desulfurization costs and would have included their costs in their
estimate for Tier 3.
3) In their project design cost estimates, engineering companies add on an additional
contingency cost beyond the best estimate for completing the project. They do so to
guarantee the project completion under budget. We asked an engineering company
what a typical cost guarantee contingency factor would be for capital cost projects. The
engineering company typically adds 15% above and beyond the estimated costs.
4) Based on an anecdotal comment by a refiner, API doubled the capital costs that it uses
for an FCC postreater over the refinery installations cost estimates for Tier 2 which
were adjusted to current year dollars.
We believe that these various factors associated with API's cost estimation
methodology led to the overestimation of API's cost estimate for an FCC postreater. We know
that API's cost estimate for an FCC postreater is high because their estimated capital costs for
an FCC postreater are higher than that for an FCC pretreater sized the same, however, FCC
pretreaters are a much higher pressure and are constructed of much more expensive metallurgy
than FCC postreater units.
API commented that vendors are incentivized to provide the lowest possible costs and
EPA failed to adjust them higher to reflect realistic capital costs. We believe that vendors can
underestimate the cost of its costs when the vendors anticipate that its costs will be published
and compared to other vendor costs. In this case, we wanted to avoid placing the vendors in
that position, so we promised the vendors that we would not publish individual vendor cost
estimates, but aggregate them together. For this reason, we believe that the vendors provided
their best estimates for the capital costs of their grassroots units and revamps. In fact, we see
no basis for this assertion when we look across the range and variety of data and sources made
available to us. Conversely, these comments ignore the very real bias of the refining industry
estimates themselves in estimating the costs of EPA fuel standards, which has been
demonstrated in cost studies in the past. The only vendor that appeared to demonstrate a clear
bias was the vendor that is also a refiner, and provided costs far out of line with other sources
of information. This information proved unusable since it assumed Tier 2 was not in place, and
not a valid estimate for the increment of sulfur control needed for Tier 3. We also believe that
we are not overestimating capital costs based on the pitfalls inherent in the API capital cost
estimation methodology.
While we have high confidence in our estimated costs, the peer review comments make
a strong case for accounting for additional costs. Three peer reviewers on our cost analysis, as
well as oil industry comments, pointed out that refiners may incur additional costs not covered
7-36
-------
Tier 3 Summary and Analysis of Comments
by the ISBL and OSBL costs. For example, a refinery may need to add pilings to establish a
solid foundation for its reactor and other major units if the soil is poor, the project planning and
construction workers may require overtime pay, the refiner will incur some costs (termed
owner costs) for planning and executing the project, and there are miscellaneous contingencies
which must be addressed. Two of the peer reviewers suggested that we add an additional 30%
contingency factor to capture such costs. One of the peer reviewers commented that one of the
vendor's costs were very similar to the costs that they use for a two stage FCC postreater
provided that a 30 percent contingency factor is applied. Another peer reviewer suggested that
we add a 20% contingency factor while another peer reviewer supported the capital costs that
we used, which did not include a contingency factor, as the costs are consistent with the capital
costs that they used in their modeling and the capital costs estimated by Mathrpo for the ICCT.
After considering the various comments, we chose to use a 20% contingency cost factor added
to our capital costs to account for costs that may not otherwise have been fully accounted for.
What Commenters Said:
Commenter: Marathon Petroleum Company LP (MFC)
Even though EPA criticized the Baker & O'Brien capital costs, they admitted that EPA
received two vendor estimates for grassroots FCCU post-treating and elected to ignore the
higher cost estimate.
Our Response:
While we agree with the comment that we did not use the vendor cost that estimates a
higher cost, we disagree with the comment that we should have considered the higher cost. We
used the vendor that provided the ISBL cost estimate for a grassroots postreater for Tier 3 that
estimated the capital costs to be $1500 per barrel per day, for a 30 thousand barrel per day unit.
We rejected the other vendor that provided an ISBL cost estimate for a Tier 2 postreater
(desulfurizing FCC naphtha from 800 ppm down to 75 ppm) with estimated capital costs to be
$2894 per barrel per day for a 10 thousand barrel per day unit. When we adjusted the second
vendor's cost estimate to reflect a 30 thousand barrel per day unit size (to be on the same basis
as the other estimate), the second vendor's capital cost decreased to $1970 per barrel per day.
The difference between the two capital cost estimates reflects the cost of installing a Tier 2
grassroots FCC postreater ($1970 per barrel per day) compared to a grassroots Tier 3 FCC
postreater ($1500 per barrel per day) as we described above. If we would have based our
capital cost estimate on the second vendor that provided the capital cost for a grassroots Tier 2
postreater, we would have overestimated the capital cost for a grassroots FCC postreater
installed for Tier 3 by about 30%.
7.2.4. Octane Cost
What Commenters Said:
7-37
-------
Chapter 7: Costs
Commenter: American Petroleum Institute (API) and the Association of Fuel &
Petrochemical Manufacturers (AFPM); Marathon Petroleum Company LP
(MFC)
EPA's inadequate modeling of octane costs: For estimating the costs of the proposal, EPA
correctly pointed out that, "the most important input is the cost making up the octane loss that
occurs with desulfurization." [DRIA 5.1.2 page 5-3] However, EPA then goes on to
immediately dismiss the octane costs by assuming "the cost of octane is expected to decrease
dramatically due to expected much larger use of ethanol under the RFS2 rulemaking". [DRIA
5.1.2 page 5-4] Specifically, EPA assumes that by 2017, 50% of the gasoline market is E10 and
the other half is El 5, a contested fuel authorized under a partial waiver for late model vehicles
only.
First, it is not appropriate to assume high market penetration of a contested fuel simply because
it makes the Tier 3 proposal more palatable from a cost perspective. It is not clear that if given
the choice, consumers would willingly consume El 5 fuel given its lower energy content and
vehicle manufacturer statements advising against its use. It is also unclear whether the vehicle
fleet composition and fueling infrastructure in RFG zones will be able to convert 75% of
consumption to E15. EPA makes the assumption that since E15 does not qualify for the 1 psi
ethanol waiver, 100% of the E15 will be sold in RFG areas, which will make E15 75% of all
gasoline in those markets (page 7-12).
Second, EPA's assumption for the FCC octane loss penalty from increased desulfurization is
overly optimistic. In Table 5-18, EPA suggests that refiners will, on average, have to reduce
sulfur in FCC naphtha from 80 ppm to 21 ppm. Then for the vendor requests to estimate this
cost, EPA only requested a 75 ppm to 25 ppm target. The stated rationale appears to be that a
50-ppm reduction is close enough to the 80 to 21 calculated in their averages, even though
octane loss is not linear with desulfurization, especially at lower sulfur levels. The final result
in Table 5-26, is that the volume-weighted average octane loss for this target is 0.49 (R+M/2).
In their model, this half octane number reduction translated into 0.38 cents per gallon of FCC
naphtha (page 5-5 and table 5-41 which is incorrectly referenced). FCC naphtha is assumed to
be 36% of the gasoline pool, making EPA's net operating cost for octane loss an astonishingly
low value of 0.14 cents per gallon of gasoline.
Finally, this low octane cost per gallon of gasoline does not align with the trend of market
octane cost. By examining the spread between premium and regular grades of gasoline, and
knowing the difference in octane value of each grade, a cost/octane gallon can be calculated.
Figure 8 [of EPA-HQ-OAR-2011-013 5-4276] shows the annual average of daily posted prices
and makes it clear that octane is getting more expensive, even as ethanol blending has saturated
the market. Unless the market is assumed to be inefficient, it would appear octane has value
and it is increasing. This is in direct contradiction to EPA's assessment that costs associated
with octane destruction through desulfurization are largely negligible as described in the prior
paragraph.
Our Response:
7-38
-------
Tier 3 Summary and Analysis of Comments
The oil industry commented that we should use the wholesale price difference between
premium and regular gasoline grades to develop an octane cost. Using this method, they
estimated that the cost of octane would be $l/octane number-barrel. To verify API's estimated
octane cost using the wholesale price difference between premium and regular gasoline grades,
we also evaluated what the octane cost would be based on bulk pricing data differences
between premium and regular gasoline, divided by six. Based on actual data from 2008 to
2012, we estimated the average cost as $1.087 octane number-barrel of gasoline.
We believe that the premium minus regular grade pricing method for estimating octane
cost considerably overstates the cost of making up lost octane in the FCC naphtha pool.
Premium grade gasoline is produced in smaller batches than regular grade and must be handled
specially and separately to avoid compromising its octane content, and is transported in smaller
batches. For example, when shipping premium gasoline in pipelines, the interface between
premium and regular grades of gasoline must be downgraded into the regular grade gasoline to
ensure that premium gasoline's high octane content is not compromised - this downgrading
increases the production cost of premium gasoline. Furthermore, the pricing between regular
and premium grades of gasoline tends to reflect higher profit margins on premium fuel. Thus,
if we used the wholesale price difference between premium and regular grade gasolines, we
would be estimating the cost of not only producing the premium gasoline, but also include the
cost for special handling of premium gasoline relative to regular grade gasoline, plus some
profit, which is irrelevant to the cost of making up lost octane at the refinery due to Tier 3
sulfur control.
Instead, we estimated the cost of making up the lost octane in the FCC naphtha. To do
this we ran a control case using the Haverly LP refinery model which was run to model the
octane loss associated with desulfurizing gasoline in the year 2018. To only capture the cost of
the octane loss we reduced the octane value of the FCC naphtha by one octane number, and this
was the sole change relative to the reference case. The control case was run with capital costs
evaluated at a 15 percent rate of return on investment (ROI) after taxes.4 Unlike for the NPRM
where commenters criticized us for assuming El5 for the FRM analysis, this case was run with
E10 (no E15) and a small volume of E85, and we substituted 2013 natural gas liquid prices
(ethane, propane and butane) which are much lower compared to the historical price
relationship from previous years (the lower natural gas prices increases the octane cost
provided by the reformer since the reformer produces natural gas liquids as byproducts). The
octane cost estimated by the LP cost model is $0.3 I/octane number-barrel (0.74 c/octane
number-gallon). Because the octane loss associated with a specific sulfur control technology
may be lower or higher than 1 octane number, we scaled the octane cost based on the relative
estimated octane loss on the FCC naphtha (i.e., a 1A octane loss of the FCC naphtha was
estimated to cost $0.155/octane number-barrel.
4 Normally we conduct the refinery modeling assuming an after-tax 15% ROI and adjust the costs to reflect a
before-tax 7% ROI to report the costs. However, in this case because the new capital investments were so
minimal, we omitted the capital cost amortization adjustment because its effect on costs was judged to be
negligible.
7-39
-------
Chapter 7: Costs
However, it was further analysis of our LP refinery modeling work which further
explained why the price differentials between premium and regular grade gasoline overstates
the cost of octane when the octane loss is occurring in the FCC naphtha. While the octane cost
for making up octane loss in the FCC naphtha is $3 I/octane number-barrel from the LP model,
the octane cost determined by the premium-regular grade differential using the LP model is
$0.50/octane number-barrel, which is 60 percent higher. We believe that the LP refinery model
is estimating a higher octane cost for the premium-regular grade differential because of the cost
of producing premium gasoline, which is typically 6 octane numbers higher than the regular
grade. And yet, the LP model is not capturing the additional cost inflating factors mentioned
above such as smaller tankage, special handling and distribution and profit. Our conclusion
from this analysis is that the premium-regular grade price differential is a poor indicator of the
cost of making up the small amount of lost octane in FCC naphtha due to desulfurization and
by using it would overstate the cost. Despite our confidence in the octane cost that we
generated, we recognize the need to quantify the impacts on our costs if octane costs are indeed
higher than what we estimated, so where we present the Tier 3 fuel cost in the Chapter 5.2 of
the RIA for the FRM, we provide a sensitivity cost estimate assuming that octane costs
$0.50/octane number-barrel.
When we had our cost model peer reviewed the first time, one of the peer reviewers
also evaluated octane cost and concluded that while recent octane costs using premium minus
regular grade price differences were in the $ I/octane number barrel range, modeled future
octane costs for 2015 and 2020 were appreciably lower. This peer reviewer's estimate for
future octane costs could be much lower and provided a range of $0.25 - $0.5/octane number-
barrel. The octane cost that we are using is well within the projected octane cost range
projected by this peer reviewer and our octane cost sensitivity case is at the high end of this
range.
Based partially on the oil industry comments as well as our own understanding of the
hurdles that El5 is facing for widespread implementation, we elected to base our LP refinery
modeling octane cost case assuming E10 and a small volume of E85, but no assumed El 5
penetration.
What Commenters Said:
Commenter: Small Business Refiners (SBR)
In addition to capital costs and increased operating expenses incurred through additional
hydrotreating, we expect the refinery yield to decrease as a result of this change. As a result of
desulfurization through hydrotreating, the resultant fuel octane is lower. To compensate for a
lower octane blend component, the catalytic reformer severity must be increased to improve
blend component octane. The net effect of this process change is a one to two percent yield loss
on the products.
Our Response:
7-40
-------
Tier 3 Summary and Analysis of Comments
We agree that to the extent that refiners rely on reformers to make up for the octane loss
due to desulfurization (instead, refiners could use alkylation units, isomerization units or blend
in isooctane or more ethanol to make up for the lost octane), that there will be some yield loss.
This occurs because some of the reformer feed is cracked to lighter hydrocarbons. The cost
impact of this is reflected in our estimated octane cost. Because the reformate (the product
from the reformer unit) is higher in energy density than the feed, the volume loss that occurs
with reformers is somewhat offset by the higher energy density of the higher octane reformate.
The U.S. fuels market is now in a mode of slacking gasoline demand relative to increasing
diesel fuel demand, thus, the small reduction in gasoline yield associated with Tier 3 will have
no perceived impact on gasoline supply.
What Commenters Said:
Commenter: Marathon Petroleum Company LP (MFC)
EPA's assumption for the FCC octane loss penalty from increased desulfurization is overly
optimistic. In Table 5-18, EPA suggests that refiners will, on average, have to reduce sulfur in
FCC naphtha from 80 ppm to 21 ppm. Then for the vendor requests to estimate this cost, EPA
only requested a 75 ppm to 25 ppm target. The stated rationale appears to be that a 50-ppm
reduction is close enough to the 80 to 21 calculated in their averages, even though octane loss is
not linear with desulfurization, especially at lower sulfur levels. The final result in Table 5-26,
is that the volume-weighted average octane loss for this target is 0.49 (R+M/2). In their model,
this half octane number reduction translated into 0.38 cents per gallon of FCC naphtha (page 5-
5 and table 5-41 which is incorrectly referenced). FCC naphtha is assumed to be 36% of the
gasoline pool, making EPA's net operating cost for octane loss an astonishingly low value of
0.14 cents per gallon of gasoline.
Our Response:
We recognized after we completed the cost analysis for the NPRM that we were not
capturing the range of desulfurization costs experienced at refineries that had higher and lower
levels of sulfur reduction than what the vendor data estimates (although on an average basis,
our method of using typical desulfurization amounts is still valid). In response to comments, a
change we made in our cost analysis for the final rulemaking is to scale the sulfur reduction
costs based on the amount of sulfur reduction in the FCC naphtha. If the actual sulfur reduction
was 60 ppm instead of 50 ppm that the vendor data was based on, we increased the octane loss
and hydrogen consumption by 20 percent (60 ppm is 20% greater than 50 ppm). We also
created a cost factor for exponentially increasing desulfurization costs above a certain
desulfurization percentage (96% desulfurization for single stage units and 99% desulfurization
for two stage units), which reduced the potential for underestimating desulfurization costs for
cases of deep desulfurization.
The inclusion of additional vendor data coupled with our adjustments increased the
volume-weighted average octane loss for a revamp of an existing FCC postreater to produce 25
ppm FCC naphtha (to produce 10 ppm gasoline) from 0.49 (R+M)/2 to 0.77 (R+M)/2.
7-41
-------
Chapter 7: Costs
One reason why the octane cost number we are estimating seems "astonishing low" to
API is API is overestimating the cost of recouping octane by basing their octane cost on the
price differential between regular and premium grade gasoline as discussed above.
7.2.5. Modeling Techniques
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel &
Petrochemical Manufacturers (AFPM); Marathon Petroleum Company LP
(MFC)
DRIA Cost/Benefit Analysis: EPA makes the assumption that mandated 10 ppm average sulfur
level will result in refineries targeting exactly 10 ppm. Refinery operations do not run smoothly
and as a result refiners must aim below the 10 ppm target, since ending the year above 10 ppm
is unacceptable. Refineries will aim for somewhere around 8 ppm, so that an upset at the end of
the year does not result in non-compliance for the entire year. Even with Averaging, Banking
and Trading (ABT) program, the refineries generating credits and those using the credits will
target around 2 ppm below the average that they must have for compliance. The ability to
create Tier 2 sulfur credits is in part caused by this under targeting of sulfur levels. EPA of
course assumes that the world is perfect and in that world everyone will on average achieve
exactly 10 ppm sulfur in gasoline. The fact is that the refining industry has almost always
overachieved when meeting mandated product specifications. EPA needs to redo their
modeling and analysis based on an 8 ppm sulfur average to reflect real world conditions.
Our Response:
We acknowledge that refiners could target a lower sulfur level than 10 ppm to generate
credits for use within the company or for sale to other refining companies. However, we
believe our assumptions are reasonable because if a refiner thought that it would not be able to
sell the credits it generated, it would not continue to produce gasoline with less than 10 ppm
sulfur since they would be overcomplying at a cost to the refiner. Since FCC postreaters do not
provide any economic return, we believe that across the entire refining industry, refiners will
target 10 ppm on average. If they were overcomplying, they would adjust their operations and
average 10 ppm going forward to avoid the excess cost.
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel &
Petrochemical Manufacturers (AFPM)
It is well known that Linear Programming (LP) modeling approaches such as EPA's refining
model over optimize and underestimate refinery costs. The refinery LP approach assumes that
the entire PADD modeled essentially has infinite trading capabilities for crude properties, final
7-42
-------
Tier 3 Summary and Analysis of Comments
product properties, intermediate stream properties and unit processing capabilities. For
example, EPA's average refinery is assumed to have a small hydrocracker when in fact most
U.S. refineries do not have hydrocrackers. This assumption allows the notional refinery a
multitude of processing options that are unavailable to most refineries.
Commenter: American Petroleum Institute (API) and the Association of Fuel &
Petrochemical Manufacturers (AFPM); Marathon Petroleum Company LP
(MFC)
DRIA Cost/Benefit Analysis - Underestimated Costs: On page 5-41, EPA admits to not using
any refinery data more recent than 2009 and therefore assuming that most refineries with sulfur
averages above 80 ppm have an average of 30 ppm sulfur. This is a poor modeling assumption
to start off a refinery by refinery analysis.
Our Response:
Consistent with our normal refinery modeling practice which is also responsive to these
comments, we updated our cost analysis to more recent 2011 refinery gasoline quality and
volume data. Since all refineries were required to comply with Tier 2 in 2011 (all the small
refiner and geographic phase-in delays had expired at the end of 2010), we were able to use
refinery data for each refinery's gasoline sulfur level and not use estimates.
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel &
Petrochemical Manufacturers (AFPM); Marathon Petroleum Company LP
(MFC)
Another mistake that EPA makes in many areas is to use average values inappropriately. EPA
assumes that refineries process crude with average properties and designs the units in the EPA
refinery by refinery model on that basis. EPA provides no data on which specific crudes are
used by the refineries in this analysis. EPA fails to recognize that refinery units must be
designed with flexibility to handle the worst crudes and the poorest quality feedstocks
anticipated in the refinery slate. A grassroots or revamp FCCU post-treating unit designed on a
minimal basis to handle a 100 ppm sulfur feed will not produce the required sulfur reduction
when a 300 ppm or 500 ppm sulfur feed is used. EPA simply assumes that the high sulfur feed
will average with the low sulfur feed but this is not the case in the real world. EPA's use of the
"average case" and failure to "design" for a range of refinery feedstocks, results in a constant
bias toward underestimating refinery unit costs.
Our Response:
We do not agree with the comment that we should not be using average crude oil
qualities since this could affect compliance with the Tier 3 sulfur standard. These comments
may have some relevance if we were finalizing a 10 ppm cap standard that applies in other
7-43
-------
Chapter 7: Costs
countries around the world. However, because the Tier 3 sulfur standard is an averaging
standard, it is entirely appropriate to use average crude oil qualities. If, for example, a refiner
refined a crude oil containing 2.0 weight percent (wt%) sulfur for half the year and 1.0 wt%
sulfur for the other half the year, the refiner could assess how to comply with Tier 3 assuming
1.5% sulfur, which is the average between the two. Assuming that the refiner had targeted a
sulfur level of 10 ppm in its gasoline pool based on the average crude oil sulfur level, when
refining the 1.0 wt% sulfur crude oil, the refinery will likely average about 7 ppm gasoline, and
when refining 2.0wt% sulfur crude oil the refinery will likely average about 15 ppm gasoline.
However, since the Tier 3 sulfur standard is an annual average standard, it would be fine for the
refinery to average 7 ppm sulfur gasoline for half the year and 15 ppm sulfur gasoline for the
other half of the year. The very high 80 ppm cap standard allows the refinery this flexibility.
There may be situations where the crude oil refined by an individual refinery will result in
higher amounts of sulfur in gasoline than what we modeled and in that case we may have
underestimated the cost of compliance. However, our cost analysis is also most likely
overestimating the desulfurization cost at another refinery which is refining a different crude oil
which tends to produce gasoline with lower amounts of sulfur in the gasoline pool than what
we modeled. Since we are estimating costs on an industry wide basis, these sorts of issues tend
to average out across our analysis.
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel &
Petrochemical Manufacturers (AFPM); Marathon Petroleum Company LP
(MFC)
DRIA Cost/Benefit Analysis - Underestimated Costs - EPA assumes that the industry cost of
capital is 7% before taxes, which would approach 3.5% for most companies. This is compared
to the 10% after taxes that Baker & O'Brien assumed. EPA mistakenly assumes that this ROI
on regulatory driven refinery projects is all profit when in fact it represents the cost of obtaining
the capital for such large projects and the opportunity cost of that capital. EPA is way off base
with respect to the true cost of capital and the opportunity cost associated with regulatory
driven investment. In the real world, a cost of capital of 3.5% over the life of the investment is
far too low.
Our Response:
The oil industry commented that the methods used for amortizing capital costs do not
correctly represent a refiner's investment cost.
'In OMB circular A-94, OMB directs federal departments and agencies to assess capital
costs using a before-tax 7% ROI because it approximates the opportunity cost of capital. Also,
OMB circular A-94 instructs federal departments and agencies to assess costs at a different ROI
as a sensitivity analysis of another discount rate. In the RIA, we provide a cost estimate for the
final Tier 3 gasoline sulfur standard based on an after-tax 10% ROI. An important reason for
using the discount rate based on a before-tax 7% ROI is that all cost analysis conducted by
7-44
-------
Tier 3 Summary and Analysis of Comments
EPA are conducted on the same basis, which allows comparison of different environmental
programs on the same basis.
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel &
Petrochemical Manufacturers (AFPM); Marathon Petroleum Company LP
(MFC)
The MathPro study conducted for ICCT reports an average cost for PADD but in fact these are
not average costs but the costs for an average notional refinery in each PADD. If the costs for
refineries of different sizes fell along a straight line, this average refinery cost would be
appropriate. However, a typical cost curve for refineries and for refinery units is exponential in
nature not linear, as demonstrated in Figure 9 [of EPA-HQ-OAR-2011-0135-4276]. EPA
recognizes the exponential nature of economies of scale in their equation 5-4 on page 5-34
which incorporates the six-tenths rule. By applying this rule, curves such as EPA's DRIA
Figure 5-4 demonstrate that the costs are not linear and the mid-point of the curve is not
representative of the average of all the points on the curve. Thus EPA's own RIA analysis
demonstrates that the MathPro approach is faulty and does not represent the real average.
It is well known that modeling approaches such as MathPro's notional model underestimate
refinery costs. The notional LP approach assumes that the entire PADD modeled essentially has
infinite trading capabilities for crude properties, final product properties, intermediate stream
properties and unit processing capabilities. For example, MathPro's notional refinery is
assumed to have a small hydrocracker when in fact most U.S. refineries do not have
hydrocrackers. This assumption allows the notional refinery a multitude of processing options
that are unavailable to most refineries.
It should also be noted that MathPro's use of an average notional refinery represents some
elements of a sulfur banking and credit trading system. While it does not represent the factors
involved in the early credit program, it does represent the final sulfur ABT system since the
notional refinery has the average sulfur removal capacity in the PADD and achieves average
compliance across the PADD.
5. MathPro Inc. Refining Economics of a National Low Sulfur, Low RVP Gasoline Standard. Performed
for The International Council For Clean Transportation. October 2011.
Commenter: Emissions Control Technology Association (ECTA)
First, EPA's methodology is sound because it appropriately used a refinery-by-refinery cost
model that permitted each refinery to embrace a different sulfur-reduction strategy based on its
particular circumstances. EPA's cost model was peer-reviewed by two refinery industry
consultants, (17) and it was built off the model developed by Mathpro to estimate the cost of
benzene control under MSAT2 (18). The EPA notes that the peer-review process "suggested
7-45
-------
Chapter 7: Costs
that there would be little to no change in [its] desulfurization cost estimate," indicating that
EPA's model complied with best practices in the profession (19).
Commenter: MathPro
These investment estimates and refining cost estimates cover de-sulfurizing the refinery
streams needed to meet the 10 ppm average, supplying the additional hydrogen needed for de-
sulfurization, replacing the small losses in both gasoline volume and gasoline octane that
accompany tight de-sulfurization, and expanding refineries' sulfur recovery in offsites is
needed to support additional sulfur control.
We conducted our analysis using four regional refinery LP models representing projected
aggregated refining capacity in pads one through four, with all major regulatory programs
affecting fuel quality in place. We have used similar refinery LP models in many previous
analyses of the refining costs of new fuel standards, including Tier 3 gasoline sulfur, ULSD,
Federal and California RFG programs, and Arizona CBG.
In this study, our models represented the U.S. refining sector maintaining regional and total
U.S. gasoline production at the volumes projected by EIA for 2015, meaning the refineries
would adjust operations to replace the small losses in volume that would accompany tighter
sulfur control. Our projected product volumes, crude oil, and natural gas prices for 2015 were
developed from EIA's AEO 2011.
Our Response:
We received comments in support of and against the LP refinery model. The oil
industry commented that it was over-optimized resulting in underestimates of refinery costs.
We also received comments in support of the model.
We disagree with the oil industry comment that EPA LP refinery modeling case was
over optimized. The most important LP refinery modeling case that we ran for the Tier 3
rulemaking was the octane cost case. When we analyzed that case post modeling, the refinery
model simply increased the charge volume of heavy naphtha to reformers which were
underutilized. This means that the refinery model only needed to estimate operational changes
in octane level and cost which are not impacted by refinery size because capital investments in
reforming capacity were not necessary. Thus there was no risk of the refinery model
overoptimizing costs due to oversized assumed refinery unit (reformer) size. The second way
that the refinery model protected against overoptimization of octane costs was by pooling
refinery streams together, such as the light and heavy FCC catalytic naphtha. Although many
refineries (but not all) separate the light cat (FCC) naptha from the heavy FCC cat naphtha for
hydrotreating which allows those refineries to optimize their blending by blending them
separately, in our refinery cost model we choose a more conservative approach by pooling the
light and heavy cat naphtha together and blending the pooled FCC naphtha stream. This results
in the underoptimizion of gasoline blending of FCC naphtha. Also, "That a model is limited or
imperfect is not, in itself a reason to remand agency decisions based upon it." Id. "It is only
when the model bears no rational relationship to the characteristics of the data to which it is
7-46
-------
Tier 3 Summary and Analysis of Comments
applied that [courts] will hold that the use of the model is arbitrary and capricious."
Appalachian Power Company v. EPA, 135 F.3d 791, 802 (D.C. Cir. 1998)(internal citations
omitted).
Moreover, the oil industry failed to provide any information or data to support the
comment that Mathpro's LP cost analysis for ICCT, which is based on a PADD- average LP
refinery models, over optimized costs. LP refinery models do average the costs over many
refineries to estimate costs, but whether any LP refinery model over or under-optimizes capital
costs depends on what unit size is chosen in the refinery modeling. If the unit size being
modeled represents a smaller size than that of the refineries in the PADD the refinery model
would tend to underoptimize costs. However, if the unit size modeled is larger than average,
then the refinery model would tend to overoptimize costs. Why we say that the commenter did
not verify whether the Mathpro refinery model overoptimized capital costs is because the
commenter did not compare the unit size modeled to the unit sizes of the modeled refineries in
each PADD to verify that the unit size selected was too large. Reading through the refinery
modeling report, Mathpro discussed overoptimization of capital costs and did model a higher
capital cost for PADD 4 refineries to capture the poorer economies of scale for those smaller
refineries. This suggests that Mathpro did assume an FCC postreater unit size to avoid
overoptimization.
The Mathpro refinery modeling study did not model an ABT program as the oil industry
comment suggested because in the LP refinery model study Mathpro assumed either grassroots
installation or revamps of FCC postreaters at each refinery. If the Mathpro cost analysis was
modeling an ABT program, the model would have been set up to allow some refineries to
reduce their gasoline sulfur further than required and generate credits to show compliance for
other refineries through credit transfers to the refineries which do not take action. In this
regard, the Mathpro refinery model explicitly did not model an ABT program.
Regardless, these comments appear directed at the Mathpro study, not at EPA's cost
analysis. Linear program models such as the one used by Mathpro have several advantages and
several disadvantages compared to other modeling techniques. We have chosen for our
analysis to develop and use a refinery-by-refinery model for our cost analysis, but which is still
informed and aided by a refinery LP model when necessary and appropriate that is also widely
used by industry.
7.2.6. ABT Program Impacts on Cost
What Commenters Said:
Commenter: Emissions Control Technology Association (ECTA)
EPA accommodates an averaging, banking, and trading (ABT) mechanism that reduces the
compliance costs by permitting "high cost" firms to meet the sulfur requirement via consuming
credits. EPA found that many refineries with both pre- and postreaters are able to achieve sulfur
reductions beyond the 10 ppm standard at a relatively low incremental cost and then sell the
7-47
-------
Chapter 7: Costs
credits to those refineries that would otherwise be faced with grassroots FCC postreater
investments (21). Assuming counterfactually that each refinery were required to produce at
exactly 10 ppm—that is, without the benefit of trading—EPA estimated the average
compliance cost across all refineries to be 0.97 cents per gallon (22). With a credit-trading
system in place, EPA estimated the average compliance cost would fall to 0.79 cents per gallon,
with 25 refineries consuming credits. Under a more limited trading program, in which credits
could only be exchanged within companies that owned multiple refineries, EPA estimated that
the average cost of compliance would be 0.89 cents per gallon, with only eight refineries
consuming credits.
EPA conservatively chose to use the estimate from this restricted, intra-firm trading regime for
its proposal (23). Yet we believe that firms would have strong incentives to trade credits with
other firms under Tier 3. In particular, so long as the cost of consuming a credit plus the
transaction costs associated with inter-firm trading was less than the cost of reducing sulfur at a
given refinery, that refinery would consume a credit from another firm. Accordingly, we expect
the true average compliance cost to be closer to 0.79 cents per gallon. EPA's analysis was also
conservative in the sense that it restricted refineries that sought to reduce sulfur to achieve
sulfur levels of precisely 10 ppm or 5 ppm, when in reality, some of those refineries could
reduce sulfur to levels above 10 ppm (say 20 ppm) and consume credits, expanding the
opportunities for further trading and further average cost reductions.
Baker and O'Brien failed to properly consider the average, banking, and trading (ABT)
provisions of EPA's Tier 3 proposal. In particular, API imposed a rigid 20 ppm cap for each
refinery, which eliminates many of the opportunities for trading credits (30). In contrast, EPA
imposed a more flexible cap of 80 ppm, which permits additional gains from trading and
thereby reduces average compliance costs (31).
20 - EPA Proposal, Fed. Reg. at 29977.
21 -Id. at29973.
22 - Id. at 29974.
23-Id. at 29975.
28 - EPA Proposal, Fed. Reg. at 29978.
29 - Id.
30 - Id. at 29977. 3 lid. at 29972.
Commenter: Environmental Defense Fund (EDF)
The averaging, banking, and trading (ABT) program will help achieve these low costs by
allowing refiners to optimize their investments. And that low cost estimate assumes that less
than a quarter of the 111 existing refiners will participate in the ABT program. [This statement
also cross-referenced with section 5.5.1]
Commenter: National Association of Clean Air Agencies (NACAA)
Opponents of Tier 3 contend that the cost of low-sulfur fuel will be not less than a penny a
gallon, but between 6 cents and 9 cents per gallon. However, in making this estimate they did
7-48
-------
Tier 3 Summary and Analysis of Comments
not account for the mitigating impacts of EPA's proposed set of flexibilities, all of which have
proven successful in prior fuel programs adopted by the agency (e.g., lowering gasoline vapor
pressure).
These include 1) an annual average sulfur standard with an ability to use a higher per-gallon
cap; 2) an averaging, banking and trading program that would allow refiners six years —
January 1, 2014 through December 31, 2019 — over which to spread out their investment and
receive credit for early compliance by over complying with the current 30-ppm average sulfur
standard from 2014 through 2016; 3) relief for small refiners and refineries — those producing
fewer than 75,000 barrels a day — in the form of a three-year delay in compliance until
December 31, 2019; and 4) relief for economic and technical hardship, available to all refiners.
Our Response:
We agree with the comments that the flexibilities inherent in the ABT program lowers
the cost of complying with Tier 3 and will ease compliance with the Tier 3 program start date.
As summarized above in Chapter 7.2.2.1 of this Summary and Analysis of Comments
document, the ABT program is estimated to reduce the cost of the Tier 3 sulfur program by
about 25%. As we discuss in Chapter 5.1.4.3 of this document, the flexibilities offered by the
ABT program will enable refiners to earn early sulfur reduction credits prior to the program
start date which will provide refiners the flexibility to start up their grassroots or revamped
FCC postreaters that fits best in the schedule for installing and tying in the capital additions at
each refinery.
7.2.7. Other Issues
What Commenters Said:
Commenter: Pennsylvania Department of Environmental Protection (DEP)
The EPA should examine whether damage will occur to seals on gas station equipment or other
equipment. It is always a risk that equipment will be affected when fuel specifications change.
These risks do not appear to have been considered in the proposed rulemaking. Lower levels of
sulfur could cause some older gaskets in engines and equipment to leak. Additionally, although
EPA does not address the 15 percent volume content of ethanol fuel directly in this proposed
rule, the agency does indicate that it believes that 15 volume percent ethanol will eventually be
in widespread use. EPA should factor in the potential cost from damage that could occur as a
result of increasing ethanol content to 15 volume percent. EPA has failed in this proposed Tier
3 rule and in its earlier ethanol rule to address potential costs and emissions that will result if
the higher ethanol content fuel causes damage to gasoline storage and dispensing equipment or
other types of highway and nonroad equipment. Alcohol can degrade plastic, rubber and some
types of metals.
Our Response:
7-49
-------
Chapter 7: Costs
Tier 3 sulfur control is not expected to cause other fuel changes that might have any
impacts on vehicles and engines. The only expected impact of the sulfur control is the benefits
to the operation of the catalytic aftertreatment systems that are intended by the sulfur standard.
Concerns with respect to El 5 ethanol blends or other fuels are outside the scope of this
rulemaking.
7.3. Cost Effectiveness of the Tier 3 Standards
What Commenters Said:
Commenter: American Petroleum Institute and American Fuel and Petrochemical
Manufacturers
In their comments, API and AFPM argued EPA's cost-effectiveness analysis is fundamentally
flawed. They state that, "in assessing cost-effectiveness, EPA simply computes the costs
incurred in the years 2017 and 2030 and divides those costs by the emission reductions it
estimates will occur in those years. This approach is simply incorrect." They go on to say that,
"under the approach used by EPA in the RIA, the compliance cost for a given year represents
only the costs for new vehicles sold in that year; those costs are then divided by the cumulative
emission reductions achieved by the regulation, which include benefits from previous model-
year vehicles."
Commenter: Marathon Petroleum Company LP (MFC)
In their comments, Marathon argues that the proposal's cost effectiveness analysis was
inconsistent with OMB guidelines. Marathon notes that OMB states that the three basic
elements of a good regulatory analysis should include: 1) a statement of the need for the
proposed action; 2) an examination of alternative approaches, and; 3) an examination of
benefits and costs of the proposed action and the main alternatives (OMB Circular A-4, page
2). EPA's draft RIA does not include an examination of alternative approaches and fails to
meet the above basic criteria.
Marathon notes that, for major rule-makings, OMB recommends providing both a benefit cost
analysis (BCA) and a cost-effectiveness analysis (CEA) (OMB Circular A-4, page 9).
Marathon argues that, even though Tier 3 is a major rule, EPA did not provide a CEA that
addresses alternative measures to reduce PM2.5 and ozone air pollution, costs of alternate
measures, and cost comparisons and, without a CEA, EPA cannot determine if the Tier 3 rule is
the most cost effective way to reduce PM2.5 or ozone air pollution.
Our Response:
We acknowledge that the analysis presented in the proposal as a cost-effectiveness
analysis was not a "cost-effectiveness" analysis in the way the commenter is using the term and
should have been described as a "cost per ton-reduction" analysis. The numbers in proposed
7-50
-------
Tier 3 Summary and Analysis of Comments
preamble Table VII-9 (see 78 FR 29979) are, in fact, the annual costs of the program in the
indicated year divided by the tons reduced by the program in that given year.
That being said, a cost-effectiveness analysis is not a requirement of the Clean Air Act
sections under which our Tier 3 regulatory authority resides. We are required to give
appropriate consideration to cost, among other factors, when issuing vehicle standards under
section 202(a) of the Clean Air Act, but the statute does not prescribe a cost-effectiveness
analysis. Similarly, section 21 l(c)(l) of the Clean Air Act, which is our statutory authority to
issue fuel standards makes no mention of a need to demonstrate the cost-effectiveness of the
fuel standards. While a cost-effectiveness analysis is not a statutory requirement, we
nonetheless have sometimes presented cost/ton-reduced numbers in the supporting analyses for
regulations that we issue. "Because section 213 does not mandate a specific method of cost
analysis, we find reasonable the EPA's choice to consider costs on the per ton of emissions
removed basis." Husqvarna AB v. EPA, 254 F. 3d 195 at 200 (D.C. Cir. 2001)).
What Commenters Said:
Commenter: American Petroleum Institute and American Fuel and Petrochemical
Manufacturers:
In their comments, API and AFPM claim that EPA has not performed the incremental cost-
effectiveness analysis required to establish that individual portions of the Tier 3 Proposal are cost
effective. They state that, "the vehicle standards and sulfur reduction elements of the Tier 3
program are severable, as either could be implemented by EPA without the implementation of the
other. Given this, EPA should have assessed the costs and benefits of each element individually so
that the need for each element could be assessed based on its relative contribution to the overall
benefits expected from the Tier 3 proposal and in light of its relative contribution to the overall
costs."
They also claim that the fuel sulfur control provisions of the Tier 3 proposal provide only a small
portion of the total expected health benefits and are therefore "expected to be much less cost
effective than the proposed vehicle standards."
Commenter: Chevron Products Company
In their comments, Chevron states that they do not believe that the reduction in annual average
gasoline sulfur from 30 ppm to 10 ppm is warranted based on tailpipe emissions and cost
effectiveness. Chevron states their belief that that the reduction in annual average gasoline
sulfur from 30 ppm to 10 ppm is not warranted based on tailpipe emissions and cost
effectiveness.
Chevron also states that EPA's cost-benefit analysis is based on an assumption that a sulfur
level of 10 ppm is necessary to obtain the claimed benefits. Since EPA did not test at an
intermediate sulfur level, there was insufficient data available at the time the NPRM was
written to properly determine whether to set the new standard at 10 ppm or some intermediate
level between 10 and 30 ppm.
7-51
-------
Chapter 7: Costs
Chevron also states that the NPRM does not discuss the cost-effectiveness of the program as a
whole (vehicles and fuels) with a sulfur level between 10 and 30 ppm and that there is no
compelling argument for why the sulfur level should be 10 ppm and not somewhere in
between.
Commenter: Countrymark
In their comments, Countrymark argues that the Tier 3 reduction in gasoline sulfur will be
significantly more expensive per pound of sulfur removal, than previous regulations. They state
that the Tier 2 sulfur reductions cost approximately $100,000 per ppm of sulfur, while Tier 3
sulfur reductions are expected to cost $900,000 to $1,000,000 per ppm of sulfur.
Our Response:
EPA disagrees with the commenters' claim that the vehicle and fuel programs are
"severable, as either could be implemented by EPA without the implementation of the other."
For both the proposal and the final rulemaking, EPA analyzed the combined fuel and vehicle
program, in keeping with the fundamental principles of the Tier 3 program: that vehicles and
fuels should be treated as an integrated system, as has been successfully done in the mobile
source emission control program for over a decade.
As described in detail in Sections LA and IV.A.6 of the preamble to the final rule, there
is a first-order relationship between fuel sulfur content and the efficiency of modern three-way
catalytic converters. The Tier 3 final rulemaking addresses the stringency of exhaust gas
standards and fuel sulfur level as a system because, as documented in the scientific literature
and more fully described in Section IV.A.6 of the preamble to the final rule, the two are
strongly linked. For this reason, it is not possible to ascribe emission reductions, air quality and
human health benefits to the vehicle program independent of the sulfur program (or vice versa),
because the new vehicle program depends on the reduced sulfur level of the fuel program.
While it is not possible to identify the air quality impacts, health effects, or benefits of a
fuel sulfur program independent of a vehicle program, the final rulemaking contains a
significant amount of analysis that distinguishes between the emission reductions associated
with the fuel program on the existing fleet (pre-Tier 3 vehicles) and the emission reductions
associated with the vehicle and fuel programs on the new fleet (Tier 3 vehicles). See preamble
Section III.B of the preamble for these emission impacts.
In addition to the impacts of fuel sulfur standards alone on the pre-Tier 3 fleet and the
impacts of fuel sulfur standards and vehicle standards combined on the Tier 3 fleet presented in
Tables III-2 through III-4 of the preamble, EPA has also conducted air quality modeling for
2018 in order to understand the immediate impacts of the Tier 3 standards. The air quality
impacts in 2018 are mostly due to the impact of the fuel sulfur program on the existing fleet,
reflecting the corresponding reduction in emissions. These air quality modeling results are
presented in detail in Section III.C of the preamble to the final rule and RIA Chapter 7. Finally,
we performed an illustrative analysis of monetized impacts associated with the final rulemaking
in 2018 (See Section VIII.E of the preamble to the final rule and RIA Chapter 8.1.4). Total
benefits in 2018 are estimated to be between $1.9 and $5.6 billion, depending on the suite of
7-52
-------
Tier 3 Summary and Analysis of Comments
health impact studies and discount rates used. Total benefits in 2030 are estimated to be
between $6.7 and $19 billion.
The final rulemaking analysis also estimates the costs of the vehicle and sulfur
programs both separately and combined, for both 2018 and 2030.
The Tier 3 final rulemaking presents substantial and transparent analysis of the impacts
of the vehicle and sulfur standards, and the results of this analysis demonstrate the estimated
benefits of the final standards clearly outweigh the projected costs.
