Control of Air Pollution From New Motor Vehicles: Proposed Tier 2 Motor Vehicle Emissions Standards and Gasoline Sulfur Control Requirements

Related Material

Other Related Documents

Note: EPA no longer updates this information, but it may be useful as a reference or resource.

[Federal Register: May 13, 1999 (Volume 64, Number 92)]
[Proposed Rules]
[Page 26003-26052]
From the Federal Register Online via GPO Access [wais.access.gpo.gov]
[DOCID:fr13my99-28]

[[Page 26003]]

_______________________________________________________________________

Part III

Environmental Protection Agency

_______________________________________________________________________

40 CFR Parts 80, 85 and 86

Air Pollution; Tier 2 Motor Vehicle Emission Standards and Gasoline
Sulphur Control Requirements; Diesel Fuel Quality Controls; Proposed
Rules

[[Page 26004]]

ENVIRONMENTAL PROTECTION AGENCY

40 CFR Parts 80, 85 and 86

[AMS-FRL-6337-3]
RIN 2060-AI23

Control of Air Pollution From New Motor Vehicles: Proposed Tier 2
Motor Vehicle Emissions Standards and Gasoline Sulfur Control
Requirements

AGENCY: Environmental Protection Agency (EPA).

ACTION: Notice of proposed rulemaking.

-----------------------------------------------------------------------

SUMMARY: Today's document proposes a major program designed to
significantly reduce the emissions from new passenger cars and light
trucks, including pickup trucks, minivans, and sport-utility vehicles.
These reductions would provide for cleaner air and greater public
health protection, by reducing ozone and PM pollution. The proposed
program is a comprehensive regulatory initiative that treats vehicles
and fuels as a system, combining requirements for much cleaner vehicles
with requirements for much lower levels of sulfur in gasoline. A list
of major highlights of the proposed program appears at the beginning of
SUPPLEMENTARY INFORMATION.
For the first time, through a phase-in, we propose to apply a
single average exhaust emission standard that would cover both
passenger cars and all light trucks operated on any fuel. The proposed
emission levels (``Tier 2 standards'') are feasible for both types of
vehicles and are appropriate since the miles traveled in light trucks
are increasing and the emissions from these vehicles are thus an
increasing problem. This approach will build on the recent technology
improvements resulting from the successful National Low-Emission
Vehicles (NLEV) program and improve the performance of these vehicles
through lower sulfur gasoline.
To enable the vehicle technology and generate emission reductions
from current vehicles we propose to significantly reduce average
gasoline sulfur levels nationwide. Refiners would generally install
refining equipment to remove sulfur in their refining processes, while
importers would be required to market only gasoline meeting the
proposed sulfur standards. The proposal outlines an averaging, banking,
and trading program to provide flexibility for refiners and ease
implementation.
This program focuses on reducing the passenger car and light truck
emissions most responsible for causing ozone and particulate matter
problems. Without today's action, we project that emissions from these
vehicles will represent 30-40 percent of nitrogen oxides and volatile
organic compound emissions in some cities, and almost 20 percent
nationwide, by the year 2020.
Our proposal would bring about major reductions in annual emissions
of these pollutants and also reduce the emissions of sulfur compounds
coming from the sulfur in gasoline. For example, we project a reduction
in oxides of nitrogen emissions of nearly 800,000 tons per year by 2007
and 1,200,000 by 2010, the time frame when many states will have to
demonstrate compliance with air quality standards. Emission reductions
would continue increasing for many years, reaching almost 2,200,000
tons per year in 2020. In addition, the proposed program would reduce
the contribution of vehicles to other serious public health and
environmental problems, including regional visibility problems, toxic
air pollutants, acid rain, and nitrogen loading of estuaries.
Furthermore, we project that these reductions, and their resulting
environmental benefits, would come at an average cost increase of less
than $100 per passenger car, less than $200 per light truck, and an
increase of less than 2 cents per gallon of gasoline (or about $100
over the life of an average vehicle).

DATES: Comments: We must receive your comments by August 2, 1999.
Hearings: We will hold four public hearings, on June 9-10, June 11,
June 15, and June 17, 1999. EPA requests that parties who want to
testify notify the contact person listed in the ADDRESSES section of
this document two weeks before the date of the hearing.

ADDRESSES: Comments: You may send written comments in paper form or by
E-mail. We must receive them by the date indicated under ``DATES''
above (August 2, 1999). Send paper copies of written comments (in
duplicate if possible) to Public Docket No. A-97-10 at the following
address: U.S. Environmental Protection Agency (EPA), Air Docket (6102),
Room M-1500, 401 M Street, SW, Washington, DC 20460. If possible, we
also encourage you to send an electronic copy of your comments (in
ASCII format) to the docket by e-mail to A-and-R-Docket@epa.gov or on a
3.5 inch diskette accompanying your paper copy. If you wish, you may
send your comments by E-mail to the docket at the address listed above
without the submission of a paper copy, but a paper copy will ensure
the clarity of your comments.
Please also send a separate paper copy to the contact person listed
below. If you send comments by E-mail alone, we ask that you send a
copy of the E-mail message that contains the comments to the contact
person listed below.
EPA's Air Docket makes materials related to this rulemaking
available for review at the above address (on the ground floor in
Waterside Mall) from 8:00 a.m. to 5:30 p.m., Monday through Friday,
except on government holidays. You can reach the Air Docket by
telephone at (202) 260-7548 and by facsimile at (202) 260-4400. We may
charge a reasonable fee for copying docket materials, as provided in 40
CFR part 2.
Hearings: We will hold four public hearings at the following
locations:

June 9-10, 1999, Top of the Tower, 1717 Arch Street, 51st Floor,
Philadelphia, PA 19103, telephone: 215-567-8787, fax: 215-557-5171
June 11, 1999, Renaissance Atlanta Hotel, 590 West Peachtree Street,
Atlanta, GA, 30308, telephone: 404-881-6000, fax: 404-815-5010
June 15, 1999, Doubletree Hotel, 3203 Quebec Street, Denver, CO, 80207,
telephone: 303-321-3333, fax: 303-329-5233
June 17, 1999, Holiday Inn Lakeside City Center, 1111 Lakeside Avenue,
Cleveland, OH 44144, telephone: 216-241-5100, fax: 216-241-7437

Additional information on the comment procedure and public hearings
can be found in SUPPLEMENTARY INFORMATION under Section VII, ``Public
Participation.''

FOR FURTHER INFORMATION CONTACT: Carol Connell, U.S. EPA, National
Vehicle and Fuels Emission Laboratory, 2000 Traverwood, Ann Arbor MI
48105; Telephone (734) 214-4349, FAX (734) 214-4816, E-mail
connell.carol@epa.gov.

SUPPLEMENTARY INFORMATION:

Highlights of the Tier 2/ Gasoline Sulfur Proposal

For cars and light trucks, the proposed program would:
Through a phase-in, apply for the first time a single
average exhaust emission standard that would cover both passenger cars
and all light trucks. The proposed emission levels (``Tier 2
standards'') are feasible for both types of vehicles and are
appropriate since the miles traveled in light trucks is increasing and
the emissions from these vehicles are thus an increasing problem.

During the phase-in, apply interim standards that match or
are more

[[Page 26005]]

stringent than current federal and California ``LEV I'' (Low-Emission
Vehicle, Phase I) standards.

Apply the same standards to vehicles operated on any fuel.

Allow auto manufacturers to comply with the very stringent
proposed new standards in a flexible way while ensuring that the
expected environmental benefits occur.

Build on the recent technology improvements resulting from
the successful National Low-Emission Vehicles (NLEV) program and
improve the performance of these vehicles through lower sulfur
gasoline.

Set more stringent particulate matter standards, primarily
affecting diesel powered vehicles.

Set more stringent evaporative emission standards.
For commercial gasoline, the proposed program would:

Significantly reduce average gasoline sulfur levels
nationwide. Refiners would generally install refining equipment to
remove sulfur in their refining processes. Importers of gasoline would
be required to import and market only gasoline meeting the proposed
sulfur limits.

Provide for flexible implementation by refiners through an
averaging, banking, and trading program.

Apply temporary, less stringent gasoline sulfur standards
to certain small refiners.

Enable the new Tier 2 vehicles to meet the proposed
emission standards, since sulfur in gasoline degrades a vehicle's
emission control performance. Lower sulfur gasoline is also important
in order to enable the introduction of advanced technologies that
promise higher fuel economy but are very susceptible to sulfur
poisoning (for example, gasoline direct injection engines).

Reduce emissions from NLEV vehicles and other vehicles
already on the road.

Regulated Entities

This proposed action would affect you if you produce new motor
vehicles, alter individual imported motor vehicles to address U.S.
regulation, or convert motor vehicles to use alternative fuels. It
would also affect you if you produce, distribute, or sell gasoline
motor fuel.
The table below gives some examples of entities that may have to
follow the proposed regulations. But because these are only examples,
you should carefully examine the proposed and existing regulations in
40 CFR parts 80, 85 and 86. If you have questions, call the person
listed in the FOR FURTHER INFORMATION CONTACT section above.

----------------------------------------------------------------------------------------------------------------
Examples of potentially regulated
Category NAICS Codes ^a SIC Codes ^b entities
----------------------------------------------------------------------------------------------------------------
Industry................................... 336111 3711 Motor Vehicle Manufacturers.
336112 .............. ...................................
336120 .............. ...................................
Industry................................... 336311 3592 Alternative fuel vehicle
converters.
336312 3714 ...................................
422720 5172 ...................................
454312 5984 ...................................
811198 7549 ...................................
541514 8742 ...................................
541690 8931 ...................................
Industry................................... 811112 7533 Commercial Importers of Vehicles
and Vehicle Components.
811198 7549 ...................................
541514 8742 ...................................
Industry................................... 324110 2911 Petroleum Refiners.
Industry................................... 422710 5171 Gasoline Marketers and
Distributors.
422720 5172 ...................................
Industry................................... 484220 4212 Gasoline Carriers.
484230 4213
----------------------------------------------------------------------------------------------------------------
^a North American Industry Classification System (NAICS).
^b Standard Industrial Classification (SIC) system code.