Further, as described in Section IV of the preamble, we believe that 10 ppm sulfur is
required to enable the Tier 3 standards. We see no reason to have analyzed the costs and
emission reductions of sulfur levels between 10 and 30 ppm, as suggested by Chevron, given
that the vehicle program would not be feasible at those intermediate sulfur levels. We received
similar comments - that EPA should have considered sulfur levels between 10 and 30 ppm -
and summarize them and respond to them in Chapter 4.1.4.3 of this Summary and Analysis of
Comments document.
Lastly, we do not see the relevance of comparisons between the Tier 2 and Tier 3
programs in terms of costs per ppm of sulfur reduction. Both programs are feasible, provide
significant public health benefits and do so at reasonable costs.
7-53
-------
Tier 3 Summary and Analysis of Comments
8. Estimated Benefits of the Proposed Rule
What We Proposed:
The comments in this chapter correspond to Section VIII of the preamble to the proposed
rule and address the estimated benefits of the Tier 3 program. A summary of the comments
received and our response to those comments are located below.
8.1. Comments on Benefits Methods and Assumptions
What Commenters Said:
Commenter: Institute for Energy Research (IER)
Second, EPA is required to publish key pieces of information when it proposes a new rule.
Specifically, EPA shall provide a summary of "the factual data on which the proposed rule is
based" and "the methodology used in obtaining the data and in analyzing the data."3 EPA has
failed to comply with this requirement. For example, the vast majority of benefits from this rule
come from reductions in particulate matter, but EPA has consistently refused to release the data
upon which it relies for these benefits calculations to the public. This failure is in clear violation
of President Obama's claim that his administration is "the most transparent administration in
history'.
Our Response:
The Tier 3 final rulemaking presents substantial and transparent analysis of the impacts
of the vehicle and sulfur standards, and the results of this analysis demonstrate the estimated
benefits of the final standards clearly outweigh the projected costs. For both the proposal and the
final rulemaking, we provide factual data on which the rulemaking is based, both in the preamble
and Regulatory Impact Analysis. We also provide extensive documentation of our data,
methods, and underlying assumptions in the docket that accompanies this rule (EPA-HQ- OAR-
2011-0135).
We also reject the commenter's claim that EPA has consistently refused to release the
underlying data that supports the PM2.5 benefits analysis. The peer-reviewed scientific evidence
that particulate matter (and ozone) are associated with significant public health impacts is robust,
well-understood, and has been documented extensively in the rulemaking package (for example,
see Chapter 6 and Chapter 8 of the RIA).
Questions have been raised in specific regard to the underlying data associated with the
PM-related mortality studies EPA uses in its risk and benefits analyses (the American Cancer
Society and Harvard Six Cities cohort studies). The data are held by the outside research
institutions that conducted these large-scale epidemiological studies, not the EPA, and the
Agency has released all of the data that we have received thus far from those institutions. The
Agency is also engaged with the relevant research institutions to get access to additional data.
8-1
-------
Chapter 8: Benefits
We emphasize, however, that the fact that some of the data is not currently public in no
way undermines the validity of the studies' results. Nor does it call into question the EPA's
reliance on those studies, along with thousands of other peer-reviewed studies, when the agency
considers the scientific foundation regarding methods used in risk and benefits assessments. In
fact, the original studies based on these cohorts already have been subject to reanalysis and
validation by the Health Effects Institute (HEI). HEI entered into confidentiality agreements
with the owners of the data to have access to the data in order to conduct a reanalysis of the study
cohorts. The reanalysis confirmed the validity of the findings and methodology.
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM); Marathon Petroleum Company LP (MFC)
Implausible health benefits due to highly conservative, unrealistic assumptions:
Using more realistic assumptions per existing references, the economic benefits of this proposed
rule would be markedly reduced lower than the costs in the proposed rule. Additionally, the
DRIA does not comply with various sections of the OMB Circular A-4 guidelines. The
references used in this section can be found in the Appendix [EPA-HQ-OAR-2011-0135-4276].
Our Response:
We base our analysis of the program's impact on human health and the environment on
peer-reviewed studies of air quality and human health effects.1'2'3 Our benefits methods are
consistent with the Regulatory Impact Analyses (RIAs) that accompanied the final revisions to
the National Ambient Air Quality Standards (NAAQS) for Particulate Matter, the 2008 final
ozone NAAQS, and the 2010 ozone NAAQS reconsideration.
These benefits methods are fully consistent with Circular A-44 and EPA guidance.5
However, we note that Circular A-4 provides guidance to Federal agencies on the development
of regulatory analyses under Executive Orders 12866 and 13563 and does not establish legally
enforceable requirements.
U.S. Environmental Protection Agency. (2012). Regulatory Impact Analysis for the Final Revisions to the
National Ambient Air Quality Standards for Particulate Matter. Prepared by: Office of Air and Radiation, EPA-
452/R-12-005. Retrieved November 22, 2013 at http://www.epa.gov/ttn/ecas/ria.html
2 U.S. Environmental Protection Agency. (2008). Final Ozone NAAQSRegulatory Impact Analysis. Prepared by:
Office of Air and Radiation, Office of Air Quality Planning and Standards. Retrieved November 22, 2013 at
http://www.epa.gov/ttn/ecas/ria.html. EPA-HQ-OAR-2009-0472-0238
3 U.S. Environmental Protection Agency. (2010). Summary of the updated Regulatory Impact Analysis (RIA) for the
Reconsideration of the 2008 Ozone National Ambient Air Quality Standard (NAAQS). Prepared by: Office of
Air and Radiation, Office of Air Quality Planning and Standards. Retrieved November 22, 2013 at
http://www.epa.gov/ttnecasl/regdata/RIAs/sl-supplemental_analysis_full.pdf.
4 Office of Management and Budget (OMB). 2003. Circular A-4: Regulatory Analysis. Washington, DC. Available
on the Internet at
5 U.S. Environmental Protection Agency (U.S. EPA). 2010. Guidelines for Preparing Economic Analyses. EPA 240-
R-10-001. National Center for Environmental Economics, Office of Policy Economics and Innovation.
Washington, DC. December. Available on the Internet at .
8-2
-------
Tier 3 Summary and Analysis of Comments
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM); Marathon Petroleum Company LP (MFC)
Inflated economic benefits based on "avoided or premature mortality":
To monetize the mortality related benefits, EPA continues to use the inappropriate metric of
"avoided deaths" and the Value of a Statistical Life (VSL). This approach is inconsistent with the
older age of the population most impacted by air pollution (Sunstein et al. 2004; Rabl et al. 2003
and 2006; Leksell and Rabl 2001). EPA VSL-only approach is also inconsistent with OMB
guidelines which call for the presentation of results of mortality valuation using both the value of
statistical life year (VSLY) or similar metric such as value of a life year (VOLY) and VSL
(OMB Circular A-4, page 30). This single factor resulted in an approximate 5-fold over
estimation of the mortality benefits versus those that would have been resulted if EPA used a
VOLY figure such as the one from Aldi and Viscusi (2007). If EPA used the VOLY or a similar
approach as recommended by OMB and held all other assumptions constant, the costs and
benefits of the rule would be about the same.
It is worth noting that EPA is the only regulatory authority worldwide that uses a VSL only
approach to assess the economic impacts of mortality attributed to air pollution. In the European
Union, both the VSL and VOLY approaches are used. The VOLY is the approach preferred by
UK Institute of Occupational Medicine, the group that develops health component of the cost
benefit analysis for the Clean Air for Europe program. The World Health Organization uses a
quality adjusted life year approach to assess impacts of air pollution.
EPA's 8.0 to 9.9 M$ VSL figures are by far the highest used worldwide and nearly an order of
magnitude higher than the 1 M Euro figure used to assess the economic impacts/ benefits of air
pollution reduction policies in Europe, a region with an economy and population characteristics
similar to the U.S.
Even though hedonic wage studies are not applicable for air pollution control policy
development, EPA inappropriately used them to derive recommended VSL for the Tier 3 DRIA.
As noted by Fraas and Lutter (2012), the inclusion of hedonic wage studies results in a two-fold
inflation of the VSL figure currently used by EPA.
EPA again used the same limited approach to assess uncertainty and did not provide a
quantitative analysis of the key uncertainties that drive the very high benefits figures. Some of
the key uncertainties that were not addressed include: ... 4) the impact of using a high end versus
a more central VSL and the impact of using the more policy relevant VOLY metric to monetize
loss of life expectancy.
Our Response:
8-3
-------
Chapter 8: Benefits
Regarding the application of a Value of Statistical Life (VSL) versus a Value of
Statistical Life Year (VSLY) approach as an appropriate metric, OMB Circular A-4 suggests that
agencies consider quantifying life years, but it does not require an agency to do so. The EPA has
explored the issue of remaining life expectancy and its potential effect on WTP in the past and
has received guidance from the Science Advisory Board (SAB) on the matter.6 To summarize,
the SAB concluded that economic theory is ambiguous on how people's willingness to pay to
reduce mortality risks should change over their lifetime, leaving it therefore as an empirical
matter. The judgment of the SAB, however, is that the empirical literature is not advanced
enough to provide clear guidance on how age and health status affect willingness to pay to
reduce mortality risks and that"... the use of a constant Value of a Statistical Life Year (VSLY),
which assumes that the VSL is strictly proportional to remaining life expectancy, is
unwarranted." (p. ii) Instead, the SAB recommended that EPA use an age-independent valuation
of mortality risk reductions (an age-independent VSL), which the EPA has done here. This is
also consistent with EPA's Economic Guidelines, which conclude that, for the present time, the
appropriate default approach for valuing the benefits of reductions in mortality risk is provided
by a central VSL estimate.
EPA's VSL approach for monetizing benefits from reductions in mortality risk (including
sources of estimates and methods for adjustment over time) has been peer reviewed by the SAB.
The 26 estimates underlying EPA's guidance for Willingness to Pay (WTP) for mortality risk
reductions come from Viscusi (1992),7 which describes the selection criteria for the studies that
-^^ &
EPA originally used in The Benefits and Costs of the Clean Air Act: 1970 to 1990. Most of the
estimates in this set of literature are based on wage differentials for workplace risks. While these
estimates cannot be taken as precise values for mortality risk reductions from environmental
policies, their usefulness for informing environmental policy analysis has been recognized
repeatedly by the Environmental Economics Advisory Committee of the SAB (SAB-EEAC).9'10
In the context of re-evaluating Agency guidance on this issue, EPA examined multiple
meta-analyses published in the economics literature and consulted with the SAB-EEAC on how
to evaluate and apply meta-analyses of mortality risk valuation studies in a policy context.
Specifically, when asked whether EPA guidance on the issue should rely on stated preference or
6 U.S. Environmental Protection Agency—Science Advisory Board (U.S. EPA-SAB). 2007. SAB Advisory on
EPA's Issues in Valuing Mortality Risk Reduction. EPA-SAB-08-001. October. Available on the Internet at
.
7 Viscusi, W.K. 1992. Fatal Tradeoffs: Public and Private Responsibilities for Risk. New York, NY: Oxford
University Press.
8 U.S. Environmental Protection Agency (EPA). 1997. The Benefits and Costs of the Clean Air Act, 1970 to 1990.
Prepared for U.S. Congress by U.S. EPA, Office of Air and Radiation/Office of Policy Analysis and Review,
Washington, DC.
9 U.S. Environmental Protection Agency—Science Advisory Board (U.S. EPA-SAB). 2000. An SAB Report on
EPA's White Paper Valuing the Benefits of Fatal Cancer Risk Reduction. EPA-SAB-EEAC-00-013. July.
Available on the Internet at
.
10 U.S. Environmental Protection Agency (U.S. EPA). 2010. Guidelines for Preparing Economic Analyses. EPA
240-R-10-001. National Center for Environmental Economics, Office of Policy Economics and Innovation.
Washington, DC. December. Available on the Internet at .
8-4
-------
Tier 3 Summary and Analysis of Comments
hedonic wage studies, the SAB-EEAC responded, "[t]he SAB believes that both stated
preference and revealed preference studies should be considered in valuing mortality risks ...
Both approaches have strengths and weaknesses in a particular context, and, as a result, we do
not believe that the Agency should rely exclusively on one or the other in all contexts."11
More recently, in the context of reviewing EPA's strategy for updating mortality risk
valuation estimates in the future, the SAB supported combining the two streams of literature to
arrive at estimates for use in regulatory analysis, stating ".. .to the extent that each literature
provides useful information about the VRR [Value of a Risk Reduction] in a particular context,
19
or the variation of VRR between contexts, it is important to combine their results."
Regarding the comment that EPA should examine the impact of using a high end versus a
more central VSL, we do indeed examine this impact in our probabilistic uncertainty analysis.
The EPA currently recommends a central VSL of $8.3m (1990$, 1990 income) based on a
Weibull distribution fitted to 26 published VSL estimates (5 contingent valuation and 21 labor
market studies). The underlying studies, the distribution parameters, and other useful information
are available in Appendix B of the EPA's Guidelines for Preparing Economic Analyses.
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM); Marathon Petroleum Company LP (MFC)
EPA also inappropriately scaled up the VSL over time based on the assumption that there will be
a continuing increase in real income in the United States. As a result, in the draft DRIA, EPA
used a VSL figure that grew from $8.0 million in 2000 to $9.9 million in 2030. EPA's
assumption defies recent economic trends as reported by the United States Census Bureau
(2012). Adjusted for inflation, the average median household income in the United States has
steadily declined each year over the last four years, falling from $54,489 in 2007 to $50,054 in
2011.
Our Response:
Our analysis accounts for expected growth in real income over time. This is a distinct
concept from inflation and currency year. Economic theory argues that Willingness to Pay
(WTP) for most goods (such as environmental protection) will increase if real incomes increase.
11 U.S. Environmental Protection Agency—Science Advisory Board (U.S. EPA-SAB). 2007. SAB Advisory on
EPA's Issues in Valuing Mortality Risk Reduction. EPA-SAB-08-001. October. Available on the Internet at
.
12 U.S. Environmental Protection Agency—Science Advisory Board (U.S. EPA-SAB). 2011. Review of Valuing
Mortality Risk Reductions for Environmental Policy: A White Paper (December 10, 2010). EPA-SAB-11-011
July. Available on the Internet at
.
8-5
-------
Chapter 8: Benefits
There is substantial empirical evidence that the income elasticity13 of WTP for health risk
reductions is positive, although there is uncertainty about its exact value. Thus, as real income
increases, the WTP for environmental improvements also increases. Although many analyses
assume that the income elasticity of WTP is unit elastic (i.e., a 10% higher real income level
implies a 10% higher WTP to reduce risk changes), empirical evidence suggests that income
elasticity is substantially less than one and thus relatively inelastic. As real income rises, the
WTP value also rises but at a slower rate than real income.
The effects of real income changes on WTP estimates can influence benefits estimates in
two different ways: through real (national average) income growth between the year a WTP
study was conducted and the year for which benefits are estimated, and through differences in
income between study populations and the affected populations at a particular time. The SAB-
EEAC advised the EPA to adjust WTP for increases in real income over time but not to adjust
WTP to account for cross-sectional income differences "because of the sensitivity of making
such distinctions and because of insufficient evidence available at present."14 An advisory by
another committee associated with the SAB, the Advisory Council on Clean Air Compliance
Analysis (SAB-Council), has provided conflicting advice. While agreeing with "the general
principle that the willingness to pay to reduce mortality risks is likely to increase with growth in
real income" and that "[t]he same increase should be assumed for the WTP for serious nonfatal
health effects," they note that "given the limitations and uncertainties in the available empirical
evidence, the Council does not support the use of the proposed adjustments for aggregate income
growth as part of the primary analysis."15 Until these conflicting advisories have been reconciled,
the EPA will continue to adjust valuation estimates to reflect income growth.
The agency continues to update its income elasticity and income growth estimates using
the available projections in growth, including GDP growth.
What Commenters Said:
Commenter: Marathon Petroleum Company LP (MFC)
Furthermore, OMB recommends that agencies use multiple measures of effectiveness (OMB
Circular A-4, page 13) including at least one integrated measure of effectiveness such as Quality
13 Income elasticity is a common economic measure equal to the percentage change in WTP for a 1% change in
income.
14 U.S. Environmental Protection Agency—Science Advisory Board (U.S. EPA-SAB). 2000. An SAB Report on
EPA's White Paper Valuing the Benefits of Fatal Cancer Risk Reduction. EPA-SAB-EEAC-00-013. July.
Available on the Internet at
.
15 U.S. Environmental Protection Agency—Science Advisory Board (U.S. EPA-SAB). 2004. Review of the Draft
Analytical Plan for EPA's Second Prospective Analysis—Benefits and Costs of the Clean Air Act, 1990-2020:
An Advisory by the Advisory Council for Clean Air Compliance Analysis. EPA-SAB-COUNCIL-ADV-04-004.
May. Available on the Internet at
.
8-6
-------
Tier 3 Summary and Analysis of Comments
Adjusted Life Year if the rule claims both mortality and morbidity benefits (Circular A-4, page
12). As mentioned above, EPA only used a VSL to monetize mortality in this rule and the draft
rule does claim morbidity benefits. Therefore, EPA's VSL only approach does not comply with
two separate sections of OMB guidelines as described in Circular A-4.
Our Response:
As we note in Chapter 7.3 of this document, a cost-effectiveness analysis is not a
requirement of the Clean Air Act sections under which our Tier 3 regulatory authority resides.
We are required to give appropriate consideration to cost, among other factors, when issuing
vehicle standards under section 202(a) of the Clean Air Act, but the statute does not prescribe a
cost-effectiveness analysis. Similarly, our statutory authority to issue fuel standards makes no
mention of a need to demonstrate the cost-effectiveness of the fuel standards (see section
21 l(c)(l) of the Clean Air Act). We do, however, conduct a Cost Benefit Analysis (CBA),
which is entirely consistent with OMB guidance.
Furthermore, Circular A-4 does not require agencies to conduct a cost-effectiveness
analysis: ".. .you should prepare a cost-effectiveness analysis ... to the extent that a valid
effectiveness measure can be developed to represent expected health and safety outcomes [p.
9]."16 According to EPA's economic guidelines:
"The fields of health economics and public health often account for health status through
the use of quality-adjusted life years (QALYs) or disability adjusted life years (DALYs).
These measures have their place in evaluating the cost-effectiveness of medical
interventions and other policy contexts, but have not been fully integrated into the
welfare economic literature on risk valuation."17
In conclusion, EPA's use of the VSL to monetize the reductions in premature mortality is
fully consistent with Circular A-4 and EPA guidance.
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM); Marathon Petroleum Company LP (MFC)
Inappropriately monetized health effects not caused by PM2.5:
In the DRIA as per all recent air related RIAs, EPA monetized a large number of health effects
that have clearly not been established to be caused by PM or ozone. The most significant of these
from an economic perspective is chronic bronchitis attributed to exposure to PM2.5.
16 Office of Management and Budget (OMB). 2003. Circular A-4: Regulatory Analysis. Washington, DC. Available
on the Internet at
17 U.S. Environmental Protection Agency (U.S. EPA). 2010. Guidelines for Preparing Economic Analyses. EPA
240-R-10-001. National Center for Environmental Economics, Office of Policy Economics and Innovation.
Washington, DC. December. Available on the Internet at .
8-7
-------
Chapter 8: Benefits
To estimate the incidence of chronic bronchitis, EPA relied on a very poor quality and outdated
study (Abbey et al. (1999). This study examined oxidative air pollution in southern California
over three decades ago. Abbey et al. did not evaluate fine PM but rather a different NAAQS
pollutant, total suspended particulates. EPA inappropriately converted the risks attributed to
course PM in Abbey et al. to fine PM. With this approach, EPA ignored well-known medical
knowledge that bronchitis is a disease of the upper airway where larger particles deposit. It is
unlikely that exposure to fine PM influence the development of bronchitis. Since Abbey et al. did
not report a statistically significant increase in bronchitis due to exposure to PM, EPA used a
non-statistically significant finding to estimate bronchitis incidence due to fine PM exposure.
Our Response:
We base our analysis of the program's impact on human health and the environment on
peer-reviewed studies of air quality and human health effects.18'19 Our benefits methods are
consistent with the RIA that accompanied the final revisions to the National Ambient Air Quality
Standards (NAAQS) for Particulate Matter and the final ozone NAAQS. As a result of this
update to be consistent with the PM NAAQS analysis, we have removed the quantification of
chronic bronchitis from our main analysis. This change is consistent with the findings of the PM
Integrated Science Assessment (ISA) that the evidence for an association between long-term
exposure to PM2.5 and respiratory effects is more tenuous.20
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM); Marathon Petroleum Company LP (MFC)
No estimates of Tier 3 impacts on urban populations.
Our Response:
This is an incorrect statement, since EPA's impacts analysis - including emissions, air
quality, and health benefits - cover the national impacts of the final rulemaking, including both
urban and rural areas (the contiguous 48 state area for the air quality modeling and benefits
analysis).
U.S. Environmental Protection Agency. (2012). Regulatory Impact Analysis for the Final Revisions to the
National Ambient Air Quality Standards for Particulate Matter. Prepared by: Office of Air and Radiation, EPA-
452/R-12-005. Retrieved August 14, 2013 at http://www.epa.gov/ttn/ecas/ria.html
19 U.S. Environmental Protection Agency. (2008). Final Ozone NAAQS Regulatory Impact Analysis. Preparedby:
Office of Air and Radiation, Office of Air Quality Planning and Standards. Retrieved March, 26, 2009 at
http://www.epa.gov/ttn/ecas/ria.html. EPA-HQ-OAR-2009-0472-0238
20 U.S. Environmental Protection Agency (U.S. EPA). 2009b. Integrated Science Assessment for Particulate Matter
(Final Report). EPA-600-R-08-139F. National Center for Environmental Assessment—RTF Division.
December. Available on the Internet at .
-------
Tier 3 Summary and Analysis of Comments
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM); Marathon Petroleum Company LP (MFC)
Insufficient quantification of key uncertainties in claimed benefits:
The quantitative treatment of uncertainty was extremely limited. As has become standard for all
recent EPA RIAs, here EPA only presented a simple range of figures based on whether they used
the results of Pope et al. (2002) or Laden et al. (2006) to assess chronic PM mortality. EPA did
not address the full range of uncertainty in this single factor, i.e. the PM mortality concentration
response relationship. As noted above, EPA excluded from consideration many other studies that
report a different range of results.
Similarly, for ozone, the only uncertainty EPA attempted to address quantitatively was the
magnitude of the acute ozone mortality risk function. Again, EPA did not address this issue
thoroughly. Rather, EPA selected two positive studies and excluded from consideration many
other studies that provided a different spectrum of results.
As pointed out by Frass and Lutter (2012), EPA's current approach to assess uncertainties
focuses narrowly on the PM and ozone mortality concentration response functions; as such it
does not meet the National Research Council recommendations (NRC, 2002). Similarly, the
approach used by EPA does not meet the requirements of OMB (OMB Circular A-4, page 40-
41). For rules for which the costs are projected to exceed $1 billion annual threshold OMB
requires a formal quantitative analysis of the relevant uncertainties including, if possible,
probability distributions.
EPA again used the same limited approach to assess uncertainty and did not provide a
quantitative analysis of the key uncertainties that drive the very high benefits figures. Some of
the key uncertainties that were not addressed include: 1) the quantitative impact of using other
studies besides the outdated ACS Pope and Harvard Six Cities Laden to estimate PM mortality;
2) the quantitative impact of using studies besides Bell and Levy to estimate ozone mortality.
Commenter: Mercatus Center at George Mason University
We find that the EPA has failed to acknowledge the high degree of uncertainty surrounding its
estimates of benefits from this regulation.
Additionally, there is a high degree of uncertainty, which the EPA itself acknowledges,
surrounding the EPA's benefits estimates. For example, the EPA describes criticisms related to
its uncertainty analysis made by the National Resource Council in a 2002 report (1). Despite the
EPA's acknowledgement of these criticisms, the EPA continues to evaluate uncertainty in a
similar manner (2).
Our Response:
8-9
-------
Chapter 8: Benefits
The EPA disagrees that it has not sufficiently quantified uncertainty in the Tier 3 benefits
analysis. Our analysis contains a complex uncertainty characterization for the estimated benefits
including a suite of quantitative and qualitative uncertainty analyses organized within a tiered
uncertainty framework. The quantitative uncertainty analyses include Monte Carlo and
probabilistic assessments, as well as sensitivity analyses using alternate assumptions. The EPA's
uncertainty analyses for benefits examine the effects of the most important methodological
choices on results. For example, the EPA estimates mortality impacts using health effect
estimates garnered from an EPA-sponsored expert elicitation,21 which provide a range of effect
estimates that span from smaller than the effect coefficient in the American Cancer Society
99 9^
cohort studies (Pope et al., 2002; Krewski et al, 2009) ' and larger than the effect coefficient in
the Harvard Six Cities cohort studies (Laden et al., 2006; Lepeule et al., 2012).
24,25
Furthermore, we characterize ozone mortality using a range across six studies, not only
the two cited by the commenter - three multi-city studies (Bell et al., 2004; Huang et al., 2005;
Schwartz, 2005) and three meta-analyses (Bell et al., 2005; Ito et al., 2005; Levy et al.,
2005).26'27'28'29'30'31 This approach is consistent with recommendations provided by the National
Research Council (NRC, 2008)32 and the studies were reviewed as part of the last ozone
Integrated Science Assessment (ISA) prepared for the ozone NAAQS.33
These approaches are fully consistent with Circular A-4 and EPA guidance.
21 Roman, Henry A., Katharine D. Walker, Tyra L. Walsh, Lisa Conner, Harvey M. Richmond, Bryan J. Hubbell,
and Patrick L. Kinney. 2008. "Expert Judgment Assessment of the Mortality Impact of Changes in Ambient Fine
Paniculate Matter in the U.S." Environ. Sci. Technol., 42(7):2268-2274.
22 Pope, C.A., III, R.T. Burnett, M.J. Thun, E.E. Calle, D. Krewski, K. Ito, and G.D. Thurston. 2002. "Lung Cancer,
Cardiopulmonary Mortality, and Long-term Exposure to Fine Paniculate Air Pollution." Journal of the American
Medical Association 287:1132-1141.
23 Krewski D, Jerrett M, Burnett RT, Ma R, Hughes E, Shi, Y, et al. 2009. "Extended follow-up and spatial analysis
of the American Cancer Society study linking paniculate air pollution and mortality." HEI Research Report, 140,
Health Effects Institute, Boston, MA.
24 Laden, F., J. Schwartz, F.E. Speizer, and D.W. Dockery. 2006. "Reduction in Fine Paniculate Air Pollution and
Mortality." American Journal of Respiratory and Critical Care Medicine 173:667-672.
25 Lepeule J, Laden F, Dockery D, Schwartz J 2012. "Chronic Exposure to Fine Particles and Mortality: An
Extended Follow-Up of the Harvard Six Cities Study from 1974 to 2009." Environ Health Perspect.
Jul;120(7):965-70.
26 Bell, M.L., et al. (2004). Ozone and short-term mortality in 95 U.S. uiban communities, 1987-2000. JAMA,
2004. 292(19): p. 2372-8. EPA-HQ-OAR-2009-0472-1662
27 Huang, Y.; Dominici, F.; Bell, M. L. (2005) Bayesian hierarchical distributed lag models for summer ozone
exposure and cardio-respiratory mortality. Environmetrics. 16: 547-562. EPA-HQ-OAR-2009-0472-0233
28 Schwartz, J. (2005) How sensitive is the association between ozone and daily deaths to control for temperature?
Am. J. Respir. Crit. Care Med. 171: 627-631. EPA-HQ-OAR-2009-0472-1678
29 Bell, M.L., F. Dominici, and J.M. Samet. (2005). A meta-analysis of time-series studies of ozone and mortality
with comparison to the national morbidity, mortality, and air pollution study. Epidemiology. 16(4): p. 436-45.
EPA-HQ-OAR-2009-0472-0222
30 Ito, K., S.F. De Leon, and M. Lippmann (2005). Associations between ozone and daily mortality: analysis and
meta-analysis. Epidemiology. 16(4): p. 446-57. EPA-HQ-OAR-2009-0472-0231
31 Levy, J.I., S.M. Chemerynski, and J.A. Sarnat. (2005). Ozone exposure and mortality: an empiric bayes
metaregression analysis. Epidemiology. 16(4): p. 458-68. EPA-HQ-OAR-2009-0472-0236
32 National Research Council (NRC). 2008. Estimating Mortality Risk Reduction and Economic Benefits from
Controlling Ozone Air Pollution. National Academies Press. Washington, DC.
33 U.S. EPA. Integrated Science Assessment of Ozone and Related Photochemical Oxidants (Final Report). U.S.
Environmental Protection Agency, Washington, DC, EPA/600/R-10/076F, 2013.
8-10
-------
Tier 3 Summary and Analysis of Comments
Since the publication of NRC (2002),34 the EPA has continued work to improve the
characterization of uncertainty in both health incidence and benefits estimates. In response to
these recommendations, the EPA expanded our previous analyses to incorporate additional
quantitative and qualitative characterizations of uncertainty. While we have not yet been able to
make as much progress towards the full, probabilistic uncertainty assessment envisioned by the
NAS as we had hoped, we have added a number of additional quantitative and qualitative
analyses to highlight the impact that uncertain assumptions may have on the benefits estimates.
In addition, for some inputs into the benefits analysis, such as the air quality data, it is difficult to
address uncertainty probabilistically due to the complexity of the underlying air quality models
and emission inputs. In cases where there is insufficient scientific evidence to support alternative
assumptions about key uncertain parameters, we provide qualitative assessments and do not
simply speculate for the purposes of generating probabilistic uncertainty distributions.
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM); Marathon Petroleum Company LP (MFC)
EPA's DRIA assumes implausible health benefits due to highly conservative, unrealistic
assumptions. In the DRIA, the overwhelming majority of the benefits are derived from a very
small calculated reduction of 0.05 ug/m3 in PM2.5 when the NAAQS standard is 12 ug/m3.
EPA assumes with 100% certainty that exposure to any level of PM and ozone, whether far
above or far below the current national standards, causes mortality. If EPA's assumptions on
mortality are not correct, and mortality does not occur at all or does not occur at lower levels, the
benefits of the Tier 3 rule would be markedly reduced and the costs of the rule would exceed the
benefits. API and AFPM detail our concerns in each of these areas, and point out the omission of
key uncertainties that EPA failed to quantify. Using more realistic assumptions per existing
references, the economic benefits of this proposed rule would be markedly lower than the costs
in the proposed rule. As explained in detail in the comments, the flawed DRIA does not comply
with OMB guidelines and should be withdrawn.
Over-estimated mortality attributed to small changes in PM2.5:
Similar to recent EPA air-related rules, most of the economic benefits of the proposed Tier 3 rule
(well over 95%) accrue from preventing "premature mortality" attributed to small reductions in
ambient PM2.5 and ozone. EPA assumes with 100% certainty that exposure to any level of PM
and ozone, whether far above or far below the current national standards, causes mortality. If
EPA's assumptions on mortality are not correct, and mortality does not occur at all or does not
occur at lower levels, the benefits of the Tier 3 rule would be markedly reduced and the costs of
the rule would exceed the benefits.
Nearly a decade ago, the National Research Council recommended that EPA discontinue using
both the American Cancer Society (ACS) and Harvard Six Cities (H6C) cohorts for decision-
34 National Research Council (NRC). 2002. Estimating the Public Health Benefits of Proposed Air Pollution
Regulations. Washington, DC: The National Academies Press. Washington, DC.
8-11
-------
Chapter 8: Benefits
making (NRC, 2004). This recommendation was based on the concern for the age of the cohorts
and fact that the individual and group factors used to adjust air pollution mortality risks were
collected over 20 years ago and were never been updated. Today, the average ages of these
cohorts is now 87 and the individual and group risk factors are over 30 years out of date.
According to data provided by the Centers for Disease Control, cardiovascular mortality rates in
the United States have been steadily decreasing (Kochanek et al. 2009; Danaei et al. 2009). In
just one year, 2008 to 2009, the age-adjusted death rate for diseases of the heart decreased by 3.7
% (Kochanek et al. 2009). The key factors responsible for this trend include reduction in
smoking rates, various efforts to control blood pressure, and changes in diets (Danaei et al.
2009). These factors for cardiovascular risk, called covariates in epidemiology studies, have
changed markedly over the last three decades. For example, when the ACS and Harvard Six
Cities cohorts were enrolled in the late 1970's to early 1980's, smoking rates were 33-37%
(CDC, 2012). Today, these rates have fallen to under 20%. Various updates and reanalysis of the
ACS and Harvard Six Cities studies inappropriately assume that the much higher rates were
constant or proportional over the 30 years of study follow-up.
Similarly, other key cardiovascular risk factors including control of high blood pressure and low
density and total cholesterol, intake of salt and omega 3 fatty acids, have markedly changed over
the last 30 years (Danaei et al. 2009). By not considering these national trends in well accepted
risk factors for cardiovascular disease, the reduction in cardiovascular mortality that has been
occurring in the U.S. has been inappropriately assigned by EPA to the coincident reduction in
ambient PM2.5 in the U.S.
Despite the recommendations of the NRC and the well-known changes in risk factors responsible
for the decline of cardiovascular mortality in the U.S., in the DRIA for the Tier 3 rule, EPA once
again exclusively relied on studies using these two aged cohorts to estimate chronic mortality
attributed to PM2.5 exposure. EPA ignored studies that used more modern cohorts or that
reported a different spectrum of results (Greven et al., 2011; Krewski et al. 2000; Enstrom 2005;
Beelen et al. 2008; Janes et al. 2007). EPA also ignored studies that reported a threshold for
mortality (Abrahamowicz et al. 2003; Nicolich and Gamble 1999; Smith et al. 2000; Stylinaou
and Nicolich 2009; Gamble and Nicolich 2006) and other studies that challenge EPA's no
threshold approach to risk assessment for PM (Koop and Tole 2006; Roberts and Martin 2006).
In the draft DRIA, over two-thirds of the economic benefits are attributed to a small (0.05
ug/m3) reduction in PM2.5 and most of these benefits are attributed to mortality. For the primary
regulatory scenario that EPA presents in the DRIA, the number of PM mortalities estimated to be
prevented ranged from 800 using Pope et al. (2002) to 2,100 using Laden et al. (2006). By
comparison, the estimated ozone mortalities avoided were 170 to 770.
EPA used a single range of mortality coefficients (Pope to Laden) for PM2.5 and applied them
across the U.S., even though significant regional heterogeneity has been reported in recent
studies (Peng et al. 2005; Dominici et al. 2006, 2007; Enstrom 2005). In this manner, EPA does
not address the issue that the benefits in various regions of the country differ widely. Since OMB
guidelines encourage exploring regional variations in benefits (Circular A-4 page 8), EPA's draft
DRIA for the Tier 3 rule is deficient in this area.
8-12
-------
Tier 3 Summary and Analysis of Comments
Over 90% of the benefits result from reducing ambient levels of PM2.5 well below the NAAQS
level [See Figure 12 on p. 39 of Docket number EPA-HQ-OAR-2011-0135-4276-A2]. Since a
large fraction of the U.S. population lives in areas where the PM2.5 levels are well below EPA's
recently revised NAAQS, EPA assumes that ambient exposures even near background present an
equal risk. EPA should not calculate mortality or other health benefits below the NAAQS for PM
and ozone, levels.
If EPA set aside the economic benefits of the Tier 3 rule that are estimated below the NAAQS
for PM and ozone, and ignored all of the other changes suggested here, the benefits of this rule
would be markedly reduced (90% reduction) and would fall in the range of $0.8-2.3 billion. In
this case, EPA's estimated costs of the rule ($3.4 billion) would exceed the benefits by about 1.5-
4 fold.
EPA again used the same limited approach to assess uncertainty and did not provide a
quantitative analysis of the key uncertainties that drive the very high benefits figures. Some of
the key uncertainties that were not addressed include: 4) the impact of extrapolating mortality
and other health effects to far below levels deemed safe by EPA to near or below background/
EPA's assumption that soluble sulfate and nitrate PM produce mortality with no threshold is
inconsistent with published scientific information. Despite numerous attempts to encourage EPA
to recognize that a long term epidemiology study has evaluated this key issue, EPA again
ignored the study by Abrahamowicz et al. (2003) which reported no mortality at sulfate
concentrations less than 12 ug/m3. This level is far above current ambient levels of sulfate PM in
the U.S. Although soluble nitrate PM has not been evaluated in long term epidemiology studies,
based on physical chemical properties, a similar result would be expected. Thus, the results of
Abrahamowicz et al. suggest reducing secondary sulfate levels further by a fraction of 1 ug/m3
as proposed in this rule will not result in any decrease in mortality. Therefore most of the
mortality related benefits that EPA claims will result from this rule are not expected to occur.
Commenter: Mercatus Center at George Mason University
A growing literature calls into question the causal link between the total concentration of
ambient paniculate matter and mortality levels, especially at the low doses that exist today in
many parts of the United States.
First, over 50 percent of the quantified benefits from the regulation are due to reductions in total
particulate matter. However, a growing literature raises doubts about the causal link between
ambient PM2.5 levels and increases in mortality.
The EPA also acknowledges uncertainty with regards to its benefits analysis, including these
points from table 8-16 of the Regulatory Impact Analysis for the rule:
The extrapolation of effect estimates is beyond the range of ozone or PM concentrations
observed in the source epidemiological study.
8-13
-------
Chapter 8: Benefits
Each point is important. First, the EPA states that its estimates go beyond those confirmed in the
epidemiological study upon which it bases its findings. This means benefits of the regulation are
based primarily upon model selection, not empirical evidence. The EPA assumes a linear-dose
response down to the origin, resulting in large benefits estimates. Selecting another model, such
as a threshold- or hormetic-dose response at low doses, would produce vastly lower benefits
estimates. Recent academic literature has suggested there may be reason to believe PM exhibits a
hormetic-dose response at low doses.
Points 1... above imply that a benefits estimate of zero is within the realm of possibility for
benefits resulting from reductions in particulate matter. Acknowledging this uncertainty should
be a part of the Regulatory Impact Analysis (RIA). Elsewhere in the RIA, the EPA
acknowledges that the "EPA estimated PM-related mortality without applying an assumed
concentration threshold (section 8-9 of RIA)." Thus, the EPA assumes PM-related health
benefits continue all the way down to very low levels. However, Cox has shown that there may
be a hermetic-, or J-shaped-, dose-response curve for PM at low dose levels. If true, this implies
there may be no negative health effects and potentially even health benefits to PM exposure at
low-dose levels, rather than the harm the EPA assumes by model selection. Elsewhere, Cox has
argued that the causal link between PM and human health benefits has not been adequately
demonstrated at low doses. The EPA appears to be pointing to correlations without assessing
whether causation is present. Fortunately, there are tests that can be done to demonstrate
causation. The EPA would benefit from running these tests with the data available and
presenting these results to the public.
Additionally, Lutter and Fraas show that uncertainties surrounding benefits estimates from PM
reductions may greatly exceed those that the EPA acknowledges in previous analyses for PM-
related rules. Lutter and Fraas give the EPA the benefit of the doubt and assume a causal
relationship exists between PM and increases in mortality. They go on to demonstrate that
benefits estimates vary greatly by modifying assumptions such as the value of reducing mortality
risk or whether the toxicity is above or below the average for fine particles.
Other experts in the field of environmental risk assessment have shown similar skepticism about
the benefits of PM-related regulations. For example, the number of lives saved may be vastly
overstated.
The EPA's methodology appears to be at odds with the very standards it applied to its own
analyses prior to 2009. The EPA even acknowledges this methodological problem in point 1
above.
Our Response:
The EPA disagrees with the commenters regarding the evidence for health impacts at low
concentrations, disagrees with the commenters' recommendations regarding the assumptions and
magnitude of benefits results that the EPA should apply in the RIA, and disagrees with the
commenters' erroneous interpretation of the NRC guidance with regard to the American Cancer
Society (ACS) and Harvard Six Cities (H6C) studies. We also disagree that zero is within the
realm of possibility for benefits resulting from reductions in parti culate matter. The EPA's
8-14
-------
Tier 3 Summary and Analysis of Comments
methods for quantifying health benefits of emission reductions are based on the best available
peer-reviewed science and methods that have withstood scrutiny from EPA's independent
Science Advisory Board (SAB), the National Academy of Sciences (NRC, 2002), and
continuous interagency review.
Regarding the NRC guidance, the citation provided is taken out of context. The NRC
commented that the EPA should not rely on the Harvard Six Cities cohort and American Cancer
Society cohort alone to the exclusion of a "new generation of cohort studies." In formulating the
decisions that informed the PM NAAQS (and also inform the Tier 3 analysis), the EPA
considered all of the available scientific evidence, including studies of new cohorts. In addition,
two separate panels of EPA's independent Science Advisory Board (SAB) recently
recommended that the EPA use these two cohorts to quantify PM2.5-related mortality risks and
benefits (i.e., CASAC (2009, 2010) and Council (2010)).35'36'37 Despite some inherent
limitations, these cohorts continue to have several advantages over other currently available
cohorts, including age and gender representativeness, geographic representativeness, study size,
consideration of confounders, and length of follow-up. EPA's approach is consistent with the
advice from NRC and SAB.
The EPA has also carefully considered the epidemiological, toxicological and clinical
evidence linking exposure to fine particles and the onset of adverse health outcomes. When
estimating PM-related risks, we rely upon studies and conclusions considered in the Integrated
Science Assessment for P articulate Matter (PM ISA)(U.S. EPA, 2009) and the Provisional
Assessment (U.S. EPA, 2012).38'39 These two comprehensive documents have assessed the entire
body of scientific evidence regarding particles, including thousands of new studies (and include
the studies mentioned by the commenter). The PM ISA received two rigorous rounds of peer
review by the independent Clean Air Scientific Advisory Committee (CASAC) and concluded
that the no-threshold model is best supported by the available data. Specifically, the PM ISA
concluded that "[o]verall, the studies evaluated further support the use of a no-threshold log-
35 U.S. Environmental Protection Agency—Science Advisory Board (U.S. EPA-SAB). 2009. Review of Integrated
Science Assessment for Paniculate Matter (Second External Review Draft, July 2009). EPA-CASAC-10-001.
November. Available on the Internet at
.
36 U.S. Environmental Protection Agency Science Advisory Board (U.S. EPA-SAB). 2010. CASAC Review of
Quantitative Health Risk Assessment for Paniculate Matter—Second External Review Draft (February 2010).
EPA-CASAC-10-008. Available on the Internet at
.
37 U.S. Environmental Protection Agency Science Advisory Board (U.S. EPA-SAB). 2010. Review of EPA's
DRAFT Health Benefits of the Second Section 812 Prospective Study of the Clean Air Act. EPA-COUNCIL-10-
001. June. Available on the Internet at
.
38 U.S. Environmental Protection Agency (U.S. EPA). 2009. Integrated Science Assessment for Paniculate Matter
(Final Report). EPA-600-R-08-139F. National Center for Environmental Assessment—RTF Division.
December. Available on the Internet at .
39 U.S. Environmental Protection Agency (U.S. EPA). 2012. Provisional Assessment of Recent Studies on Health
Effects of Particulate Matter Exposure. EPA/600/R-12/056A. National Center for Environmental Assessment—
RTF Division. December.