Access to Rulemaking Documents Through the Internet

Today's document is available electronically on the day of
publication from the Environmental Protection Agency Internet Web site
listed below. Electronic copies of the preamble, regulatory language,
Draft Regulatory Impact Analysis, and other documents associated with
today's proposal are available from the EPA Office of Mobile Sources
Web site listed below shortly after the rule is signed by the
Administrator. This service is free of charge, except any cost that you
already incur for connecting to the Internet.
Environomental Protection Agency Web Site:

http://www.epa.gov/fedrgstr/epa-air/

(Either select a desired date or use the Search feature.)
Office of Mobile Sources (OMS) Web Site:

http://www.epa.gov/omswww/

(Look in ``What's New'' or under the ``Automobiles'' topic.)
Please note that due to differences between the software used to
develop the document and the software into which the document may be
downloaded, changes in format, page length, etc. may occur.

Outline of This Preamble

I. Introduction
A. What Are the Basic Components of Today's Proposal?
1. Vehicle Emission Standards
2. Gasoline Sulfur Standards
B. What Is EPA's Statutory Authority for Proposing Today's
Action?
1. Light-Duty Vehicles and Trucks
2. Gasoline Sulfur Controls
C. The Tier 2 Study and the Sulfur Staff Paper
II. Proposed Tier 2 Determination
A. There Is a Substantial Need for Further Emission Reductions
in Order to Attain and Maintain National Ambient Air Quality
Standards
B. More Stringent Standards for Light-Duty Vehicles and Trucks
Are Technologically Feasible
C. More Stringent Standards for Light-Duty Vehicles and Trucks
Are Needed and Cost Effective Compared to Available Alternatives
III. Air Quality Need for and Impact of Today's Proposal
A. Americans Face Serious Air Quality Problems That Require
Further Emission Reductions

[[Page 26006]]

B. Ozone
1. Ozone Levels Have Declined, but Unhealthy Levels of Ozone
Persist
2. Cars and Light-Duty Trucks Are a Big Part of the VOC and
NO_X Inventory, and Today's Proposal Would Reduce This
Contribution Substantially
3. Tier 2/Sulfur Ozone Benefits and the Post Tier 2/Sulfur Ozone
Problem
C. Particulate Matter
1. Particulate Matter Presents Substantial Public Health Risks
2. Reducing Emissions From Cars and Light Trucks Would Reduce
Ambient Levels
3. Today's Proposal Would Limit the Potential Health Risks From
Increased Diesel Engine Use in Cars and Light Trucks
4. Today's Proposal Would Have Substantial PM Benefits
D. Other Criteria Pollutants: Carbon Monoxide, Nitrogen Dioxide,
Sulfur Dioxide
E. Visibility
F. Air Toxics
G. Acid Deposition
H. Eutrophication/Nitrification
I. Conclusion: Cleaner Cars and Light Trucks Are Critically
Important to Improving Air Quality
IV. What Are We Proposing and Why?
A. Why Are We Proposing Vehicle and Fuel Standards Together?
1. Feasibility of Stringent Standards for Light-Duty Vehicles
and Light-Duty Trucks
a. Gasoline Fueled Vehicles
b. Diesel Vehicles
2. Gasoline Sulfur Control Is Needed to Support the Proposed
Vehicle Standards
a. How Does Gasoline Sulfur Affect Vehicle Emission Performance?
b. How Large Is Gasoline Sulfur's Effect on Emissions?
3. A Comprehensive Vehicle/Fuel Approach Is Therefore Necessary
B. Our Proposed Program for Vehicles
1. Overview of the Proposed Vehicle Program
a. Introduction
b. Corporate Average NO_X Standard
c. Tier 2 Emission Standard ``Bins''
d. Schedules for Implementation
i. Implementation Schedule for LDVs and LLDTs
ii. Implementation Schedule for HLDTs
e. LDVs and LDTs Not Covered by Tier 2
i. Interim Standards for LDV/LLDTs
ii. Interim Standards for HLDTs
iii. Interim Programs Would Provide Reductions over Previous
Standards
iv. Alternative Approach for Interim Standards
f. Generating, Banking, and Trading NO_X Credits
2. Why Are We Proposing the Same Set of Standards for Tier 2
LDVs and LDTs?
3. Why Are We Proposing the Same Standards for Both Gasoline and
Diesel Vehicles?
4. Key Elements of the Proposed Vehicle Program
a. Basic Exhaust Emission Standards and ``Bin'' Structure
i. Why Are We Proposing Extra Bins?
b. The Proposed Program Would Phase in the Tier 2 Vehicle
Standards over Several Years
i. Primary Phase-in Schedule
ii. Alternative Phase-in Schedule
c. Manufacturers Would Meet a ``Corporate Average''
NO_X Standard.
d. Manufacturers Could Generate, Bank, and Trade NO_X
Credits.
i. General Provisions
ii. Averaging, Banking and Trading of NO_X Credits
Would Fulfill Several Goals.
iii. How Manufacturers Would Generate and Use NO_X
Credits.
iv. Manufacturers Could Earn and Bank Credits for Early
NO_X Reductions.
v. NO_X Credits Would Have Unlimited Life.
vi. NO_X Deficits Could Be Carried Forward.
e. Interim Standards
i. Interim Standards for LDV/LLDTs
ii. Interim Standards for HLDTs
f. More Stringent Proposed Light-Duty Evaporative Emission
Standards
C. Our Proposed Program for Controlling Gasoline Sulfur
1. Oil Industry Proposal
2. Why EPA Believes the Gasoline Sulfur Program must Be
Nationwide
a. Sulfur's Negative Impact on Tier 2 Catalysts Is Irreversible.
b. Sulfur Has Negative Impacts on OBD Systems and I/M Programs.
c. Sulfur Reductions Would Ensure Lower Emissions of Many
Pollutants.
d. The Refining Industry Can Control Gasoline Sulfur.
e. Other Stakeholders Support National Gasoline Sulfur Control.
3. Proposed Gasoline Sulfur Standards
a. Standards for Refiners and Importers
i. Why Begin the Program in 2004?
ii. How Did We Arrive at the 80 ppm Cap and 30 ppm Average
Standards?
iii. Should a Near-Zero Gasoline Sulfur Standard Be Considered?
iv. Why Are We Proposing Less Stringent Standards for 2004 and
2005?
b. Standards for Small Refiners
i. What Standards Would Small Refiners Have to Meet Under
Today's Proposal?
ii. Application for Small Refiner Status
iii. Application for a Small Refiner Sulfur Baseline
iv. Volume Limitation on Use of a Small Refinery Standard
v. Hardship Extensions Beyond 2007 for Small Refiners
vi. What Alternative Provisions for Small Refiners Are Possible?
4. Compliance Flexibilities
a. Sulfur Averaging, Banking, and Trading (ABT) Program
i. Why Are We Proposing a Sulfur Averaging, Banking, and Trading
Program?
ii. How Would Refiners Establish a Sulfur Baseline?
iii. How Would Refiners Generate Credits?
iv. How Would Refiners Use Credits?
v. Could Small Refiners Participate in the ABT Program?
vi. What Alternative Implementation Approaches Are Possible?
b. Refinery Air Pollution Permitting Requirements
i. New Source Review Program
ii. Title V Operating Permit Program
iii. EPA Assistance to Explore Permit Streamlining Options and
Solicitation of Comment
c. Should Hardship Relief Be Available?
5. Consideration of Diesel Fuel Control
D. What Are the Economic Impacts, Cost Effectiveness and
Monetized Benefits of the Proposal?
1. What Are the Estimated Costs of the Proposed Vehicle
Standards?
2. What Are the Estimated Costs of the Proposed Gasoline Sulfur
Standards?
3. What Are the Aggregate Costs of the Tier 2/Gasoline Sulfur
Proposal?
4. How Does the Cost-Effectiveness of this Program Compare to
Other Programs?
a. What Is the Cost Effectiveness of this Program?
b. How Does the Cost Effectiveness of this Program Compare with
Other Means of Obtaining Mobile Source NO_X + NMHC
Reductions?
c. How Does the Cost Effectiveness of this Proposed Program
Compare with Other Known Non-Mobile Source Technologies for Reducing
NO_X + NMHC?
5. Does the Value of the Benefits Outweigh the Cost of the
Proposed Standards?
a. What Is the Purpose of this Benefit-Cost Comparison?
b. What Was Our Overall Approach to the Benefit-Cost Analysis?
c. What Are the Significant Limitations of the Benefit-Cost
Analysis?
d. How Did We Perform the Benefit-Cost Analysis?
e. What Were the Results of the Benefit-Cost Analysis?
f. What Additional Efforts Will Be Made Following Proposal?
E. Other Program Design Options We Have Considered
1. Corporate Average Standards Based on NMOG or
NMOG+NO_X
2. More Stringent Tier 2 NO_X and Gasoline Sulfur
Standards
V. Additional Elements of the Proposed Vehicle Program and Areas for
Comment
A. Other Vehicle-related Elements of the Proposal
1. Proposed Tier 2 CO, HCHO and PM Standards
a. Carbon Monoxide (CO) Standards
b. Formaldehyde (HCHO) Standards
c. Particulate Matter (PM) Standards
2. Useful Life
a. Mandatory 120,000 Mile Useful Life
b. 150,000 Mile Useful Life Certification Option
3. Light Duty Supplemental Federal Test Procedure (SFTP)
Standards
4. LDT Test Weight
5. Test Fuels
6. Changes to Evaporative Certification Procedures to Address
Impacts of Alcohol Fuels
7. Other Test Procedure Issues
8. Small Volume Manufacturers
9. Compliance Monitoring and Enforcement
a. Application of EPA's Compliance Assurance Program, CAP2000
b. Compliance Monitoring

[[Page 26007]]