8-15
-------
Chapter 8: Benefits
linear model." In 2010, the Health Effects Subcommittee of the EPA's independent Advisory
Council on Clean Air Compliance Analysis (Council) "fully supports EPA's decision to use a
no-threshold model to estimate mortality reductions. This decision is supported by the data,
which are quite consistent in showing effects down to the lowest measured levels. Analyses of
cohorts using data from more recent years, during which time PM concentrations have fallen,
continue to report strong associations with mortality. Therefore, there is no evidence to support a
truncation of the CRF [concentration-response function]."40 A summary of the scientific review
statements regarding the lack of a threshold in the PIVb.s-mortality relationship is documented in
a technical support document.41
Our approach to estimating PM-related health impacts below the lowest measured level
(LML) is consistent with this advice. In general, we are more confident in the magnitude of the
risks we estimate from simulated PM2.5 concentrations that coincide with the bulk of the
observed PM concentrations in the epidemiological studies that form the basis of those
estimates. Likewise, we are less confident in the risk we estimate from simulated PM2.5
concentrations that fall below the bulk of the observed data in these studies. For this reason, the
EPA includes statistics in the final RIA showing the percentage of health impacts occurring
above these levels for readers to understand the effect of this methodological decision. While
this assessment provides some insight into the level of uncertainty in the estimated PM mortality
benefits, EPA does not view the LML as a threshold and continues to quantify the PM-related
mortality impacts using the full range of modeled air quality concentrations as the best estimate
of the benefits.
We disagree with the comment that the RIA is deficient because the analysis does not
address the issue that the benefits in various regions of the country differ widely, even though
significant regional heterogeneity has been reported in recent studies.
With respect to understanding the nature and magnitude of PM2.s-related risks, the EPA
agrees that epidemiological studies evaluating health effects associated with long-term PM2.5
exposures have reported heterogeneity in responses between cities and effect estimates across
geographic regions of the U.S. (U.S. EPA, 2009, sections 6.2.12.1, 6.3.8.1, 6.5.2, and 7.6.1; U.S.
EPA, 2011, p. 2-25).42'43 However, the rationale that heterogeneity in risk estimates presents a
potential bias as posed by the commenters is simplistic and does not account for a number of
40 U.S. Environmental Protection Agency Science Advisory Board (U.S. EPA-SAB). 2010. Review of EPA's
DRAFT Health Benefits of the Second Section 812 Prospective Study of the Clean Air Act. EPA-COUNCIL-10-
001. June. Available on the Internet at
.
41 U.S. Environmental Protection Agency (U.S. EPA). 2010. Technical Support Document: Summary of Expert
Opinions on the Existence of a Threshold in the Concentration-Response Function for PM2.5-related Mortality.
Research Triangle Park, NC. June. Available on the Internet at
.
42 U.S. Environmental Protection Agency (U.S. EPA). 2009. Integrated Science Assessment for Paniculate Matter
(Final Report). EPA-600-R-08-139F. National Center for Environmental Assessment—RTF Division.
December. Available on the Internet at .
43 U.S. Environmental Protection Agency (U.S. EPA). 2011. Policy Assessment for the Review of the Particulate
Matter National Ambient Air Quality Standards. EPA-452/D-11-003. April. Available on the Internet at
.
8-16
-------
Tier 3 Summary and Analysis of Comments
factors that have been shown to influence city-specific risk estimates in epidemiologic studies.
As discussed in the ISA, the EPA recognizes that there are compositional differences in PM2.5
across the country and that the county-level air quality data used in epidemiological studies may
result in exposure error, which could in part account for variability in city-specific risk estimates
(U.S. EPA, 2009, section 2.3.2).
Overall, the EPA recognizes that uncertainties still remain regarding various factors that
contribute to heterogeneity observed in epidemiological studies. Nonetheless, the EPA
recognizes that this heterogeneity could be attributed, at least in part, to differences in PM2.5
composition across the U.S., as well as to exposure differences that vary regionally such as
personal activity patterns, microenvironmental characteristics, and the spatial variability of
PM2.5 concentrations in urban areas (U.S. EPA, 2009, section 2.3.2; 77 FR 38910).
The current epidemiological evidence and the limited amount of city-specific speciated
PM2.5 data does not allow conclusions to be drawn that specifically differentiate effects of PM in
different locations (U.S. EPA, 2011, p. 2-25). Furthermore, the PM ISA concluded, "that many
constituents of PM2.5 can be linked with multiple health effects, and the evidence is not yet
sufficient to allow differentiation of those constituents or sources that are more closely related to
specific health outcomes" (U.S. EPA, 2009, p. 2- 17). CASAC thoroughly reviewed the EPA's
presentation of the scientific evidence indicating heterogeneity in PM2.5 effect estimates in
epidemiological studies and concurred with the overall conclusions presented in the ISA.
For all of the reasons listed above, the EPA believes that the approach used to estimate
and present the human health benefits of the Tier 3 rule is appropriate and consistent with the
currently accepted best practices for conducting national-level health impacts and benefits
analyses.
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM); Marathon Petroleum Company LP (MFC)
Dismissed chemistry differences in PM toxicity potential:
Similar to EPA's air-related rules, most of the chemistry of PM reduction from the Tier 3 rule is
soluble secondary inorganic sulfate and nitrate PM that exhibits low toxicity potential in animal
and human clinical studies (Schlesinger et. al. 2003; Amdur et al. 1998). By 2017, EPA
estimated this rule will result in reduction of 284,381 tons of nitrogen oxides and 16,261 tons of
sulfur oxides. EPA predicted much of these gaseous emissions will be converted to PM. By
comparison, in the executive summary of the DRIA, EPA estimated the rule will produce a small
(121 ton) reduction in directly emitted PM2.5 whereas in another section 7.1.3.2.2, EPA predicts
the rule will produce an increase in directly emitted PM2.5. In either case, the rule will have little
impact on ambient levels of directly emitted PM2.5 in the U.S. Thus, the large economic benefits
EPA estimated for this rule are highly dependent on the assumption that soluble sulfate and
nitrate PM exhibit the same remarkably high mortality producing potential as claims for PM2.5,
the form of PM evaluated in the underlying health studies that EPA relied on.
8-17
-------
Chapter 8: Benefits
EPA again used the same limited approach to assess uncertainty and did not provide a
quantitative analysis of the key uncertainties that drive the very high benefits figures. Some of
the key uncertainties that were not addressed include: ... 3) the quantitative impact of varying
particle chemistry on the PM mortality risk functions and benefits.
Commenter: Mercatus Center at George Mason University
The EPA also acknowledges uncertainty with regards to its benefits analysis, including these
points from table 8-16 of the Regulatory Impact Analysis for the rule:
Direct causal agents within the complex mixture of PM have not been identified.
Next, the EPA fails to address whether the concentration of total particulate mass or the
composition of those particulates are the cause of the health effects found in the cited studies.
Moreover, because the composition of rural parti culates is different from urban parti culates, the
health effects are likely to be different than those estimated. In order to provide a causal link, the
EPA should be able to determine which components of parti culate matter are the sources of the
higher morbidity and mortality rates. For instance, Bell finds that higher concentrations of PM2.5
Nickel are associated with higher rates of cardiovascular or respiratory hospitalizations.
As Bell et al. conclude: Because of these limitations, health risks could be associated with the
true concentrations of a component or set of components that co-varies with PM2.5 total mass,
even if measured concentrations in this data set do not co-vary with PM2.5 total mass because of
measurement error. Further, we did not investigate the possibility that observed PM2.5 health
effects could result from a set of components with a collective concentration that co-varies with
PM2.5 total mass, although individual component concentrations do not.
Without a clear link between the chemical components of PM2.5 that are associated with health
effects, the EPA is left to assume that the overall level of PM2.5 is the source of health risks,
rather than particular components of the total PM2.5. If the health effects are due to a particular
component, a more targeted, lower-cost, and potentially higher-benefit air pollution regulation
might be warranted. The EPA does acknowledge that the causal connection between PM and
human health outcomes is uncertain, as point 2 above demonstrates.
Our Response:
While there remain uncertainties about the role and relative toxicity of various
components of fine PM, the current evidence continues to support the view that fine particles
should be addressed as a group for purposes of public health protection and that "many
constituents of PM can be linked with differencing health effects and the evidence is not yet
sufficient to allow differentiation of those constituents or sources that are more closely related to
specific health outcomes" such that it is inappropriate to remove any constituent of PM2.s from
8-18
-------
Tier 3 Summary and Analysis of Comments
the standard (U.S. EPA, 2009, 2-17; U.S. EPA, 2011, p. 2-25).44'45 EPA therefore disagrees that
uncertainty in this assumption would eliminate almost all of the benefits associated with the Tier
3 standards.
While most epidemiological studies continue to be indexed by PM2.5 mass, several recent
epidemiological studies included in the ISA have used PM2.5 speciation data to evaluate health
effects associated with fine particle exposures. In the ISA, the EPA thoroughly evaluated the
scientific evidence that examined the effect of different PIVb.s components and sources on a
variety of health outcomes (U.S. EPA, 2009, section 6.6) and observed that the available
information continues to suggest that many different chemical components of fine particles and a
variety of different types of source categories are all associated with, and probably contribute to,
effects associated with PIVb.s. The ISA concluded that the current body of scientific evidence
indicated that "many constituents of PM can be linked with differing health effects and the
evidence is not yet sufficient to allow differentiation of those constituents or sources that are
more closely related to specific health outcomes" (U.S. EPA, 2009, p. 2-26 and 6-212).
Furthermore, the ISA concluded that the evidence is not sufficient to support eliminating any
component or group of components associated with any specific source categories from the mix
of fine particles included in the PM2.5 indicator (U.S. EPA, 2009, p. 2-56). The Clean Air
Science Advisory Committee (CASAC) agreed that it was reasonable to retain PM2.5 as an
indicator for fine particles, as "[tjhere was insufficient peer-reviewed literature to support any
other indicator at this time" (p. 12).46
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM); Marathon Petroleum Company LP (MFC)
Over-estimated mortality assumptions from small changes in ozone:
To estimate acute mortality attributed to ozone exposure, EPA used the study by Bell et al.
(2004) and Levy et al. (2005) to characterize the low end and high end range, respectively, of the
available literature. In this manner, EPA only used studies that reported a positive association
between ozone and mortality. EPA excluded other studies that reported no clear association and
confounding by particular matter (Smith et al. 2009, Franklin et al. 2007; Katsouyianni et al.
2009). EPA also excluded epidemiology studies that report a threshold for ozone mortality
(Stylianou et al. 2009) as well as other studies that conclude it is not possible to determine
44 U.S. Environmental Protection Agency (U.S. EPA). 2009. Integrated Science Assessment for Paniculate Matter
(Final Report). EPA-600-R-08-139F. National Center for Environmental Assessment—RTF Division.
December. Available on the Internet at .
45 U.S. Environmental Protection Agency (U.S. EPA). 2011. Policy Assessment for the Review of the Particulate
Matter National Ambient Air Quality Standards. EPA-452/D-11-003. April. Available on the Internet at
.
46 U.S. Environmental Protection Agency Science Advisory Board (U.S. EPA-SAB). 2010. CASAC Review of
Quantitative Health Risk Assessment for Particulate Matter—Second External Review Draft (February 2010).
EPA-CASAC-10-008. Available on the Internet at
.
8-19
-------
Chapter 8: Benefits
whether or not thresholds exist at low ozone levels with observational data given the poor
ambient to personal exposure correlation and high exposure misclassification (Brauer et al.
2002). It is hard to understand how EPA could use anything other than zero for the low end.
Our Response:
The range of ozone benefits associated with the final Tier 3 rule is estimated based on
risk reductions derived from several sources of ozone-related mortality effect estimates. This
analysis presents six alternative estimates for the association based upon different functions
reported in the scientific literature. We use three multi-city studies,4 '48'49 including the Bell,
2004 National Morbidity, Mortality, and Air Pollution Study (NMMAPS) that was used as the
primary basis for the risk analysis in the ozone Staff Paper5 and reviewed by the Clean Air
Science Advisory Committee (CASAC).51 We also use three studies that synthesize ozone
mortality data across a large number of individual studies.52'53'54 This approach is consistent with
recommendations provided by the NRC in their ozone mortality report (NRC, 2008),55 "The
committee recommends that the greatest emphasis be placed on estimates from new systematic
multicity analyses that use national databases of air pollution and mortality, such as in the
NMMAPS, without excluding consideration of meta-analyses of previously published studies."
The NRC goes on to note that there are uncertainties within each study that are not fully captured
by this range of estimates, but that analyses should "give little or no weight to the assumption
that there is no causal association between estimated reductions in premature mortality and
reduced ozone exposure."
47 Bell, M.L., et al. (2004). Ozone and short-term mortality in 95 US urban communities, 1987-
2000. Jama, 2004. 292(19): p. 2372-8.
,10
Huang, Y.; Dominici, F.; Bell, M. L. (2005) Bayesian hierarchical distributed lag models for
summer ozone exposure and cardio-respiratory mortality. Environmetrics 16: 547-562.
49 Schwartz, J. (2005) How sensitive is the association between ozone and daily deaths to control
for temperature? Am. J. Respir. Crit. CareMed. 171: 627-631.
50 U.S. EPA (2007) Review of the National Ambient Air Quality Standards for Ozone, Policy
Assessment of Scientific and Technical Information. OAQPS Staff Paper.EPA-452/R-07-003.
This document is available in Docket EPA-HQ-OAR-2003-0190. Retrieved on April 10,
2009, from http:www.epa.gov/ttn/naaqs/standards/ozone/s_o3_cr_sp.html
51 CASAC (2007). Clean Air Scientific Advisory Committee's (CASAC) Review of the
Agency's Final Ozone Staff Paper. EPA-CASAC-07-002. March 26.
52 Bell, M.L., F. Dominici, and J.M. Samet. (2005). A meta-analysis of time-series studies of
ozone and mortality with comparison to the national morbidity, mortality, and air pollution
study. Epidemiology, 16(4): p. 436-45.
53 Ito, K., S.F. De Leon, and M. Lippmann. (2005). Associations between ozone and daily
mortality: analysis and meta-analysis. Epidemiology. 16(4): p. 446-57.
54 Levy, J.I., S.M. Chemerynski, and J.A. Sarnat. (2005). Ozone exposure and mortality: an
empiric bayes metaregression analysis. Epidemiology. 16(4): p. 458-68.
55 National Research Council (NRC), 2008. Estimating Mortality Risk Reduction and Economic
Benefits from Controlling Ozone Air Pollution. The National Academies Press: Washington,
D.C.
8-20
-------
Tier 3 Summary and Analysis of Comments
The point raised by the commenters that the Smith et al. 2009, Franklin et al. 2007; and
Katsouyianni et al. 2009 studies do not provide evidence of a clear association with ozone and
mortality is incorrect. It is important to note that the scientific evidence available at the time of
the 2008 Ozone NAAQS review forms the basis of this analysis. As such, these studies
mentioned by the commenter were not considered in this analysis. However, these studies were
thoroughly evaluated in the recently completed Ozone (Os) ISA, which will form the scientific
basis of the ongoing Oi NAAQS review. 6 Overall, these studies add to the growing body of
scientific evidence for short-term Oj, exposures and mortality and contributed to the ISA
concluding that there is "likely to be causal relationship between short-term O^ exposures and
mortality."
Finally, regarding the comment about thresholds, the conclusions in EPA's recently
completed ozone Integrated Science Assessment summarizes our position:
"Overall, the studies evaluated support a linear Os-mortality C-R relationship and
continue to support the conclusions from the 2006 O3 Air Quality Criteria Document
(AQCD), which stated that "if a population threshold level exists in Os health effects, it is
likely near the lower limit of ambient Oj concentrations in the United States" (U.S. EPA,
2006)"57
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM); Marathon Petroleum Company LP (MFC)
EPA's reliance on the meta analysis study by Levy et al. is not appropriate and significantly
inflated the ozone mortality values. EPA's use of this study is inconsistent with the conclusion
the Agency reached in Integrated Science Assessment (ISA) for ozone. In the ISA, EPA
recognized the risk estimates reported by Levy were biased high. Levy et al. only considered
positive studies and there is inherent positive study publication bias in the time-series literature.
Our Response:
It is unclear what the commenters are referring to because the Levy study was not re-
evaluated in the 2013 ISA.58 It is possible the commenters are referring to the 2006 Ozone Air
56 U.S. EPA. Integrated Science Assessment of Ozone and Related Photochemical Oxidants (Final Report). U.S.
Environmental Protection Agency, Washington, DC, EPA/600/R-10/076F, 2013.
57 U.S. EPA. Air Quality Criteria for Ozone and Related Photochemical Oxidants (2006 Final). U.S. Environmental
Protection Agency, Washington, DC, EPA/600/R-05/004aF-cF, 2006.
58 U.S. EPA. Integrated Science Assessment of Ozone and Related Photochemical Oxidants (Final Report). U.S.
Environmental Protection Agency, Washington, DC, EPA/600/R-10/076F, 2013.
8-21
-------
Chapter 8: Benefits
Quality Criteria Document (AQCD).59 In the AQCD, EPA never states that the Levy et al. results
are biased high. In fact, we state they are consistent with the Bell and Ito meta-analyses:
"As stated earlier, the combined 63 excess mortality risk estimates from the meta-
analyses by Bell et al. (2005), Ito et al. (2005), and Levy et al. (2005) were all very
consistent. Although the analyses were conducted independently, there was considerable
overlap among the risk estimates used in the three meta-analyses; thus, the agreement in
the combined risk estimates was not unexpected... The positive 63 effect estimates,
along with the sensitivity analyses in these three meta-analyses provide evidence of a
robust association between ambient 03 and mortality. The combined effect estimates
from the various meta-analyses ranged from 1.5 to 2.5% excess risk in all-cause mortality
(p. 7-96 to 7-97; 2006 O3 AQCD)."
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM); Marathon Petroleum Company LP (MFC)
In the draft DRIA, EPA calculated U.S. wide ozone mortality using a single range of "national"
risk coefficients (Bell to Levy) and applied this range across the United States. EPA excluded
published studies that report clear regional heterogeneity in the ozone mortality association
(Smith et al. 2009; Bell and Dominici 2008; Bell et al. 2007). In many regions of the U.S., no
ozone mortality association is observed. Therefore, there is no valid national risk coefficient. In
this manner, EPA avoided the issue that the benefits in various regions of the country will differ
widely. Again, EPA's approach did not conform to OMB guidelines that encourage exploring
regional variation in benefits (OMB Circular A-4 page 8).
Our Response:
The EPA recognizes that various individual- and community-level factors may influence
regional heterogeneity (see p. 2-34 to 2-35 of the 2013 Ozone Integrated Science Assessment).60
We also recognize that there is currently no one currently agreed upon national risk coefficient to
estimate ozone-related premature mortality. Instead, we present risk reductions derived from
several sources of ozone-related mortality effect estimates, each with its own within-study
probabilistic distribution of uncertainty. This approach is consistent with other measures of
health impacts where multiple mortality effect estimates are available in the literature (e.g., PM-
related mortality risk). We therefore disagree that we did not conform to OMB guidelines when
assessing the national-level health impacts associated with the Tier 3 rulemaking.
Furthermore, the OMB A-4 guidelines cited by the commenter (pg. 8, "Different
Requirements for Different Geographic Regions) are not applicable to the methods cited in the
59 U.S. EPA. Air Quality Criteria for Ozone and Related Photochemical Oxidants (2006 Final). U.S. Environmental
Protection Agency, Washington, DC, EPA/600/R-05/004aF-cF, 2006.
60 U.S. EPA. Integrated Science Assessment of Ozone and Related Photochemical Oxidants (Final Report). U.S.
Environmental Protection Agency, Washington, DC, EPA/600/R-10/076F, 2013.
8-22
-------
Tier 3 Summary and Analysis of Comments
comment. The guidelines cited by the commenter refer to alternative regulatory approaches (or
the form of the standard), not to national-level health impact assessment methods.
As mentioned earlier in this comments section, the range of ozone benefits associated with the
final Tier 3 rule is estimated based on risk reductions derived from several sources of ozone-
related mortality effect estimates. This analysis presents six alternative estimates for the
association based upon different functions reported in the scientific literature. We use three
multi-city studies,'62'63 including the Bell, 2004 National Morbidity, Mortality, and Air
Pollution Study (NMMAPS) that was used as the primary basis for the risk analysis in the ozone
Staff Paper64 and reviewed by the Clean Air Science Advisory Committee (CASAC).65 We also
use three studies that synthesize ozone mortality data across a large number of individual
studies.66'67'68 This approach is consistent with recommendations provided by the NRC in their
ozone mortality report (NRC, 2008),69 "The committee recommends that the greatest emphasis
be placed on estimates from new systematic multicity analyses that use national databases of air
pollution and mortality, such as in the NMMAPS, without excluding consideration of meta-
analyses of previously published studies." The NRC goes on to note that there are uncertainties
within each study that are not fully captured by this range of estimates, but that analyses should
"give little or no weight to the assumption that there is no causal association between estimated
reductions in premature mortality and reduced ozone exposure."
At this time, as noted in the 2013 Ozone ISA (p. 2-35), "studies have not consistently
identified specific community characteristics that explain the observed heterogeneity," making it
difficult to isolate and account for regional heterogeneity in a national-level rulemaking analysis.
EPA continues to review the literature and will update our analyses as appropriate.
61 Bell, M.L., et al. (2004). Ozone and short-term mortality in 95 US urban communities, 1987-
2000. Jama, 2004. 292(19): p. 2372-8.
62 Huang, Y.; Dominici, F.; Bell, M. L. (2005) Bayesian hierarchical distributed lag models for
summer ozone exposure and cardio-respiratory mortality. Environmetrics 16: 547-562.
63 Schwartz, J. (2005) How sensitive is the association between ozone and daily deaths to control
for temperature? Am. J. Respir. Crit. CareMed. 171: 627-631.
64 U.S. EPA (2007) Review of the National Ambient Air Quality Standards for Ozone, Policy
Assessment of Scientific and Technical Information. OAQPS Staff Paper.EPA-452/R-07-003.
This document is available in Docket EPA-HQ-OAR-2003-0190. Retrieved on April 10,
2009, from http:www.epa.gov/ttn/naaqs/standards/ozone/s_o3_cr_sp.html
65 CASAC (2007). Clean Air Scientific Advisory Committee's (CASAC) Review of the
Agency's Final Ozone Staff Paper. EPA-CASAC-07-002. March 26.
66 Bell, M.L., F. Dominici, and J.M. Samet. (2005). A meta-analysis of time-series studies of
ozone and mortality with comparison to the national morbidity, mortality, and air pollution
study. Epidemiology, 16(4): p. 436-45.
67 Ito, K., S.F. De Leon, and M. Lippmann. (2005). Associations between ozone and daily
mortality: analysis and meta-analysis. Epidemiology. 16(4): p. 446-57.
68 Levy, J.I., S.M. Chemerynski, and J.A. Sarnat. (2005). Ozone exposure and mortality: an
empiric bayes metaregression analysis. Epidemiology. 16(4): p. 458-68.
69 National Research Council (NRC), 2008. Estimating Mortality Risk Reduction and Economic
Benefits from Controlling Ozone Air Pollution. The National Academies Press: Washington,
D.C.
8-23
-------
Chapter 8: Benefits
What Commenters Said:
Commenter: Mercatus Center at George Mason University
Additionally, the EPA failed to consider other adverse effects of its rules, such as impacts on
low-income individuals and adverse employment effects that were overlooked in the EPA's scant
employment impact analysis.
Finally, the EPA should go further to estimate unintended effects of its rules on employment and
on low-income populations.
Our Response:
As discussed in Preamble Section X.D and RIA Chapters 9.2.2 and 9.3.2, we expect the
Tier 3 standards will have little effect on both the economy and jobs. Some commenters argue
that the standards may increase, rather than decrease, employment (see Chapter 9.3 of this
Response to Comments). In both the vehicle and the refinery sectors, reduced sales will tend to
reduce employment, but the cost of the program is only about $72 per vehicle and less than a
penny per gallon of gasoline. At the same time, the technologies required to comply with the
standards will tend to contribute to increased employment, for example creating jobs at refineries
and at high-tech companies developing and deploying emission control equipment for vehicles.
EPA does not quantify most of the employment impacts of the standards and thus does not
evaluate these countervailing impacts. However, because these two effects are likely to be small
and counterbalance each other, we expect net employment impacts to be small.
Regarding impacts on low-income individuals, we note that the costs of these standards
are estimated to be small - about $72 per vehicle, and less than one cent per gallon of gasoline.
Because low-income households rarely buy new vehicles, we expect the impacts of the vehicle
standards on low-income households to be very small.
What Commenters Said:
Commenter: Chevron Products Company
There is more uncertainty around the non-economic components of the modeling used in the
rule. This uncertainty stems from uncontrolled or poorly controlled confounding by co-pollutants
and other risk factors when estimating an individual pollutant's effect which likely overestimates
the risk. Thus, in many studies, the data are likely compatible with a risk estimate of zero,
particularly when reference is made to ozone mortality at ambient levels.
Our Response:
8-24
-------
Tier 3 Summary and Analysis of Comments
It is important to note for studies that examine the relationship between health effects and
short-term air pollution exposures the only potential confounders are those variables that can also
fluctuate on a day-to-day basis, such as other air pollutants, temporal trends, and weather. Within
the 2006 Oj, AQCD,70 specifically sections 7.1.3 and 7.6.4, the EPA thoroughly reviewed issues
related to confounding and the evidence of potential confounding by copollutants. EPA
recognizes that a major methodological issue affecting Os epidemiologic studies concerns the
evaluation of the extent to which other air pollutants may confound or modify (Vrelated effect
estimates, and that the changing relationship between Oj, and copollutants across seasons further
complicates the issue. The use of multipollutant regression models is the prevailing approach for
controlling potential confounding by copollutants in Os health effects studies (2006 Criteria
Document, p.7-24). In section 7.6.4.2, the 2006 Criteria Document reviews the evidence from
studies that use multipollutant models to evaluate confounding by copollutants for effects
ranging from mortality and respiratory hospitalizations to lung function measures and symptoms.
EPA concluded in the 2006 Criteria Document that multipollutant regression analyses indicated
that 63 risk estimates, in general, were not sensitive to the inclusion of copollutants, including
PM2.5 and sulfate. As can be seen in Figure 1 below (Figure 7-22 from the 2006 Criteria
Document) the O^ effect estimates for mortality are generally unchanged upon inclusion of PM
in the models. These results suggest that the effect of Os on respiratory health outcomes appears
to be robust and independent of the effects of other copollutants (2006 Criteria Document, p.7-
154). Additionally, though the 63 effect was shown to be slightly diminished and did not retain
statistical significance in multipollutant models with NC>2, SC>2 and PMio, there was considerable
overlap in the 95% confidence intervals between the single-pollutant and multipollutant model
results, leading EPA to conclude that the association was generally robust.
70 U.S. EPA. Air Quality Criteria for Ozone and Related Photochemical Oxidants (2006 Final). U.S. Environmental
Protection Agency, Washington, DC, EPA/600/R-05/004aF-cF, 2006.
8-25
-------
Chapter 8: Benefits
£= <"
210
O
^. C/)
~ CD
O> CD
o
3~ CD
— -S
1:8
CO
CD
% Change in Mortality
-10
Samet et al. (20OO, reanalysis
Dominici et al., 2003): U.S. 9O cities —
with PM10 _
Schwartz (2O05): U.S. 14 cities _
with PM10 _
Kinney and Ozkaynak (1991):
Los Angeles County, CA —
with KM —
Kinney et al. (1995): Los Angeles County, CA —
with PM10 —
Fairley (2OO3): Santa Clara County, CA —
with PM, « —
Gamble (1998): Dallas, TX —
with PM10 —
Ito and Thurston (1996): Cook County, IL —
with PM10 —
Lipfert et al. (2OOOa): 7 counties in PA and NJ —
with PM2 s —
with PM10 —
Chock et al. (2OOO): Pittsburgh, PA (age O-74) —
with PM10 —
Touloumi et al. (1997): 4 European cities
(Athens, Barcelona, London, Paris) •
with BS -
Anderson et al. (1996): London, England -
with BS .
Hoek (200O, reanalysis Hoek, 20O3):
The Netherlands -
with PM10 -
Verhoeff et al. (1996):
Amsterdam, the Netherlands -
with PM10 -
Borja-Aburto et al. (1997): Mexico City —
with TSP —
Borja-Aburto et al. (1998): SW Mexico City —
with PM25 —
Ostro et al. (1996): Santiago, Chile —
with PM10 —
Kim et al. (2004): Seoul, Korea —
with
-5
I
1O
I
15
[_
lag O
I Latin America |
lag O
lag O
lag 1
lag 1 -2
lag 1-2
O
lad O-1 • f ^^ no* S'ven* significant at p = O.O55
i -^ SE not given; not
significant at p = O.O55
lag O
-© Best of lag O to 3
^ by city
lag 2
lag O
lag 1 -2
lag 1
lag 1
est i m a | Ff i i r
unless otherwise noted.
Additional epidemiologic studies that examine the relationship between short-term 63
exposures and mortality, published since the completion of the 2006 O^ AQCD, were evaluated
in the 2013 Os Integrated Science Assessment (ISA) (Section 6.6.2.1)71 Unlike previous studies
that were limited to primarily examining the confounding effects of PMio, the newer O^-
mortality studies expanded their analyses to include multiple PM indices (e.g., PMio, PM2.5, and
71 U.S. EPA. Integrated Science Assessment of Ozone and Related Photochemical Oxidants (Final Report). U.S.
Environmental Protection Agency, Washington, DC, EPA/600/R-10/076F, 2013.
8-26
-------
Tier 3 Summary and Analysis of Comments
PM components). An examination of copollutant models reported evidence that associations
between Os and all-cause mortality were robust to the inclusion of PMio or PM2.5 (Stafoggia et
al.. 2010: Katsouvanni et al.. 2009: Bell et al.. 2007X72'73'74 while other studies reported evidence
of a modest reduction (-20-30%) when examining PMio (Smith et al., 2009).75 An additional
study that examined PM components, specifically sulfate, reported evidence for reductions in
Os-mortality risk estimates in copollutant models (Franklin and Schwartz, 2008).76 However,
those studies that reported evidence of reductions in the Os-mortality associations had limited
PM or PM component datasets as a result of the every-3rd- and 6th-day PM sampling schedule
instituted in most cities, limiting the overall sample size employed to examine whether PM or
one of its components confounds the Os-mortality relationship. Overall, the results from
copollutant models in studies evaluated in the 2013 Os ISA are consistent with and generally
support the total body of epidemiologic evidence indicating that the Os-mortality risk estimates
remain positive in copollutant models.
In addition to examining copollutant confounding, the 2006 Os AQCD evaluated
sensitivity to model specification and concluded that Os effect estimates "were generally
more sensitive to alternative weather models than to varying degrees of freedom for
temporal trend adjustment" (2006 Criteria Document, p.7-176). Model specification was
further assessed in the 2013 Os ISA, specifically for mortality, within the Air Pollution and
Health: a European and North American Approach (APHENA) study. In APHENA the
authors examined the influence of the extent of temporal smoothing on Os-mortality risk
estimates. Katsouvanni et al. (2009) found that the extent of smoothing or the methods used
for adjustment can influence Os risk estimates when not applying enough degrees of
freedom to control for temporal/season trends. This is because of the opposing seasonal
trends between Os and mortality. Overall, the results of the APHENA study are consistent
with and support the conclusions of the 2006 Os AQCD.
What Commenters Said:
Commenters: American Fuel & Petrochemical Manufacturers (AFPM), American Petroleum
Institute (API), Marathon Petroleum Company LP (MFC), Chevron and Mercatus Center at
George Mason University
72 Stafoggia, M; Forastiere, F; Faustini, A; Biggeri, A; Bisanti, L; Cadum, E; Cernigliaro, A; Mallone, S;
Pandolfi, P; Serinelli, M; Tessari, R; Vigotti, MA; Perucci, CA. (2010). Susceptibility factors to ozonerelated
mortality: A population-based case-crossover analysis. Am J Respir Crit Care Med 182: 376-384.
http://dx.doi.org/10.1164/rccm.200908-1269OC
73 Katsouyanni, K; Samet, JM; Anderson, HR; Atkinson, R; Le Tertre, A; Medina, S; Samoli, E; Touloumi, G;
Burnett, RT; Krewski, D; Ramsay, T; Dominici, F; Peng, RD; Schwartz, J; Zanobetti, A. (2009). Air
pollution and health: A European and North American approach (APHENA). (Research Report 142).
Boston, MA: Health Effects Institute. http://pubs.healtheffects.org/view.php?id=327
74 Bell, ML; Kim, JY; Dominici, F. (2007). Potential confounding of paniculate matter on the short-term
association between ozone and mortality in multisite time-series studies. Environ Health Perspect 115:
1591-1595. http://dx.doi.org/10.1289/ehp.10108
75 Smith, RL; Xu, B; Switzer, P. (2009). Reassessing the relationship between ozone and short-term mortality in
U.S. urban communities. Inhal Toxicol 21: 37-61. http://dx.doi.org/10.1080/08958370903161612
76 Franklin, M; Schwartz, J. (2008). The impact of secondary particles on the association between ambient ozone
and mortality. Environ Health Perspect 116: 453-458. http://dx.doi.org/10.1289/ehp.10777
8-27
-------
Chapter 8: Benefits
EPA Overstates Air Quality Benefits by Citing Benefits for Reducing Pollutants for Areas that
are Already in Attainment:
Carbon Monoxide - In Table III-l of the Tier 3 proposal, EPA projects CO reductions of
746,683 tons in 2017 and 5,765,362 tons in 2030, but EPA does not even bother to discuss these
benefits in the Tier 3 emissions reduction section. This is because EPA knows that nationwide
CO emissions are averaging less than 25% of the CO NAAQS and have decreased 73% over the
past 20 years. Reductions of pollutants below the established NAAQS safe levels do not count as
benefits. All CO non-attainment areas are now in attainment and are now in maintenance mode.
EPA needs to eliminate throughout this proposal all CO references that imply there are benefits
from further CO as part of Tier 3, including its mention in Table III-l.
Estimated Air Quality Benefits - The Clean Air Act requires EPA to set NAAQS at a level that
protects public health with an adequate margin of safety. Therefore, there is no basis to look for
or estimate possible effects below the level of the standard. The analysis should be a comparison
of effects between the reference situation (i.e., no Tier 3) and the control situation (with Tier 3 in
place) for counties not in attainment for a given standard at the onset of the Tier 3 program.
EPA cites benefits for reducing pollutants for areas that are already in attainment—like reducing
CO in a CO attainment area. Monetizing these reductions and attributing the economic benefit to
Tier 3 is overstating the true value of the program. The Tier 3 program should not claim
additional economic benefits for pollutant reductions when the base pollutant level is already in
attainment with an existing standard. Only PM and ozone reductions are relevant for the
proposed Tier 3 rule—all other criteria species are either attainment nationwide or attainment
practically everywhere.
We have identified several examples where we believe air quality benefits have been overstated:
- Ozone benefits are stated for counties that are already in attainment. Out of 676 nationwide
counties modeled by EPA, only 11 non-California counties are forecast to exceed the 2008 ozone
NAAQS in 2030 and therefore benefit from ozone precursor reductions. EPA included benefits
for all counties modeled.
- Similarly, annual Average Particulate Matter air quality levels are not significantly impacted by
the Tier 3 rulemaking. Of 864 counties modeled, only 12 counties require PM2.5 precursor
controls. By 2030, the Tier 3 rulemaking will actually increase PM2.5 levels in 10 nonattainment
counties. This 'disbenefif effect is due to reductions in reactive precursor species (like NOx)
which can often result in increases in the secondary pollutants ozone and PM.
- Carbon monoxide reductions in emissions are not needed since all counties experience healthy
air concentration levels—monetized benefits attributed to CO reductions are therefore
inappropriate, except perhaps as an ozone or PM precursor.
- Mobile Source Air Toxics reductions are appropriately described but EPA is precluded by the
Clean Air Scientific Advisory Committee from using current tools to monetize or justify benefits
because human health data and ambient measurements are not credible. Therefore, MSAT
emissions reductions can be acknowledged but not as a quantifiable benefit.
Based on these observations, we conclude that EPA has overstated the monetary value of the
Tier 3 air quality benefits.
8-28
-------
Tier 3 Summary and Analysis of Comments
It is also odd that the EPA would seek to reduce PM levels beyond those the EPA just finalized
in new NAAQS standards for PM in January. If there are so many unclaimed benefits from
reducing PM, why not make the NAAQS standards even lower when the EPA had the chance?
Our Response:
EPA does not agree that it has overstated the monetary value of the Tier 3 air quality
benefits. Chapter 8 of the RIA describes the methodology for quantifying and monetizing
benefits. The quantified benefits associated with this rule are based on reductions in emissions,
for the entire contiguous United States, that affect ambient concentrations of ozone and PM2.s.
EPA does not quantify or monetize health benefits associated with CO or mobile source air
toxics. To be consistent and transparent in its assessment of Tier 3 impacts, EPA presents
changes in emissions and ambient concentrations for all criteria pollutants (including CO) and
some mobile source air toxics. EPA also presents changes in all modeled ozone and PM2.s
design values due to the Tier 3 standards. Design values are projected only for counties with
monitored data available.
EPA disagrees with the statement that "reductions of pollutants below the established
NAAQS safe levels do not count as benefits." The NAAQS are not set at levels that eliminate
the risk from air pollution. Instead, EPA sets the NAAQS at a level requisite to protect public
health with an adequate margin of safety, taking into consideration effects on susceptible
populations found in the scientific literature. For instance, the risk analysis prepared in support of
the PM NAAQS reported risks below these levels, while acknowledging that the confidence in
those effect estimates is higher at levels closer to the standard (U.S. EPA, 2010).7? Furthermore,
both the PM and Ozone Integrated Science Assessments (U.S. EPA, 2009 and U.S. EPA,
2013)78'79 have determined that there is no evidence of a population-level threshold (i.e., a level
below which there is no risk for adverse health effects) in PM- and ozone-related health effects
in the epidemiological literature. While estimates of PM- and ozone-related health impacts that
occur below the standards are more uncertain than those occurring above the standards, the EPA
considers these to be legitimate components of measures of criteria pollutant health risk.
EPA also states in Section II of the preamble that as of September 27, 2010 all CO
nonattainment areas were redesignated to maintenance areas.
What Commenters Said:
11 U.S. Environmental Protection Agency (U.S. EPA). 2010. Quantitative Health Risk Assessment for Paniculate
Matter—Final Report. EPA-452/R-10-005. Office of Air Quality Planning and Standards, Research Triangle
Park, NC. September. Available on the Internet at
.
78 U.S. Environmental Protection Agency (U.S. EPA). 2009. Integrated Science Assessment for Paniculate Matter
(Final Report). EPA-600-R-08-139F. National Center for Environmental Assessment—RTF Division.
December. Available on the Internet at .
79 U.S. Environmental Protection Agency (U.S. EPA). 2013. Integrated Science Assessment for Ozone and Related
Photochemical Oxidants (Final Report). EPA 600/R-10/076F. National Center for Environmental Assessment—
RTF Division. December. Available on the Internet at
.
8-29
-------
Chapter 8: Benefits
Commenter: Chevron Products Company
As noted by Environ in their assessment of the impacts of nationwide adoption of the LEV III
program and a 10 ppm sulfur standard, (1) the Tier 2 program is estimated to result in substantial
emission reductions. In 2022, Environ estimated that the Tier 2 program would result in a 66%
reduction in NOx (11,342 Mg/day) and a 55% reduction in VOC (3787 Mg/day) for the on-road
motor vehicle fleet relative to the Tier 1 program. On the other hand, the LEV 111/10 ppm S
controls (similar to EPA's Tier 3 proposal) would result in a 6% reduction from both NOx (628
Mg/day) and VOC (179 Mg/day)relative to Tier 2 in 2022 (greater reductions would accrue as
the vehicle standards are fully implemented). EPA should also report the benefits attributable to
the Tier 2 program to allow an assessment of the incremental benefits of the Tier 3 proposal
relative to the Tier 2 program.
It is not clear exactly how EPA estimated the health and economic benefits. Without a better
understanding, it is difficult to understand how the extremely modest estimated impact of the
Tier 3 programs had on pollutant levels by 2030 translated into the EPA reported health and
economic benefits. For example, Chapter 7.2.4 of the draft RIA estimates that, in 2030, Tier 3
will result in a reduction of the weighted annual average values for ozone and PM2.5 that are
each less than 1 percent of the current annual standards.
Our Response:
The benefits presented in the RIA are based on the incremental reduction in PM- and
ozone-related pollutant emissions related to the final Tier 3 standards compared to a world
without the Tier 3 standards in place. This baseline, or reference case, includes all finalized
regulations at the time of the analysis. The baseline therefore includes Tier 2 emission controls;
the benefits of the Tier 3 rule are incremental to those related to the Tier 2 rule. We refer the
commenter to the Tier 2 RIA for information about the estimated impacts of that rulemaking
(http ://www.epa. gov/tier2/).
Chapter 8 of the RIA describes the methodology involved in quantifying and monetizing
the benefits of the Tier 3 rule. We base our analysis of the program's impact on human health
SO 81 S'7
and the environment on peer-reviewed studies of air quality and human health effects. ' ' Our
benefits methods are consistent with the Regulatory Impact Analyses (RIAs) that accompanied
the final revisions to the National Ambient Air Quality Standards (NAAQS) for Particulate
Matter, the 2008 final ozone NAAQS, and the 2010 ozone NAAQS reconsideration.
U.S. Environmental Protection Agency. (2012). Regulatory Impact Analysis for the Final Revisions to the
National Ambient Air Quality Standards for Particulate Matter. Prepared by: Office of Air and Radiation, EPA-
452/R-12-005. Retrieved November 22, 2013 at http://www.epa.gov/ttn/ecas/ria.html
U.S. Environmental Protection Agency. (2008). Final Ozone NAAQS Regulatory Impact Analysis. Preparedby:
Office of Air and Radiation, Office of Air Quality Planning and Standards. Retrieved November 22, 2013 at
http://www.epa.gov/ttn/ecas/ria.html. EPA-HQ-OAR-2009-0472-0238
'' U.S. Environmental Protection Agency. (2010). Summary of the updated Regulatory Impact Analysis (RIA) for
the Reconsideration of the 2008 Ozone National Ambient Air Quality Standard (NAAQS). Prepared by: Office
of Air and Radiation, Office of Air Quality Plannin and Standards. Retrieved November 22, 2013 at
http://www.epa.gov/ttnecasl/regdata/RIAs/sl-supplemental_analysis_full.pdf.
8-30
-------
Tier 3 Summary and Analysis of Comments
What Commenters Said:
Commenter: Private Citizen
It is my belief that the supposed cost savings in health costs will never materialize as suggested
by proponents of the bill. What we do know from past experience is that 1) the government
always under estimates the cost of everything they foist upon the public 2) that the performance
of light and medium vehicles will diminish 3) the cost to implement and comply with these
regulations will be much greater than proposed 4) this is but another attack on the free enterprise
system. Those in favor of this regulation point out that a large number of people in this country
have asthma and apparently it is caused by the exhaust from automobiles. Is that true? If
automobile exhaust is the cause of asthma why doesn't everyone suffer from asthma? If this
regulation is implemented how many asthma suffers will be cured? How many will be prevented
from becoming asthmatics? I see this as another unneeded intrusion by government into the free
enterprise system.
Our Response:
Chapter 7 of the RIA that accompanies this rulemaking describes the methods and
assumptions that inform our cost analysis. Our cost estimates reflect the most up-to-date and
scientifically sound data available, and we believe our estimates are therefore accurate
predictions of future costs associated with the final Tier 3 rulemaking. Furthermore, as presented
in Chapter 8 of the RIA, the final Tier 3 rulemaking will provide the public with very significant
health benefits that clearly outweigh the costs. For the final rulemaking, we estimate that by
2030, the annual emission reductions of the Tier 3 standards will annually prevent between 660
and 1,500 PM-related premature deaths, between 110 and 500 ozone-related premature deaths,
2,200 hospital admissions and asthma-related emergency room visits, 19,000 asthma
exacerbations, 30,000 upper and lower respiratory symptoms in children, and 1.3 million lost
school days, work days, and minor restricted activity days.
Asthma prevalence in the United States is widespread and well-documented, but there is
currently no cure for asthma. Asthma symptoms can be triggered by a number of different
causes, including air pollution emitted from vehicles. The health impact assessment (HIA) in the
Tier 3 RIA quantifies the changes in the incidence of adverse health impacts resulting from
changes in human exposure to both PM2.5 and ozone air quality. HIAs are a well-established
approach for estimating the change in adverse health impacts expected to result from population-
level changes in exposure to pollutants. They rely on the published scientific literature to
ascertain the relationship between PM2.5, ozone, and adverse human health effects. For asthma-
related endpoints included in this analysis, we estimate the number of exacerbated asthma attacks
and the number of asthma-related emergency room visits that will be avoided as a result of the
PM2.5 and ozone reductions associated with the final Tier 3 rulemaking. Please refer to Chapter
8 of the RIA for more information.
8-31
-------
Tier 3 Summary and Analysis of Comments
9 Economic Impact Analysis
What We Proposed:
The comments in this chapter correspond to Section X of the preamble to the proposed
rule and address the economic impacts of the program. A summary of the comments received
and our response to those comments are located below.
What Commenters Said:
Commenter: BlueGreen Alliance
The Tier 3 standard will only bolster the auto industry's ability to meet a strong fuel efficiency
standard and generate these net positive economic outcomes. Implementation of Tier 3 will
also have a positive economic impact in other sectors as well.
Additionally, the program includes critical elements that will allow refiners to deliver
cleaner fuels to the market in an economical and timely manner, as well as provisions that
ensure smaller manufacturers and refiners can reach compliance without unfair burden. It is
critical to provide a mechanism like this to help industries transition to a cleaner regulatory
framework. For example, the advanced technology vehicle manufacturing loan program has
created or sustained more than 40,000 jobs in the auto sector advancing fuel saving
technologies.
As we have seen in the automobile and other industries, investment in pollution reduction
technologies leads to gains in efficiency and job creation.
Commenter: Boulder County Board of Commissioners and Boulder County Board of Health
Boulder County applauds the EPA for proposing the Tier 3 package and reiterates its strong
support for this program because it will directly benefit the air quality in our community and
across the nation. These are substantial, cost-effective emission reductions from proven
technologies that will create jobs and improve public health and the environment.
Commenter: Business for Innovative Climate and Energy Policy (BICEP)
In sum, the proposed Tier 3 standards represent a unique opportunity for the United States to
strengthen the economy and create jobs, as well as prevent illness and reduce healthcare
expenditures.
Commenter: California Air Resources Board (CARB)
CARB staff supports the approach used to evaluate the economic impacts of U.S. EPA's Tier 3
proposal.
9-1
-------
Chapter 9: Economic Impact Analysis
Due to the differences in regulatory scope, timeframes, and geography, a direct comparison to
CARB's LEV III economic impact analysis is incompatible. However, given that the proposal's
compliance costs and cost-effectiveness metrics are on the same order of magnitude as those
estimated by CARB in previous related rulemakings, CARB staff concurs with U.S. EPA's
overall conclusions on the direction and magnitude of economic impacts to the auto and
refinery sectors. However, some time has passed since the initial evaluation of technology costs
detailed in the draft Regulatory Impact Analysis. Therefore, it may be appropriate for U.S. EPA
to revisit those costs estimates to determine if any revisions are warranted.
Commenter: Ceres Investor Network on Climate Risk
Furthermore, Tier 3 will result in net economic gains and job growth, and help ensure the
United States' leadership position in vehicle and emission control technology. As widely
diversified investors with exposure across virtually all industries, sectors and markets the
performance of member portfolios is linked in no small part to overall economic performance.
Consequently, we believe the economic benefits described in this letter will also be beneficial
to the performance of our investments.
Further, Tier 3 also presents an opportunity to ensure the United States continues to be a
globally competitive leader in the auto and emission control industries. The standards will send
a clear regulatory signal to suppliers and manufacturers, encouraging technological innovation
and investment.
Commenter: Emissions Control Technology Association (ECTA)
ECTA has long supported the adoption of emission control standards reflected in EPA's Tier 3
proposal, but opponents have continuously tried to undermine Tier 3 by making claims that it
would have negative economic consequences. It has been claimed that the sulfur content
requirement of Tier 3 will dramatically increase the cost of refining gasoline, increase the price
of gasoline at the pump, cause job loss, and reduce economic activity. To address these
criticisms, ECTA engaged Navigant Economics [See Docket number EPA-HQ-OAR-2011-
0135-4293-A2] to evaluate the economic impact of the pending rule (1). The study concluded
that the pending rule would increase the average cost of refining by about one cent per gallon,
in all likelihood not affect gasoline prices, increase employment, and generate economic
activity. A copy of the study is attached with this comment.
The rule had not been published when this study was completed in June, 2012. But, it was well
known in June, 2012 that the substance of the proposal would be similar to LEV III, which was
adopted by California in 2012.
Commenter: Kentucky Division for Air Quality
EPA's documents state that the increase in gas and Vehicle prices would cause an initial
decrease in the sales of gas and vehicles. Historically, the reduced travel was determined to be
negligible; therefore, it would have little impact. It is also for this reason, we do not see EPA's
Tier 3 proposed rule affecting KYTC in an adverse way.
9-2
-------
Tier 3 Summary and Analysis of Comments
Commenter: Manufacturers of Emission Controls Association (MECA)
EPA's proposed Tier 3 emissions and fuel standards will provide additional support for the
continued development of a thriving U.S. industry focused on a wide range of technologies that
can reduce vehicle criteria emissions.
Commenter: Motor & Equipment Manufacturers Association (MEMA)
As the process moves forward, the EPA must maintain regulatory clarity and a coherent
national pathway that affords the vehicle manufacturers and vehicle parts suppliers with critical
planning and investment certainty. MEMA recognizes the need for appropriate, sensible
regulation. On the other hand, regulations should not place an undue burden on free enterprise,
should not be arbitrary and capricious in their requirements or implementation, and should not
hinder the global competitiveness of the industry. Rather, the regulations should be based on
sound science and data and balance the public interest objectives with economic realities of
manufacturers. The effectiveness of the proposed regulations are projected figures and should
be periodically evaluated and reviewed with recorded, real-world data to see if the cost/benefit
projections are being realized.
Commenter: PennFuture
In addition, Tier 3 will drive the advances in emissions control technology developed here in
the U.S. and exported globally.
Commenter: United Steelworkers Union (USW)
The USW has a diverse membership that will, in several cases, benefit from the proposed rule.
For example, USW members employed at Corning Inc. who produce advanced catalytic
converters will see the benefits of high tech jobs building pollution-reducing vehicle
components.
Commenter: Wespath Investment Management
Wespath has reviewed positions on both sides of Tier 3, and has considered the potential long-
term financial implications to our investments, which is material to fulfill our promises to our
stakeholders. In our view, Tier 3 standards will be beneficial to several sectors where we hold
investments, including the auto manufacturers, auto parts manufacturers, trucking companies,
health care providers, and health insurance companies. We also invest in many oil refining
companies, which have expressed concern over the impact of Tier 3 on their operations.
Overall, we believe the potential for negative impact in the oil sector is limited, and outweighed
by the collective benefit to other sectors.
Job creation: The fact that a wide range of stakeholders have spoken in favor of the proposed
standards speaks to the broad economic benefits. The standards are supported by trade
associations such as the Alliance of Automobile Manufacturers, Global Automakers, the
9-3
-------
Chapter 9: Economic Impact Analysis
American Trucking Association, and the Manufacturers of Emission Controls Association.
Public companies whose shares we hold that have also made supportive statements include
General Motors, Honda, Toyota, Daimler Chrysler, Ford, Donaldson Company, Tenneco,
BorgWarner, and Corning. These companies believe that the standards are in their long-term
best interests and will promote their future growth, and we agree.
Many companies have told us that without clear policy signals from Washington, innovation
stalls and capital sits on the sidelines. Setting strong standards will be the signal companies are
looking for to begin production that incorporates new technologies. This will ripple throughout
the supply chain, allowing manufacturers to bring new and more efficient products to market.
Further, many in the industry have said that the Tier 3 standards are necessary to enable auto
companies to meet the new fuel economy and greenhouse gas standards, which will spur
innovation and growth in the auto industry.
Our Response:
Some of these comments express support for the Tier 3 standards based on the positive
impacts and innovation that these standards will spur; others provide statements of caution
about the tradeoffs between "public interest objectives [and] economic realities of
manufacturers." Some comments suggest that the standards will increase economic activity
and employment, while others suggest that the effects will be small. Chapters 9.1, 9.2, and 9.3
below provide more detailed responses related to the impacts on vehicle sales, oil production,
and employment. As discussed in Section X. of the Preamble and Chapter 9 of the RIA, EPA
has concluded that the effects of the standards on sales in both the auto and refinery sectors are
likely to be small, because the cost impacts to consumers are expected to be small. These small
impacts imply small effects on employment as well.
We agree, as several commenters note, that the standards are likely to encourage firms
to innovate in order to find less expensive ways to achieve the standards. The certainty
provided by the standards is expected to encourage investment both to achieve the standards
and to find less expensive ways to meet them.
We agree with MEMA that "the regulations should be based on sound science and data
and balance the public interest objectives with economic realities of manufacturers." The
Preamble and RIA provide our full discussion of these tradeoffs. RIA Chapter 8.1.3 provides a
comparison of the benefits and costs, and finds that the benefits in 2030 (the year in which air
quality benefits are monetized) are expected to outweigh the costs: for a 3% discount rate,
benefits less costs are estimated to be $5.9-$18 billion; for a 7% discount rate, benefits less
costs are estimated to be $5.2-$ 17 billion. EPA follows trends in the auto and fuels industries
relating to costs, and it monitors compliance with the standards and the impacts on air quality.
9.1. Vehicle Sales Impacts
What Commenters Said:
9-4
-------
Tier 3 Summary and Analysis of Comments
Commenter: Attorney General of Connecticut, et.al
The Tier 3 standards would also benefit our economy. Automakers strongly support the
standards, which would enable them to sell vehicles meeting California's Low Emission
Vehicle III standards in all 50 states and at the same time achieve EPA greenhouse gas
emission standards for the model years 2017-25.
Commenter: Attorney General of New York, et.al:
The Tier 3 standards would also benefit our economy. Automakers strongly support the
standards, which would enable them to sell vehicles meeting California's Low Emission
Vehicle III standards in all 50 states and at the same time achieve EPA greenhouse gas
emission standards for the model years 2017-25.
Commenter: Consumers Union
New car buyers will also benefit. Starting in 2017, new cars will have tighter limits on tailpipe
emissions, including carbon monoxide and benzene, which can linger in garages and even
attached residential living space (4). The proposed rule also offers automakers an incentive to
go beyond the minimum 8-year/ 80,000-mile warranty currently required for emissions control
systems, and extend it to 15-years/150,000 miles for new vehicles. This move could improve
reliability and lower costs to maintain emissions control systems.
Low sulfur gasoline also enables auto makers to develop a greater array of technology, such as
lean burn, to meet emissions and fuel economy standards more creatively and at a lower cost.
4 - International Journal of Ventilation Volume 2 No 3 , "Air and Pollutant Transport from Attached
Garages to Residential Living Spaces - Literature Review and Field Tests,", National Institute of
Standards and Technology, available at: http://fire.nist.gov/bfrlpubs/build03/PDF/b03067.pdf.
Commenter: Manufacturers of Emission Controls Association (MECA)
MECA agrees with EPA staffs assessment that achieving the proposed Tier 3 exhaust and
evaporative emission standards and associated emission reductions are both technically feasible and
cost-effective. This fact is clearly demonstrated by the more than two million SULEV- and
PZEV-compliant light-duty vehicles that have been sold in the U.S. market since these near-
zero emission, gasoline vehicles were first introduced more than ten years ago.
Commenter: Consumers Union
New car buyers will also benefit. Starting in 2017, new cars will have tighter limits on tailpipe
emissions, including carbon monoxide and benzene, which can linger in garages and even
attached residential living space.4 The proposed rule also offers automakers an incentive to go
beyond the minimum 8-year/ 80,000-mile warranty currently required for emissions control
systems, and extend it to 15-years/150,000 miles for new vehicles. This move could improve
reliability and lower costs to maintain emissions control systems.
9-5
-------
Chapter 9: Economic Impact Analysis
4 International Journal of Ventilation Volume 2 No 3 , "Air and Pollutant Transport from Attached
Garages to Residential Living Spaces - Literature Review and Field Tests,", National Institute of
Standards and Technology, available at: http://fire.nist.gov/bfrlpubs/build03/PDF/b03067.pdf.
Commenter: U.S. Coalition for Advanced Diesel Cars
The standards will also encourage the development of innovative new technologies and provide
American consumers with more and cleaner vehicle choices.
Our Response:
These comments express support for the standards based on their impacts on vehicle
sales and innovation. The auto industry is expected to take advantage of having one set of
standards for the U.S.; this program contributes to that goal. We agree with MECA that the
new standards are feasible. We agree as well that new vehicle owners as well as the general
public will benefit from reduced tailpipe and evaporative emissions. And, as discussed in the
introductory part of Chapter 9 (above) of this S&A, we agree that the standards will encourage
innovation in technologies.
9.2. Impacts on Petroleum Industry
What Commenters Said:
Commenter: Clean Air Watch
I'll just note that they're wrong. You're going to hear today, I believe, from Navigant, who's
going to critique some of the oil industry's current cost projections on this study, so I will hope
that you'll listen carefully and then leave it to them. But I'd like to look at some of the past
examples of what I would call crying wolf by the oil industry. Let's consider a few examples
where they've been wrong.
When EPA proposed a phase out of lead in gasoline in the 1970s, the oil industry claimed there
was no public health advantage to be gained, and that this would cause supply shortages. Not
only was this sky is falling prediction dead wrong, we know now it was one of the best steps
for public health ever made by the EPA. Taking lead out of gas reduced the risk of lead
toxicity for hundreds of thousands of children. It also decreased the risk of heart attack and
stroke. Oh, and we didn't have those supply shortages.
In 1999, when EPA proposed the so-called Tier 2 standards in allied sulfur and gas reductions,
once again the oil industry claimed that clean up wasn't needed, and it would cause shut downs
and supply shortages. In fact, in this case they even said it would threaten national security.
Well, of course these problems didn't happen either. It was sheer nonsense.
Let's take a third example. In the year 2000 after EPA proposed clean diesel fuel standards, the
oil industry again claimed - now, can you guess - that the cleanup wasn't needed, and it was
cause refinery shutdowns and supply shortages. Guess what? The exact opposite happened.
The U.S. now exports clean diesel fuel because EPA stuck to its guns.
9-6
-------
Tier 3 Summary and Analysis of Comments
So I would just like to say let's learn from the past. This is not the United States of Amnesia, I
hope. Let's remember the oil industry has cried wolf so many times in the past, always been
wrong. It's doing it again here. Let's consider the massive health benefits of this plan that a lot
of other witnesses have already talked about. We know that they dwarf the very modest costs.
Commenter: Emissions Control Technology Association (ECTA)
Tier 3 is Unlikely to Increase Retail Gasoline Prices. We disagree with EPA's conclusion that
the proposed rule would increase the cost of gasoline to consumers.
EPA concludes that:
"We do not estimate the price increase that consumers are likely to face, though we note that it
should be positive and up to the increase in manufacturer's cost of gasoline production" (2).
This conclusion is drawn from: (1) the observation that in a perfectly competitive industry the
increased cost of refining would be passed along to consumers; and (2) the assumption, based
upon a report published by the Federal Trade Commission, that the refining industry is
relatively competitive (3).
While we do not comment on the competitiveness of the U.S. refining industry, we believe that
the proposed rule is unlikely to result in an increase in the price of gasoline to consumers. We
draw this conclusion from an empirical analysis previously prepared by Navigant Economics
(4), which examined the effect of EPA sulfur content regulation on the monthly average price
of gasoline during the period from January 2001 to January 2012.
This statistical evidence leads one to conclude that, in all probability, the sulfur-content
regulation of Tier 2 did not increase gas prices. Given this conclusion, it is unlikely that the
sulfur content regulation of Tier 3 would result in higher gas prices for two reasons.
First, relative to Tier 2, Tier 3 requires a much smaller reduction in the average sulfur of
content of the fuel. Tier 3 requires a reduction of 20 ppm (from an average of 30 ppm to 10
ppm), while Tier 2 required a reduction of 270 ppm (from 300 ppm to 30ppm). Tier 3 requires
a reduction in sulfur that is 93% lower than that required of Tier 2.
Second, relative to Tier 2, Tier 3 imposes a much smaller cost of compliance on refiners. EPA
estimates that the sulfur requirement of Tier 3 will increase the cost of refining by 0.89 cents
per gallon, which is considerably less than the agency's prior cost estimate of Tier 2 (1.7 to 1.9
cents per gallon). From the perspective of the refiners, Tier 3 will impose half the costs of Tier
2.
It has been noted by some commenters in this proceeding that Tier 3 would "...increase the
price for gasoline paid at the pump by 6 to 9 cents per gallon (5). These comments cite a study
prepared by Baker and O'Brien for the American Petroleum Institute (6). Although Baker &
O'Brien concluded that Tier 3 would increase the marginal cost of refining by 6 to 9 cents per
gallon, their study was silent on the associated retail price effect. Advocates for the refining
industry often characterize the Baker and O'Brien study as concluding that Tier 3 will increase
gasoline prices, but even the study's authors were not willing to embrace this position.
2 - EPA, Control of Air Pollution From Motor Vehicles: Tier 3 Motor Vehicle Emission and Fuel
Standards, May 21, 2013, Proposed Rules (hereafter EPA Proposal). Fed Register pp. 29988-29989.
3 - Id. at footnote: 499, p. 29989.
9-7
-------
Chapter 9: Economic Impact Analysis
4 - See Navigant Economics, Economic Analysis of the Implications of Implementing EPA's Tier 3
Rules, prepared for The Emissions Control Technology Association, June 14, 2012, p. 17-23 and
Technical Annex (hereinafter "Navigant Study").
5 - May 23, 2013 letter to Gina McCarthy from the Institute For Energy Research, p. 2.
6 - See Baker & O'Brien, Inc. Potential Supply and Cost Impacts of Lower Sulfur, Lower RVP
Gasoline, prepared for the American Petroleum Institute, July 2011 (hereinafter "Baker & O'Brien 2011
Study"), and The Baker & O'Brien, Inc., Addendum to Potential Supply and Cost Impacts of Lower
Sulfur, Lower RVP Gasoline, prepared for the American Petroleum Institute, March 2012 (hereinafter
"Baker & O'Brien 2012 Study").
Commenter: Mercatus Center at George Mason University
REGRESSIVE EFFECTS
While the EPA estimates that that this regulation will have a minimal impact (approximately
one penny) on fuel costs, others have estimated the impact to be significantly greater, perhaps
as much as 9 cents per gallon (18). Exactly how much gasoline prices may rise is difficult to
say, but given that energy costs consume a higher proportion of low-income individuals'
income relative to high-income individuals, it is worth acknowledging that this regulation may
impose a disproportionate impact on low-income individuals. Similarly, recent academic
research has shown that regulations, such as the proposed regulation, are often more in line
with risk preferences of wealthy households (19). It is unlikely that poor households are
worried about the risks posed from PM2.5 and ozone when they face much larger risks
elsewhere in their lives. The income they lose to comply with this regulation may be better
utilized toward other risk mitigation.
18 - David Tamm and Kevin Milburn, "Addendum to Potential Supply and Cost Impacts of Lower
Sulfur, Lower RVP Gasoline," (American Petroleum Institute, Houston, TX, March 2012).
19 - Diana Thomas, "Regressive Effects of Regulation." (Working Paper No. 12-35, Mercatus Center at
George Mason University, Arlington, VA, November 2012), http://mercatus.org/publication/regressive-
effects-regulation.
Commenter: Northeast States for Coordinated Air Use Management (NESCAUM)
To put the Tier 3 low sulfur impacts on the petroleum refining industry into perspective, it is
instructive to summarize what has been previously required in other fuel programs. Over the
past 13 years, EPA undertook three rulemakings, requiring significant reductions in the sulfur
content of petroleum fuels. The Tier 2 regulation, promulgated in 2000, reduced the
sulfur content of gasoline from a pre-regulation average of about 330 ppm down to 30 ppm and
included an interim 120 ppm step. The highway diesel regulation, promulgated in 2001,
reduced the sulfur content of diesel from 500 ppm down to 15 ppm. The non-road diesel
regulation, promulgated in 2004, reduced the sulfur content of non-road fuel from a pre-
regulation average of 3000 ppm down to 15 ppm, with an interim 500 ppm step. Compared to
what is now proposed under Tier 3, these previous initiatives required very ambitious
reductions. The refining industry met these challenges, supplying the market with compliant
fuel while continuing to prosper economically. EPA's structuring of these regulations played a
critical role in facilitating the transition from high to low sulfur fuels in each case.
9-8
-------
Tier 3 Summary and Analysis of Comments
Specifically, EPA built several very important regulatory flexibilities for refiners into the
gasoline and diesel fuel sulfur standards, including 1) providing several years of lead time for
all refiners to add and/or enhance desulfurization capabilities, 2) ABT programs to encourage
early compliance where possible and provide means for extending compliance dates where
needed, 3) provisions for smaller refiners to further extend compliance deadlines and credit
generation opportunities, 4) opportunities for refiners to integrate their desulfurization
infrastructure planning processes across all three fuels programs, 5) interim sulfur limits to
allow refiners to phase their operations into compliance with the final standards, and 6) various
hardship waiver provisions to provide a means to address unexpected circumstances. Most of
these same flexibilities are built into the proposed Tier 3 program structure and in conjunction
with comparatively modest requirements for reduction in sulfur will make for a relatively
smooth transition from a 30 ppm average down to a 10 ppm average sulfur content.
Some in the refining sector have made projections that the Tier 3 sulfur standards will force a
number of refineries to close because the costs of compliance will be too high and/or the
deadlines are too soon to make the necessary equipment and operational changes at the
refinery. Consequently, the U.S. fuel markets allegedly will become increasingly more
dependent on competitive foreign imports of products. Similar projections were made during
promulgation of the previous rulemakings for low sulfur fuels. It is instructive to briefly review
what actually occurred in the U.S. refining sector over the phase-in period of these regulations
in order to gain a sense of what the likely outcomes will be under this more modest Tier 3
scenario.
Refining activity in the U.S. increased over the same period as indicated in Table 2 [p. 10 of
EPA-HQ-OAR-2011-0135-4283-A1]. Particularly noteworthy is the fact that desulfurization
capacity increased by 40 percent from 2000 to 2010, indicating that the domestic refining
industry responded positively to the regulatory challenge and succeeded in dramatically
reducing the sulfur content of fuels.
Refiners' original projections of very high compliance costs were principally based on the
assumption that conventional energy-intensive hydrotreating technology would have to be
deployed almost exclusively to achieve the low sulfur levels in the final products. In actual
practice, refiners opted for a combination of technologies and facility efficiency improvements
to cost-effectively remove fuel sulfur. At around the time of the first of the three rulemakings
(Tier 2 gasoline), the National Petroleum Council issued a report (16) identifying several more
cost-effective desulfurization technology improvements, classified either as "demonstrated" or
"near commercial status." By the time refiners had to select equipment for meeting Tier 2
gasoline standards, there were several additional, less energy-intensive technology choices
available to them (17). Despite industry predictions of insufficient lead time and manufacturing
resources for timely compliance with the standards for all three rulemakings, early compliance
was widespread, as evidenced by the surplus of credits under the ABT programs (18,19).
[T]here are strong economic incentives for refiners to design and purchase the most energy-
efficient process equipment to minimize the cost of production. For example, most of the new
or modified units expected to be involved in refinery projects designed to meet the proposed
Tier 3 standards are fuel combustion units (e.g., process heaters). Because fuel cost (direct cost
9-9
-------
Chapter 9: Economic Impact Analysis
in the case of purchased natural gas and opportunity cost in the case of refinery-generated fuel
gas) represents a significant component of total operating cost for such units, refineries will
strive to maximize energy efficiency based on available technologies as part of their project
design (20).
Regarding the issue of foreign imports of fuels, the discussion in the proposed rule points out
that "despite refining industry projections that previously imposed diesel rules would lead to
greater U.S. reliance on imports through major negative impacts on domestic refining, the
reverse has actually occurred. Over the last 8 years, imports of gasoline and diesel fuel have
continued to be the marginal supply, and have even dropped precipitously so that the U.S. is
now a net exporter of diesel fuel and is importing half the gasoline that it did at its peak in
2006. With the projected decline in future gasoline demand in the U.S. as vehicle fuel
efficiency improves, gasoline imports are expected to continue to decline' (21). According to
the U.S. Energy Information Administration's 2013 Annual Energy Outlook, the U.S. is
projected to be a net exporter of petroleum products at least through 2040.
In response to favorable long-term economic outlooks, the North American refining industry is
taking the initiative to make significant new investments in its capacity to increase the supply
of petroleum products, including low sulfur products. Several refineries in the Northeast that
recently were in danger of permanently closing have reopened or remained in business due to
ownership changes and favorable developments related to petroleum and product supply. In
2010, PBF Energy purchased the Valero refinery in Delaware City, DE. In conjunction with the
purchase of the Delaware City facility, PBF Energy announced plans to invest $500 million to
enhance its desulfurization capacity (22). In 2011, PBF Energy announced plans to invest $1
billion at the Delaware City refinery to boost distillate output and heavy crude capacity.
In 2010, Marathon Petroleum completed a $3.9 billion expansion, nearly doubling the capacity
of its Garyville, LA refinery (23). This refinery is now the fourth largest in the U.S. and
increases the ability of Gulf Coast refiners to augment the petroleum product needs in the
Northeast. In 2009, Irving Oil Refinery (Saint John, New Brunswick) completed a $220 million
upgrade. The majority of the work focused on improving its yield of ultra-low sulfur products.
The Saint John Refinery is Canada's largest and exports more than 80 percent of its products to
the U.S. (24).
16 - U.S. DOE. 2000. U.S. Petroleum Refining: Assuring the Adequacy and Affordability of Cleaner
Fuels, June 2000.
17 - MathPro Inc. 2003. Evolution of Process Technology for FCC Naphtha Desulfurization: 1997-
2003; An Example of Technical Progress Induced by Environmental Regulation, March 2003.
18 - U.S. EPA, Summary and Analysis of the 2005 Highway and Nonroad Diesel Fuel Pre-Compliance
Reports, EPA 420-R-06-012, June 2006.
19 - U.S. EPA. 2010. Summary and Analysis of the 2010 Nonroad Diesel Fuel Pre-Compliance Reports,
EPA 420-R-10-028, December 2010.
20 - 78 FR 29934, May 21, 2013
21 - 78 FR 29992, May 21, 2013.
22 - heatingoil.com. Delaware Refinery to Reopen with Plans to Produce Low-Sulfur Heating Oil and
Biofuels, Posted June 2, 2010.
23 - The Times-Picayune, Marathon Completes $3.9 Billion Expansion in Garyville, March 25, 2010.
9-10
-------
Tier 3 Summary and Analysis of Comments
24 - Irving Oil Company. 2009. Press Release: Irving Oil Refinery Completes $220 Million Investment
Project, November 17, 2009.
Commenter: Pennsylvania Department of Environmental Protection (DEP)
DEP recognizes the important health benefits and cost-effectiveness of this proposed rule.
Nevertheless, DEP has concerns about the impacts of costs to refiners of gasoline, projected
gasoline cost increases to be passed along to the consumer.
Commenter: United States Congress, House of Representatives, Pennsylvania, 7th District
High gas prices have stretched Pennsylvania families thin already, and in the Philadelphia area,
the refining industry is showing new promise after weathering a very tough economy for the
sector. Adding new regulations without appropriate consultation with stakeholders— including
the oil workers whose jobs could be put at risk through new mandates — and a thorough
evaluation of the regulation's economic and consumer consequences is the wrong approach.
Our Response:
These comments reflect two views: one, that the standards will have significant adverse
effects; the other, that the standards will have negligible or no effects.
EPA estimates that the national average refinery costs incurred to comply with the fully
phased-in Tier 3 sulfur control program with ABT program will be 0.65 cents per gallon, averaged
over all gasoline. Section VII.B of the Preamble, Chapter 5 of the RIA, and Chapter 7.2 of this
document discuss EPA's fuel cost estimates in detail. As discussed in Preamble Section X.C.
and RIA Chapter 9.3, EPA considers this impact to be small, and thus to be a barely measurable
contributor to gasoline prices. We do not consider this view to contradict the study by
Navigant (cited by ECTA) that was not able to find any impacts from the larger Tier 2
standards; in that case, and likely in the case of Tier 3 as well, the small increases may be
swamped by the general variability in fuel prices. We agree with NESCAUM that the refinery
sector on the whole has weathered the Tier 2 standards and is likely to face less additional
difficulty with the Tier 3 standards. Because the impact of the standards on gasoline prices is
expected to be small, we do not believe that the standards will contribute in a measurable way
to household fuel costs. We expect this impact to be small even for low-income households.
An increase of 0.65 cents per gallon for gasoline is less than 0.2 percent. The lowest quintile of
U.S. households in 2012 spent $1,222 on gasoline and motor oil (Bureau of Labor Statistics,
Consumer Expenditure Survey, Quintiles of income before taxes,
http://www.bls.gov/cex/#tables); an increase of 0.2 percent is under $3 per year, an amount that
is unlikely to cause significant additional hardship to low-income households.
Households will, in addition, receive the benefits of cleaner air. As stated in Preamble
Section XII. J, "This final rule will reduce emissions from vehicles across the nation, both new
vehicles (beginning in model year 2017, when the vehicle standards start to apply) and existing
vehicles (as soon as the lower-sulfur gasoline becomes available in 2017). As such, this rule
increases the level of environmental protection for all populations. As discussed in Section
9-11
-------
Chapter 9: Economic Impact Analysis
III.C.7, there is evidence that minority populations and low-income populations live
disproportionately near high-traffic roadways, where concentrations of many air pollutants are
elevated. We expect this final rule to increase the level of environmental protection for these
populations."
In addition to receiving many public comments, both in public hearings and submitted
in writing, EPA has conducted significant consultation with stakeholders in developing these
standards; see Preamble Section XI. Section V.L of the Preamble, Chapter 4 of the RIA, and
Chapter 5 of this document discuss the feasibility for refineries to meet these standards. As
discussed in those, we believe that refineries will not be put at undue risk. We agree with
Clean Air Watch that previous rulemakings have not led to significant adverse impacts in the
refining sector, nor do we expect these standards to have significant adverse impacts. EPA
agrees that the Tier 3 standards, as well as the other standards cited by Clean Air Watch, have
produced significant public health benefits. The benefits of Tier 3 are discussed in Preamble
Section VIII and RIA Chapters 6-8.
Preamble Section X.D, RIA Chapter 9.3.2, and Chapter 9.3.2 of this document provide
EPA's examination of employment impacts on refineries. EPA does not quantify most of the
employment impacts of the standards. Most likely, employment will be gained in some sectors
or locations, and it may be reduced in other sectors or locations; the net effect is expected to be
small. We note that the United Steel, Paper and Forestry, Rubber, Manufacturing, Energy,
Allied Industrial and Service Workers International Union (USW) and the International Union,
United Automobile, Aerospace & Agricultural Implement Workers of America (UAW)
provided comments supportive of Tier 3, including mention of expected positive job impacts
from the standards.
9.3. Employment Impacts
What Commenters Said:
Commenters: American Petroleum Institute (API) and the Association of Fuel &
Petrochemical Manufacturers (AFPM)
Jobs. Mandatory government regulations requiring capital expenditures and the creation of
temporary jobs for unjustified new specifications constitute inefficient uses of market financial
resources. As such, it makes little sense for Navigant or EPA to count the mandated costs of a
regulation, i.e. Tier 3, as "benefits" when in reality economic growth is slowed from what it
could have been with more efficient use of capital.
Commenter: Marathon Petroleum Company LP (MFC)
Jobs. An objective economist would be aware that mandatory government regulations requiring
significant capital expenditures and the creation of jobs to perform unnecessary gasoline sulfur
reductions is an inefficient use of market financial resources, pulling capital from more
efficient uses. As such it makes little sense for Navigant or EPA to count the mandated costs of
9-12
-------
Tier 3 Summary and Analysis of Comments
a regulation as benefits when in reality economic growth is slowed from what it could have
been with more efficient use of capital.
Commenter: Care 2 (comment campaign)
The new standards will also create jobs as automakers apply advanced technologies and
refineries upgrade their equipment.
Commenter: Union of Concerned Scientists (comment campaign)
I support Tier 3 standards that will: create new jobs at refineries, installing and operating sulfur
reduction equipment, and in the automotive sector, developing and manufacturing cleaner
vehicles.
Commenter: Medical Advocates for Healthy Air
Financial Benefits: In purely economic terms, the implementation of cleaner fuel standards will
create 5,300 permanent jobs in the operation and maintenance of new refining equipment as
well as over 24,000 new jobs, over a three year period, for equipment installation at the
nation's refineries. The industry which manufactures emissions control technology generated
$12 billion of economic activity and accounted for 65,000 US jobs in 2010. Lower traffic
pollution will lead to improved health outcomes and greatly reduce absenteeism from work and
school, thus increasing productivity.
Commenter: Mercatus Center at George Mason University
EMPLOYMENT EFFECTS
The EPA cites three different ways in which regulations may impact employment: demand
effects, cost effects, and factor-shift effects. The first method reduces employment in a
regulated industry, the second increases it, and the last effect is ambiguous. However, the EPA
should be careful to acknowledge that both jobs gained and jobs destroyed are costs of the
proposed regulation. Ideally, compliance jobs should be minimized, not viewed as a benefit of
a regulation (15).
The EPA should consider some of the long-term effects of its rules on those who lose their jobs
at this unusual moment in time, with unemployment at a level that exceeds historical norms.
The literature shows that these effects can be significant. For example, recent estimates of
earning losses resulting from job loss range from 1.4 years of earnings in times of low
unemployment to 2.8 years during times of high unemployment (16). Similarly, research shows
that after reemployment it can take as long as 20 years for workers to catch up on lost earnings,
largely due to skill mismatches between the jobs lost and the new jobs created in the economy
(17). Additionally, the EPA should acknowledge that those who lose jobs and those who gain
them are not in fact the same individuals; vastly different skill sets may be required in one type
of employment relative to the other. Moving from one area of employment to the other may
require significant skills retraining if those laid off are to be reemployed in the new line of
9-13
-------
Chapter 9: Economic Impact Analysis
compliance work.
15 - See for example Keith Hall, 'Goodbye to Green Jobs, You Won't Be Missed," Forbes, April 4,
2013, http://www.forbes.com/sites /realspin/2013/04/04/goodbye-to-green-jobs-you-wont-be-missed/.
16 - Keith Hall, 'The Employment Costs of Regulation." (Working Paper No. 13-06, Mercatus Center at
George Mason University, Arlington, VA, March 2013), http://mercatus.org/publication/employment-
costs-regulation.
17-Ibid.
Commenter: Natural Resources Defense Council (NRDC)
The Tier 3 new vehicle and fuel standards will add thousands of jobs to the U.S. economy by
requiring new technologies to clean up gasoline and clean vehicle exhaust.
New Vehicle and Fuel Standards will Create Thousands of Jobs. The Tier 3 standards will add
thousands of jobs to the U.S. economy by requiring new technologies to clean up gasoline and
clean vehicle exhaust. Importantly, Baker and O'Brien found that the Tier 3 regulations would
not result in refinery closures. With continued operation, some refineries will have to deploy
new technologies. Similarly, jobs will be created in the automobile manufacturing and
automobile parts sectors to add new technologies. Total long-term job estimates in both the
refining and automotive sectors range from at least about 2,000 jobs in EPA's low-end
assessment to nearly 5,300 jobs estimated by Navigant Economics (29). In the short-term, EPA
and Navigant agree that thousands of workers are expected be employed to design and build the
needed modifications to refineries.
29 - Schink, op. cit.
Commenter: Northeast States for Coordinated Air Use Management (NESCAUM)
In addition to critical air quality, public health, and environmental benefits, Tier 3 would
promote economic growth and create jobs throughout the U.S. According to the Manufacturers
of Emission Controls Association, the emission control technology industry provides 65,000
domestic jobs and accounted for $12 billion in economic activity in the U.S. in 2010 (7).
7 - Manufacturers of Emission Controls Association. 2011. Press release: MECA Highlights Economic
Benefits of Mobile Source Emissions Control Industry. Available online at
http://www.meca.org/galleries/default-
file/MECA%20economic%20benefits%20press%20release%20031111 .pdf.
Commenter: Sierra Club, Clean Air Watch, Respiratory Health Association
Further, the new standards will bring more American jobs as parts suppliers develop and build
new technology and refiners and automakers install it. According to a study by Navigant
Economics, implementation of cleaner tailpipe standards would create nearly 5,300 permanent
jobs in the operation and maintenance of new refining equipment, as well as more than 24,000
new jobs over a three-year period as refineries install new equipment.3 [EPA-HQ-OAR-2011-°135-4308-A1
9-14
-------
Tier 3 Summary and Analysis of Comments
p. 2-3]
3 Navigant Economics. Economic Analysis of the Implications of Implementing EPA's Tier 3 Rules.
June 2012. http://www.ectausa.com/061212-Economic-Analysis-of-the-Implications-of-Tier-3-Sulfur-
Reduction-Final embargoed.pdf
Commenter: United Steelworkers Union (USW)
In fact U.S. refiners who currently export to lOppm sulfur standard countries have already
made efforts to upgrade assets to compete. As indicated on the April second 2013 tier three call
organized by EPA, out of the one hundred and eleven (111) U.S. refineries impacted by the
proposed rule, twenty-nine (29) can already meet the new standards or need to make
operational changes. Sixty-six (66) will need to make equipment modifications and sixteen (16)
will need to acquire additional equipment. These modifications will improve the health and
welfare of the communities our members live and work in and provide economic development
in the areas where the refineries operate.
The modifications and upgrades required under the proposed rule will also provide sustained
employment for our members employed at refineries as well as create additional jobs. The
Emissions Control Technology Association (ECTA) contracted with Navigant Economics to
review the impacts of the EPA Tier 3 standards and found installation of the refinery
modifications produces almost 24,500 jobs for full-time equivalent employees in the upgrading
of facilities to comply with the rule and the continuing annual operation of the refinery
modifications produces almost 5,300 jobs for full time equivalent employees (1).
As stated above, the proposed rules will allow domestic refining capacity to remain globally
competitive and will provide significant employment opportunities for workers who
manufacture, install, and operate the necessary refinery equipment to produce lOppm sulfur
gasoline.
1 - http://www.ectausa.com/061212-Economic-Analysis-of-the-Implications-of-Tier-3-Sulfur-
Reduction-Final embargoed.pdf
Our Response:
Most of these comm enters cite estimates of job increases associated with the Tier 3
standards. As discussed in Preamble Section X.D and RIA Chapters 9.2.2 and 9.3.2, EPA does
not quantify most of the employment impacts of the standards and thus does not evaluate these
estimates. Most likely, employment will be gained in some sectors or locations, and it may be
reduced in other sectors or locations; the net effect is expected to be small.
The Mercatus Center recommends that "the EPA should be careful to acknowledge that
both jobs gained and jobs destroyed are costs of the proposed regulation. Ideally, compliance
jobs should be minimized, not viewed as a benefit of a regulation." This assertion is confusing,
in that it implies that any effect on employment, positive or negative, is costly. MFC and API
and AFPM argue that employment impacts should not be counted as benefits. The discussion
in Preamble Section X.D of the employment impacts of the standards is intended to identify
additional ways that these standards will affect the public. Employment clearly provides
9-15
-------
Chapter 9: Economic Impact Analysis
benefits to the people who are employed, and reducing unemployment is considered a desirable
outcome in times of high unemployment. EPA disagrees that minimizing compliance jobs
should be an independent objective of this rulemaking. The objective of the program is to
improve air quality through reductions in vehicle emissions; we provide information
concerning employment impacts as part of our overall analysis of economic impacts from the
rule.
The Mercatus Center also requests that we consider the long-term effects of
unemployment on workers who may lose jobs, and that we acknowledge that those who lose
jobs may not be the same people who gain jobs. We recognize there are costs to workers who
shift from one job to another,]!] but we also note, as discussed in Preamble Section X.D., that
we expect very small employment impacts from the standards.
We disagree with API and AFPM that the standards are "unjustified," and with MFC
that these standards are "unnecessary;" see Preamble Section VIII and RIA Chapter 8 for an
assessment of the benefits of the standards and the benefit-cost analysis, which show
significantly positive net benefits. In sum, as earlier explained in Chapter 2.1 of this document,
we believe that the final rule implements the Tier 3 program in a manner consistent with our
legal obligations, with sound science, and with sound environmental, energy, and economic
policy, and we disagree with comments suggesting that the Tier 3 rule is not justified. The Tier
3 program, including the gasoline sulfur standards, will reduce ambient levels of air pollution
that endanger public health and welfare and will provide important benefits to the public such
as preventing PM- and ozone-related premature deaths. The Tier 3 Preamble very clearly
explains the need for the Tier 3 standards in Section II, our technical justification for the
vehicle emission controls in Section IV, and technical justification for the fuel standards in
Section V. A vast body of underlying technical analyses supporting the Tier 3 standards can be
found in our Regulatory Impact Analysis, which reflects the best methods, data and
assumptions available at the time of the rulemaking analysis. We thus disagree that
investments to meet the standards are an inefficient use of resources.
[11 E.g., Jacobson, Louis S., Robert J. LaLonde, and Daniel G. Sullivan, "Earnings Losses of Displaced
Workers." American Economic Review 83(4) (1993): 685-709.
9.3.1. Employment Impacts in the Auto Sector
What Commenters Said:
Commenter: BlueGreen Alliance
The automotive industry has added more than 200,000 jobs since 2009, and the U.S. auto
market has had three consecutive years of double digit growth due in large part to rising
demand for these cleaner cars in the market. 'Gearing Up,' our report last year by the BlueGreen
Alliance and the American Council for an Energy Efficient Economy, estimated that meeting
the 54.5 mile per gallon rule will create more than 500,000 jobs through the U.S. economy by
2030. Building cleaner, more efficient cars creates jobs by sending money otherwise spent on
9-16
-------
Tier 3 Summary and Analysis of Comments
fuel back into the U.S. economy, and also through the development and production of new,
more efficient vehicle components.
Americans are working in a variety of industrial sectors in the Midwest and across the country.
Members of the auto workers and steel workers will manufacture vehicle emission control
technologies to meet these standards. As the BlueGreen Alliance's Regional Program Manager
for Illinois and Indiana, I am frequently in contact with people working in the automotive
industry.
Indiana ranks third in the country and Illinois seventh for jobs in the automotive parts industry,
and they are also both in the top 10 for States with jobs in automotive assembly. The Tier 3
standard will bolster the auto industry's ability to meet a strong fuel efficiency standard and
generate net positive economic outcomes.
Commenter: Business for Innovative Climate and Energy Policy (BICEP)
Additional benefits and jobs in the auto and emission control industries, which are significant
drivers of the economy, will result from the development and production of Tier 3 technology.
Commenter: California Air Resources Board (CARB)
Additionally, we believe that given the magnitude of compliance costs relative to the size of the
domestic economy, U.S. EPA's more qualitative approach for estimating employment effects
using the Morgenstern et al. framework instead of relying on economy-wide modeling is
reasonable.