c. Relaxed In-Use Standards for Tier 2 Vehicles Produced During
the Phase-in Period
d. Enforcement of the Tier 2 and Interim Corporate Average
NO_X Standards.
10. Miscellaneous Provisions
B. Other Areas on Which We Are Seeking Comment
1. LDV/LDT Program Options
a. Alternatives to Address Stringency of the Standards
i. Alternative Standards and Implementation Schedules
ii. Use of Family Emission Limits (FELs) Rather than Bins
iii. Use of Different Averaging Sets
iv. Different Standards for Different Categories of Vehicles
v. Consideration of Special Provisions for the Largest LDTs and
Advanced Technology
vi. Measures to Prevent LDT Migration to Heavy-Duty Vehicle
Category
vii. Use of Non-conformance Penalties (NCPs)
viii. Additional NO_X Credits for Vehicles Certifying
to Low NO_X Levels
ix. Incentives for Manufacturers to Bank Additional Early
NO_X Credits
x. Flexibilities for Small Volume Manufacturers and Small
Businesses
xi. Adverse Effects of System Leaks
xii. Consideration of Other Corporate Averaging Approaches
2. Tighter Evaporative Emission Standards
3. Credits for Innovative VOC, NO_X and Ozone
Reduction Technologies Not Appropriately Credited by EPA's Emission
Test Procedures
4. Need for Intermediate Useful Life Tier 2 Standards
VI. Additional Proposed Elements and Areas for Comment: Gasoline
Program
A. Other Areas for Comment
1. Would States Be Preempted from Adopting Their Own Sulfur
Control Programs?
2. Potential Changes in Gasoline Distillation Properties
B. Gasoline Sulfur Program Compliance and Enforcement Provisions
1. Overview
2. What Requirements Is EPA Proposing for Foreign Refiners and
Importers?
a. What Are the Proposed Requirements for Small Foreign Refiners
with Individual Refinery Sulfur Standards?
b. What Are the Proposed Requirements for Truck Importers?
3. What Standards Would Apply Downstream?
4. What Are the Proposed Testing and Sampling Methods and
Requirements?
a. What Is the Primary Test Method for Gasoline?
b. What Is the Proposed Test Method for Sulfur in Butane?
c. Is EPA Proposing a Requirement to Test Every Batch of
Gasoline Produced or Imported?
d. What Sampling Methods Are Proposed?
e. What Are the Proposed Gasoline Sample Retention Requirements?
5. What Federal Enforcement Provisions Would Exist for
California and When Could California Test Methods be Used to
Determine Compliance?
6. What Are the Proposed Recordkeeping and Reporting
Requirements?
a. What Are the Proposed Product Transfer Document Requirements?
b. What Are the Proposed Recordkeeping Requirements?
c. What Are the Proposed Reporting Requirements?
d. What Are the Proposed Attest Requirements?
7. What Are the Proposed Exemptions for Research, Development
and Testing?
8. What Are the Proposed Liability and Penalty Provisions for
Noncompliance?
9. How Would Compliance with the Sulfur Standards Be Determined?
VII. Public Participation
A. Comments and the Public Docket
B. Public Hearings
VIII. Administrative Requirements
A. Administrative Designation and Regulatory Analysis
B. Regulatory Flexibility Act
1. Potentially Affected Small Businesses
2. Small Business Advocacy Review Panel and the Evaluation of
Regulatory Alternatives
C. Paperwork Reduction Act
D. Intergovernmental Relations
1. Unfunded Mandates Reform Act
2. Executive Order 12875: Enhancing Intergovernmental
Partnerships
3. Executive Order 13084: Consultation and Coordination with
Indian Tribal Governments
E. National Technology Transfer and Advancement Act
F. Executive Order 13045: Children's Health Protection
IX. Statutory Provisions and Legal Authority

I. Introduction

Since the passage of the 1990 Clean Air Act Amendments significant
progress has been made in reducing emissions from passenger cars and
light trucks. The National Low-Emission Vehicle (NLEV) and Reformulated
Gasoline (RFG) programs are important examples of control programs that
will continue to help reduce car and truck emissions into the near
future.
Nonetheless, due to increasing vehicle population and vehicle miles
traveled, passenger cars and light trucks will be significant
contributors to air pollution inventories into the indefinite future.
In fact, the emission contribution of light trucks and sport utility
vehicles will likely surpass that of passenger cars within the next
year. (This is occurring because of the combination of growth in miles
traveled by light trucks and their less stringent emission standards
compared to passenger cars). The program we describe below builds on
the NLEV and RFG Phase II programs to develop a strong national program
to protect public health and the environment well into the next
century. The program while reducing VOC emissions focuses especially on
NO_X because that is where the largest air quality gains can
be achieved.
We have followed several overarching principles in developing this
proposal:

Design a strong national program to assist states in every
region of the country in meeting their air quality objectives.

View vehicles and fuels as an integrated system. Define a
program that continues to ensure that car and truck emission reductions
are part of the solution to our nation's air quality problems.

Establish a single set of emission standards that apply
regardless of the fuel used and regardless of whether the vehicle is a
car or a light truck.

Provide compliance flexibilities that allow vehicle
manufacturers and oil refiners to adjust to future market trends and
honor consumer preferences.

Encourage the development of advanced low emission, fuel
efficient technologies such as lean-burn engines.

Ensure sufficient leadtime for phase-in of the Tier 2 and
gasoline sulfur program.
With these principles as background, we turn now to an overview of
the vehicle and fuel aspects of the proposal. Sections I and II of this
preamble will give you a brief overview of our proposal and the basics
of our rationale for proposing it. Subsequent sections will expand on
the air quality need, the economic impacts, and provide a more detailed
description of the specifics of the proposal. The final sections deal
with several subjects, including opportunities for public participation
that you may wish to take advantage of. You may also want to review our
Draft Regulatory Impact Analysis (RIA), found in the docket and on the
Internet, where we present more detailed analyses and discussions of
many topics raised in this preamble.