Commenter: Ceres Investor Network on Climate Risk
Additional benefits and jobs in the auto and emission control industries, which are significant
drivers of the economy, will result from the development and production of Tier 3 technology.
Commenter: Emissions Control Technology Association (ECTA)
Although Navigant did not estimate the impact of the rule on the auto and parts industries, we
disagree with EPA's conclusion that the associated employment effect would be negligible. We
draw two conclusions from the REVIS model. The model shows that for every $1 million
investment in new automotive parts, the REVIS multipliers imply 5.9 jobs (14).
We believe that these positive short-term employment effects are quite important as a matter of
policy. The U.S. economy is currently operating under its full potential as witnessed by the
high unemployment rate of 7.6 percent, as well as testimony by the Congressional Budget
Office (CBO) that the rate will stay above 7.5 percent through 2014. CBO also stated in its
testimony that economic output will stay below its potential level until 2017 (15). The
simulative effect of new investment is most pronounced during periods of less-than-full
employment. In the case of Tier 3, labor and other resources that are currently underutilized
9-17
-------
Chapter 9: Economic Impact Analysis
will be utilized by the refining and automobile makers as they comply with the new rule. This
will increase economic activity and employment at a time when the nation desperately needs it.
14-
http://www.cityofVentura.net/files/file/commdevelop/Economic%20Development/Economic%20Inform
ation-Data%20Book/18-Regional%20Input-Output%20Modeling%20System.pdf
15 - CBO Testimony, CBO Budget and Economic Outlook, February 15, 2013,
http: //www .cbo .gov/publication/43 94.
Commenter: Sierra Club (comment campaign)
The new standards will also create jobs as automakers apply advanced technologies and
refineries upgrade their equipment.
Commenter: Sierra Club
The new standards will not only reduce air pollution, they will create jobs as auto part suppliers
manufacturer the technologies that reduce vehicle emissions. Auto makers apply advanced
vehicle technologies and refineries upgrade their equipment. This is a win-win for public
health and the economy, and EPA should move forward to finalize these standards without
delay.
Commenter: United Automobile Workers
These proposed rules are part of a comprehensive approach to creating the next generation of
clean vehicles. They go hand-in-hand with the new CAFE standards that were passed to
regulate the fuel efficiency of our nation's vehicles over the next twelve years. Automakers are
investing billions of dollars in retooling and expanding to produce the cleaner cars that
consumers are demanding. Manufacturing advanced vehicles and their fuel-efficient
components here in the United States is a crucial opportunity to develop domestic jobs that are
sustainable in the long-term.
Commenter: Wespath Investment Management
It is also anticipated that the development and production of Tier 3 technology will result in
additional jobs in the auto and emission control industries.
Our Response:
All these comments suggest that these standards will increase employment in the auto
sector. EPA believes that this outcome is possible, though we do not expect a large effect,
because the standards are not expected to have a large effect on auto production or costs.
ECTA argues that job impacts will be not only positive in the auto and auto parts sectors, but
substantially more positive than EPA states, based on Regional Input-Output Modeling System
(REVIS) multipliers that find increases of 5.9 jobs for every $1 million investment in new auto
parts. Multiplier impacts trace the entire chain of job impacts associated with an expenditure,
including factors such as changes in employment in retail establishments due to changes in
9-18
-------
Tier 3 Summary and Analysis of Comments
workers in the auto sector. Because the expenditures will occur when unemployment is still
high, ECTA argues, it is important to consider the multiplier effects of the expenditures that are
embedded in the RIMS multipliers. As discussed in Preamble Section X.D, the multiplier
impacts of expenditures depend heavily on the state of the macroeconomy. Because of
uncertainty over the state of the economy when the increased expenditures will occur, EPA has
not quantified the multiplier impacts due to changes in employment associated with these
standards. As discussed again below, some expenditures, perhaps especially in the refinery
sector, will occur before the standards take effect, to get the manufacturing processes in place
to meet the standards. These near-term expenditures may have some multiplier effects, because
they are more likely to have their impacts during the current period of above-average
unemployment. In contrast, the ongoing costs of complying with the standards, which
contribute most directly to price increases, are likely to come in future years, when it is
expected that unemployment rates will be lower.
9.3.2. Petroleum Industry Employment Impacts
What Commenters Said:
Commenters: Attorney General of Connecticut, et.al; Attorney General of New York, et.al;
BlueGreen Alliance; Business for Innovative Climate and Energy Policy
(BICEP); Ceres Investor Network on Climate Risk; Clean Air Council; League
of Conservation Voters (comment campaign); Sierra Club; Wespath Investment
Management
The refinery upgrades necessary to meet the lower sulfur gasoline standards are expected to
generate 24,500 installation jobs during the first three years of the program and 5,300
permanent jobs.
Commenter: Emissions Control Technology Association (ECTA)
Tier 3 Is Likely to Have a Positive Impact on Employment and the Economy: We disagree with
EPA's conclusion that the proposed rule would have an ambiguous effect on employment in the
refining sector. We find that the employment effect will be positive and significant.
EPA concludes that:
"...it is not evident whether the proposed rule would increase or decrease employment in the
refining sector. However, given the small anticipated reduction in quantities sold, it appears
that the rule would not have major employment consequences for [the refining] sector" (7).
We disagree with this conclusion and believe that the employment impact on the refining
industry will be positive and potentially quite significant. It could result in the creation of
24,500 jobs (i.e., full time equivalents or FTEs) over the three years during the installation of
desulfurization equipment and 5,300 jobs for the operation and maintenance of the equipment
over the longer term. We draw this conclusion from a study by Navigant Economics on the job
impact of Tier 3 on the refining industry (8).
9-19
-------
Chapter 9: Economic Impact Analysis
We agree with EPA's use of the "demand effect," the "cost effect," and the "factor shift effect"
in analyzing the impact of the rule on employment in the refining and auto sectors (9). But, we
draw a sharply different conclusion when applying this analytical approach to the refining
sector.
First, EPA concludes that the demand effect is expected to be negative; that is, the cost of Tier
3 to refiners will be passed on to consumers in the form of higher gasoline prices, which will
reduce the demand for gasoline and hence output by the refining sector. As explained in detail
in Section I of this submission, we find that the increased cost to refiners is unlikely to be
passed on to consumers; accordingly, we do not expect any negative demand effect arising
from the rule. Even if a portion of the rather small costs to refiners were passed along, the
demand effect would be minimal because the demand for gasoline is relatively inelastic as
consumers have no close substitutes for gasoline.
Second, EPA concludes that in the short-term the "cost effect" is likely to be positive; that is,
the increased investment by the refining industry to comply with Tier 3 will generate increased
employment over the short-run. We agree with this conclusion, but we find that the cost effect
will be much larger than EPA's estimate.
EPA expects that refiners will invest $2.2 billion between 2014 and 2019, which will increase
employment by 7,000 jobs for design, engineering, and construction of the desulfurization
equipment (10). We conclude from the Navigant study that the refining industry's initial
investment of $3.9 billion (relating to installation expenditures) will create 24,500 FTEs for the
first three years of implementation. The Navigant estimate is generated through the use of the
REVIS II input-output model developed and maintained by the Bureau of Economic Analysis. In
addition to a significant positive employment effect, this model also projects that the increased
investment by the refining industry will generate over $2 billion in economic value-added per
year over three years, $1.2 billion annually in increased employment compensation, and $0.5
billion per year in tax revenue for federal, state, and local government (11).
We also conclude from the Navigant study that a further 5,300 jobs will be created from the
$0.5 billion per year in operation and maintenance costs associated with the desulfurization
equipment after it is installed. This will generate $0.6 billion per year in economic value added,
$0.3 billion in employment compensation annually, and $0.1 billion in increased tax revenue
per year (12). EPA gave no consideration to the economic impact of the expenditures that are
related to the operation and maintenance of the new equipment.
Finally, like EPA, we did not analyze the "factor shift effect"; that is, the changes in labor
intensity associated with a regulation that drives demand for new technology that is more or
less labor intensive than the "old technology." Because Tier 3 would not require refineries to
deploy a new technology— instead, the same technology used to reduce sulfur would be
intensified—we think the concept of factor shift effect is not applicable here.
EPA concludes:
9-20
-------
Tier 3 Summary and Analysis of Comments
"...while we do not have an estimate of the direction of the overall effect of the proposed rule
on auto-sector employment, we expect it to be imperceptible (13).
7 - EPA Proposal, Fed. Reg. at p. 29992.
8 - Navigant Study at pp. 25-27.
9 - EPA Proposal, Fed. Reg. at p. 29991.
10 - EPA Proposal, Fed. Reg. at p. 29999.
11 - Navigant Study, pp. 25-27 and footnote 42.
12 - Navigant Study, pp. 25-27.
13 - EPA Proposal, Fed. Reg. at 29991.
Commenter: National Association of Clean Air Agencies (NACAA)
Tier 3 will also provide new employment opportunities. In its draft RIA, EPA projects that the
work refineries will need to undertake to satisfy the Tier 3 requirements will create about 1,000
front-end design and engineering jobs and 6,000 construction jobs for a total of 7,000 new jobs.
The agency notes that the petroleum sector employed about 65,000 workers in 2009. Therefore,
the new jobs to be created as a result of Tier 3 would increase employment in the petroleum
sector by over 10 percent compared to 2009 levels.
Our Response:
Most of these comments cite Navigant Economics for positive and substantial job
impacts; NACAA cites EPA's estimates from the NPRM.
As discussed in Chapter 9.3.1 of this Summary and Analysis of Comments document,
EPA is not using multiplier effects for its estimates of employment impacts here, as does the
Navigant Economics study, because of uncertainty over the state of the macroeconomy at the
time that these expenditures will occur. As discussed in Chapter 9.2 of this document, we do
not consider our finding of a small demand or output effect to contradict Navigant Economies'
finding of no demand effect; it is possible that the effect of the standards on price, and thus on
fuel sales, is too small to measure.
EPA has revised its employment estimates for refineries for this final rule. As
discussed in Preamble Sections V.L.3.e and X.D.2, to meet the Tier 3 sulfur standards, EPA
now estimates that refiners are expected to invest $2 billion between 2012 and 2019 and utilize
approximately 250 front-end design and engineering jobs and 1,500 construction jobs. As the
petroleum sector employed approximately 71,000 workers in 2011, this temporary increase in
employment will be small when compared to 2011 levels.
Commenters: American Petroleum Institute (API) and the Association of Fuel &
Petrochemical Manufacturers (AFPM); Marathon Petroleum Company LP
(MFC)
Refining sector jobs:
We recommend that EPA removes the reference and benefits from 1,600 "jobs gained in the
refinery sector". First, it should be pointed out that these are temporary constructions jobs and
9-21
-------
Chapter 9: Economic Impact Analysis
not permanent jobs gained in the refinery sector. Second, this estimate was based on a single
study by Morgenstern et al. "Jobs versus the environment: an industry level perspective. J. Env.
Econ. andManag. 43:412-436." EP A used a figure of 2.17 jobs created per M$ of
expenditures (1987 dollars scaled to 1.289 G$ costs in 2017.) The applicability of the
Morgenstern study, which focuses on jobs associated with refinery emission reductions, to Tier
3 gasoline sulfur reduction is questionable.
Commenter: Chevron Products Company
EPA does not account for the economic hardship and resulting impact on local and regional
economies incurred as a result of the above average cost of production for some of the
refineries. This analysis also alludes to increases in employment in the refining sector resulting
from past regulations without any reference to the economic context of that time or the global
competitive forces. This leads the reader to conclusions without any fact or basis. EPA makes
the argument that 'While closing refineries has a negative effect on industry employment, it is
likely that the increased refining capacity at many of the remaining plants had a positive effect
on industry employment.' Increases in production may not result in any significant increases in
employment to offset the impact of closing refineries.' If throughput is increased without the
addition of new equipment trains to manage, the labor needed to keep the refinery running
would be about the same.
Our Response:
In the NPRM, EPA presented the Morgenstern et al.-based estimate, not as a conclusive
estimate, but rather as one estimate that we too consider of unclear applicability to the
standards. As discussed in Section X.D of this final rule, we are now using estimates from
Berman and Bui ("Environmental Regulation and Labor Demand: Evidence from the South
Coast Air Basin." Journal of Public Economics 79(2) (2001): 265-295) that "employment
effects are very small, generally positive, but not significantly different from zero." As
discussed above, to meet the Tier 3 sulfur standards, EPA now estimates that refiners are
expected to invest $2 billion between 2012 and 2019 and utilize approximately 250 front-end
design and engineering jobs and 1,500 construction jobs. As the petroleum sector employed
approximately 71,000 workers in 2011, this temporary increase in employment will be small
when compared to 2011 levels.
Chevron Products Company states that we do not account for the hardship on local and
regional economies due to the increase in refinery costs. Because we do not consider
employment impacts to be large, even at local and regional scale, we do not expect large
impacts on the economies at that scale. Chevron also states that our employment analysis does
not account for economic context or global competitive forces. These are both considered in
Preamble Section X.D and RIA Chapter 9.3. As discussed there, we recognize that
employment impacts depend on the overall state of the macroeconomy. We also discuss the
effects of Tier 3 in a global context: Canadian and Caribbean refineries are expected to have
costs similar to those of U.S. refineries, while European refineries are already producing
gasoline to a 10-ppm sulfur cap for Europe. EPA and Chevron appear to agree that minor
changes in production are not expected to lead to significant changes in employment.
9-22
-------
Tier 3 Summary and Analysis of Comments
10. Statutory Process and Executive Orders
What We Proposed:
The comments in this chapter correspond to Section XII of the preamble to the proposed
rule and address statutory process and executive orders as they relate to the Tier 3 rule. A
summary of the comments received and our responses to those comments are located below (and
in Section XII of the preamble to the final rule).
10.1. Administrative Procedure
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM)
EPA is pursuing a flawed process in developing this rule. It is outrageous for EPA to hold public
hearings as the docket continues to be populated with supporting information. It is invalid for
EPA to hold these hearings in advance of publishing the rule in the Federal Register.
EPA must follow the Clean Air Act's procedural requirements. Section 307 sets clear
requirements for the rulemaking process that begins with EPA first publishing a proposed rule in
the Federal Register. In this publication, EPA shall specify the comment period. EPA cannot set
a deadline for public comments on a proposal that has not yet been published, and the
supplemental publication that merely identifies the hearing dates and locations is a clear attempt
to circumvent public participation in the rulemaking process.
Clean Air Act Section 307 also requires EPA must include with the Federal Register publication
factual data on which the proposed rule is based, the methodology used in obtaining the data and
in analyzing the data, and the major legal interpretations and policy considerations underlying
the proposed rule. In short, this means all the information EPA uses as the basis for the rule must
be available. As of last week, over 700 new documents had been added to the docket, leaving not
time to analyze the information to provide meaningful input.
Once the proposed rule is published, the comment deadline needs to be revised. Another public
hearing should be scheduled, and it must be held long enough after publication to allow for
public review of all of EPA's data and analysis. The availability of a pre-publication rule cannot
be used as a reason to truncate the review period.
Commenter: Chevron Products Company
Chevron does not agree with the notice and comment process that EPA followed for publishing
the Tier 3 NPRM. The delayed publication of the NPRM and the accelerated timeline for review
and comment establishes a poor precedent for future rulemakings. A proposal of this importance
10-1
-------
Chapter 10: Statutory Process and Executive Orders
deserves to be presented with a transparent administrative process and sufficient time for review
and comment.
Regarding the administrative process for the Tier 3 NPRM, Chevron is disappointed by the
procedure that EPA has followed for publishing the proposal and soliciting comments. The
NPRM was published in the Federal Register on May 21, 2013, and EPA established a comment
submittal deadline of July 1, 2013. This is a 41 day comment period for a proposal that
comprises several thousand pages of regulatory text, analysis, test data, and various other docket
submissions. A proposal of this length and complexity deserves to be presented with a
transparent administrative process and sufficient time for review and comment.
We understand from prior discussion with EPA that the Tier 3 NPRM has been under
development and was largely completed for perhaps a year or more prior to its release. There has
been ample time to inform the public of the proposal and conduct the notice and comment
process without resorting to an unrealistic, accelerated schedule. We are concerned by the
precedent that this establishes for future rulemakings. EPA owes it to all Tier 3 stakeholders and
the American public to conduct these proceedings with the highest regard for transparency,
fairness, and due diligence.
Commenter: Chrysler Group LLC
EPA's Procedural Obligations Under the Clean Air Act:
Chrysler observes that EPA failed to comply with its procedural obligations under the Clean Air
Act (CAA) in issuing the proposed rule. Specifically, CAA section 307(d)(3) states that a "notice
of proposed rulemaking [must] be published in the Federal Register" and must "specify the
period available for public comment" (40). The CAA further requires EPA to "ensure a
reasonable period for public participation of at least 30 days" (41). Here, EPA published a
Federal Register notice on April 8, 2013, announcing not the rule, but that public hearings for the
proposed Tier 3 rule would be conducted on April 24, 2013 and April 29, 2013. Without having
published the actual rule, that federal register notice also announced that the comment period for
the proposed Tier 3 rulemaking would end on June 13, 2013. Curiously, then, the proposed Tier
3 rule itself was not published in the Federal Register until May 21, 2013, less than 30 days prior
to June 13, 2013, and EPA did not "refresh" the comment period in the proposal in order to
comply with the CAA.
EPA cannot, consistent with the CAA, establish a public comment period prior to publishing the
actual proposed rule on which the public is to participate/comment, especially where the
resultant comment period does not comport with the CAA. As such, EPA's initial decision to
close the public comment period on June 13, 2013—24 days after the proposed rule was
issued—violated the CAA's minimum 30-day public comment requirement. Only after EPA
received correspondence pointing out these procedural flaws did the agency extend the comment
deadline to July 1, 2013. Chrysler commends EPA for extending the comment period and
complying with the statutory obligation and urges the Agency to refrain from taking such an
approach in future notice-and-comment rulemaking actions.
40-42U.S.C. §7607(d)(3).
-------
Tier 3 Summary and Analysis of Comments
41 -42U.S.C. §7607(h).
Commenter: Marathon Petroleum Company LP (MFC)
EPA has also taken several questionable steps in the process that it has followed for this
proposed rulemaking. Most notably, EPA conducted field hearings for this proposal prior to
official publication of the proposal in the Federal Register. EPA also failed to provide the
affected stakeholders with sufficient time to comment on a rulemaking of this magnitude. And
on June 25, 2013 (less than one week before comments were due) the Agency was still adding
large supporting documents to the docket.
As a result of the overall lack of justification for these fuel changes and the significant time
needed for review of EPA documents, MFC asserts that finalizing this proposal would be
arbitrary and capricious and unlawful. EPA should either drop the current proposal or restart the
proposal process in a manner that allows all stakeholders to have sufficient time to analyze and
understand all the documents and issues involved in this complex rulemaking and to make
appropriate comments. EPA must strictly re-examine the purported benefits of this rule and its
real costs.
My comments today aren't really on the rule - the proposal itself, but the process that EPA is
using in this rulemaking, and the fact that they've created a very unusual and unacceptably
shortened comment period, which won't allow adequate public comment on the rule. And it's a
rule that has significant impact on both my industry and the consumer.
It's taken EPA at least two years to get to this point, and the administrative record right now
reflects a very complex set of documents consisting of thousands of pages of text. EPA has
chosen to hold only two field hearings on this proposal, one today and one last week, and unless
they actually printed the proposal in the Federal Register today, we haven't got an official copy
of the proposal yet.
The two field hearings are occurring way too early, and simply stated, EPA has provided the
stakeholders with insufficient time to analyze the overwhelming quantity of data and documents
offered in support of the Agency's proposed action. We are still searching the docket trying to
find several key documents that EPA references in the draft Regulatory Impact Analysis, and we
have no assurance that the proposal we're working from is going to be the final proposal since
each page has got a memo attached to it that says this is not an official version.
The public and the regulated community should be allowed at least the typical 90 days to review
and comment on any proposal, especially this proposal. And the EPA shouldn't shorten the
process that it has historically used in developing the proposed rules.
Affected parties deserve and should be granted adequate time to review the hundreds of
documents that the Agency has placed in the docket in the last two weeks, and those that EPA is
still placing in the docket. According to EPA - or according to us, we recommend that EPA
schedule a third hearing that's a minimum of 30 days after the publication of the Proposed
Notice of Rulemaking in the Federal Register, and after all the documents are put in the docket.
10-3
-------
Chapter 10: Statutory Process and Executive Orders
This would allow all the interested parties to have time to review the entire proposal and
documentation before combining their comments.
Additionally, the comment period should not close sooner than 90 days after the rule is published
in the Federal Register and all the documents are put in the docket. This is a significant rule, and
it warrants and complete and thoughtful input from all the affected parties.
We in our industry have gone to great lengths to try to understand the implications for our
industry and for consumers in lowering sulfur levels to 10 ppm. As of last week, despite
requests over the last two-year period, the Agency still hasn't given us sufficient justification for
lowering sulfur levels. We're now facing a forced march to comment based on a pre-publication
announcement that simply references sources of justification that have just started to be placed in
the docket.
This absence of information and a truncated comment period basically fly in the face of the
Administration's promise to have regulatory transparency. Proceeding in this fashion may even
be unlawful under the Administrative Procedures Act and the Clean Air Act.
In summary, it's not reasonable, fair, or lawful for EPA to expect us to come up with an
intelligible set of comments in such a short period of time to the complicated and incomplete
regulatory package that EPA has provided. We ask that EPA proceed fairly, lawfully, and issue
the entire proposal, all the background documents, publish the proposal in the Federal Register,
and put all the other documents in the docket, and all stakeholders 30 days to prepare for a public
hearing, and 90 days to enter all their comments.
Commenter: Environmental Defense Fund (EDF)
EPA's Public Participation Procedures for the Tier 3 Proposal Met All Legal Requirements:
In a letter to EPA dated May 7, 2013, API alleged that the Agency did not follow the Clean Air
Act's procedural requirements in proposing Tier 3. As EPA set forth in its May 23, 2013
response to API, however, the Agency's Tier 3 procedures met all legal requirements, including
requirements for publication of the proposed rule in the Federal Register and public participation
opportunities such as the period for written and oral comments. As EPA noted, the Tier 3
proposal was prominently posted and available for public review upon signature, on March 29,
2013. In addition, EPA widely disseminated notice of the proposal's availability at that time and
specifically notified stakeholders, including API and the American Fuel & Petrochemical
Manufacturers. With the extension of the comment period until July 1, 2013, this means that
commenters have had, as a practical matter, over 90 days to review the proposed rulemaking. In
addition, EPA provided notice in the Federal Register on April 8, 2013 of the location of the
docket, the dates for the public hearings, and the duration of the comment period. In summary,
EDF believes that the Agency met all legal requirements in proposing Tier 3, and strongly urges
the Agency to move expeditiously to finalize the Tier 3 standards by December 31, 2013.
Commenter: Union of Concerned Scientists (UCS)
-------
Tier 3 Summary and Analysis of Comments
Finally, the procedures that EPA followed for the Tier 3 proposal met, and exceeded, all legal
requirements, including requirements in the Clean Air Act for publication of the proposed rule in
the Federal Register and public participation activities such as the public comment period and the
public hearings. The rule was prominently posted and available for public review when it was
signed on March 29th. With comments due on July 1st, commenters will have had more than 90
days to comment on the rule, including more than 30 days after the proposal was published in the
Federal Register.
Commenter: Flint Hills Resources, LP (FHR)
Proposed Rules for Fuel Used in Flexible-Fuel Vehicles Should Be Published in the Federal
Register:
EPA's preamble to the proposed Tier 3 regulations references a separate memorandum (1) that
includes additional proposed regulations for which EPA seeks comments. These additional
regulations include prescribing new requirements for E51- E83 for key fuel properties such as
sulfur, RVP, and benzene. While FHR may support EPA's effort to develop regulations for E51-
E83, the process EPA used to inform the regulated community runs afoul of the federal
Administrative Procedures Act (APA) (2) and the federal Clean Air Act (CAA).
The APA is clear that EPA's forum for informing the public of substantive rules, amendments,
and revisions is through the Federal Register (3). Further, the APA states 'except to the extent
that a person has actual and timely notice of the terms thereof, a person may not in any manner
be required to resort to, or be adversely affected by, a matter required to be published in the
Federal Register and not so published.'
Section 307(d)(3) of the CAA echoes the requirement for proposed rules to be published in the
Federal Register. Among other things, the notice must 'specify the period available for public
comment.' In addition, at the time of publication, there must be a publicly available docket that
includes: (1) the factual data on which the proposal is based; (2) an explanation of how EPA
collected and analyzed the data; and (3) an explanation of the major legal interpretations and
policy considerations underlying the proposal.
In the instant matter, EPA describes and seeks comment on very significant potential changes to
the regulations. EPA is required, by law, to publish such a proposal in the Federal Register.
Posting on the EPA website or including within the rulemaking docket is not an option —
publication in the Federal Register means publication in the Federal Register. EPA must include
all proposed changes in the proposed rule. References to documents external to the proposed rule
do not give the regulated community and public the opportunity to which it is entitled to generate
informed and timely comments regarding the proposed rule.
1- Herzog, J. (January 2012). Possible Approach to Fuel Quality Standards for Fuels Used in Flexible-
Fuel Automotive Spark-Ignition Vehicles (FFVs), Memorandum to the docket. The memorandum is
referenced in footnote 388 of the pre-publication proposed rule; however, the footnote reference is to a
memorandum dated January 2012 but the memorandum in the docket is dated April 8, 2013. Further, this
memorandum was not in the docket at the time the proposed rules were announced on EPA's website.
2 - United States Code, Title 5, Chapter 5
10-5
-------
Chapter 10: Statutory Process and Executive Orders
Our Response:
With regard to the public comment period, EPA met all legal requirements, including
requirements in the Clean Air Act for publication of the proposed rule and notice of hearings and
the public comment period in the Federal Register. EPA provided early notice of the Tier 3
proposal by posting it on EPA's website upon signature (March 29, 2013), and the docket was
opened early to allow the public access to materials in the docket. Further, EPA held two public
hearings for the Tier 3 rulemaking, and also provided an extension of the comment period. All
of these items were announced in Federal Register notices—and posted on EPA's website—
going beyond the procedural steps required by law, to promote transparency and public
participation.
As explained in Section V.H of the preamble to the final rule, we are deferring final action
on our proposed requirements for E51-83 at this time. We appreciate the valuable input provided
by commenters and will take this into consideration in our final actions on these standards. We
will continue to work with stakeholders in developing in-use fuel quality standards for higher
level ethanol blends. We may issue a supplementary proposal prior to issuing a final rule if the
additional information we receive from stakeholders warrants such an action.
10.2. Other Executive Orders
Environmental Justice
What Commenters Said:
Commenter: Concerned Citizens Around Murphy
The U.S. Environmental Protection Agency has made good changes to the gasoline we use in our
vehicles.
One of the first changes was to mandate unleaded gasoline, which has provided a healthier
environment for us and our children. Then came low-sulfur gasoline, which also introduced
ethanol into the product. Then, the EPA required a low-benzene gasoline. The Tier 3 proposal
will require reductions once again in the sulfur content of gasoline. The reduction of volatile
organic compounds and sulfur from our vehicles is an important goal to which we should all do
our part to be part of the solution, but it should not come at the cost of the little good air that is
left in our neighborhoods around refineries.
The air quality in fenceline neighborhoods can be so poor at times that residents shelter
themselves in place or hold their noses when they are on the roads that pass through the refinery
facilities.
Clean fuel shouldn't cost air quality. The EPA should do more to control emissions from the
refineries that make the lower sulfur content gasoline. Lowering vehicle emissions is important,
but too much emphasis has been placed on achieving that goal through manufacturing and
-------
Tier 3 Summary and Analysis of Comments
refining of lower sulfur content fuels and not enough enforcement and compliance at the
refineries that make the gasoline.
EPA's Tier 3 proposal fails to consider that the reductions at the tailpipe will be offset by the by
the increased emissions in the neighborhoods around those refineries that do not simultaneously
install upgraded pollution controls.
EPA's Tier 3 proposal fails to consider the increased emissions at the processing plants (the
refineries). At a recent St Bernard Parish Quality of Life Commission meeting, representatives
from Louisiana DEQ made a presentation on the non-attainment status for the 2010 sulfur
dioxide standard. They were asked how the anticipated State SIP for SO2 attainment would
address Tier 3 mandates. Officials were asked if the refineries lower sulfur in gasoline (for EPA
Tier 3 mandates), if they take more SO2 out of the gasoline, where will it go? Louisiana DEQ
responded to assure residents that 'certainly we will not let it go into the air." Yet, neither EPA
nor LDEQ have the authority to require zero sulfur emissions from Tier 3 projects. That sulfur is
going to go somewhere. That somewhere is our neighborhoods. Whether it happens from
everyday processing emissions or from switching to a higher content sulfur crude oil or sand oil
or from emergency flaring or shutdown and startup emissions, the sulfur emissions at the plants
will offset the reductions at the tailpipes.
While EPA's proposed Tier 3 is said to prevent up to 2,400 premature deaths, 23,000 cases of
childhood respiratory ailments, and 1.8 million lost school days, work days, and days when
activities would be restricted due to air pollution, the proposal does not account for the health, air
quality, and quality of life in the refinery fenceline neighborhoods. Those sacrifice zones, areas
that are overburdened and underserved, where residents are mainly low income people of color,
will be traded for improvements in other areas that do not have refineries. Air quality in
fenceline communities will be sacrificed for lower emissions at the tailpipe.
While EPA's proposed Tier 3 is said to prevent up to 2,400 premature deaths, 23,000 cases of
childhood respiratory ailments, and 1.8 million lost school days, work days, and days when
activities would be restricted due to air pollution, the proposal does not account for the health, air
quality, and quality of life in the refinery fenceline neighborhoods. Those sacrifice zones, areas
that are overburdened and underserved, where residents are mainly low income people of color,
will be traded for improvements in other areas that do not have refineries. Air quality in
fenceline communities will be sacrificed for lower emissions at the tailpipe.
We urge EPA to reconsider how the emissions at the refmery(s) will offset the emissions from
the tailpipes. We urge EPA to require simultaneous installation of state-of-the-art control
technology at the refineries that manufacture Tier 3 gasoline products. Our health is depending
on it.
Commenter: Mercatus Center at George Mason University
While the EPA estimates that that this regulation will have a minimal impact (approximately one
penny) on fuel costs, others have estimated the impact to be significantly greater, perhaps as
much as 9 cents per gallon (18). Exactly how much gasoline prices may rise is difficult to say,
10-7
-------
Chapter 10: Statutory Process and Executive Orders
but given that energy costs consume a higher proportion of low-income individuals' income
relative to high-income individuals, it is worth acknowledging that this regulation may impose a
disproportionate impact on low-income individuals. Similarly, recent academic research has
shown that regulations, such as the proposed regulation, are often more in line with risk
preferences of wealthy households (19). It is unlikely that poor households are worried about the
risks posed from PM2.5 and ozone when they face much larger risks elsewhere in their lives. The
income they lose to comply with this regulation may be better utilized toward other risk
mitigation.
18 - David Tamm and Kevin Milburn, "Addendum to Potential Supply and Cost Impacts of Lower Sulfur,
Lower RVP Gasoline," (American Petroleum Institute, Houston, TX, March 2012).
19 - Diana Thomas, "Regressive Effects of Regulation." (Working Paper No. 12-35, Mercatus Center at
George Mason University, Arlington, VA, November 2012), http://mercatus.org/publication/regressive-
effects-regulation.
Commenter: WE ACT for Environmental Justice
Thank you for the opportunity to provide comments on the proposed Tier 3 Motor Vehicle
Emission and Fuel Standards. My name is Dr. Jalonne L. White-Newsome and I am a federal
policy analyst with WE ACT for Environmental Justice, a 25-year old environmental justice
organization with offices in Harlem, New York and Washington, DC. WE ACT's mission is to
build healthy communities by assuring that people of color and/or low income participate
meaningfully in the creation of sound and fair environmental health and protection policies and
practices. The inception of our organization, in fact, was fueled by the battle for clean, healthy
air in Northern Manhattan. In an effort to protect and address the health concerns related to air
emissions from a sewage treatment plant that was plopped into the Harlem community, as well
as the 6 diesel bus depots, 4 garbage truck depots and many other sources of pollution... WE ACT
continues to educate, engage community residents, academic and health partners to demand that
the protection of human health and welfare be afforded to all - despite the hue of your skin, the
amount of money and resources in your pockets, and the zip code you happen to live in. Since
1988, we have had the opportunity to be that voice, not only in Harlem, but on the national stage
to help influence the conversation around several domains of environmental work, with clean air
always being a priority concern.
I'm sure you know that the residents of Harlem, NY and many other similar cities and towns
across our great nation are not breathing the same air - some of clean, and some is pretty dirty.
So as many organizations across the country are wrapping up their Earth Day celebrations, just
yesterday, I was reminded of why we do the work we do.
I went to pick up my 4 and 6 year old from school yesterday, going through the usual motions of
signing out my girls, saying HI to other Moms and Dads, when I noticed something different this
time. As I peered behind the secretary's desk, a cute little girl, with two little pig-tails and
beautiful bright eyes, caught my attention - first because she reminded me of my little one, but,
more importantly in addition to her school uniform, she was wearing a breathing mask,
connected to a small machine, receiving a breathing treatment. I think the secretary saw me
looking with a concerned look on my face and I gave the little girl a slight smile. I walked to my
car, thinking about this young lady, and the many other children that I know that have to receive
-------
Tier 3 Summary and Analysis of Comments
breathing treatments, or take medications, miss school, etc. to survive and try and live a kid's
life. Due to no fault of their own - other than playing outside, going to school and living - their
quality of life is compromised because of the state of the air they breathe. A child.
And then I thought about some of the senior citizens I have worked with, specifically in Detroit,
Michigan, a part of Wayne County and one of the areas discussed in the rule documentation and
I thought about my great Aunt, who is in her mid-80s that suffers with severe asthma, constantly
having to be hospitalized because she has lived the majority of her life in various neighborhoods
in Detroit, my hometown - unfortunately in areas that continue to be overwhelmed with multiple
social stressors - crime, poverty, etc. - but also air toxins from both stationary and mobile
sources that continue to negatively impact her health status.
You know, cumulative risk assessment or cumulative impact assessment, however you want to
phrase it, is a way of understanding the real and total impacts on communities. We know that
environmental pollution doesn't act in silos, and that's what we really need to consider as we
look at these different rules.
When you look at EJ communities, they're hot spots. They're areas where there's a lot of things
going on, and most of the time the rules and the things that we use to frame these rules don't
capture everything that's happening. And so, what I would suggest is that, I guess, really offer a
competent suggestion is that keep in mind that having these more stringent standards will work
towards minimizing the burden on communities that are already compromised.
So if we focus on really the most vulnerable, I truly believe that more people will benefit overall,
and having the rules, like the soot rule, Tier 3, and others, work in parallel is great. And so my
suggestion, however, is that in future analyses, a cumulative impact strategy be employed.
I know that several of your colleagues in the EPA are working towards this as well as developing
an environmental justice guidance that will be incorporated throughout the Agency. But I hope
that this is something that you will strongly consider in future assessments to make sure that we
get a real glimpse of the impacts and the benefits that come from these proposed rules.
I'm sure you are aware of the Urban Air Toxics report that is supposed to be prepared for
Congress per Section 112(k) of the Clean Air Act. Briefly, the purpose of this report was to
provide an update on the strategies taken on by the EPA to address the cumulative health risks in
urban areas, specifically around urban toxics. Unfortunately, the first report was issued back in
2000 but there has not been an update since then. From what I have heard, the EPA is working
on getting this report out in the near future. I bring this report up because the Tier 3 rules
proposed will work to address - over time - some of the air toxins identified in the urban air
toxics strategy. It is so critical that we can document the levels of air toxins in our urban
neighborhoods, but also quantify the health impacts of potential carcinogens in our EJ
communities. Having the data is the only way we can understand the present levels in our most
compromised communities, and evaluate, over time, if the variety of programs and other
mitigation efforts are actually working. So I hope, that in addition to these proposed standards,
that the proper and adequate monitoring of our roadways, as well as refining the specificity of
urban air toxics monitoring and urban communities will be an issue that is addressed, so we can
10-9
-------
Chapter 10: Statutory Process and Executive Orders
truly understand the exposure and, hopefully, the decrease in urban air toxins in the communities
that are already disproportionately impacted.
Urban air toxics are a significant factor in cumulative risk assessment or 'Cumulative impacts' as
those in the environmental justice community like to capture this ideal. Because pollution -
whether it be air, waste, water - does not act in a silo. Consequently, we believe that in order to
properly characterize and environmental justice community, the SUM of the exposures must be
considered - together, and not just individual pollutants. I do believe that the word 'cumulative
impacts' was mentioned in the proposed rule. And I simply want to offer a comment and
suggestion. Keep in mind that the more stringent standards will work towards minimizing that
burden on communities that are already compromised. So, if we can focus on making life better
for the "most vulnerable", I truly believe that more people will benefit, overall. Having the rules
- soot rule, Tier 3 and others - work in parallel is great. My suggestion, however, is that in future
analysis, a cumulative impacts strategy is employed. I know several of your colleagues are
working to develop environmental justice guidance as well as guidance for preparing a
cumulative risk assessment strategy so I would encourage you to potentially look into that work
and how it can be integrated this time or in the next iteration of proposed rules.
Our Response:
Regarding the Mercatus Center comment, please see Chapter 9.2 of this Summary and
Analysis of Comments document.
Regarding the comments from Concerned Citizens Around Murphy, please see Chapter
3.3.1 of this Summary and Analysis of Comments document.
Regarding WE ACT for Environmental Justice's comments, we acknowledge the
concerns expressed and agree that these standards will help to reduce the burden on communities
affected by multiple stressors. While the air quality improvements associated with this final rule
affect the nation as a whole, locations in close proximity to major roadways will experience the
most direct changes in a way that reduces their exposures. With regard to the comment urging
analysis of cumulative impacts, EPA notes that the quantitative air quality analyses used to
evaluate the health impacts of this rule include multiple air pollutants, including PM2.5, ozone,
and air toxics. Furthermore, the RIA reviews available literature on the health effects associated
with living or attending school in close proximity of traffic. In describing how this rule affects
environmental justice, the RIA summarizes studies that evaluate the cumulative impacts of air
pollution and social stressors such as neighborhood violence. EPA agrees that consideration of
the full range of impacts of the rule is important. We acknowledge the comments concerning
updating the Urban Air Toxics report and urging proper monitoring of air quality, but they are
beyond the scope of this rulemaking.
10.3. Comment Period
What Commenters Said:
-------
Tier 3 Summary and Analysis of Comments
Commenter: American Fuel & Petrochemical Manufacturers (AFPM)
As you know, the Agency released a proposal for Tier 3 motor vehicle and fuel standards on
March 29, 2013. This proposal has not yet been published in the Federal Register.
In a hearing notice published in the Federal Register on April 8, 2013, EPA announced that the
public comment period would end on June 13, 2013. The length of this comment period is
inadequate for a broad technical rulemaking such as this, especially one that includes diverse
topics of vehicle emissions standards, fuel standards and the distinct unrelated issue of
Performance Based Measurement System. Each of these issues requires different technical
expertise to analyze.
The docket was opened on April 15. There are over 600 documents now in the Tier 3 docket in
the Supporting and Related Material category. It is virtually impossible to review and analyze
this volume of supporting technical information in the time allotted for public comment.
Accordingly, we request additional time to prepare informed comments on the proposal.
AFPM requests that the public comment period be extended to 90 days after publication of the
proposal in the Federal Register.
The AFPM has requested that the public comment period for the Tier 3 rulemaking be extended
beyond June 13. There are hundreds of documents in the docket, and additional time is
necessary for a comprehensive review. AFPM recommends 90 days after publication of this
proposal in the Federal Register.
Commenter: Institute for Energy Research (IER)
The Institute for Energy Research is deeply concerned about EPA's proposed Tier 3 rulemaking.
The public comment period deadline has been set for June 13th, and yet EPA only published the
proposed rule in the Federal Register on May 21st. This allows the public less than a month's
time examine the significant economic implications of the rule. Moreover, questions remain
about the data used by EPA to derive the costs and benefits of the Tier 3 proposal, and the rule
has also included several non-germane regulations such as the El5 certification fuel. Given these
complexities and unanswered questions, we request at least a 90-day extension for the comment
period.
Closing the public comment period on June 13, 2013 is likely in violation of the Clean Air Act
that requires "notice of proposed rulemaking shall be published in the Federal Register" and that
"[the Administrator] shall ensure a reasonable period for public participation of at least 30 days."
The abbreviated period of 24 days does not give the American people sufficient time to comment
on this complex and expensive regulation, which will impact each and every American.
Third, the impact of Tier 3 will be widespread and detrimental to the U.S. economy, and as such,
the rule warrants greater consideration. A study by Baker & O'Brien Inc., prepared for the
American Petroleum Institute, estimated that Tier 3 would impose initial compliance costs to oil
refineries of $9.7 billion dollars. The ongoing costs associated with this rule would also increase
10-11
-------
Chapter 10: Statutory Process and Executive Orders
the price paid for gasoline at the pump by 6-9 cents per gallon, which does not capture what the
full "ripple effect" would be on the greater economy (6). Given the large costs, and the fact that
EPA has failed to conduct whole-economy modeling for the rule, an extension of 90 days does
not seem unreasonable.
We appreciate EPA's concern and actions to ensure legal and transparent public participation in
the rulemaking process. Strict legal compliance with rulemaking procedures as outlined in the
Clean Air Act will ensure this rule is fully evaluated and given proper consideration by all
concerned parties in an effort to identify and craft legislation that best serves the public interest
of this country. It is therefore critical that EPA extend the comment period and release all of the
data upon which it relies. Such an action would be in keeping with the President's pledge to
foster openness and transparency in the workings of the federal government, particularly when
they affect the public so profoundly.
6 - Tamm, David C. and Milburn, Kevin P. (2012) Addendum to Potential Supply and Cost Impacts of
Lower Sulfur, Lower RVP Gasoline. Houston, TX. Retrieved May 17th, 2013 from http://api.org/news-
and-media/news/newsitems/2012/mar-2012/~/media/Files/News/2012112-March/Addendum -Potential-
Impacts-of-Lower-Sulfur-Lower-RVP-Gasoline-Report.ashx
Commenter: National Propane Gas Association (NPGA)
NPGA urges the Agency to extend the comment period beyond the stated lune 13, 2013
deadline.
The current 23 day comment period (based on the May 21, 2013 publication in the Federal
Register) simply does not grant adequate time to complete a detailed analysis of this complex,
377-page proposed rule. The additional time is needed to develop the substantive comments
sought by EPA.
NPGA believes a 60-day comment period should be established, based on the Federal Register
publication date, which is standard for a proposal of this complexity. This would extend the
deadline to luly 21, 2013, and NPGA strongly urges EPA to take the action of extending the
comment period.
Our Response:
EPA's procedures for the Tier 3 rulemaking have met and exceeded all legal
requirements, including requirements in the Clean Air Act for publication of the proposed rule in
the Federal Register and opportunities for public participation. In addition to holding two public
hearings on the proposed rule, we provided early public notice of the proposal (prominently
posted on EPA's website and available for public review immediately after signature on March
29, 2013) as well as early public access to the Tier 3 rulemaking docket.
In response to requests, we also extended the comment deadline from June 13, as
originally set, to July 1, 2013 to provide the public with more time to prepare and submit
comments. With the extension, we provided over 90 days to review and submit comments on the
-------
Tier 3 Summary and Analysis of Comments
proposed rule, including more than 30 days after the May 21, 2013 publication of the proposed
rule in the Federal Register. We concluded that the extension to July 1 was sufficient in light of
the time period provided to review and submit comments on the proposal. We were also mindful
that additional delay could affect the timely finalization and implementation of the Tier 3
program and its close harmonization with the California LEV III program, the need to provide
adequate compliance lead time for regulated parties, and the importance of assisting states by
achieving emission reductions to help attain and maintain the existing National Ambient Air
Quality Standards (NAAQS).
We continue to believe that these actions complied with and went beyond the procedural
steps required by law, to promote transparency and public participation in the Tier 3 rulemaking.
10-13
-------
Tier 3 Summary and Analysis of Comments
11. Comments on Proposed Regulatory Text
What We Proposed:
The comments in this chapter correspond to the proposed regulations, and are specific
edits to the regulatory text that were suggested by commenters. As described below, many of
these regulatory comments (such as those that were editorial or otherwise very specific) do not
fit well within the structure of this document and are addressed elsewhere. This is because the
majority of the comments received on this rule were comments on the proposed policy rather
than the proposed regulatory text. Thus, the other chapters of this Summary and Analysis of
Comments document were structured to address such policy comments rather than the type of
comments addressed by this Chapter. While summaries of the regulatory comments received
and our responses to those comments are discussed briefly below, readers are referred to other
documents in the public docket for this rulemaking for a more complete analysis of these
comments.
11.1. Engine and Vehicle Testing Regulations (40 CFR Parts 86,1065, and 1066)
What Commenters Said:
We received detailed comments from manufacturers and the State of California on the
regulatory text describing the test procedures. These comments and our responses to them have
been summarized in detail in 5 memoranda to the docket,1 and are not repeated here. The
comments addressed issues such as temperature tolerances for testing, design specifications for
laboratory cooling fans, and preconditioning sample systems prior to testing, along with dozens
of similar comments.
Commenters:
Alliance of Automobile Manufacturers (Alliance) and Association of Global Automakers
(Global)
BMW
California Air Resources Board (CARB)
Emissions Control Technology Association (ECTA)
Manufacturers of Emission Controls Association (MECA)
Truck and Engine Manufacturers Association (EMA)
Volkswagen Group of America, Inc.
Volvo Car Group
11.2. Fuels Regulations (40 CFR Parts 78 and 80)
What Commenters Said:
EPA-HQ-O AR-2011-0135.
11-1
-------
Chapter 11: Comments on Proposed Regulatory Text
We received comments from several parties in the fuel industry with detailed edits on the
proposed regulatory text in 40 CFR parts 79 and 80. These edits, which are not repeated or
summarized here, addressed various areas of the regulations including suggested changes to
definitions, deadlines, and additions/deletion of text to add clarity to the regulations.
Commenters:
American Petroleum Institute (API) and the Association of Fuel & Petrochemical Manufacturers
(AFPM)
Chevron Products Company
Growth Energy
Magellan Midstream Partners, L.P.
Marathon Petroleum Company LP (MFC)
POET, LLC
Renewable Fuels Association (RFA)
Shell Oil Products for Shell and Motiva
Weaver and Tidwell LLP
Our Response:
We have responded to some of the issues raised by the edits to the fuel-related regulatory
text in Chapters 5.3, 5.5, 5.6, and 6.1 of this Summary and Analysis of Comments document.
Further, we have made changes to the regulations, where appropriate, based on these suggested
edits or new information received. A summary of our responses to these comments is contained
in a memorandum to the docket entitled "Summary of Responses to Comments Regarding
Detailed Edits of the Proposed Fuel Regulations in the Tier 3 NPRM".
11-2
-------
Tier 3 Summary and Analysis of Comments
12. Comments Unrelated to the Proposed Rule
What We Proposed:
The comments in this chapter do not necessarily correspond to a specific section of the
preamble to the proposed rule, but rather are comments that span various areas and issues. A
summary of the comments received and our responses to those comments are located below.
12.1. Boutique Fuels
Organizations: National Association of Clean Air Agencies (NACAA)
Mid-Atlantic Regional Air Management Association Inc. (MARAMA)
When promulgating this final rule, EPA should provide a clear path forward for state and local
agencies with respect to transitioning from current boutique fuel programs to the federal Tier 3
program including a streamlined SIP process and related guidance.
Our Response:
This comment appears to request that the Agency work with them to develop a
streamlined process whereby States may end their current State-run fuel programs without
interfering with their SIP compliance. The commenters appear to believe that the emission
reductions resulting from Tier 3 may obviate the need for their existing "boutique fuel programs"
which are almost all State or Local RVP control standards. The EPA would be willing to
discuss options for doing so, but this comment is otherwise outside the scope of this rulemaking.
12.2. Comments Related to the Renewable Fuels Standard
What Commenters Said:
Commenter: Pennsylvania Department of Environmental Protection (DEP)
The EPA should explore opportunities and implement program changes for relieving the burden
placed on refiners to meet the Tier 3 requirements.
A practical way to relieve the burden on refiners to meet the Tier 3 standards would be for EPA
to reduce or stop increasing the national quota for biofuel production. The EPA sets this
increasing biofuel mandate or Renewable Portfolio Standard which costs refiners resources that
refiners could use to meet the Tier 3 standards. While reduced sulfur in gasoline, which will
occur as a result of Tier 3, allows for better catalytic converter performance for reducing NOx
emissions, the annually increasing Renewable Fuel Portfolio standard, which relies heavily on
ethanol use, increases NOx emissions from automobiles. Given the huge health benefits
associated with lowering NOx emissions and ozone levels as stated in this proposed rulemaking,
12-1
-------
Chapter 12: Comments Unrelated to the Tier 3 Rulemaking
ail EPA priority should be to reduce NOx emissions from all programs, and correspondingly,
ozone levels across the nation.
Commenter: Clean Fuels Development Coalition (CFDC)
We also applaud EPA for recognizing its key role in the implementation of the federal
Renewable Fuel Standard. EPA identified in its Regulatory Impact Analysis that accompanied
the final rule governing the RFS, there are substantial benefits that can result from full
implementation of the RFS and specific and quantifiable positive health impacts. Because of our
broad base of membership and interests as well as our longevity in working on these issues, both
our organizations regularly communicate with other stakeholders and like-minded organizations.
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM); Marathon Petroleum Company LP (MFC)
RFS2 Q&A 7.8 (diesel blendstock RVO issue) (44) suggests that any diesel blendstock or
heating oil that meets the qualities of MVNRLM diesel should be included in an obligated party's
obligated volume. This is in direct contradiction to § 80.1407 (e) and (f), in Preamble IIF 2 (75
FR 14720 and 14721) which state that diesel fuel that is designated as heating oil, jet fuel, or any
designation other than MVNRLM or a subcategory of MVNRLM will not be subject to the
applicable percentage standard and will not be used to calculate the RVOs. EPA should strike
RFS2 Q&A 7.8 (diesel blendstock RVO issue).
The regulations specify that when a refiner designates gasoline or diesel which it produces, for
export, there is a commensurate reduction in that refiner's RFS RVO. This limitation will likely
result in obligated parties having to purchase RINs for exported gasoline and diesel since it often
happens that the party that designates product for export will be other than the refiner of that
product. EPA should allow any obligated party that "designated for export" gasoline or diesel to
reduce their RVO regardless of whether that product was so designated when it was produced.
Likewise, a company that changes the use of a fuel "designated for export" to domestic use
would incur a RVO obligation for the volume that the use designation was changed. This ensures
industry wide volume obligations are properly accounted for and attainable. This system allows a
refiner to claim the RVO benefit without unnecessary tracking by designating product for export.
It also ensures appropriate accounting for fuels designated for export that are later used
domestically.
Our Response:
These comments are related to specific elements of the RFS program, and are thus
outside of the scope of the Tier 3 program.
The issue involving exports and RVOs is outside the scope of this proposed rulemaking.
However, it was initially addressed in the RFS Renewable Identification Number (RIN) Quality
Assurance Program proposed rule (78 FR 12158 (February 21, 2013)). It will be fully addressed
in the final rule for the RIN Quality Assurance Program which is expected to be finalized in the
coming months.
12-2
-------
Tier 3 Summary and Analysis of Comments
12.3. Vapor Pressure Reduction of In-Use Fuels
What Commenters Said:
Commenter: Ford Motor Company (Ford)
Meeting the more stringent evaporative emissions standards of Tier 3 is made possible through
the promotion of the "fuels and vehicle are a system" messaging communicated in the Tier 3
NPRM. The reduction in market fuel vapor pressure affects the entire fleet of vehicles on the
road, giving benefits immediately after introduction of the new fuel. Additionally, any gasoline
powered off-road vehicle or machinery would also immediately benefit from reduced
evaporative emissions due to an appropriate reduction in vapor pressure.
Our Response:
A reduction in the vapor pressure of in-use fuels as discussed by the commenter was not
part of the proposed or final Tier 3 evaporative emission standards.
12.4. Comments Related to Ethanol and Aromatics Control
What Commenters Said:
Commenter: POET, LLC
As noted above, ethanol cost-effectively provides a high-octane, clean-burning fuel that is
important for meeting recently-tightened CAFE and EPA greenhouse gas emissions standards.
Ethanol, including from cellulosic and other next generation biofuel sources, can dramatically
reduce GHG emissions and even further increase the environmental benefits of ethanol. In 2012
renewable fuel use reduced greenhouse gas emissions by over 33 million metric tons. (9)
Furthermore, EPA estimates that by 2022, the RFS will reduce greenhouse gas emissions by 138
million metric tons or the equivalent of taking 27 million passenger vehicles off the road. (10)
Numerous companies, including POET and DSM, have made significant investments in
cellulosic ethanol production that can provide particularly significant greenhouse gas reductions.
(11)
Regarding emissions, clean-burning ethanol is key to Tier 3 emission reduction goals for sulfur
and paniculate matter, two pollutants around which the Tier 3 rule is anchored. Ethanol contains
sulfur in de minimis amounts, so increasing ethanol blends in gasoline can help bring about the
sulfur-related benefits that Tier 3 seeks to accomplish. Ethanol can also reduce highly harmful
particulate matter emissions, particularly in next-generation engines. (5)
Clean burning ethanol is also critical to reducing air toxics, a subset of particularly harmful
substances. The Clean Air Act fuel 'air toxics' regulations in Section 202(1) specifically identify
12-3
-------
Chapter 12: Comments Unrelated to the Tier 3 Rulemaking
three pollutants including benzene, 1,3 butadiene, and formaldehyde, and Section 21 l(k)(10)
further identifies as 'toxic air pollutants' polycyclic organic matter (which includes polycyclic
aromatic hydrocarbons known as PAHs), and acetaldehyde. In contrast to gasoline, ethanol does
not contain benzene, a key air toxic. Increased ethanol use can also reduce PAHs (6) and 1,3-
butadiene. (7) The Harvard Center for Risk Analysis has found that gasoline aromatic
hydrocarbons, including benzene, lead to 'approximately 3800 predicted premature mortalities
nationwide.' (8) Clean-burning ethanol provides a ready means of reducing toxic aromatics,
including benzene and others.
Oddly, EPA in the Proposed Rule mischaracterizes (without clear explanation) ethanol as an air
toxic; EPA must avoid this mischaracterization. As noted above, ethanol is not listed as one of
the specifically enumerated air toxics of concern in Sections 202(1) or 21 l(k)(10). Furthermore,
the Agency did not include ethanol among the 177 air toxics compounds that were evaluated in
the most recent National-Scale Air Toxics Assessment. (12) Nor does the Proposed Rule provide
any reasoned grounds for its passing reference to ethanol as an air toxic.
It should also be recognized that the Tier 3 rule and the RFS are mutually-reinforcing. In
particular, EPA recognizes that the RFS 'will result in significant amounts of ethanol-blended
gasoline in the implementation timeframe of the proposed Tier 3 program.' (18) Ethanol can help
to obtain Tier 3's emissions reduction goals, while a properly-designed Tier 3 rulemaking can
help to provide a market for, and enable the deployment of, clean-burning biofuels to ensure that
Congress' biofuels use targets in the RFS are met.
In summary, clean-burning ethanol is key to Tier 3 emission reduction goals for sulfur,
particulate matter, air toxics, and other key pollutants, and improves our nation's economy and
energy security. However, to achieve these emissions reductions, appropriate regulations are
necessary for the widespread use of low-emitting, domestically-sourced biofuels.
5 - See e.g., M. Maricq et al., The Impact of Ethanol Fuel Blends on PM emissions from a Light-Duty
GDI Vehicle, 46 Aerosol Science & Technology 576 (January 2012), available at
http://www.tandfonline.com/doi/pdf/10.1080/02786826.2011.648780; see also, Robert A. Stein etal., An
Overview of the Effects of Ethanol-Gasoline Blends on SI Engine Performance, Fuel Efficiency, and
Emissions, AVL Powertrain Engineering Inc. and Ford Motor Company (April 8, 2013), regarding the
ability of increased ethanol blends to improve engine efficiency and reduce a variety of pollutants,
including particulate matter.
6 - See, e.g., M.A. Costagliola et al., Combustion efficiency and engine out emissions of a S.I. engine
fueled with alcohol/gasoline blends, Applied Energy (2012), available at
http://www.sciencedirect.comj science/article/pii/S0306261912006836.
7 - While ethanol may increase emissions of acetaldehyde and formaldehyde, due to the much higher
toxicity weighting of benzene and 1,3-butadiene, ethanol nevertheless results in significant health benefits
due to reductions in these pollutants. See Stein et al., An Overview of the Effects of Ethanol-Gasoline
Blends on SI Engine Performance, Fuel Efficiency, and Emissions, supra.
8 - See Katherine von Stackelberg, et al., Public health impacts of secondary particulate formation from
aromatic hydrocarbons in gasoline, 12 Environmental Health 19 (February 20, 2013), available at
http://www.ehjournal.net/content/12/l/19 -/abstract.
9 - Renewable Fuels Association, Battling for the Barrel: 2013 Ethanol Industry Outlook (February
2013), p.18, available athttp://ethanolrfa.org/page/-/PDFs/2013%2ORFA%200utlook.pdf?nocdn=l.
12-4
-------
Tier 3 Summary and Analysis of Comments
10 - EPA final rule, Regulation of Fuels and Fuel Additives: Changes to Renewable Fuel Standard
Program, 75 Fed. Reg. 14,670, 14,683 (March 26, 2010).
11 - POET-DSM's first commercial ethanol production facility, slated to begin operating in early 2014,
has been predicted by a third-party analysis to reduce GHGs by 111% compared to gasoline—i.e., the
cellulosic ethanol will more than offset the GHG emissions of gasoline. See Air, Inc., Lifecycle
Emissions of POET's LIBERTY Cellulosic Ethanol Plant, available at http://poet.com/media/LCA-exec-
summary.pdf These significant GHG reductions are achieved by eliminating the need for fossil fuel both
at the cellulosic plant and at an adjacent, grain-based ethanol facility.
12 - See, e.g., the list of Air Toxics in the 2005 NATA Assessment released March 11, 2011, available at
http://www.epa.govittn/atwinata2005.
18-78 Fed. Reg. 29,820.
Commenter: American Coalition for Ethanol (ACE)
The combination of the Renewable Fuel Standard (RFS), 2017-2025 GHG/CAFE standards, and
Tier 3 proposal by EPA make up the best option for substantially further reducing GHG
emissions from the transportation sector because collectively these programs will result in the
use of higher blends of ethanol. These higher ethanol blends will replace the marginal gallon of
petroleum (that has significantly higher GHG emissions) and are much more practical in the near
term than electric cars (powered by coal and natural gas) and LNG vehicles that also have
significant infrastructure issues. Taken in combination, the RFS and proposed Tier 3 standards
should help signal to oil companies and automakers that future vehicle technologies will need to
be capable of taking advantage of ethanol's clean octane and the many other benefits that
ethanol-blended fuels have to offer.
ACE believes it is proper for EPA to use the Tier 3 rule to limit sulfur and harmful gasoline-
based aromatics and for public health reasons, but we also take this opportunity to encourage the
Agency to address other public health concerns stemming from petroleum as well.
First, EPA needs to revisit its models with respect to data on emissions from today's petroleum.
We encourage the Agency to place a priority on making certain that the Community Multiscale
Air Quality (CMAQ) model recognizes the dominant role gasoline aromatics play in urban
particulate matter (PM2.5) secondary aerosol (SOA) emissions. A recent Harvard study confirms
that EPA's models under-predict SO A emissions by at least a factor of two to four times, and
that they are responsible for tens of billions of dollars worth of avoidable public health impacts.2
Second, EPA should be more transparent about the air and water pollution health risks from
highly toxic polycyclic aromatic hydrocarbons (PAHs), which the Agency acknowledges in its
regulatory impact analysis, but for some reason does not currently regulate for mobile sources.
Recent studies by automakers show that midlevel blends of ethanol (such as E30) are able to
reduce these particle-borne PAH emissions by 45 — 60 percent or more, and we encourage the
Agency to include these findings in the final rule.3
Finally, it is critically important that EPA recognize it unfairly penalizes ethanol's field-to-wheel
carbon footprint. We encourage the Agency to correct deficiencies in the GHG rule regarding
ethanol's significantly smaller carbon footprint compared to gasoline. Doing so would help
provide more sufficient incentives for automakers to manufacture additional FFVs and future
12-5
-------
Chapter 12: Comments Unrelated to the Tier 3 Rulemaking
vehicle technologies which can take greater advantage of the high and clean octane benefits of
midlevel ethanol blends.
Fully valuing the impacts of petroleum and advantages of higher ethanol blends as discussed
above will significantly underscore the benefit in EPA's implementation of Tier 3, CAFE-GHG,
and the RFS and further motivate more rapid expansion of higher octane ethanol blends in the
United States.
2 Katharine von Stackelberg, et al., 12 Public Health Impacts of Secondary Particulate Formation from
Aromatic Hydrocarbons in Gasoline, Environmental Health (2013), available at
http://www.ehjournal.net/content/12/l/19.
3 See Koichiro Aikawa, et al., Development of a Predictive Model for Gasoline Vehicle Particulate Matter
Emissions, SAE International (Oct. 25, 2010); MA. Costagliola, et al., Combustion Efficiency and
Engine Out Emissions of a S.I. Engine Fueled with Alcohol/Gasoline Blends, Applied Energy (2012); M.
Matti Maricq, et al., The Impact of Ethanol Fuel Blends on PM Emissions from a Light-Duty GDI
Vehicle, 4 6 Aerosol Sci. & Tech. 57 (2012).
Commenter: American Council on Renewable Energy (ACORE)
There is a major policy option that will significantly reduce our dependence on oil, and improve
the nation's health and economy. In Section 202 of the 1990 Clean Air Act Amendments, EPA
was required to reduce aromatics in gasoline to the maximum extent possible. EPA designed
questionable models (and still relies on these out-of-date computer models). Consequently the
EPA determined the most cost-effective approach, and then established a 25% ceiling on
benzene, toluene, and xylene and continues to reduce the benzene level, with little mention of
other chemicals in the aromatics group. It soon became clear that aromatic levels could be
appreciably lowered, providing important economic and human health benefits; and at the same
time boosting the market share in the light-duty transportation sector for sustainable alcohols.
But, EPA has not taken action, even though there is a plethora of information to fully justify this
action. This information is available to the Committee, and in some cases has already been or
being supplied by several forces, specifically now in the Coalition/Urban Air Initiative
(EFC/UAI) comments.
Commenter: Clean Fuels Development Coalition (CFDC)
In particular we had an opportunity to review and contribute to comments submitted to you on
this Rule by the Urban Air Initiative and the Energy Future Coalition. CFDC and 25 x 25 have
been studying the impacts of aromatic hydrocarbons for several years now and completely agree
with the UAI/EFC comments with regard to calling on EPA to recognize and quantify the public
health benefits that can be achieved by reducing toxic aromatic compounds in gasoline. The UAI
comments and supporting documentation does, in our view provide an excellent case for
reducing the use of aromatics in transportation fuels and importantly ties it to numerous other
public policy initiatives EPA is grappling with ranging from overall CO2 greenhouse gas
emissions to black carbon and particulates. Therefore, we urge EPA to focus on the UAII/EFC
comments as they are representative of the views of our organizations as well.
12-6
-------
Tier 3 Summary and Analysis of Comments
The other huge impediment to getting more high-ethanol capable vehicles on the road was
created through the Corporate Average Fuel Economy/Greenhouse Gas Rule last year. In that
Rule vehicles operating on blends even as high 85% receive a paltry 1% GHG Credit due to
outdated and inaccurate modeling. We believe ethanol's GHG advantages over petroleum are
severely under estimated and when combined with some of these other initiatives could provide a
pathway to increased biofuel use, successfully implementing the RFS.
Commenter: E.I. du Pont de Nemours and Company
In enacting the RFS, the Congress set a forward looking 15-year policy intended to incent the
large scale development of renewable fuels in the U.S. It succeeded ahead of schedule for grain
ethanol, and has to date met its overall volume targets successfully. The existing suite of biofuels
policies, of which the RFS is a significant component, has been very successful in standing up a
U.S. biofuels industry that is making a material contribution to U.S. energy security and reducing
the environmental footprint of transportation.
Specifically, biofuels have made significant contributions to reducing GHG emissions but also
provide a low sulfur, aromatic-free, olefin-free gasoline volume which assists in meeting criteria
pollutant goals.
We have to remember that we started down the road of alternative transportation fuels because of
the variety of security, environmental and economic ramifications of our dependence on
petroleum. Those challenges have only grown more acute and we are making solid progress.
Today, biofuels production in the U.S. offsets over 10 billion gallons of petroleum demand each
year.
Commenter: Energy Future Coalition and Urban Air Initiative
Air quality and public health: Aromatic hydrocarbons—which are air toxics in their own right—
make up approximately 20%-30% of standard motor vehicle fuel in the United States. On
combustion, these compounds produce emissions of benzene, toluene, ethylbenzene, xylene, 1,3-
butadiene, polycyclic aromatic hydrocarbons (PAHs), and a host of other hazardous air
pollutants, including dangerous levels of fine and ultra-fine particulate matter (PM2.5 and
UFPs). Ethanol is a substitute for octane-enhancing aromatics; the average aromatics content in
gasoline has dropped 16 percent over the past decade due to ethanol blending. (2)
PM2.5 is widely viewed as the most lethal air pollutant in the United States. It causes thousands
of deaths every year, as well as a range of other health and environmental effects. (54) As EPA
noted in its 2007 MSAT rule, "[hjealth effects associated with short-term exposures (e.g., hours
to days) in ambient PM2.5 include mortality, increased hospital admissions, heart and lung
diseases, increased cough, adverse lower-respiratory symptoms, decrements in lung function and
changes in heart rate rhythm and other cardiac effects." (55) Studies also show associations
between long-term exposure to PM2.5 and cardiorespiratory and lung-cancer mortality. (56)
54 - See C. Arden Pope III, Epidemiology of Fine Participate Air Pollution and Human Health: Biologic
Mechanisms and Who's at Risk?, 108 Environ. Health Perspectives Supplements (2000); Pope, et al.,
12-7
-------
Chapter 12: Comments Unrelated to the Tier 3 Rulemaking
Lung Cancer, Cardiopulmonary Mortality, and Long-term Exposure to Fine Particulate Air Pollution, 287
JAMA 1132 (2002), available at http://www.epw.senate.gov/107th/Levy_l.pdf.
55 - 2007 MSAT Rule, 72 Fed. Reg. at 8443; accord Regulation of Fuels and Fuel Additives: Changes to
Renewable Fuel Standard Program, 75 Fed. Reg. 14670, 14808-09 (March 26, 2010) [hereinafter
"RFS2"] ("Health effects associated with short-term exposures (hours to days) to ambient PM include
premature mortality, aggravation of cardiovascular and lung disease (as indicated by increased hospital
admissions and emergency department visits), increased respiratory symptoms including cough and
difficulty breathing, decrements in lung function, altered heart rate rhythm, and other more subtle changes
in blood markers related to cardiovascular health.").
56 - See RFS2, 75 Fed. Reg. at 14808-09 ("Long-term exposure to PM2.5 and sulfates has also been
associated with mortality from cardiopulmonary disease and lung cancer, and effects on the respiratory
system such as reduced lung function growth or development of respiratory disease. A new analysis
shows an association between long-term PM2.5 exposure and a subclinical measure of atherosclerosis.").
A recent study led by the Harvard Center for Risk Analysis, with participation by EPA,
estimated the public health impacts from exposure to only one of these hazards—PM2.5
originating from aromatic hydrocarbons in gasoline—at approximately 3800 premature
mortalities nationwide and total social costs of $28.2 billion. (3) A straight-line extrapolation of
aromatics reduction from ethanol replacement suggests that a 30% ethanol blend would save on
the order of $9 billion annually for just this one of many public health benefits—more than the
low end of the total monetized health benefits of the Proposed Rule. (4) Of particular concern,
emission reduction technology in newer vehicles may worsen PM2.5 pollution if aromatics are
not reduced in fuel. (5) Reduction of mobile source air toxics, required by the Clean Air Act
Amendments of 1990, has been too long neglected by EPA.
Achieving reductions in PM2.5 from mobile sources would also reduce regulatory pressure on
stationary sources to find ever more costly ways to meet the National Ambient Air Quality
Standard for concentrations of PM2.5 in the air. Electric utilities already face annual costs of
$9.6 billion to reduce such hazardous air pollutants. (6)
We also recommend that EPA take this opportunity to lead a nationwide transition to cleaner fuel
in order to meet the requirements of the Clean Air Act Amendments of 1990 and ensure "the
greatest degree of [air toxic] emissions reduction achievable" from motor vehicle exhaust. (15)
Shifting to a mid-level ethanol blend is achievable with currently available technology and a
sensible plan of action; however, the regulatory regime currently governing certification fuel and
fuel marketability has thus far prevented automobile manufacturers and fuel refiners from
initiating such a change. As EPA recognized in the Proposed Rule, meaningful emissions
reductions can only come about "as part of a systems approach in addressing the impacts of
motor vehicles and fuels on air quality and public health." (16) EPA has both the power and legal
authority to craft such an approach. With time—and the removal of regulatory disincentives for
alternative fuels—the nation's motor vehicles could all be running on cleaner, safer fuel and
benefiting from the performance enhancement that a higher-octane blend would make possible.
Accordingly, we respectfully request that EPA take the following steps to facilitate a nationwide
transition to a cleaner, safer mid-level ethanol blend vehicle fuel:
- Approve a splash-blended, mid-level ethanol blend as a certification fuel;
12-8
-------
Tier 3 Summary and Analysis of Comments
- Adjust the greenhouse gas emissions calculation for all vehicles certifying on this fuel to
account for ethanol's life-cycle carbon advantage over gasoline;
- Extend to this fuel the one-pound waiver for Reid vapor pressure (RVP) that currently applies
toElO;
- Beginning with a future model year, require all new gasoline vehicles to be certified on this
fuel; and
- Consistent with the pace of adoption of both fuel and vehicles, limit the aromatic content of all
light-duty motor vehicle fuel to the greatest degree achievable.
EPA Must Reduce Mobile Source Air Toxics in Motor Vehicle Fuel: What goes into motor
vehicle fuel directly affects the emissions that come out the tailpipe. By transitioning to a mid-
level ethanol blend, EPA would fulfill its statutory duty to regulate air toxics from mobile
sources. In the 1990 Clean Air Act Amendments, Congress gave EPA the duty to "promulgate
(and from time to time revise)" regulations governing emissions of "hazardous air pollutants
from motor vehicles and motor vehicle fuels." (25) Furthermore, Congress required that these
regulations "reflect the greatest degree of emission reduction achievable through the application
of technology which will be available" (26); in other words, the mandate is strict and technology-
forcing. "Hazardous air pollutants," as that term is used in section 202(1) of the Act, are
synonymous with "mobile source air toxics" (MSATs) and include the aromatic compounds
which currently make up approximately 20%-30% of light-duty motor vehicle fuel in the United
States. On combustion, these aromatic hydrocarbons—which are air toxics in their own right—
produce benzene, toluene, ethylbenzene, xylene, 1,3-butadiene, polycyclic aromatic
hydrocarbons (PAHs), and a host of other hazardous air pollutants. (27) In addition, the
combustion of aromatics in motor vehicle engines produces dangerous levels of fine and ultra-
fine particulate matter (PM2.5 and UFPs), causing a range of environmental and human health
effects, including thousands of deaths every year. (28) Finally, the combustion of aromatics in
motor vehicle engines produces emissions of black carbon, one of the most powerful agents of
climate change. Perhaps counterintuitively, in gasoline direct injection (GDI) engines (which are
likely to become standard equipment in response to stricter fuel economy and greenhouse gas
standards in the recent CAFE rule), emissions of black carbon and ultra-fine particles (UFPs)
will increase if the level of aromatics in gasoline is not reduced. (29)
EPA Has Not Met its Obligation to Regulate MSATs: EPA has acknowledged that the
"hazardous air pollutants" that must be regulated under section 202(1) of the Clean Air Act
include aromatics, yet, apart from requiring a small decrease in benzene content in 2007, (30) the
Agency has not yet substantially reduced aromatics in motor vehicle fuel. In a study required by
section 202(1)(1), EPA discussed aromatics, including "benzene, formaldehyde, and 1,3-
butadiene," as well as the "gasoline particulate" that they produce. (31) In its 2001 MS AT Rule,
the Agency included in its list of MSATs the aromatics benzene, toluene, xylene, 1,3-butadiene,
naphthalene, and polycyclic organic matter (POM), a subclass of air toxics that includes all
PAHs. (32) Nevertheless, EPA did not require any reductions in emissions beyond those already
required by other existing or proposed mobile source regulations. (33) In its 2007 MSAT rule,
EPA required a slight reduction in benzene content in vehicle fuel, but—based on several factual
predicates that have since changed, (34) the Agency declined to regulate aromatics further. (35)
12-9
-------
Chapter 12: Comments Unrelated to the Tier 3 Rulemaking
Today, EPA has an opportunity to correct this deficiency and reduce the dangers of motor
vehicle exhaust to public health. As noted in the Proposed Rule, more than 158 million
Americans are currently experiencing unhealthy levels of air pollution, which are linked with
respiratory and cardiovascular problems and other adverse health impacts that lead to increased
medication use, hospital admissions, emergency department visits, and premature mortality.
Motor vehicles are a particularly important source of exposure to air pollution, especially in
urban areas. (36)
Because current motor vehicle fuel is a major contributor to air toxics and particulate matter in
the atmosphere, (37) and because benzene, toluene, ethylbenzene, and xylene (BTEX) (the class
of aromatics added in the highest quantities to motor vehicle fuel) all produce benzene on
combustion, EPA has the authority, and indeed the obligation, to reduce the aromatic content of
motor vehicle fuel under sections 202(1) and 211 of the Clean Air Act. EPA must promulgate
regulations that "reflect the greatest degree of emissions reduction achievable through the
application of technology which will be available, taking into consideration the availability and
costs of the technology, and noise, energy, and safety factors, and lead time." (38) While
declining to regulate aromatics in 2007 because of cost and other factual predicates that are no
longer applicable, the Agency acknowledged that "[tjhere may be compelling reasons to consider
aromatics control in the future, especially regarding reduction in secondary PM2.5 emissions, to
the extent that evidence supports a role for aromatics in secondary PM2.5 formation." (39) The
Agency now acknowledges the important role of aromatics in PM2.5 formation, (40) and the
time for cleaner gasoline has come; indeed, it is long overdue. EPA leadership to initiate a
nationwide switch to a cleaner, less aromatics-intensive, mid-level ethanol blend would
constitute a great stride toward a cleaner energy future and produce widespread environmental
and health benefits.
MSATs Have a Significant Impact on Human and Environmental Health, Especially in Urban
Areas: Regulation of gasoline aromatics under Clean Air Act sections 211 and 202(1) is urgently
needed because aromatics cause thousands of deaths each year, along with a host of other serious
health and environmental problems. All of these problems would be greatly ameliorated through
the use of a cleaner, mid-level ethanol blend.
Benzene, toluene, ethylbenzene, and xylene, known collectively as "BTEX," are all aromatic
compounds included in gasoline to increase its octane rating. These air toxics, which belong to a
class known as volatile organic compounds (VOCs), are emitted in tailpipe exhaust. Recent
studies show that the majority of BTEX (83%-100%) on and near roadways is the result of
gasoline tailpipe emissions. (41) As a result, BTEX from gasoline exhaust has significant health
and developmental implications, especially for urban populations (42)—pertinent to EPA's
environmental justice criterion in the Proposed Rule. (43)
On their own, VOCs cause many harmful effects on the human nervous system and can
contaminate soil and groundwater with lasting effects. (44) Benzene, for example, is a known
human carcinogen and causes a range of health effects, including blood disorders and
immunotoxicity. (45) Furthermore, once in the ambient air, VOCs combine through complex
chemical pathways to produce significant amounts of fine particle pollution, including
particularly dangerous ultra-fine particles (UFPs), which can become coated by toxic polycyclic
12-10
-------
Tier 3 Summary and Analysis of Comments
aromatic compounds (PAHs), allowing these harmful cytotoxic and mutagenic chemicals to
penetrate deep inside the human body. Each of these dangerous forms of pollution caused by
aromatics is discussed below.
Fine Particle Pollution (PM2.5): Aromatics in gasoline are among the most efficient
anthropogenic precursors of secondary organic aerosol (SOA), (46) a species of fine particulate
matter (PM2.5)—i.e., particles less than 2.5 micrometers in diameter. It is widely acknowledged
that EPA's current models do not adequately capture the SOA-forming potential of gasoline
aromatics, (47) nor do motor vehicle emissions tests capture the formation of SOA. (48)
However, aromatics' SOA formation potential is of deadly import, because SOA is a major
contributor to the PM2.5 burden throughout the United States, particularly in urban areas. (49) In
addition to SOA, gasoline aromatics produce direct PM in the form of black carbon and PAHs,
both discussed below. (50) In fact, the entire PM2.5 formation potential of gasoline comes from
aromatics. (51) Ethanol, by contrast, does not produce PM2.5, as EPA's own investigations have
concluded. (52) Thus, a reduction in the aromatic content of motor vehicle fuel would
significantly reduce PM2.5 levels in urban areas and near roadways. (53)
Because PM2.5 is particularly lethal, power plants and other stationary sources are strictly
regulated to reduce emissions, and States must meet a National Ambient Air Quality Standard
(NAAQS) that limits concentrations of PM2.5 in the air.
Nevertheless, States are currently unable to obtain routine State Implementation Plan credit
toward meeting the NAAQS by reducing SOA from motor vehicles, (57) and EPA's own PM2.5
regulations focus mainly on stationary sources such as power plants. As a result, costly
reductions are required in remote areas, while relatively little is done to reduce PM2.5—
particularly SOA—from motor vehicles, even in densely populated areas, where the benefits of
reducing mobile source PM would be significantly higher on a per-ton basis. (58) This anomaly
is particularly perverse considering that, unlike stationary source reductions, which cost billions
of dollars per year to implement, (59) reducing SOA from motor vehicle fuel could save
consumers billions of dollars per year at the pump. (60) Focusing exclusively on stationary
sources thus places an unnecessary burden on state regulators, who must require unnecessarily
costly ways to reduce PM pollution in order to meet their NAAQS obligations. This regulatory
disparity between mobile and stationary sources effectively grants refiners a "hidden subsidy,"
since aromatics' considerable health costs are both borne directly by affected citizens, taxpayers,
and health providers and transferred indirectly to other businesses by state environmental
authorities—even though aromatics could be significantly reduced by ethanol at far lower cost
than other means of control.
When it comes to fine parti culate pollution, size and type matter. Because of their small size,
ultra-fine particles (UFPs)—that is, particles less than 100 nanometers in diameter—are an
especially dangerous byproduct of high-distillate aromatics, (61) including naphthalene, in
gasoline. Unlike larger particles, which can sometimes deposit in the mouth and throat after
inhalation, UFPs are likely to travel deep into the alveolar region of the lungs, (62) which allows
them to function as vehicles on which dangerous toxic compounds such as PAHs (63) (also
present in large quantities in gasoline exhaust as a result of aromatics combustion) can enter
bodily tissues and wreak havoc on a cellular level. (64) Once in the alveolar tissue, UFP-borne
12-11
-------
Chapter 12: Comments Unrelated to the Tier 3 Rulemaking
PAHs (and the participate matter itself) can be absorbed into the bloodstream and enter other
bodily organs, such as the liver. (65) UFPs in gasoline exhaust exhibit a relatively high number
of these so-called "redox-active" compounds. Paired with greater bioavailability and lung
retention than other forms of particulate, this makes UFPs highly pathogenic. (66) Studies have
linked UFPs to myriad adverse health effects, including asthma and other respiratory conditions;
cardiovascular disease; preterm births; a wide range of cancers; DNA damage; even autism and
brain disorders. (67) Furthermore, the small size of UFPs means that—beyond the added danger
posed by particle type—each ton of UFPs is more dangerous than a ton of PM2.5 because UFP
pollution contains many more particles per unit mass (this property is known as "particle
number," or "PN"). (68) However, because EPA currently regulates only mass, but not particle
number or type, of parti culate pollution, (69) the Agency has yet to separately measure or
regulate these highly pathogenic UFPs. Of particular concern, emission reduction technology in
newer vehicles tends to produce more high-particle-number emissions, even as it reduces
parti culate emissions by mass. (70) Thus, the health impact of UFP pollution will only grow as
the next generation of direct injection vehicles replaces older port injection vehicles. Perhaps
counterintuitively, newer vehicles tend to emit more UFPs than legacy vehicles by as much as
fourfold. (71)
PAHs are a chemical subset of polycyclic organic matter (POM) and are a particularly
dangerous, yet largely unregulated, air toxic largely attributable to mobile sources generally, (72)
and gasoline-fueled vehicles in particular. (73) As with UFPs, PAHs are produced by aromatics
in gasoline, (74) particularly high-distillate aromatics. (75) PAHs can exist in both gaseous and
particle form. They are oxidative derivatives of gasoline aromatics and result from incomplete
combustion of fused aromatic rings. Many aromatics are themselves PAHs. Naphthalene, for
instance, is a dangerous aromatic in motor vehicle fuel—and it is the simplest of the PAHs.
However, the most harmful PAHs are the high-molecular-weight PAHs that coat UFPs. (76)
Urban UFPs disproportionately carry PAHs. (77) As EPA noted in its 2007 MSAT rule, "[rjecent
studies have found that maternal exposures to PAHs in a population of pregnant women were
associated with several adverse birth outcomes, including low birth weight and reduced length at
birth, as well as impaired cognitive development at age three." (78) Laboratory experiments
show that fetal nervous systems may be particularly sensitive to PAHs, (79) which can cross the
placenta and blood-brain barrier from mother to child. (80) PAHs have been linked to a range of
adverse health outcomes in infants and children, including developmental delay, reduced IQ,
anxiety/depression, and possible endocrine disruption. (81) Motor vehicle emissions are a major
source of PAHs found to contaminate groundwater and aquatic organisms. (82) As with other
hazardous air toxics from gasoline, the worst effects of PAHs are found in urban areas. (83)
Although PAHs and their associated health risks are better understood today than in the recent
past, EPA restates verbatim its six-year-old statement of uncertainty about their prevalence. (84)
The temperature at which a given percentage of fuel, by volume, will evaporate is an important
characteristic from a pollution perspective. Gasoline contains a wide range of components,
including aromatics of various molecular weights and chemical properties, which boil at
different temperatures. High-distillate aromatics evaporate only at very high temperatures and
often do not vaporize during gasoline combustion. (85) "T50," "T90," and "endpoint" numbers
for a given fuel refer to the temperatures at which 50%, 90%, and 100% of the fuel will
evaporate, respectively. Since compounds that do not evaporate during engine combustion are
12-12
-------
Tier 3 Summary and Analysis of Comments
often emitted from the tailpipe in the form of dangerous pollution, the more high boilers present
in a fuel blend (and thus the higher the T50, T90, and endpoint numbers of the fuel), the more
pollution the fuel will cause. If T50, T90, and endpoint numbers are reduced, on the other hand,
pollution will decrease.
High-distillate aromatics raise the T50, T90, and endpoint numbers of a fuel blend because they
evaporate only at very high temperatures. As discussed above, these "heavy" aromatics,
including naphthalene, are the primary source for dangerous UFPs and the PAHs that bind to
them. (86) It is no surprise, then, that fuel blends with higher levels of high-distillate aromatics
produce more of nearly every type of pollution. In a 2010 SAE paper, Honda reports that
predictive modeling indicated that aromatics with a high boiling point and a high double bond
equivalent (DBE) value tended to produce more PM emissions. . . . [A]ll of the additional PM is
considered a PAH . . . with a high boiling point or soot. The higher the boiling point hydrocarbon
added, the more the PM increases. This trend is particularly notable with aromatic substances.
(87)
EPA's EPAct model corroborates this observation. To produce the fuels used in the EPAct study,
modelers created new types of gasoline "blendstock" to mix with ethanol, so that the final blends
would maintain the same T50 and T90 numbers as the non-ethanol fuels used in the study.
Because ethanol has a relatively low boiling point, mixing ethanol with standard gasoline (i.e.,
"splash blending") results in final blends with significantly lower T50 and T90 numbers.
Therefore, in order to maintain the original T50 and T90 numbers, modelers had to increase the
high boiler components of the gasoline blendstock when adding ethanol—resulting in blends
with a much larger proportion of high-distillate aromatics than would be permitted in market
gasoline. (For a discussion of the significance of this methodology, see the comments at Part
IV.A.).
Adding high-distillate aromatics caused nearly every type of tailpipe pollution to increase. For
example, the EPAct study found hot-running NOx emissions to be "dominated by the effect of. .
. ethanol, followed closely by aromatics." (88) This result can only be explained by the addition
of high-distillate aromatics to compensate for ethanol's favorable effect on the T50 and T90
numbers of blended fuel. (89) Not surprisingly, these harmful aromatics increased NOx
emissions. That effect cannot be attributed to ethanol. When ethanol is simply splash-blended
into ordinary gasoline, it has the opposite effect—it lowers NOx emissions. (90)
In addition to producing traditional air pollutants, aromatics in motor vehicle fuel contribute to
global climate change.
Black carbon (BC), commonly referred to as "soot," is the "carbonaceous component of PM that
absorbs all wavelengths of solar radiation." (91) The term "BC" is often used interchangeably
with elemental carbon (EC), but the two terms refer to different measurement techniques that
capture roughly the same substance. (92) BC emissions result from incomplete engine
combustion, which is worsened by the presence of high-distillate aromatics in motor vehicle fuel.
(93) BC is considered either PM2.5 or UFP, depending on its size. Apart from being a form of
direct PM with various health and environmental effects described above, BC has an established
impact on climate change. (94)
12-13
-------
Chapter 12: Comments Unrelated to the Tier 3 Rulemaking
After CO2, BC is considered the second largest contributor to global warming, and it causes net
radiative forcing through a number of processes. (95) For instance, its "high capacity for light
absorption and its role in key atmospheric processes link it to a range of climate impacts,
including increased temperatures, accelerated ice and snow melt, and disruptions in precipitation
patterns." (96) When deposited on snow and ice, BC darkens the surface, thereby decreasing
reflectivity and increasing absorption; this raises surface temperatures and accelerates melting.