A. What Are the Basic Components of Today's Proposal?

The nation's air quality, while certainly better than in the past,
will continue to expose tens of millions of Americans to unhealthy
levels of air pollution well into the future in the absence of
significant new controls on emissions from motor vehicles. EPA is
therefore proposing a major, comprehensive program designed to
significantly reduce emissions from passenger cars and light trucks
(including sport-utility vehicles, minivans, and pickup trucks) and
reduce sulfur in gasoline. Under the proposed program, automakers would
produce vehicles designed to have very low emissions when operated on
low-sulfur gasoline, and oil refiners would

[[Page 26008]]

provide that cleaner gasoline nationwide. In this preamble, we refer to
the proposed comprehensive program as the ``Tier 2/Gasoline Sulfur
Control Program'' or simply as the ``Tier 2 Program.''
1. Vehicle Emission Standards
Today's action proposes new federal emission standards (``Tier 2
standards'') for passenger cars and light trucks. The program is
designed to focus on reducing the emissions most responsible for the
ozone and particulate matter (PM) impact from these vehicles--nitrogen
oxides (NO_X) and non-methane organic gases (NMOG),
consisting primarily of hydrocarbons (HC) and contributing to ambient
volatile organic compounds (VOC). The program would also, for the first
time, apply the same federal standards to passenger cars and all light
trucks (``light light-duty trucks'' (or LLDTs), rated at less than 6000
pounds gross vehicle weight and ``heavy light-duty trucks'' (HLDTs),
rated at more than 6000 pounds gross vehicle weight).
The proposed Tier 2 standards would reduce new vehicle
NO_X levels to an average of 0.07 grams per mile (g/mi). For
new passenger cars and light LDTs, these standards would phase in
beginning in 2004, with the standards to be fully phased in by
2007.¹ For heavy LDTs, the proposed Tier 2 standards would
be phased in beginning in 2008, with full compliance in 2009. During
the phase-in period from 2004-2007, all passenger cars and light LDTs
not certified to Tier 2 standards would have to meet an interim average
standard of 0.30 g/mi NO_X, equivalent to the current NLEV
standards for LDVs.² During the period 2004-2008, heavy LDTs
not certified to Tier 2 standards would phase in an average standard of
0.20
g/mi NO_X. Those not covered by the phase-in would be
required to meet a traditional (non-averaging) standard of 0.60 g/mi
NO_X.
---------------------------------------------------------------------------

\1\ By comparison, the NO_X standards for the National
Low Emission Vehicle (NLEV) program, which will be in place
nationally in 2001, range from 0.30 g/mi for passenger cars to 0.50
g/mi for medium-sized light trucks. For further comparison, the
standards met by today's Tier 1 vehicles range from 0.60 g/mi to
1.53 g/mi.
\2\ There are also NMOG standards associated with both the
interim and Tier 2 standards. The NMOG standards vary depending on
which of various individual sets of emission standards manufacturers
choose to use in complying with the average NO_X standard.
This ``bin'' approach is described more fully in section IV.B.
---------------------------------------------------------------------------

Manufacturers would be allowed to comply with the very stringent
proposed new standards in a flexible way, assuring that the average
emissions of a company's production met the target emission levels
while allowing the manufacturer to choose from several more- and less-
stringent emission categories for certification. The proposed
requirements also include more stringent PM standards, which primarily
affect diesel vehicles, and more stringent hydrocarbon controls
(exhaust NMOG and evaporative emissions standards).
We are also proposing stringent particulate matter standards that
would be especially important if there were substantial future growth
in diesel sales. Even under an assumed scenario where diesel sales grew
to represent 50 percent of all light-duty trucks by 2010, the PM
standards being proposed today would result in a steady decrease in
total direct PM <INF>2.5</INF> from cars and light trucks. For this
scenario of a 50 percent share for diesel light trucks, direct PM
emissions in 2020 with today's proposal would be less than they are at
present. Therefore, we believe that today's proposal accommodates
environmental concerns about such vehicles in a way that insures
positive environmental results.
2. Gasoline Sulfur Standards
The other major part of today's proposal would significantly reduce
average gasoline sulfur levels nationwide. These reductions could begin
to phase in as early as 2000, with full compliance by 2006. Refiners
would generally install advanced refining equipment to remove sulfur
during the production of gasoline. Importers of gasoline would be
required to import and market only gasoline meeting the proposed sulfur
limits. Temporary, less stringent standards would apply to a few small
refiners.
EPA is proposing that gasoline produced by refiners and sold by
gasoline importers generally meet an average sulfur standard of 30 ppm
and a cap of 80 ppm in 2004. The proposed program builds upon the
existing regulations covering gasoline content as it relates to
emissions performance. It includes provisions for trading of sulfur
credits, increasing the flexibility available to refiners for complying
with the new requirements. We intend the proposed credit program to
ease compliance uncertainties by providing refiners the flexibility to
phase in early controls in 2000-2003 and use credits gained in these
years to delay some control to as late as 2006. As proposed, the
program would achieve expected environmental benefits while providing
substantial flexibility to refiners. The effect of the credit program
is that those refiners that participate would have the opportunity for
more overall leadtime to reach the final sulfur levels.