(97) BC deposits have been linked to accelerated snow and ice melting in certain areas, (98)
including Himalayan glaciers (99) and in the Western United States, where BC deposition on
mountain glaciers causes early spring melting and impacts freshwater resources. (100) EPA, in
the Regulatory Impact Assessment accompanying last year's proposed PM implementation rule,
cites research showing that "[p]er unit of mass in the atmosphere, BC can absorb a million times
more energy than CO2." (101) Because BC is such a powerful agent of climate change, and yet
is relatively short-lived, controlling its emission from fossil fuel has been identified as one of the
most effective and immediate ways of slowing global climate change. (102)
Mobile sources are "the dominant contributor to total BC emissions in the United States." (103)
Indeed, EPA estimates that mobile sources are responsible for a majority—52.3%— of BC
emissions in the USA. (104) Although heavy-duty diesel vehicles have been thought to be the
main contributor of BC emissions, recent studies show that light-duty gasoline emissions form a
very significant part of the overall BC burden and that past research seriously underestimated
this contribution. (105) In fact, recent tests performed by the California Air Resources Board
(CARB) estimate that—in stark contrast to prior understanding—EC accounts for approximately
70% of PM mass emissions from gasoline-powered light duty vehicles. (106) What is worse,
newer, gasoline direct injection (GDI) engines emit significantly more black carbon than legacy
port fuel injection (PFI) engines, (107) so the contribution of gasoline vehicles to ambient soot is
likely to significantly worsen in future years as auto manufacturers gravitate to GDI technology
in order to meet more stringent fuel economy and CO2 emissions standards in years 2017 and
beyond.
In sum, soot formed by aromatics in motor vehicle fuel has a significant impact on climate
change, putting it at direct odds with EPA's goal in the recent 2017 and Later Model Year Light
Duty Vehicle Greenhouse Gas Emissions and Corporate Average Fuel Economy Standards (108)
and with the Administration's strong commitment to slowing and preventing global climate
change. As noted above, black carbon, like all gasoline PM, is attributable solely to the aromatic
content of the fuel. (109)
Beyond black carbon, current motor vehicle fuel negatively impacts climate change in several
additional ways. SOA, produced in significant quantities by gasoline aromatics, has a direct
climate forcing effect. (110) In addition, current motor vehicle fuel blends, and aromatics in
particular, produce significant greenhouse gas pollution that would be mitigated in mid-level
ethanol blends. (Ill)
Splash blending ethanol with gasoline blendstock, at a level between 20% and 45% on a
volumetric basis, would produce a high-octane, clean fuel with significantly less aromatic
content, thereby reducing numerous forms of tailpipe pollution, including BTEX, (112) UFPs,
(113) PAHs, (114) SOA, (115) and black carbon, (116) in accordance with EPA's statutory
12-14
-------
Tier 3 Summary and Analysis of Comments
MSAT obligation. (117) In addition, because of ethanol's oxygenate quality, a mid-level ethanol
blend would reduce pollution, particularly carbon monoxide (CO), beyond what could be
achieved by simple reduction of toxic aromatics. (118) Other significant co-benefits would
include increased U.S. energy security, decreased retail fuel prices, and—as discussed below—
reduced CO2 emissions on a life-cycle basis.
Ethanol is a motor vehicle fuel derived from renewable biomass, which removes carbon dioxide
from the air as it grows. As a result, tailpipe CO2 emissions from ethanol are offset by upstream
reductions. The majority of ethanol in the U.S. market is so-called "conventional biofuel,"
derived from corn starch. (119) However, the Renewable Fuel Standard (RFS) requires
significant increases in the use of advanced and cellulosic biofuels in coming years. (120)
"Advanced biofuel," often made from sugar cane, must be at least 50% less carbon-intensive
than gasoline, while ethanol qualifying as "cellulosic" must be at least 60% less carbon-
intensive.
According to EPA's indirect land use analyses, corn-based ethanol is responsible for 21% less
greenhouse gas pollution than gasoline on a life-cycle basis. (121) However, recent studies show
that EPA models may significantly underestimate the carbon sequestration effects of corn, and
one study estimates that corn ethanol is actually 48% to 59% less carbon-intensive than gasoline.
(122) Similarly, a 2007 study by Argonne National Laboratory found life-cycle greenhouse gas
reductions of 19% to 52% from the use of corn ethanol over gasoline, depending on the source of
energy used during ethanol production. (123) This result is almost certainly conservative, as a
nine-year USD A study of no-till corn (124) subsequently found that over half the increase in soil
organic carbon (SOC) (used in the Argonne study to measure the carbon-sequestration effect of
corn) occurs more than 30 cm below the surface, and that the model used in the Argonne study—
which relied on samples near the soil surface—significantly underestimated corn's carbon-
sequestration benefits. (125)
Corn is part of a small subset of plants known as C4 plants, meaning that the first product of
carbon fixation in corn is a molecule with four carbon atoms, in contrast to most other plants,
which produce a molecule having three carbon atoms (C3). The C4 pathway was discovered in
the 1960s, (126) and it represents a major evolutionary efficiency gain over the more common
C3 photosynthesis pathway. (127) Because C4 plants such as corn more efficiently absorb CO2
in photosynthesis, they sequester more greenhouse gases proportionally than the more common
C3 plants: In fact, although only 3% of plants use C4 photosynthesis, these plants are responsible
for roughly 20% to 30% of total soil-based carbon fixation. (128) Corn's carbon sequestration
characteristics give ethanol a significant environmental advantage over fossil fuels, whose
extraction and refinement result in significant carbon dioxide emissions, in addition to that
produced when gasoline is burned as fuel.
Ethanol's carbon sequestration advantage over gasoline will only grow as higher levels of next-
generation and cellulosic biofuels enter the market, in response to RFS requirements, while the
increasing share of tar sands, oil shale, and fracking technologies will result in substantially
larger carbon footprints for the same amount of gasoline.
12-15
-------
Chapter 12: Comments Unrelated to the Tier 3 Rulemaking
Finally, EPA should use its authority under sections 202 (1) and 211 to require a phase-down in
the aromatic content of all gasoline blends, thereby ensuring that a cleaner, mid-level ethanol
blend becomes widely available in the marketplace. As described above, (164) EPA is not
currently meeting its statutory obligation to reduce air toxics in motor vehicle exhaust to the
greatest extent achievable. However, no practical barrier remains, and a high-octane mid-level
ethanol blend offers a ready alternative. Meanwhile, the extensive body of credible science
detailing the serious environmental and health effects of aromatics in motor vehicle fuel makes
the problem one of deadly import. EPA should require all new gasoline vehicles to certify on this
new, cleaner fuel blend and then use its statutory authority to reduce the allowable aromatics
content in gasoline blends—thus cutting both air toxics-related pollution and greenhouse gas
pollution.
2 - Proposed Tier 3 Rule, 78 Fed. Reg. at 29909.
3 - Katherine von Stackelberg, et al., Public Health Impacts of Secondary Particulate Formation from
Aromatic Hydrocarbons in Gasoline, Environmental Health 2013 12:19, Feb. 20, 2013,
http://www.ehjournal.net/content/pdf/1476-069X-12-19.pdf.
4 - Proposed Tier 3 Rule, 78 Fed. Reg. at 29827.
5 - See infra Part III.B.3 & note 71.
6 - National Emission Standards for Hazardous Air Pollutants from Coal- and Oil-Fired Electric Utility
Steam Generating Units and Standards of Performance for Fossil-Fuel-Fired Electric Utility, Industrial -
Commercial-Institutional, and Small Industrial-Commercial-Institutional Steam Generating Units, 77 Fed.
Reg. 9306 (Feb. 16, 2012), http://www.gpo.gov/fdsys/pkg/FR-2012-02-16/pdf/2012-806.pdf.
15-42U.S.C. §7521(1)(2).
16 - Proposed Tier 3 Rule, 78 Fed. Reg. at 29995.
25 - Clean Air Act § 202(1), 42 U.S.C. § 7521(1).
26 - Id.
27 - Elsewhere, the Clean Air Act provides that regulation of emissions of "toxic air pollutants" includes
regulation of the "aromatic hydrocarbon content" of that fuel. 42 U.S.C. § 7545(k)(l)(A), (3)(A)(ii)
(pertaining to reformulated gasoline); see also 42 U.S.C. § 7547(k)(10)(c) (defining "toxic air pollutants"
to include benzene, 1,3 butadiene, and polycyclic organic matter (POM) [a term which includes PAHs
such as naphthalene]). Another section of the Act explicitly includes the aromatics benzene "including
benzene from gasoline," toluene, and xylenes-in a list of "hazardous air pollutants" subject to regulation.
Clean Air Act § 112, 49 U.S.C. § 7412 (pertaining to emissions from stationary sources).
28 - See von Stackelberg, et al., supra note 3, at 1 (estimating a "baseline" of up to 4,700 deaths per year
due to secondary organic aerosol (SOA) from gasoline, or over 6,300 deaths per year assuming 100% of
aromatic-produced SOA in urban areas is from motor vehicles. These estimates did not take into account
deaths from direct PM, the enhanced pathogenic qualities of UFPs, or PAH-related deaths); see infra Part
III.B.6.a, III.B.3, III.B.4.
29-See infra notes 71, 107.
30 - Control of Hazardous Air Pollutants from Mobile Sources, 72 Fed. Reg. 8428, 8477 (Feb. 26, 2007)
[hereinafter "2007 MSAT Rule"].
31 - Control of Emissions of Hazardous Air Pollutants from Mobile Sources, 66 Fed. Reg. 17230, 17234
(Mar. 29, 2001) [hereinafter "2001 MSAT Rule"].
32-Id. at 17235.
33 - Id.
34 - See Appendix II.
35 - 2007 MSAT Rule, 72 Fed. Reg. at 8478-79; see Appendix II for a complete discussion of the
changes in factual predicate since 2007.
36 - Proposed Tier 3 Rule, 78 Fed. Reg. at 29819.
12-16
-------
Tier 3 Summary and Analysis of Comments
37 - See California Office of Health Hazard Assessment, Summary of Scientific Meeting Held June, 2000
on Approaches to Assessing Health Impacts of Gasoline-Related Exposures in California, 5 [hereinafter
"2000 Assessment of Gasoline-Related Health Impacts"], available at
http://www.oehha.ca.gov/public_info/pdf/GasOEHHA.pdf
38-42U.S.C. §7521(1)(2).
39 - 2007 MSAT Rule, 72 Fed. Reg. at 8479.
40 - Clean Air Fine Particle Implementation Rule, 72 Fed. Reg. 20,586, 20,593 (Apr. 25, 2007)
[hereinafter "PM Implementation Rule"].
41 - See, e.g., Eric M. Fujita, et al., Concentrations of Air Toxics in Motor Vehicle-Dominated
Environments, Health Effects Institute Research. Rep. No. 156, at 2 (2011); see also 2000 Assessment of
Gasoline-Related Health Impacts, supra note 37, at 5 ("[I]t appears that light-duty gasoline vehicle
particulate matter emissions are significantly underestimated in current inventories. Light-duty gasoline
vehicles are the major source of 1,3-butadiene and BTEX.").
42 - K.P. Wyche, et al., Gas Phase Precursors to Anthropogenic Secondary Organic Aerosol: Detailed
Observations of 1,3,5-Trimethylbenzene Photooxidation, 9 Atmos. Chem. Phys. 635 (2009) ("[Aromatic
VOCs] make up as much as 40% of the total mass of anthropogenic hydrocarbon emissions in the city
environment."), available athttp://www.atmos-chem-phys.net/9/635/2009/acp-9-635-2009.pdf; see also
Frederica P. Perera, Prenatal Poly cyclic Aromatic Hydrocarbon (PAH) Exposure and Child Behavior at
Age 6-7 Years, 120 Environ. Health Perspectives 921, 921 (2012) ("Urban, minority populations in the
United States often have disproportionate exposure to air pollution and are at greater risk for adverse
health and developmental outcomes.").
43 - Proposed Tier 3 Rule, 78 Fed. Reg. at 29849.
44 - See John S. Zogorski, et al., The Quality of Our Nation's Waters: Volatile Organic Compounds in
the Nation's Ground Water and Drinking-Water Supply Wells, U.S. Department of the Interior, U.S.
Geological Survey, Circular 1292, at 8, 9 (2006), available at
http://pubs.usgs.gov/circ/circ 1292/pdf/circular 1292 .pdf.
45 - 2007 MSAT Rule, 72 Fed. Reg. at 8435 (citing EPA, Integrated Risk Information System
File:Benzene (2000), http://www.epa.gov/iris/subst/0276.htm); see generally EPA National Center for
Environmental Assessment, Rep. No. EPA/635/R-02/001F, Toxicological Review of Benzene
(Noncancer Effects) (2002), available at http://www.epa.gov/iris/toxreviews/0276tr.pdf
46 - E.Z. Nordin, et al., Secondary Organic Aerosol Formation from Gasoline Passenger Vehicle
Emissions Investigated in a Smog Chamber, 12 Atmos. Chem. & Phys. Discussions 31,725, 31,749
(2012) ("As shown in this study gasoline exhaust readily forms secondary organic aerosol with a
signature aerosol mass spectrum with similarities to the oxidized organic aerosol that commonly
dominates the OA mass spectra in and downwind urban areas. This substantiates recent claims that
gasoline SOA is a dominating source to SOA in and downwind large metropolitan areas."), available at
http://www.atmos-chem-phys-discuss.net/12/31725/2012/acpd-12-31725-2012-print.pdf; L. Hildebrandt,
et al., High Formation of Secondary Organic Aerosol from the Photo-oxidation of Toluene, 9 Atmos.
Chem. Phys. 2973, 2973 (2009), available athttp://www.atmos-chem-phys.net/9/2973/2009/acp-9-2973-
2009.pdf ("Toluene and other aromatics have long been viewed as the dominant anthropogenic secondary
organic aerosol precursors."); id. at 2984 ("The SOA yields from the photo-oxidation of toluene are
higher than previously reported values."); JA. de Gouw, et al., Sources of Particulate Matter in the
Northeastern United States in Summer: 1. Direct Emissions and Secondary Formation of Organic Matter
in Urban Plumes, 113 J. Geophys. Research D08301, at 8 (2008) ("Particulate yields [of OM] ... are ...
highest for aromatic compounds." (citation omitted)); see also id. at 1 ("Approximately 37% of the
secondary formation [of aerosol organic matter] can be accounted for by the removal of aromatic
precursors.").
47 - One recent study found that EPA's CMAQ v5.0 model underestimates gasoline's PM2.5 contribution
by a factor of 3.8. Stackelberg, et al., supra note 3, at 5; see also Kenneth S. Docherty, et al.,
Apportionment of Primary and Secondary Organic Aerosols in Southern California During the 2005
Study of Organic Aerosols in Riverside (SOAR-1), 42 Environ. Sci. Tech. 7655, 7660 (2008) ("With
12-17
-------
Chapter 12: Comments Unrelated to the Tier 3 Rulemaking
regard to SOA yields, it has been shown that model predictions based on simulation chamber yields
dramatically underestimate SOA formation in the polluted atmosphere." (citing R. Volkamer, et al.,
Secondary Organic Aerosol Formation from Anthropogenic Air Pollution: Rapid and Higher than
Expected, 33 Geophys. Research Letters 17 (2006))); Draft Regulatory Impact Analysis: Tier 3 Motor
Vehicle Emission and Fuel Standards, at 7-66 (March 2013) [hereinafter "Tier 3 DRIA"] ("Studies have
indicated that ambient OC levels may be underestimated by current model parameterizations. While the
treatment of new precursors has likely reduced the model/measurement bias, underestimates can
persist."); Q. Zhang, et al, Ubiquity and Dominance of Oxygenated Species in Organic Aerosols in
Anthropogenically-Influenced Northern Hemisphere Midlatitudes, 34 Geophys. Research Letters L13801,
at 1 (2007) ("[Secondary organic aerosols (SOA), formed by chemical transformation and condensation
of volatile and semivolatile species, are underestimated by an order of magnitude or more by current
models when applied in and downwind of urban areas/polluted regions.").
48 - See generally Nordin, et al., supra note 46, at 31726 ("[T]here is a lack of laboratory studies to
systematically investigate SOA formation in real-world exhaust.").
49 - See sources cited supra note 46; see also Nordin, et al., supra note 46, at 31726 ("Gasoline vehicles
have elevated emissions of volatile organic compounds during cold starts and idling and have recently
been pointed out as potentially the main source of anthropogenic secondary organic aerosol (SOA) in
megacities."); PM Implementation Rule, 72 Fed. Reg. at 20593 ("Aromatic compounds such as toluene,
xylene, and trimethyl benzene are considered to be the most significant anthropogenic SOA precursors
and have been estimated to be responsible for 50 to 70 percent of total SOA in some airsheds."); Wyche,
et al., supra note 42, at 635; see also Elizabeth A. Stone, et al., A Comparison of Summertime Secondary
Organic Aerosol Source Contributions at Contrasting Urban Locations, 43 Environ. Sci. Technol. 3448,
3454 (2009) (finding spatial differences in the primary and secondary sources of OC between cities and
noting that they "demonstrate the need to treat geographically distinct source regions individually in the
study of source contributions to SOA").
50 - See infra Parts III.B.6.a, III.B.4.
51 - See J.R. Odum, et al., The Atmospheric Aerosol-Forming Potential of Whole Gasoline Vapor, 276
Science 96, 96 (1997) ("[T]he atmospheric organic aerosol formation potential of whole gasoline vapor
can be accounted for solely in terms of the aromatic fraction of the fuel."), available at
http://www.unc.edu/courses/2007fall/envr/416/001/OdumScience97.pdf
52 - See Tier 3 DRIA, supra note 47, at 7-64.
53 - See, e.g., R. Bahreini, et al., Gasoline Emissions Dominate Over Diesel in Formation of Secondary
Organic Aerosol Mass, 39 Geophys. Research Letters L06805, at 1 (2012) ("[Substantial reductions of
SOA mass on local to global scales will be achieved by reducing gasoline vehicle emissions."); Michael J.
Kleeman, et al., Source Apportionment of Secondary Organic Aerosol During a Severe Photochemical
Smog Episode, 41 Atmos. Environ. 576 (2007) (finding that gasoline engines represented the greatest
anthropogenic contributor of SOA in Los Angeles during a smog event); 2000 Assessment of Gasoline-
Related Health Impacts, supra note 37, at 18 ("For some of the high molecular weight PAH, gasoline
appears to be the dominant source. Naphthalene levels are higher in gasoline than in diesel."); id. at 26
("Data from the Northern Front Range Study showed that even moderately clean vehicles emitted a lot
more or comparable or higher emission rates of both gas phase and particle phase PAHs-than diesel
vehicles."); Shang Liu, et al., Secondary Organic Aerosol Formation from Fossil Fuel Sources Contribute
Majority of Summertime Organic Mass at Bakersfield, 117 J. Geophys. Research DOOV26, at 1 (2012)
("In urban areas, the major source [of the organic fraction of atmospheric particles] is fossil fuel
combustion from gasoline- and diesel-powered vehicles and other industrial activities (e.g., oil
burning)."), available at http://onlinelibrary.wiley.com/doi/10.1029/2012JDO18170/pdf
57 - In order to obtain SIP credit for mobile source reductions, States would have to undertake costly
studies to, in effect, prove to EPA that credit is warranted, and even then credit would not be ensured. See
PM Implementation Rule, 72 Fed. Reg. at 20594.
58 - EPA, Environmental Benefits Mapping and Analysis Program (BenMAP): RSM-based Benefit Per
Ton Estimates, (estimating the national average benefit in per ton reduction of PM at $630,000 for mobile
12-18
-------
Tier 3 Summary and Analysis of Comments
sources and only $520,000 for point sources such as EGUs),
http://www.epa.gov/airquality/benmap/bpt.html; see also Neal Fann, et al., The Influence of Location,
Source, and Emission Type in Estimates of the Human Health Benefits of Reducing a Ton of Air
Pollution, 2 Air Qual. Atmos. Health 169, 170, 173-74 (2009) (finding that "the size of the population
exposed to PM2.5 and the susceptibility of that population to adverse health outcomes" is an important
factor in calculating the benefit of pollution reduction, and that "area source and mobile source [as
opposed to 'point source VEGU] carbonaceous particle emissions, in particular, show the highest [benefit
per ton]—suggesting that the emissions and population centers exposed are colocated").
59 - See, e.g., National Emission Standards for Hazardous Air Pollutants from Coal- and Oil-Fired
Electric Utility Steam Generating Units and Standards of Performance for Fossil-Fuel-Fired Electric
Utility, Industrial-Commercial-Institutional, and Small Industrial-Commercial-Institutional Steam
Generating Units, 77 Fed. Reg. 9304, 9305-06 & table 2 n.b (Feb. 16, 2012) [hereinafter "Utility
MACT"], (estimating annual social costs of $9.6 billion to implement, with over 90% of monetized
benefits coming from reduction in premature fatalities due to PM), http://www.gpo.gov/fdsys/pkg/FR-
2012-02-16/pdf/2012-806.pdf
60 - A recent study at Louisiana State University found that "each additional billion gallons of ethanol
reduces gasoline prices as much as $0.06 per gallon. Considering U.S. ethanol production in 2010, which
was more than 13 billion gallons, ethanol production could result in having lower gasoline prices of up to
$0.78 per gallon." Hassan Marzoughi, The Impact of Ethanol Production on the U.S. Gasoline Market 16
(presented Feb. 4-7, 2012),
http: //www .ethanol. org/pdf/contentmgmt/The_Impact_of_Ethanol_Production_on_the_U S_Gasoline_Ma
rket.pdf Likewise, a study at Iowa State University suggests "that increases in ethanol production over
the last decade have reduced overall blended fuel prices." Notice of Decision Regarding Requests for a
Waiver of the Renewable Fuel Standard, 77 Fed. Reg. 70752, 70766 (Nov. 27, 2012) [hereinafter RFS
Waiver Denial] (citing Xiaodong Du & Dermot J. Hayes, The Impact of Ethanol Production on U.S. and
Regional Gasoline Markets: An Update to 2012 (May 2012), available at
http://www.card.iastate .edu/publications/synopsis.aspx?id= 1166)).
61 -See infra Part III.B.5.
62 - See, e.g., Morton Lippmann & Roy E. Albert, The Effect of Particle Size on the Regional Deposition
of Inhaled Aerosols in the Human Respiratory Tract, 30 Am. Indus. Hyg. Ass'n J. 257 (1969), available at
http://legacy.library.ucsf.edU/documentStore/b/p/e/bpe89cOO/Sbpe89cOO.pdf
63 - See infra Part III.B.4.
64 - Gwami Shrestha, et al., Black Carbon's Properties and Role in the Environment: A Comprehensive
Review, 2 Sustainability 294, 307 (2010), available at http://www.mdpi.eom/2071-1050/2/l/294/pdf;
2000 Assessment of Gasoline-Related Health Impacts, supra note 37, at 4 (Because of their small size,
UFPs "can serve as an important vehicle for carrying other toxicants into the lung.").
65 - Shrestha, et al., supra note 64, at 307-08; 2000 Assessment of Gasoline-Related Health Impacts,
supra note 37, at 4.
66 - Jesus A. Araujo & Andre E. Nel, Particulate Matter and Atherosclerosis: Role of Particle Size,
Composition and Oxidative Stress, 6 Particle & Fibre Toxicology (2009), available at
http://www.particleandfibretoxicology.eom/content/6/l/24/ref (also noting that "[i]t will be important to
determine whether all these factors confer greater toxicity to UFP in human subjects since they may imply
the need for adjusting the metrics of exposure to take account of particle number, surface area, and
oxidant potential").
67 - Lee Hotz, The Hidden Toll of Traffic Jams, The Wall Street. Journal, Nov. 8, 2011, at Dl, available
at http://online.wsj.com/article/SB10001424052970203733504577024000381790904 .html.
68 - See Araujo & Nel, supra note 66 ("[P]articles 85-90% of the total PM2.5 particle number. Therefore,
it is conceivable that larger particle numbers in UFP atmospheres, despite a smaller mass, could result in
larger biological effects. In our study, development of larger atherosclerotic lesions in the UFP exposures
correlated with increased particle numbers rather than with PM mass." (footnote call omitted)); Zhi Ning
& Constantinos Sioutas, Atmospheric Processes Influencing Aerosols Generated by Combustion and the
12-19
-------
Chapter 12: Comments Unrelated to the Tier 3 Rulemaking
Inference of Their Impact on Public Exposure: A Review, 10 Aerosol & Air Quality 43, 54-55 (2010)
("The emission of these smaller particles in the atmosphere may pose a greater threat to public health,
since they deposit deeper in the human respiratory systems and their chemical composition appears to be
intrinsically more toxic than the non-labile PM. . . . Given the increased toxicity of these semi-volatile
species [e.g., PAHs], efforts should be made to reduce their emissions from newer vehicles, including
reductions in their gas-phase precursors formed during the combustion process."), available at
http ://aaqr .org /VOL 10_No l_February2010/6_AAQR-09-05 -IR-003 6_43 -5 8 .pdf.
69 - See 40 C.F.R. § 86.096-8(a)(l) (setting emissions standards in grams per mile).
70 - See Ning & Sioutas, supra note 68, at 54 ("Advanced vehicle emission control technologies are
effective in reducing solid, non-labile PM emissions by means of filtration. However, recent
investigations have shown substantial increases (by one order of magnitude and often more) of particle
number emissions from retrofitted-vehicles due to the formation of nucleation mode particles from
organic vapors in the exhaust."); Walter Piock, et al., Strategies Toward Meeting Future Particulate
Matter Emission Requirements in Homogeneous Gasoline Direct Injection Engines, SAE International
(2011), available at http://delphi.com/pdf/techpapers/201 l-01-1212.pdf.
71 - See Health Effects Institute Special Committee on Emerging Technologies, The Future of Vehicle
Fuels and Technologies: Anticipating Health Benefits and Challenges 3 (2011) ("Because of the less
complete mixing of fuel vapor and air [in a direct-injection engine] . . ., the particulate emissions of the
engine increase, including the number of ultrafine particles."); Felix Leach, et al., The Effect of Fuel
Volatility and Aromatic Content on Particulate Emissions from GDI Engines (presented at 16th
Combustion Generated Nanoparticles, Zurich, 2012), available at
http://www.lav.ethz.ch/nanoparticle_conf/Former/Posters_I-L.pdf; Constantinos Sioutas, et al., Final
Technical Report: Physical and Chemical Characteristics of PM in the LAB (Source Receptor Study):
Topic C: Studies of the Effects of Varying Spatial and Temporal Patterns of Ambient Particulate Matter
(PM) and Co-pollutants and Resulting Health Effects with Emphasis on the Role of Atmospheric
Chemistry (Mar. 30, 2005) ("Compared to previous studies at the Caldecott tunnel, less particle mass but
more particle numbers (by factors of 2-4 fold) are emitted by vehicles than was the case 7 years ago. As
the emissions of carbonaceous PM of newer engines decreases, the formation of nucleation mode
particles is favored due to the reduction of the available surface for adsorption of the semi-volatile
material. The resulting supersaturation of the mostly organic vapor increases the production of nano-
particles by nucleation."), available at http://www.epa.gov/ncer/reports/r827352C014fr.pdf; see also infra
note 107 & accompanying text (discussing increased black carbon emission from direct injection
engines).
72 - See Control of Hazardous Air Pollutants from Mobile Sources, Regulatory Impact Analysis, at 3-112
(2007) [hereinafter "2007 MSAT RIA"] ("Major sources of PAHs include mobile sources."); Ning &
Sioutas, supra note 68, at 50 ("PAH concentrations were consistently higher when the nearby freeway
was busy with traffic during morning rush hours.").
73 - 2000 Assessment of Gasoline-Related Health Impacts, supra note 37, at 5 ("Light-duty gasoline
vehicles emit particulate matter containing higher fractions of particulate PAHs [than heavy-duty diesel
vehicles]. Light-duty gasoline vehicles are significant sources of gas phase PAHs. Cold starts, high
accelerations, and high emitters account for most of the [LDV] particulate matter emissions. . . . [I]t
appears that [LDV] particulate matter emissions are significantly underestimated in current inventories.
Light-duty gasoline vehicles are the major source of 1,3-butadiene and BTEX.").
74 - See generally id.; Memorandum from Robert Harley, Department of Civil and Environmental
Engineering, U.C. Berkeley, to Steve Brisby, Stationary Source Division, California Air Resources Board
(Apr. 26, 2007) ("[L]owering heavy aromatics in gasoline could help to reduce polycyclic aromatic
hydrocarbons (PAH), another class of toxic compounds that are not included currently in the predictive
model's definition of toxic pollutant emissions.").
75 -See infra Part III.B.5.
76 - See Yuling Jia, et al., Estimated Reduction in Cancer Risk due to PAH Exposures If Source Control
Measures during the 2008 Beijing Olympics Were Sustained, 119 Environ. Health Perspect. 815, 820
12-20
-------
Tier 3 Summary and Analysis of Comments
(2011) (noting the importance of including high-molecular-weight PAHs in pollution studies despite their
low environmental concentrations because of the disproportionately high cancer risk associated with
them), available at http://www.ncbi.nlm.nih.gov/pmc/articles/PMC31148167; see also supra Part III.B.3
(discussing UFPs).
77 - Araujo & Nel, supra note 66 ("Urban UFP contain a higher content per unit mass of polycyclic
aromatic hydrocarbons (PAH).").
78 - 72 Fed. Reg. at 8439.
79 - Perera, supra note 42, at 921 (citing LA. Brown, et al., Down-Regulation of Early lonotrophic
Glutamate Receptor Subunit Developmental Expression as a Mechanism for Observed Plasticity Deficits
Following Gestational Exposure to Benzo(a)pyrene, 28 Neurotoxicology 965 (2007); M.M. McCallister,
et al., Prenatal Exposure to Benzo(a)pyrene Impairs Later-Life Cortical Neuronal Function, 29
Neurotoxicology 946 (2008); D.D. Wormley, Environmental Contaminant-Mixture Effects on CNS
Development, Plasticity, and Behavior, 197 Toxicol. Appl. Pharmacol. 49 (2004)).
80 - Perera, supra note 42, at 921 (citing Brown, supra note 79).
81-Id.
82 - See EPA, Technical Factsheet on Polycyclic Aromatic Hydrocarbons (PAHs) ("If released to soil
[benzo(a)pyrene] will be expected to adsorb very strongly and will not be expected to leach to the
groundwater. However, its presence in some groundwater samples indicates that it can be transported
there by some mechanism."), available at http://www.epa.gov/ogwdw/pdfs/factsheets/soc/tech/pahs.pdf;
2007 MSAT Rule, 72 Fed. Reg. at 8444 ("[RJecent studies have reported gasoline and diesel vehicles as
major contributors in the atmospheric deposition of PAHs to [various watersheds].").
83 - Kanae Bekki, et al., Evaluation of Toxic Activities of Polycyclic Aromatic Hydrocarbon Derivatives
Using In Vitro Bioassay, 55 J. Health Sci. 601, 606-07 (2009) ("PAH ketones and quinines originating
from gasoline and diesel engine exhausts are present at comparatively high concentrations ... in urban
air."), available at http://dspace.lib.kanazawa-u.ac.jp/dspace/bitstream/2297/19415/l/PH-PR-BEKKI-K-
601.pdf
84 - Tier 3 DRIA, supra note 47, at 6-25 ("Trends in PAH deposition levels are difficult to discern
because of highly variable ambient air concentrations, lack of consistency in monitoring methods, and the
significant influence of local sources on deposition levels."); 2007 MSAT RIA, supra note 72, at 3-112
(same). Despite EPA's recognition of the dangers of PAHs, the Agency excludes PAHs from the
performance evaluation of its CMAQ model. See Tier 3 DRIA, supra note 47, at 7-58.
85 - See Tier 3 DRIA, supra note 47, at 3-11 ("[G]asoline containing a large fraction of heavier aromatics
compounds with high [double bond equivalent] values result in greater vehicle PM emissions.").
86 - See supra Parts III.B.3 and III.B.4; see also generally 2000 Assessment of Gasoline-Related Health
Impacts, supra note 37, at 5 ("Light-duty gasoline and heavy-duty diesel vehicles emit significant
numbers of ultrafine particles.").
87 - Koichiro Aikawa, et al., Development of a Predictive Model for Gasoline Vehicle Particulate Matter
Emissions, 3 SAE Int'l J. Fuels & Lubricants, 610, 610-11 (Oct. 25, 2010).
88 - EPAct/V2/E-89: Assessing the Effect of Five Gasoline Properties on Exhaust Emissions from Light-
Duty Vehicles Certified to Tier 2 Standards: Final Report 221, Rep. No. EPA-420-R-13-002 (2013)
[hereinafter "EPAct Results Report"], available at
http://www.epa.gov/otaq/models/moves/documents/420rl3002.pdf; but see id. at 221 (finding that
aromatics—not ethanol—"is the single most important effect [on cold-start NOx emissions] by a wide
margin"); id. (finding RVP to be the most important term for hot-start NOx emissions, but suggesting
measurement error); EPAct/V2/E-89: Assessing the Effect of Five Gasoline Properties on Exhaust
Emissions from Light-Duty Vehicles Certified to Tier 2 Standards: Final Report on Program Design and
Data Collection 76, Rep. No. EPA-420-R-13-004 (2013) [hereinafter "EPAct Design Report"], available
at http://www.epa.gov/otaq/models/moves/documents/420rl3004.pdf (noting as one "source of
measurement error and variability" that "dilute gaseous emission measurements such as NOx may have
sample concentrations similar in magnitude to background for some portions of the test").
12-21
-------
Chapter 12: Comments Unrelated to the Tier 3 Rulemaking
89 - EPA has acknowledged that the increased NOx emissions it has predicted for some ethanol blends
"dependf] on how other fuel properties change." Partial Grant of Clean Air Act Waiver Application
Submitted by Growth Energy To Increase the Allowable Ethanol Content of Gasoline to 15 Percent, 76
Fed. Reg. 4662, 4672 (Jan. 26, 2011) [hereinafter "2011 E15 Waiver"].
90 - See M. Marti Maricq, et al., The Impact of Ethanol Fuel Blends on PM Emissions from a Light-Duty
GDI Vehicle, 46 Aerosol Sci. & Tech. 580 (2011) (finding decreases in NOx emissions of "about
20%"when the ethanol content of fuel is increased from 0% to 17% or higher).
91 - California Ai
Commenter: Governors' Biofuels Coalition
EPA must ensure that its fuel models are updated to reflect new science and the realistic
emissions from the combustion of commercial gasoline (as opposed to unrealistic certification
fuels).
One of the top priorities should be for EPA to make sure that its Community Multiscale Air
Quality (CMAQ) model recognizes the dominant role gasoline aromatics play in urban PM2.5
secondary organic aerosol (SOA) emissions. A recent Harvard study has confirmed that EPA's
models under-predict SOA emissions by at least a factor of two to four times, and that they are
responsible for tens of billions of dollars worth of avoidable public health impacts.3
EPA's final rule should recognize the air and water pollution health risks from highly toxic
polycyclic aromatic hydrocarbons (PAHs), which EPA acknowledges in its regulatory impact
analysis, but for some reason does not currently regulate for mobile sources. We are surprised
and disappointed that EPA has failed to consider recent studies by automakers which show that
E30+ blends are able to reduce these particle-borne PAH emissions by 45 to 60 percent or more,
and we urge the EPA to include these findings in the final rule.4
EPA should take this opportunity to correct the deficiencies in last year's GHG-CAFE rule, and
recognize ethanol's carbon reduction benefits. Proper recognition of ethanol's lower carbon
footprint compared to gasoline — and especially gasoline aromatic compounds —would lead
EPA to incentivize automakers' post-2016 model years to manufacture E30 capable and
optimized vehicles able to take full advantage of the high octane, high performance, ultra-low
emissions of E30 blends.
A final rule that fails to improve U.S. transportation fuel standards by reducing aromatic
compounds is the wrong policy for America. On behalf of the Coalition, I respectfully urge you
to modify the proposed rule so as to provide market-based incentives and encourage the cost-
effective substitution of domestic clean octane alternatives for toxic aromatic compounds derived
largely from imported crude oil.
3 Katherine von Stackelberg, et al., 12 Public Health Impacts ofSecondary Paniculate Formation from
Aromatic Hydrocarbons in Gasoline, Environmental Health (2013), available at
http: //www .ehj ournal .net/content/12/1/19
4 See Koichiro Aikawa, et al., Development of a Predictive Model for Gasoline Vehicle Paniculate
Matter Emissions, SAE International (Oct. 25, 2010); M.A. Costagliola, et al., Combustion Efficiency and
Engine Out Emissions of a S.I. Engine Fueled with Alcohol/Gasoline Blends, Applied Energy (2012); M.
12-22
-------
Tier 3 Summary and Analysis of Comments
Matti Maricq, et al., The Impact ofEthanol Fuel Blends on PM Emissions from a Light-Duty GDI Vehicle,
46 Aerosol Sci. & Tech. 576 (2012).
Commenter: Growth Energy
While Aromatic Gasoline Components Substantially Increase PM Emissions, Ethanol Improves
Air Quality: As we noted in previous comments to the Agency, numerous studies have looked at
the impact of fuel quality on emissions from vehicles. These studies have found that the heavier
gasoline components-aromatics - substantially increase particulate matter (PM) emissions
(lizuka, M., Kirii, A., Takeda, H., Watanabe, H. (2007). Effect of Fuel Properties on Particulate
Matter Emissions from a Direct Injection Gasoline Vehicle. JSAE Technical Paper 20074414,
2007, http://www.pecj.or.jp/iapanese/overseas/asian/asia svmp 5th/pdf 5th/l5-
Masashilizuka.pdf; letter, J. (2010). Effect of Fuel Composition on PM Emissions, LEV III
Workshop, May 18, 2010, El Monte, California; Aikawa, K., Sakurai, T., letter, J.J. (2010).
Development of a Predictive Model for Gasoline Vehicle Particulate Matter Emissions. SAE
Technical Paper 2010-01-2115. DOI: 10.4271/2010-01-2115; Khalek I.A., Bougher T., letter, J.
(2010). Particle Emissions from a 2009 Gasoline Direct Injection Engine Using Different
Commercially Available Fuels. SAE Technical Paper 2010-01-2117. DOI: 10.4271/2010-01-
2117.)
Alternatively, there is substantial evidence that increased ethanol will actually reduce particulate
matter mass and number emissions (PM and PN) from the vehicle fleet. Szybist et al.,(J. Szybist,
A. Youngquist, T. Barone, J. Storey, W. Moore, M. Foster, and K. Confer, Ethanol Blends and
Engine Operating Strategy Effects on Light-Duty Spark Ignition Engine Particle Emissions,
Energy and Fuels, vol. 25, pp. 4977-4985, 2011) also summarize recent literature on how
ethanol affects production engines:
A number of investigations have examined the effect of ethanol content on particle emissions in
vehicles. Storey et al. found that blends of 10 and 20% ethanol in gasoline (E10 and E20)
decreased particle number emissions during vehicle drive cycles, with the 20% blend decreasing
particles by about 40% during the high-load US06 vehicle drive cycle. In comparison to
gasoline, He et al. found a 20% reduction in particle emissions with E20 but no change with E10.
Khalek and Bougher showed that E10 increased particle emissions compared to two different
gasoline formulations, both with higher volatility than the E10. This works showed the
importance of the hydrocarbon fraction of the E10 blend and suggests that the heavier
hydrocarbons used to control vapor pressure of E10 may also increase particulate emissions.
Aakko and Nylund found that the particle mass emissions from 85% ethanol (E85) were
comparable to those with gasoline in a PFI vehicle but that DI (direct injection) fueling with
gasoline produced particle emissions that were an order of magnitude higher.
The Szybist et al. study investigated the effects of fuel type, fueling strategy, and engine
breathing strategy on particle emissions in a flexible spark ignited engine that was designed for
optimization with ethanol. They report:
When DI fueling is used for gasoline and E20, the particle number emissions are increased by 1
to 2 orders of magnitude compared to PFI fueling, depending upon the fuel injection timing. In
12-23
-------
Chapter 12: Comments Unrelated to the Tier 3 Rulemaking
contrast, when DI fueling is used with E85, the particle number emissions remain low and
comparable to PFI fueling. Thus, by using E85, the efficiency and power advantages of DI
fueling can be gained without generating the increase in particle emissions observed with
gasoline and E20. The main finding of the study is that use of E85 results in 1 to 2 orders of
magnitude reduction in particle emissions relative to sDI (spray-guided DI) fueling with gasoline
and E20. Furthermore, sDI particle emissions with E85 are similar to that for PFI fueling with
gasoline. Thus, an increase in particle emissions beyond that of PFI engines can be prevented
while gaining the efficiency of DI engines using E85.
Storey et al., 2010 (J. Storey, T. Barone, K. Norman, and S. Lewis, Ethanol Blend Effects On
Direct Injection Spark-Ignition Gasoline Vehicle Particulate Matter Emissions, SAE publication
2010-01-2129) characterized the emissions, including PM and aldehydes, from a U.S. legal
stoichiometric direct injected spark-ignited (DISI) vehicle operating on EO, E10, and E20. The
PM emissions were characterized for mass, size, number concentration and OC-EC (organic
carbon-elemental carbon) content. The DISI particle number-size distribution curves were
similar in shape to light-duty diesel vehicles without Diesel Particle Filters, but had lower overall
particle number and mass emissions. The aggressive US06 transient cycle had much higher PM
mass emissions in comparison to the PM mass emission observed for the FTP. With respect to
added ethanol, Storey et al. concluded:
Ethanol blends reduced the PM mass and number concentration emissions for both transient and
steady-state cycles. By increasing the ethanol blend level from EO to E20, the average mass
emissions declined 30% and 42% over the FTP and US06, respectively. Measurements during
hot cycle transient operation demonstrated that E20 also lowered particle number concentrations.
The adoption of small displacement, turbocharged DISI engines into the U.S. fleet is likely to
continue in the future, and the results of this study suggest that increasing ethanol blend levels in
gasoline will lower DISI PM emissions. In addition, increasing ethanol content significantly
reduced the number concentration of 50 and 100 nm particles during gradual and wide open
throttle (WOT) accelerations.
Maricq et al., 2012 (M. Maricq, J. Szente, and K. Jahr: The Impact of Ethanol Fuel Blends on
PM Emissions from a Light Duty GDI Vehicle, Aerosol Science and Technology, 465, 576-583)
tested a light-duty truck equipped with a 3.5-L V6 gasoline turbocharged direct injection engine
that is representative of current GDI products, but contained prototype elements that allowed
changes in engine calibrations. Because PM formation in GDI engines is sensitive to a number of
operating parameters, two engine calibrations were examined to gauge the robustness of the
results. The study used four fuels: certification test gasoline (EO), a commercial E10 fuel similar
to that expected for future certification, a commercial pump grade E10, and a commercial E100
fuel used for blending. El00 and EO were splash-blended to produce El7, E32, and E45 fuels.
Maricq et al. report:
As the ethanol level in gasoline increases from 0% to 20%, there is possibly a small (<20%)
benefit in PM mass and particle number emissions, but this is within test variability. When the
ethanol content increases to >30%, there is a statistically significant 20%-45% reduction in PM
mass and number emissions observed for both engine calibrations.
12-24
-------
Tier 3 Summary and Analysis of Comments
The results reported by Zhang (Zhang et al, "A Comparison of Total Mass, Particle size
Distribution and Particle Number Emissions of Light Duty Vehicles tested at Haagen-Smit
Laboratory from 2009 to 2010," In Proceedings of 21st CRC Real World Emissions Workshop,
San Diego, CA, USA, 20-23 March 2011) are also particularly informative. In this testing, a
2008 FFV was tested on a hot Unified Cycle on E6, E35, E65, and E85. Ethanol appears to have
caused a large reduction in PM emissions (and particularly PN) from E6 to E35, with further PM
reductions as ethanol concentration increased. However, the most significant PM and PN
reductions are between E6 and E35.