B. What Is EPA's Statutory Authority for Proposing Today's Action?

1. Light-Duty Vehicles and Trucks
We are proposing the motor vehicle emission standards under the
authority of section 202 of the Clean Air Act. Sections 202(a) and (b)
of the Act provide EPA with general authority to prescribe vehicle
standards, subject to any specific limitations otherwise included in
the Act. Sections 202(g) and (h) specify the current standards for LDVs
and LDTs, which became effective beginning in model year 1994 (``Tier 1
standards'').
Section 202(i) of the Act provides specific procedures that EPA
must follow to determine whether standards more stringent than Tier 1
standards for LDVs and certain LDTs ³ are appropriate
beginning in the 2004 model year. ⁴ Specifically, we are
required to first issue a study regarding ``whether or not further
reductions in emissions from light-duty vehicles and light-duty trucks
should be required * * *'' (the ``Tier 2 study''). This study ``shall
examine the need for further reductions in emissions in order to attain
or maintain the national ambient air quality standards.'' It is also to
consider (1) the availability of technology to meet more stringent
standards, taking cost, lead time, safety, and energy impacts into
consideration, and, (2) the need for, and cost effectiveness of, such
standards, including consideration of alternative methods of attaining
or maintaining the national ambient air quality standards. A certain
set of ``default'' emission standards for these vehicle classes is
among those options for new standards that EPA is to consider.
---------------------------------------------------------------------------

\3\ LDTs with a loaded vehicle weight less than or equal to 3750
pounds.
\4\ Section 202(b)(1)(C) forbids EPA from promulgating mandatory
standards more stringent than Tier 1 standards until the 2004 model
year.
---------------------------------------------------------------------------

After the study is completed and the results are reported to
Congress, EPA is required to determine by rulemaking whether (1) there
is a need for further emission reductions; (2) the technology for more
stringent emission standards from the affected classes is available;
and (3) such standards are needed and cost-effective, taking into
account alternatives. If EPA answers ``yes'' to these questions, then
the Agency is to promulgate new, more stringent motor vehicle standards
(``Tier 2 standards'').
EPA submitted its report to Congress on July 31, 1998. Today's
proposal considers and proposes affirmative responses to the three
questions above (see section II below) and sets forth new proposed
standards that are more

[[Page 26009]]

stringent than the default standards in the Act.
EPA is also proposing standards for larger light-duty trucks under
the general authority of section 202(a)(1) and under section 202(a)(3)
of the Act, which requires that standards applicable to emissions of
hydrocarbons, NO_X, CO and PM from heavy-duty vehicles
⁵ reflect the greatest degree of emission reduction
available for the model year to which such standards apply, giving
appropriate consideration to cost, energy, and safety.
---------------------------------------------------------------------------

\5\ LDTs that have gross vehicle weight ratings above 6000
pounds are considered heavy-duty vehicles under the Act. See section
202(b)(3). For regulatory purposes, we refer to these LDTs as
``heavy light-duty trucks'' made up of LDT3s and LDT4s.
---------------------------------------------------------------------------

2. Gasoline Sulfur Controls
We are proposing gasoline sulfur controls pursuant to our authority
under section 211(c)(1) of the Clean Air Act.⁶ Under section
211(c)(1), EPA may adopt 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 significantly impair emissions control
systems in general use or which would be in general use were the fuel
control to be adopted.
---------------------------------------------------------------------------

\6\ We currently have regulatory requirements for conventional
and reformulated gasoline adopted under sections 211(c) and 211(k)
of the Act, in addition to the ``substantially similar''
requirements for fuel additives of section 211(f). These
requirements directly or indirectly control sulfur levels in
gasoline. See the Draft RIA for more details.
---------------------------------------------------------------------------

We are proposing to control sulfur levels in gasoline based on both
of these criteria. Under the first criterion, we believe that emissions
products of sulfur in gasoline used in Tier 1 and LEV technology
vehicles contribute to ozone pollution, air toxics, and PM. Under the
second criterion, we believe that gasoline sulfur in fuel that will be
used in Tier 2 technology vehicles will significantly impair the
emissions control systems expected to be used in such vehicles. Please
refer to section IV.C. below and to the Draft Regulatory Impact
Analysis (RIA) for more details of our analysis and findings. The Draft
RIA includes a more detailed discussion of EPA's authority to set
gasoline sulfur standards, including a discussion of our proposed
conclusions relating to the factors required to be considered under
section 211(c).

C. The Tier 2 Study and the Sulfur Staff Paper

On July 31, 1998, EPA submitted its report to Congress containing
the results of the Tier 2 study.⁷ The study indicated that
in the 2004 and later time frame, there will be a need for emission
reductions to aid in meeting and maintaining the National Ambient Air
Quality Standards (NAAQS) for both ozone and PM. Air quality modeling
showed that in the 2007-2010 time frame, when Tier 2 standards would
become fully effective, a number of areas would still be in
nonattainment for ozone and PM even after the implementation of
existing emission controls. EPA also found ample evidence that
technologies would be available to meet more stringent Tier 2
standards. In addition, the study provided evidence that such standards
could be implemented at a similar cost per ton of reduced pollutants as
other programs aimed at similar air quality problems. Finally, the
study identified several additional issues in need of further
examination, including the relative stringency of car and light truck
emission standards, the appropriateness of identical versus separate
standards for gasoline and diesel vehicles, and the effects of sulfur
in gasoline on catalyst efficiency.
---------------------------------------------------------------------------

\7\ On April 28, 1998, EPA published a notice of availability
announcing the release of a draft of the Tier 2 study and requesting
comments on the draft. The final report to Congress included a
summary and analysis of the comments EPA received.
---------------------------------------------------------------------------

In addition, on May 1, 1998, EPA released a staff paper presenting
EPA's understanding of the impact of gasoline sulfur on emissions from
motor vehicles and exploring what gasoline producers and automobile
manufacturers could do to reduce sulfur's impact on emissions. The
staff paper noted that gasoline sulfur is a catalyst poison and that
high sulfur levels in commercial gasoline could affect the ability of
future automobiles to meet more stringent standards in use. It also
pointed out that sulfur control would provide additional benefits by
lowering emissions from the current fleet of vehicles.