EPA's analysis of EPACT data shows the possibility of slightly higher PM emissions with
increasing ethanol concentration. However, the EPACT test fleet was all Tier 2 vehicles, with
few if any vehicles with direct injection. Direct injection vehicles can experience higher baseline
PM emissions than ported fuel injected vehicles, which is the predominate fuel system
technology utilized for Tier 2 vehicles. It is likely that increasing ethanol content would reduce
PM on DI vehicles, as shown in the previously referenced studies in these comments. Also, the
fuels used in the EPACT testing program were match-blended. The higher ethanol blend used in
the EPACT study - an E20 fuel - is not likely to be representative of a higher ethanol blend that
would have higher octane. Therefore, the EPACT testing results are not applicable with respect
to PM emissions for the 2017 and later Tier 3 fleet.
Thus, there are now a substantial number of studies showing that ethanol blends of 20 percent
and higher reduce PM mass and number emissions in a variety of engines and vehicles. EPA
should review these studies in detail and give serious consideration to further regulate the
harmful components of gasoline as well as to encourage the use of additional ethanol as a means
to improve the nation's air quality.
Commenter: INEOS Bio
Aromatics and Toxics: INEOS Bio urges the EPA to reconsider the potential health effect
benefits associated with these higher ethanol blends. Ethanol is a high octane, low cost and non-
toxic alternative fuel that can be used to displace the aromatics in gasoline. These are highly
toxic chemicals that are the subject of multiple regulatory constraints under the CAA. EPA's
own discussion of air toxics makes clear that auto exhaust remains the primary source of air toxic
exposure for individuals, especially those living near roadways. In fact it is estimated that 1 in 3
people still live in areas that exceed at least one of the health-based National Ambient Air
Quality Standards (NAAQS), and residing where ozone (also known as smog) levels exceed the
federal standard.
EPA is required to set pollution standards for new light-duty vehicles under section 202 of the
Clean Air Act. They are also required to set standards at a level that protects public health and
welfare. The original CAA covered mobile source pollution standards for smog-forming, toxics,
and other emissions for decades. Following the Supreme Court's decision in Massachusetts v.
EPA that ruled that greenhouse gas emissions are pollutants under the CAA and that those gases
do endanger public health, the agency was required to set global warming pollution standards for
vehicles under the CAA. Another provision of the 1990 CAA - the mobile source air toxics
(MSAT) provision, requires EPA to reduce toxics as far as technology will permit. EPA has not
12-25
-------
Chapter 12: Comments Unrelated to the Tier 3 Rulemaking
enforced this section except to reduce benzene by only small amounts in 2007, notwithstanding
that the US is abundant in non-toxic alternatives such as ethanol. It seems that it would be
appropriate to consider enforcing this section as increased volumes of ethanol become available.
Biofuels and Their Supportive Policies Are Important to the U.S. Fuel Supply: The federal RFS,
enacted in 2005 and updated in 2007, is an important tool in achieving the objectives of energy
independence and pollution reduction. The RFS is the single most important federal policy
driving investment and commercialization of conventional and advanced biofuels. Biofuel
production under the RFS has already displaced nearly 10 percent of gasoline consumption and
will account for more than 20 percent of U.S. transportation fuel by 2022. Biofuel production
under the RFS reduced the need for imported oil by more than 462 million barrels in 2012.
Investment in biofuels, largely spurred by the RFS, has led to the development of facilities like
our 1st commercial biorefinery in Vero Beach, Florida, and KiOR's in Columbus, Mississippi,
which represent several hundred million dollars of investment in the United States and are poised
to begin production of the next generation of renewable fuel from non-food feedstocks this year.
Dozens more advanced biofuel projects are planned or under construction, illustrating the visible
success the RFS has had in driving development of highly skilled, well-paying jobs in rural
America.
The "Blend Wall": INEOS Bio firmly believes that the limits to market access for biofuels
commonly referred to collectively as the "blend wall" represent a series of barriers contrived by
obligated parties (4) to prevent biofuels from gaining access to the marketplace. (5) Multiple
avenues exist for blending additional volumes of biofuel into the nation's fuel supply. For
instance, as the proposed rule recognizes, El5 blends are approved and ready for use, and
production of flex fuel vehicles continues to increase. These options provide a suite of
opportunities for the growth of the entire biofuels industry and RFS compliance.
The Agency should encourage the development of biofuels, including E15 and higher octane
blends of ethanol, and the engines and infrastructure to support them. In addition, EPA should
allow the 1# RVP waiver to apply to El 5 to be consistent with its current proposal for ethanol
and butanol blends and that it currently allows for E10. This development can readily grow the
supply of biofuels in the market and overcome the blend wall by allowing Renewable
Identification Numbers (RINs) to truly reflect their market value. It will also help drive the
market and encourage retailers to adopt new infrastructure. INEOS Bio is supportive of the
proposed rule and believes the final Tier 3 rule has the potential to continue the progress of the
biofuels industry and help alleviate the "blend wall" as described above.
EPA should also seek to identify opportunities to grow biofuel markets, including adoption of a
higher-octane blend as the gasoline base fuel. Reconsideration of the gasoline base fuel would
enable engine manufacturers to optimize beneficial characteristics of higher ethanol blends in
engine design.
Commenter: National Corn Growers Association (NCGA)
12-26
-------
Tier 3 Summary and Analysis of Comments
We also commend the agency on its recognition of the benefits of ethanol in transportation fuel,
and its continued support for the Renewable Fuel Standard regulation, which is significantly
reducing GHG emissions and improving energy security in the United States through the use of
domestically produced ethanol.
Our Response:
The emission analyses in this rule take into account impacts of the ethanol mandate in the
Energy Independence and Security Act (EISA), on both emissions from vehicles and upstream
sources. Impacts of ethanol on criteria pollutant and toxic emissions, as well as air quality, were
previously addressed in EPA's Renewable Fuel Standard (RFS2) rule. The Tier 3 standards will
result in significant reductions in air toxics, including aromatic compounds, as indicated in
Chapter 7 of the RIA.
Several commenters urge EPA to use its authority under Section 202(1)(2) of the Clean
Air Act to reduce aromatics in gasoline. Those comments are beyond the scope of the current
rule. Further, comments related to comparison of air quality impacts of different ethanol blends
is beyond the scope of the Tier 3 rule.
A commenter states that EPA mischaracterized ethanol as an air toxic. In EPA's 2007
mobile source air toxics rule (72 FR 8428, February 26, 2007) the term "mobile source air
toxics" was used to define compounds that are emitted by mobile sources and have the potential
to cause serious adverse health effects. The RIA makes no statements about whether ethanol is
an air toxic of concern at environmental exposure levels.
A number of comments state that aromatics in gasoline create higher emissions of
numerous pollutants, including PAHs and other toxics, BC, and precursors of SOA and UFPs.
The Agency continues to research how various fuel properties and combustion processes affect
emissions of PAHs, other toxics, BC, UFPs, and SOA precursors. A number of commenters
urged EPA to improve the modeling of SOA in CMAQ. Commenters also stated that current
emission measurements and models do not account for the full contribution to SOA from
gasoline exhaust. EPA continues to improve the modeling capabilities of CMAQ. Many factors
affect formation of secondary organic aerosols, including emissions of PAHs and other aromatic
compounds from diesel and gasoline motor engines, as well as other SOA precursors, such as
isoprene and a-pinene. NOx also plays an important role in SOA formation. EPA is conducting
ongoing research to better understand relationships between fuel properties and emissions of
aromatics and other SOA precursors, as well as their reactions in the atmosphere under different
conditions.
Some commenters also urged EPA to encourage the use of ethanol, including by
revisiting the greenhouse gas emissions standards and fuel efficiency standards, by extending the
1-psi RVP waiver for E10 to E15 and butanol blends, or specifying different certification fuels.
Discussion of similar comments can be found in Chapter 4.5.1.5 of this Summary and Analysis
of Comments document, and are otherwise outside the scope of this rulemaking.
12.5. Nonroad Diesel Engines
12-27
-------
Chapter 12: Comments Unrelated to the Tier 3 Rulemaking
What Commenters Said:
Commenter: Dresser Trap Rock, Inc.
The most fuel efficient engines we have operated since 1979 were the pre-Tier 1 engines. As we
had to transition to Tier 1 engines, none of our operators wanted to ruin the new loaders as they
were too slow, particularly at loading railcars and off road haul trucks.
With the advent of Tier 2 engines, it was exactly the same story all over again; the operators
could not load off road haul trucks and particularly railcars with rock as quickly as they did with
the Tier 1 engines.
Again, with the advent of Tier 3 engines, we found through observation of Tier 1 powered end
loaders and Tier 3 powered loaders that a Tier 1 loader could make 4 passes to the Tier 3
powered loader's 3.
Now, you may say that there is more fuel economy and less emissions with Tier 3 engines
relative to the Tier 1 engines, but if the operator has to run the loader at full throttle 25% longer
to accomplish the identical task as a Tier 1, there cannot possibly be a reduction fuel
consumption or emissions.
Our Response:
This Tier 3 motor vehicle emissions rulemaking does not pertain to the nonroad engines
of concern to the commenter (nonroad diesel engines are, coincidentally, also regulated by EPA
in tiers—Tiers 1,2, 3, and 4).
12.6. Anti-backsliding Study
What Commenters Said:
Commenter: American Petroleum Institute (API) and the Association of Fuel & Petrochemical
Manufacturers (AFPM); Marathon Petroleum Company LP (MFC)
Anti-backsliding - EPA has not provided Congress with analyses required by legislation. We
believe that the emissions benefits from the Tier 3 rule should be counted as offsets for the
environmental impacts of the Renewable Fuel Standard since the RFS was effective long before
Tier3.
EPA has not provided Congress with analyses required by legislation. The Agency is required to
conduct anti-backsliding studies per section 1506 of EPActOS (draft for public comment was due
summer 2009 and a final report was due summer 2010) and section 209 of EISA (due summer
2009).
12-28
-------
Tier 3 Summary and Analysis of Comments
In letters to Charles Drevna (AFPM) and Jack Gerard (API) dated March 6, 2012, EPA
emphasized that the Tier 3 rule is independent of this upcoming anti-backsliding study. EPA
repeated this in a letter to Senator Inhofe dated March 19, 2012. Gina McCarthy restated this
independence on page 113 of her answers to Senator Vitter in May 2013.
We believe that the emissions benefits from the Tier 3 rule should be counted as offsets for the
environmental impacts of the Renewable Fuel Standard since the RFS was effective long before
Tier3.
Our Response:
The anti-backsliding study is required by Clean Air Act sections 21 l(q) and (v) and is
outside the scope of this rulemaking. EPA's analysis and modeling to support the Clean Air Act
section 21 l(q) and 21 l(v) "anti-backsliding" study is still underway. As required by Clean Air
Act section 21 l(q), EPA will release a draft report for public comment before publishing it in
final form. Thus, there will be an opportunity to review and comment on EPA's analysis and
conclusions.
12.7. Alternative Fuels
What Commenters Said:
Commenter: Advanced Biofuels USA (ABFUSA)
Allow Optional Renewable Higher Octane Additives: While biomass,
cellulose/hemicellulose/pectin, ("cellulosic ethanol" in RFS2) derived ethanol is the current
primary octane enhancing component foreseen for Higher Octane, Higher Ethanol Fuel, final
Tier 3 regulations should provide flexibility for other biofuel derived additives provided the
following four criteria are met:
1. Life Cycle GHGs of the alternative additives are within 110% of industry wide life-cycle
GHG emissions of the biomass ethanol used at the time of the introduction of the alternative
additives.
2. The composition of the biofuel derived fuel components do not exceed EPA or CARB,
whichever is lower, aromatic or other fuel composition standards in effect at the time of their
introduction.
3. The use of such fuel components would not materially effect the cost of vehicle certification
testing procedures.
4. The use of such fuel components would not materially effect any fuel system or engine
components of the vehicles it was used in.
Commenter: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global); General Motors LLC (GM)
While OEMs discussed the interconnection of fuels and vehicle issues with EPA during the
negotiations leading to the 2017 CAFE/GHG requirements, EPA declined to include fuel issues
12-29
-------
Chapter 12: Comments Unrelated to the Tier 3 Rulemaking
in that rulemaking. However, at that time EPA gave the industry assurance that these issues
would be addressed promptly within the Tier 3 rulemaking in order to facilitate the needed
product changes to support the CAFE/GHG standards, as well as Tier 3 requirements. The OEMs
are spending billions of dollars (34) on compliance with the CAFE/GHG rules and require the
cleanest fuels possible to meet these obligations. We urge EPA to support the progress toward
cleaner vehicles by ensuring that the proper fuels are widely available.
To date, EPA has only agreed to propose changes to the existing sulfur market fuel standards.
The Alliance and Global note that there are a number of other time-sensitive market gasoline
changes that warrant federal standards and that should be addressed by EPA in a supplemental
rulemaking. We reiterate that the regulations for engines and fuels must be integrated to ensure
environmental progress, and to ensure that new engine designs, vehicle exhaust and evaporative
after-treatment systems do not become stranded investments or opportunities, from a lack of
commitment to improve fuel quality to achieve cleaner, more efficient vehicle operation.
34 - As noted in Alliance testimony for this rulemaking, according to EPA, automakers will spend about
$15 billion in just ten years to meet the requirements of this proposed rule, and effectively remove tailpipe
emissions from the environmental equation. As EPA further states, combined with the fuel economy and
greenhouse gas program, this rule would bring auto investment in the vehicle emission reduction
programs through 2025 to more than $216 billion -about 100 times the investment that the oil industry is
being asked to make under the NPRM.
Our Response:
These comments are outside the scope of the Tier 3 rulemaking. There are already in
place a set of regulations associated with the introduction into commerce of new fuels and fuel
additives in keeping with our CAA responsibilities. Any other fuel standards would have to go
through a 21 l(c) rulemaking process similar to Tier 3.
12.8. Detergency Requirements
What Commenters Said:
Commenter: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
Detergency Requirements Need Improvement: Our members support EPA's proposed
requirements that clarify the treatment of gasoline additives (82). However, additional regulatory
improvements are needed with regard to detergency additives for motor vehicle fuel.
Lowest Additive Concentration (LAC) Detergency: Due to the ability of detergents to inhibit the
formation of engine deposits (83) and decrease emissions, the addition of detergents is required
in the U.S. under the CAA (84). The implementing regulations (40 C.F.R. Part 80, Subpart G)
allow refiners and marketers to meet the requirements by adding the lowest additive
concentration (LAC) that minimizes deposits, based on tests conducted by the additive supplier
using an EPA-defmed fuel that reflects a (one-time) determination of the 65th percentile highest
12-30
-------
Tier 3 Summary and Analysis of Comments
value for four fuel properties (sulfur, aromatics, olefins, and T90).85 This additive concentration
can then be used by any fuel supplier. The shortcoming of this practice is that there are fuels in
the marketplace that produce more deposits than the 65th percentile fuel, and therefore will not
be adequately treated to control deposits.
Automaker experience has shown that the required evaluations for fuels meeting LAC are
insufficient to assure consistency and efficacy across the fuels marketplace for today's nearly
250 million vehicles. The LAC principle has resulted in gasoline refiners and marketers reducing
the concentrations of detergent additives in their fuels by as much as half of levels observed in
the 1990s. This degradation of detergency is shown in Figure 3 [Figure 3 can be found on p. 91
of Docket number EPA-HQ-OAR-2011-013 5-4461-A1]. The trending downward of unwashed
gum levels (a marker for the amount of detergent additive present) suggests lower detergent
additive doses and downgraded quality (albeit that in some newer products, lower concentrations
can produce the same effect). At the same time, vehicle emissions and durability requirements
have become increasingly more stringent.
Recommendation: Auto manufacturers remain concerned about induction system deposits in
U.S. marketed vehicles. These concerns have been raised on multiple occasions with EPA. In the
past, the EPA was provided with data which demonstrated on-going problems with U.S. gasoline
detergency compliance showing that the program was not effective at retail. We urge EPA to
refocus on this issue, which has emissions impacts and should be a Tier 3-related market fuel
issue. Because of these adverse trends, as apparent in Figure 3, EPA must assure that all fuels
include sufficient detergent concentration. We would be happy to work with the Agency and the
other stakeholders on a regulatory solution.
82 - 78 Fed. Reg. 29908 at 29929 (May 21, 2013).
83 - Deposits are formed within the engine and/or fuel system during normal vehicle operation over a
period of time. Certain gasoline components such as di-olefins are relatively unstable and they thermally
break down during engine operation, leading to formation of deposits on fuel injectors, inlet valves, and
within the combustion chamber. Combustion chamber deposits (CCD) can result in a higher effective
compression ratio, which then increases the need for a higher octane rated fuel. Additionally, CCDs can
absorb and release hydrocarbons, leading to an increase in hydrocarbon emissions. See also current
edition Worldwide Fuel Charter recommendations, available from the Alliance of Automobile
Manufacturers.
84 - The text from the CAA requirement in Sec. 211(1) reads as follows: "Effective beginning January 1,
1995, no person may sell or dispense to an ultimate consumer in the United States, and no refiner or
marketer may directly or indirectly sell or dispense to persons who sell or dispense to ultimate consumers
in the United States any gasoline which does not contain additives to prevent the accumulation of deposits
in engines or fuel supply system. Not later than 2 years after the date of the enactment of the Clean Air
Act Amendments of 1990, the Administrator shall promulgate a rule establishing specifications for such
additives."
85 - The LAC requirement is defined in 40 C.F.R Part 80, Subpart G, Sec. 80.141(c)(3) as follows:
"Minimum recommended concentration: (i) The lower boundary of the recommended range of
concentration for the detergent additive package in gasoline, which the additive manufacturer must report
pursuant to the registration requirements in § 79.21(d) of this chapter, must equal or exceed the minimum
concentration which the manufacturer has determined to be necessary for the control of deposits in the
associated fuel type, pursuant to paragraph (e) of this section. The minimum recommended concentration
shall be provided to EPA in units of gallons of detergent additive package per thousand gallons of
12-31
-------
Chapter 12: Comments Unrelated to the Tier 3 Rulemaking
gasoline or PRC, reported to four digits. This concentration is the lowest additive concentration (LAC)
referred to elsewhere in this subpart."
Commenter: Ford Motor Company (Ford)
Increased Detergency Requirements for Market Fuel: Gasoline additives used to protect vehicles
from damage caused by deposits are known as detergents. Deposits are formed within the engine
and/or fuel system during normal vehicle operation over a period of time. Certain gasoline
components have higher boiling points, such as di-olefins, and during the combustion process,
they thermally breakdown during engine operation, leading to formation of deposits on fuel
injectors, inlet valves, and within the combustion chamber. Combustion chamber deposits (CCD)
can result in a higher effective compression ratio, which then increases the need for a higher
octane rated fuel. Additionally, CCDs allow for a medium to which liquid hydrocarbons can
absorb and release, leading to an increase in hydrocarbon emissions. Another consequence of
CCD is 'carbon knock' which can result from the mechanical interference between the piston top
and the cylinder head. Fuel injector deposits are of particular concern because the fuel passages
are so small that deposits can prevent normal flow of fuel, resulting in degraded driveability,
decreased power and fuel economy and increased exhaust emissions. Intake valve deposits also
result in reduced power, cold-start and warm-up driveability issues and increased exhaust
emissions.
The lowest additive concentration (LAC) regulation allows for refiners to meet the requirements
by adding the lowest additive concentration that prevents deposits based on a selected batch
sample from the refinery. The shortcoming of this practice is that the batch sample selected may
be a sample representative of the "best-case" fuel formulation that results in minimal deposit
formation. This allows obligated parties to choose an under representative sample from the
population of fuels that the obligated party produces for the marketplace. Common practice
results in an incorrectly calculated LAC that is then employed across all fuels produced.
Experience has shown that fuels meeting LAC are insufficient for meeting the needs across the
marketplace, which includes nearly 250 million vehicles. Auto manufacturers remain concerned
about induction system deposits in U.S. marketed vehicles. Despite continued messaging about
the need for more stringent detergency requirements, the LAC program has received no
modifications since its inception, over two decades ago. Since the Tier 3 program acknowledges
that "vehicles and fuels [are] a system," the detergency requirements of fuels must be adjusted to
ensure adequate deposit control protection for all finished gasoline sold in the U.S. market, so
that meeting the increased durability requirements of Tier 3 can be assured.
Recommendation: Ford supports the Alliance's request to require all fuels meet deposit control
tests such that enough detergents are added to meet the needs of the lowest quality product batch
rather than just the cleanest stream. Ford supports the Alliance's request that the Lowest
Additive Concentration (LAC) requirement for gasoline detergency be eliminated and an
appropriate rule be introduced in its place.
Our Response:
12-32
-------
Tier 3 Summary and Analysis of Comments
EPA's consideration of whether changes to the gasoline deposit control program are
needed is beyond the scope of the Tier 3 rule. There is no data available that indicates the
current levels of deposit control protection are insufficient to support the maintenance of the Tier
3 vehicle emissions standards in-use. EPA continues to be open to working with industry to
evaluate the potential emissions benefits of implementing updated deposit control requirements.
Data on the potential emissions benefits is vital to this evaluation. As discussed in Chapter 6.1.4
of this Summary and Analysis of Comments document and Section VI.A.4 of the preamble to the
Tier 3 final rule, we are finalizing provisions to accept "Top Tier" deposit control test data in the
certification of deposit control additives.
12.9. Methylcyclopentadienyl Manganese Tricarbonyl (MMT)
What Commenters Said:
Commenter: Alliance of Automobile Manufacturers (Alliance) and Association of Global
Automakers (Global)
EPA Should Undertake Regulatory Review of MMT/Metallic Additive: All vehicle
manufacturers have long opposed the use of any metallic based additive in gasoline, including
compounds of iron, silicon, phosphorus, and manganese such as MMT (methylcyclopentadienyl
manganese tricarbonyl). Their use results in damage to engine parts and emissions control
components.
Manganese-based additive can be used as an octane enhancer for gasoline; however, when
combusted in an internal combustion engine, manganese is transformed into manganese oxide
particles. As documented in numerous studies and scientific papers, the prolonged use of
gasoline with manganese-based additives results in such particles contributing to the plugging
and failure of catalytic converters, which have been used on vehicles for decades to control
emissions of carbon monoxide, hydrocarbons, and oxides of nitrogen.
Over the years, catalytic converter technology has improved dramatically, with conversion
efficiencies today of more than 95%, to keep pace with increasingly more stringent vehicle
emission standards. The residual amount, technically speaking, corresponds to the vehicle's
tailpipe emission level. As emission standards become more stringent, this residual amount
becomes even smaller. For today's advanced equipment, such plugging and failure problems -
which reduce conversion efficiencies - are especially problematic. Degradation and failure of
catalytic converters cause vehicles to have higher emissions, degrading air quality and
potentially causing vehicles to exceed emissions standards, and to fail state/local vehicle mission
inspections. Damaged catalytic converters are very expensive to replace.
The manganese oxide particles also cause deposits to build up on internal engine parts, spark
plugs, and oxygen sensors. This can also cause vehicle performance and emission problems,
including loss of fuel economy and illumination of dashboard "check engine" lights, and
premature and additional vehicle repairs for consumers.
12-33
-------
Chapter 12: Comments Unrelated to the Tier 3 Rulemaking
Because of these risks, all major manufacturers selling vehicles in the U. S. recommend in their
vehicle owner manuals that any fuel containing manganese-based additives NOT be used in
today's vehicles.
Metallic additives, including those with manganese, are prohibited in Federal reformulated
gasoline (RFG), which represents about one-third of the U.S., non-California gasoline pool and
is prohibited under State law in all gasoline sold in California. EPA allows the use of MMT in
non-RFG gasoline at a maximum of 1/32 gram Mn as MMT/gallon (80) following a long
stakeholder controversy beginning that had resulted in an earlier EPA ban of MMT.
For a number of years there have been no MMT or other metallic additives detected in the annual
(Summer/July and Winter/January) Alliance of Automobile Manufacturers North American Fuel
Surveys for the U.S. In the past several years, however, in several States, including Nevada and
neighboring states, and also in the Southeast, there has been renewed local interest in using
MMT additive, raising concerns among various stakeholders. An issue so fundamental for all
vehicles is not suited for State by State response - the same costly and inconsistent dynamic
which led to CAA additive provisions and a federal EPA standard in the first instance.
It would be appropriate for EPA to again consider a comprehensive prohibition for metallic
additives for any gasoline sold in the US, at the earliest possible time. EPA should formally
consider studies and data that have become available on adverse effects of MMT on vehicle
equipment since EPA's last action, for example, the findings in the 2008 Sierra Research Report
"Impacts of MMT(R) Use in Unleaded Gasoline on Engines, Emissions Control Systems, and
Emissions (copy attached as Appendix 6 to these comments) (81).
Many other organizations are also on record opposing the use of manganese-based gasoline
additives for both technical and health reasons, including the Manufacturers of Emission
Controls Association, The International Council for Clean Transportation, Truck and Engine
Manufacturers Association, and the National Resources Defense Council. See also, the current
4th edition and pending 5th edition of the Worldwide Fuel Charter, available from the Alliance
of Automobile Manufacturers website (www.autoalliance.org).
CAA §21 l(c) addresses EPA's ability to prohibit or limit the sale of fuel additives that are
already approved if they will "impair to a significant degree the performance of any emission
control device or system which is in general use." The Agency should undertake an updated
evaluation of MMT use as a gasoline additive.
Recommendation: EPA should promptly initiate a formal review of MMT use with notice and
comment to prevent putting all gasoline engines and emission reductions at risk from metallic
additives.
80 - See http://www.epa.gov/otaq/regs/fuels/additive/mmt cmts.htm; and 65 Fed. Reg. 44775 (July 19,
2000).
81 - Sierra Research, Report No. SR2008-08-01, August 29, 2008, prepared for the Canadian Vehicle
Manufacturers' Association and Association of International Automobile Manufacturers of Canada. Also
available from the Alliance website under Fuels, www.autoalliance.org.
12-34
-------
Tier 3 Summary and Analysis of Comments
Commenter: American Honda Motor Co., Inc.
Honda wishes to voice strong support for the recommendation made in the joint Association of
Global Automakers-Alliance of Automobile Manufacturers comments regarding the issue of
MMT and other metallic fuel additives. Honda sees an urgent need for, and strongly
recommends, that EPA once again consider a comprehensive prohibition of metallic additives for
any gasoline sold in the United States.
Commenter: Ford Motor Company (Ford)
Elimination of the use of Metallic Additives in Market Fuel: The use of manganese (also known
as methylcyclopentadienyl manganese tricarbonyl or MMT) began with the phase-out of leaded
fuel as it is marketed as an octane booster. When used in gasoline as an octane booster, MMT
does not burn completely, as does its ethanol counterpart. Instead, the manganese in MMT reacts
during the combustion process to form solid manganese oxide particles that are reddish-brown in
color. These particles form deposits in the engine as well as accumulate on the front inlet face of
catalytic converters and throughout the exhaust system. Manganese oxide deposits also appear
on engine components including deposits on spark plugs (leading to misfire, which leads to
increased emissions, increased fuel consumption and poor performance) and on the oxygen
sensor (causing malfunction and again increasing emissions and resulting in poor performance).
Additionally, manganese oxide deposits on the catalytic converter cause plugging of the catalyst,
increased back pressure, poor vehicle operation and increased fuel consumption, as well as
increased tailpipe emissions due to diminished emission control function.
In order to meet the more stringent emissions standards of Tier 2, the vast majority of vehicle
manufacturers implemented the use of high density close coupled catalysts (HDCC) which
contain more catalyst cells per unit area than older technology catalysts. The increase in the
number of cells per unit area results in an increase in the overall catalytic active surface area and
also decreases the actual size of each individual channel of the substrate. This advanced
technology allows for reduced time needed for the catalyst to reach operating temperatures, but
due to the smaller channel sizes, it also results in increased susceptibility to front-face catalyst
blockage from manganese oxide deposits. Many manufacturers cite that no other technologies
exist that are as effective as HDCC catalysts. Further, the need to meet the more stringent
emissions standards of the future will drive the catalysts closer to the engine, resulting in
increased catalyst inlet and front-face temperatures. With the inevitable manganese oxide
deposits that the combustion of MMT in gasoline creates, these HDCC catalysts can be rendered
useless for emissions control and can create vehicle performance issues once contaminated.
The following images, extracted from the 2008 Sierra Research report (9), illustrate the typical
severity of the plugging observed on a catalyst exposed to MMT. The first image (Figure 1) is a
catalyst obtained from a vehicle operated with MMT-free fuel for approximately 100,000 miles
while the second image (Figure 2) shows a catalyst from a vehicle after accumulating only
40,000 miles using gasoline containing MMT. The considerable amount (almost 95% of the
catalyst is covered) of manganese oxide deposits in Figure 2 demonstrates the severity of
manganese-based deposits coating catalysts (and therefore rendering them useless). Furthermore,
the additional emissions durability requirements of 150,000 miles will demand that vehicles do
12-35
-------
Chapter 12: Comments Unrelated to the Tier 3 Rulemaking
not use fuel which may damage the catalyst. Thus, considering the known abilities of MMT
containing gasoline to damage vehicle catalysts, MMT and similar metallic additives must be
prohibited from market fuels.
Recommendation: Ford continues to remain opposed to the use of any metallic based gasoline
additive (lead, iron, silicon, phosphorus, and manganese) and strongly supports the Alliance
comments which ask for the EPA to perform a regulatory review of the use of metallic additives
and manganese (also known as MMT). Ford strongly supports the Alliance comments submitted
in regard to MMT and seconds their recommendation for the ban of manganese based gasoline
additives.
9 Sierra Research Inc. "Impacts of MMTR Use in Unleaded Gasoline on Engines, Emission Control
Systems, and Emissions," August 29, 2008.
Our Response:
Comments on metallic fuel additives are outside the scope of this Tier 3 rulemaking.
12.10. Other
What Commenters Said:
Commenter: Peoples Republic of China
We noticed that in the measurement of limits of evaporative pollutants, Remote Sensing Device
(RSD) testing technology is adopted, while in the measurement of PM emission level, a new FTP
PM testing standard is adopted. In order to achieve the target specified by USA' Tier 3 emission
standard, as specified in Article 11.3 of TBT Agreement, the Proposer shall grant other
Members, especially the developing country Members, guidance on the testing technology and
technical resistance, help them fully apprehend new testing technology and improve their
production process scientifically to accommodate themselves to new technical requirement in
time.
Environment protection is a common responsibility to be borne by all countries. Superior or
inferior environment quality in any region is just partial and temporary. With a view to realize
effective environmental protection, we believe that in accordance with Article 2.6 of TBT
Agreement, WTO Members that are advanced in science and technology shall, giving priority to
appropriate international standardizing bodies, play a full part in and promote preparation of
international standards, lead actively the world to take concerted actions to realize
comprehensive and scientific environmental protection.
Commenter: Algenol Biofuels Inc.
The national policy particularly favors biofuels such as Algenol's ethanol which has an
especially low life-cycle output of greenhouse gas emissions. A life-cycle analysis conducted by
12-36
-------
Tier 3 Summary and Analysis of Comments
the company along with the Georgia Institute of Technology's Strategic Energy Institute
demonstrates at least a 60%, and in many scenarios 80%, reduction in greenhouse gas emissions
when compared to regular gasoline. (See Luo, Hu, et al, "Life Cycle Energy and Greenhouse
Gas Emissions for an Ethanol Production Process based on Blue-Green Algae," Environmental
Science & Technology. 2010, 44 (22), pp. 8670-8677.)
Commenter: Private Citizen
Subsidize electric vehicles instead! When they become less costly and services more widely
available, people will easily change to electric.
Commenter: Private Citizen
This may also decrease the emission of cadmium which is present in all combustion but not
measured adequately since lead was removed from gasoline. All the problems associated with
exposure to motor vehicle emissions can occur with cadmium exposure alone. Cadmium lowers
25 OH vit D3. Low levels are found in all serious diseases. This phenomenon has occurred since
lead was removed from gasoline. Lead blocks cadmium uptake and lead chloride formed a
complex with cadmium fumes making it less bioavailable. Blood, urine, and hair levels of
cadmium do not necessarily reflect cadmium exposure and toxicity because if there is not
sufficient metallothionein to bind it, cadmium quickly disappears from blood into the lining cells
of blood vessels and into endocrine cells and the choroid plexus.
These new standards need to address cadmium as well by developing new technologies to look at
cadmium fumes from motor vehicle emissions.
It is really necessary to re-examine the effects of lead in gasoline. Lead is the lesser toxin. By
removing lead from gasoline RICE which contains arsenic is more toxic. Lead, cadmium and
arsenic block each other's toxicity while cadmium and arsenic are synergistically toxic.
We know that these new standards are highly cost-effective - costing about a penny per gallon of
gasoline. In return, Americans would save over $10 billion per year on health care costs by 2030.
They would save even more if cadmium were addressed.
Commenter: Open Joint-Stock Company
Currently, we also found a way to solve the above-mentioned major environmental problems. To
fulfill this goal, we have developed a comprehensive high-tech multi-functional devices of new
generation for the petrochemical industry, for example, CS 'TORNADO'. Moreover, on the
basis of CS 'TORNADO', developed, such as CS 'Mercury' and other complex systems and
installations for the treatment and disposal of toxic and output of toxic gases, as well as a variety
of solid particles produced by different factories and mills of various sources of pollution.
The main purpose of the above systems, a complete cleaning of exhaust gases and obtaining
alternative fuel, for example, synthesis gas, and other gas mixtures for combustion of
hydrocarbon fuel, and for cleaning of gases. The use of CS 'TORNADO' for crude oil and other
12-37
-------
Chapter 12: Comments Unrelated to the Tier 3 Rulemaking
chemicals, will raise the refining and petrochemical sector in Latvia to a new level of high-tech
development and production standards.
We offer you a brief look at some key features and specifications of CS 'TORNADO', which
will be applied in the 21st century, and possibly the 22nd century. Next, read the applications ...
(See Appendix 4).
Commenter: Robert Bosch GmbH, Gasoline Systems, Germany
Also it needs to be assured the levels for silicone compounds from biomethane originating from
waste or sewage sludge are fixed at 0,1 mg/m3 to ensure proper working of exhaust gas treatment
systems.
Commenter: Private Citizen
I would like to add one thing. Please create a stop pollution hotline. Create a 1- 800-CLE-ANUP
hotline so that drivers and pedestrians may report vehicles that seem to be polluting, whether it is
from tail pipes or uncovered loads of materials being transported. Let such a violation result in
the situation being improved and heavy fines being collected. Perhaps the fines could be used to
reimburse hospitals that work on asthma patients. Although you will receive this message from
other citizens who care about the atmosphere, I will leave it in the letter. The facts are important
and clearly stated. Let's not emulate the pollution in China.
Commenter: Children's Environmental Health Network (CEHN)
EPA should issue a public annual compliance "report card" on each automaker. This level of
transparency is in the spirit of right-to-know laws and facilitates informed consumer choices.
Commenter: Textile Industries, Inc.
I am admittedly not a scientist or an expert on sequestration of greenhouse gases, but due to my
profession, I am required to keep myself apprised of all things having to do with our
environment. As such, I can say with some confidence that there is enough scientific evidence
now to show that air pollution, and especially greenhouse gases, are mainly responsible for
global warming, also known as climate change, in the world we currently live in. And this is the
only world we've got.
Some scientific evidence indicates that we have perhaps already passed a tipping point beyond
which we cannot prevent the domino effect of increased greenhouse gases being released from
beneath the ocean bottoms in the form of methane, which is worse even than the CO2 which has
been primarily responsible for climate change until now.
Commenter: Thomas Jefferson University Hospital
12-38
-------
Tier 3 Summary and Analysis of Comments
I have spent over 150 hours a year for the last 20 years as a volunteer in trying to decrease the
burden of asthma on children in this city and State. I've been an advocate on many levels. I was
one of the founding members of the Pennsylvania Allies Against Asthma, now deceased because
of funding, but a great organization in its time. And I'm now the Medical Director of the
Pennsylvania Asthma Partnership.
Why am I spending all this time? That's about three weeks of my year that I spend doing this,
and, trust me, no one pays me for it. Well, I do it because asthma is the most common cause of
medical school absence nationally, and, in Philadelphia, which has an even larger burden, it
reflects almost half the days lost in school to medical reasons. I do it because it's the most
common reason that people from our emergency rooms are admitted to the hospital.
But as a group, we can cause an even more important effect, and that's the effect of diesel
vehicle's emissions on asthma. If you live on a street that has lots of trucks and buses going
down it, your chances of having either chronic or acute asthma is almost double that of someone
who lives in the same city a few blocks away on a quiet street that doesn't have trucks on it.
So we know that truck emissions, even though the diesel vehicles are much cleaner than they
ever were, we know we can do more. And we can, once we have a problem, solve it. When I
started in pediatrics 34 years ago, the average lead in a child in Philadelphia was a 20
micrograms per deciliter. It is now a three. There's no safe level of lead in children, but three is
a lot better than 20.
In this same way, if we can actually stop trucks and buses from idling, if we can install systems
that turn them off when they come to a red light, if we can take the city light system, stop light
system, and get them to be coordinated so there are less stops and less traffic jams, and on a few
streets put some traffic officers to stop traffic jams from happening, we can reduce the number of
children I see in the ER. We can reduce the number of children admitted, and we can have
children go to school a lot more days than they're going to school right now. And from the
world of pediatrics, we would thank - all of us would thank you if you could do anything to help
us out on that.
Commenter: National Corn Growers Association (NCGA)
We also provide comments on EPA's stated plan to develop an assessment of the health effects
of exposure to ethanol in Appendix A.
The proposed rule indicates that: EPA is planning to develop an assessment of the health effects
of exposure to ethanol, a compound which is not currently listed on EPA's IRIS database.
Extensive health effects data are available for ingestion of ethanol, while data on inhalation
exposure effects are sparse. In developing the assessment, EPA is evaluating pharmacokinetic
models as a means of extrapolating across species (animal to human) and across exposure routes
(oral to inhalation) to better characterize the health hazards and dose-response relationships for
low levels of ethanol exposure in the environment.
12-39
-------
Chapter 12: Comments Unrelated to the Tier 3 Rulemaking
The potential for health effects due to ethanol exposures associated with its use in motor vehicle
fuel were extensively evaluated a number of years ago by several organizations. The California
EPA, (21) the Health Effects Institute (HEI), and several other groups evaluated ethanol as a
replacement when problems associated with the gasoline additive, methyl tertiary-butyl ether
(MTBE), became apparent. All these analyses have concluded that the exposures to ethanol from
its use as a motor vehicle fuel are very unlikely to have any adverse consequences.
Although it is well established that relatively high doses of ingested ethanol and high blood
levels of ethanol associated with the consumption of alcoholic beverages have toxic effects in
humans, the amount of alcohol absorbed from the low doses of environmental ethanol is
extremely small in comparison. In fact, the small doses that may be experienced are the same
order of magnitude as naturally-occurring levels of ethanol in blood. In addition, ethanol is
readily degraded in the environment, both in air and in water so it will not accumulate to levels
of concern. These findings provide strong evidence that environmental exposures to ethanol will
have no adverse health impact.
The California analysis included detailed modeling of the ambient ethanol exposures for several
fuel variations and comparisons of the results to levels of concern developed by the Office of
Environmental Health Hazard Assessment (OEHHA). The modeled incremental ambient ethanol
concentrations were in the range of 5 to 100 ppb. The OEHHA health protective concentration
(HPC) was 53 ppm. It was derived from a study that reported a lowest effect level of 5300 ppm
for sensory irritation from inhaled ethanol in normal healthy subjects. OEHHA applied an
uncertainty factor of 100 to derive the 53 ppm (53,000 ppb) health protective level.
When the modeled ambient levels were compared to the acute and chronic non-cancer HPCs,
OEHHA concluded that modeled concentrations are at least 500-fold below the HPCs. This led
OEHHA to conclude that health effects due to ethanol exposure are not expected to occur.
OEHHA also evaluated the potential for cancer and concluded that there is no evidence that
ethanol is carcinogenic via the inhalation route even though heavy consumption of alcoholic
beverages is known to be associated with increased incidences of some cancers, including those
of the oral cavity and of the liver. OEHHA noted that ethanol was considered by the Proposition
65 Science Advisory Panel, who reviewed the evidence as to the carcinogenic and co-
carcinogenic effects of ethanol in humans and animals. The Panel specified the listing of
"alcoholic beverages, when associated with alcohol abuse" as carcinogenic in July 1988. The
1999 OEHHA report followed this assessment in concluding that, whereas high levels of chronic
exposure to ethanol are carcinogenic to humans, the levels of ethanol predicted to occur in air or
water as a result of its use in gasoline were unlikely to result in a cancer risk to the exposed
population.
The Health Effects Institute carried out a thorough review of the information on exposures to and
health effects from various oxygenates including ethanol. In agreement with CalEPA, the HEI
Committee concluded that it was unlikely that the low exposures to ethanol would cause any
health effects. For example, the Health Effects Institute review included estimates of incremental
blood level increases for two refueling exposure: a typical refueling scenario, 1 ppm for 3
minutes, and an extreme exposure scenario, 10 ppm for 15 minutes. The HEI Panel concluded:
12-40
-------
Tier 3 Summary and Analysis of Comments
Even the extreme scenario is expected to result in an incremental blood level that is insignificant
compared with the endogenous blood levels resulting from normal metabolism.
The HEI Committee concluded that in exposure scenarios encountered by the general public, it is
unlikely that an increase in ethanol blood levels will be measurable and that health effects from
exposure to ambient levels of ethanol are unlikely because endogenous blood ethanol levels are
not predicted to increase significantly from inhaling ethanol in fuel.
In addition to the findings of the extensive reviews by HEI, CalEPA and others, the U. S. EPA
has not considered ethanol as an air toxic of concern in its recent evaluations and rulemakings.
For example, the Agency did not include ethanol in a group of 177 air toxics compounds that
were evaluated in the most recent National-Scale Air Toxics Assessment. Ethanol was
not explicitly considered in the February 26, 2007 Mobile Source Air Toxics Rulemaking. When
EPA first announced its interest in removing MTBE, the Agency understood that.
Given that ethanol is formed naturally in the body at low levels, inhalation exposure to ethanol at
the low levels that human are likely to be exposed are generally not expected to result in adverse
health effects.
Although the EPA IRIS database does not currently include ethanol, there has been a 1,000 ppm
occupational standard for ethanol for decades. The 1,000 ppm 8-hour time-weighted average
standard was set to provide a no-effect level for irritation in the occupational setting. An
additional relevant study is available. Nadeau, et al., 2003 specifically evaluated possible
neuromotor effects of six-hour exposures to 0, 250, 500, and 1,000 ppm ethanol to healthy non-
smoking adult males in a controlled environment exposure chamber. Nadeau et al. report no
significant differences in body sway, hand tremor, or reaction time between exposed and non-
exposed conditions. In addition, ethanol was not detected in blood or in alveolar air when
volunteers were exposed to 250 and 500 ppm. At 1,000 ppm, the blood alcohol concentration
was 0.4 mg/dl. For comparison, the legal limits for drivers are usually 80 to 100 mg/dl and the
peak blood alcohol for an adult male consuming a typical alcoholic beverage containing 12 g of
alcohol is on the order of 25 mg/dl. Given the extremely low doses of alcohol from inhalation of
ambient ethanol, the health impacts of ethanol should not be a serious concern in the Tier 3
Rulemaking.
Our Response:
These comments are all outside of the scope of the Tier 3 rulemaking.
12-41
------- |