II. Proposed Tier 2 Determination

Based on the statutory requirements described above and the
evidence provided in the Tier 2 Study, as updated in this document, EPA
proposes its determination that new, more stringent emission standards
are indeed needed, technologically feasible, and cost effective.

A. There Is a Substantial Need for Further Emission Reductions in Order
To Attain and Maintain National Ambient Air Quality Standards

We believe that there is a clear air quality need for new emission
standards, based on the continuing air quality problems predicted to
exist in future years. As the discussion in section III.B. illustrates,
our modeling shows that in 2007 approximately 80 million Americans will
be living in areas that are in nonattainment for the 8-hour ozone
NAAQS, even with all other expected controls in place. Another 49
million people will live in attainment areas that are within 15% of
being reclassified as nonattainment areas. This is a total of nearly
130 million people, which represents about 48 percent of the population
of the United States.
In addition to these ozone concerns, our models indicate that by
2010, 45 areas, with 18 million people, will be in nonattainment for
the original PM₁₀ NAAQS and 11 areas with 10 million people
will be in nonattainment for the revised PM₁₀ NAAQS. While
not a specific driving factor in today's findings, our models also
project that 102 areas with about 55 million people will be in
nonattainment with the new PM<INF>2.5</INF> NAAQS by 2010. We also must
recognize that nonattainment areas remain for other criteria pollutants
(e.g., CO) and that non-criteria pollution (e.g., air toxics and
regional haze) also contributes to environmental and health concerns.
Clearly there is a critical need for reductions in the emissions
being projected for future years. Furthermore, mobile sources are
important contributors to the emission problem. As we will explain more
fully later in this preamble, in the year 2007, the cars and light
trucks that are the subject of today's proposal are projected to
contribute nearly 40 percent of the total NO_X and VOC
inventory in some cities, and 20 percent of nationwide NO_X
and VOC emissions. This situation would have been considerably worse
without the NLEV program created by vehicle manufacturers, EPA, the
Northeastern states, and others. We therefore believe that reductions
in these source categories are an essential part of the reductions
needed to attain and maintain the NAAQS. As we explain below, we
propose to find that major reductions in future emissions from light-
duty vehicles and trucks are both feasible and cost effective compared
to available alternatives.

[[Page 26010]]

B. More Stringent Standards for Light-Duty Vehicles and Trucks Are
Technologically Feasible

We believe that emission standards more stringent than current Tier
1 and National Low Emission Vehicle (NLEV) levels are technologically
feasible. We believe this to be true both for the LDVs and LDTs
specifically covered in section 202(i) and for the heavier LDTs also
included in today's proposal. Manufacturers are currently producing
NLEV vehicles that meet more stringent standards than similar Tier 1
models. Our analysis shows that mainly through improvements in engine
control software and catalytic converter technology, manufacturers can
and are building durable vehicles and trucks, including heavy light-
duty trucks, which have very low emission levels.⁸
---------------------------------------------------------------------------

\8\ The Draft RIA contains an extended analysis, Section IV.A.
below has more discussion of the technological feasibility of our
proposed standards including detailed discussions of the various
technology options that we believe manufacturers may use to meet
these standards.
---------------------------------------------------------------------------

For light duty vehicles, certified NO_X levels for 1999
reveal that NO_X levels representing full-life, deteriorated
emissions in the 0.01 to 0.10 g/mi range are already being seen on some
production vehicles. Similarly, light-duty trucks up to 8500 lbs. GVWR,
also included in today's proposal, have some very low 1999
certification levels for NO_X, with NO_X levels of
as low as 0.04 g/mi for some of the largest LDTs. These levels are well
below Tier 1 and NLEV standards. Manufacturers have also certified LDVs
and LDTs to NMOG and CO levels as much as 80 percent below Tier 1
standards.
As discussed in more detail below and in the Draft RIA, we believe
that, by the 2004-2009 time frame proposed for the Tier 2 standards,
manufacturers would be fully able to comply with the proposed new
standard levels. In addition, to facilitate manufacturers' efforts to
meet these new standards, the Tier 2 regulations would include a
corporate fleet average, which would allow manufacturers to optimize
the deployment of technology across their product lines. Our analysis
of the available technology improvements and the very low emission
levels already being realized on these vehicles leads us to propose a
finding that today's proposed standards are fully feasible for LDVs and
LDTs.

C. More Stringent Standards for Light-Duty Vehicles and Trucks Are
Needed and Cost Effective Compared to Available Alternatives

In this document, we propose that Tier 2 motor vehicle standards
are both necessary and cost effective. We have already described our
belief that substantial further reductions in emissions are needed to
help reduce the levels of unhealthy air pollution that millions of
people are being exposed to. (We describe this further below and in the
Draft RIA.) In its analyses supporting the new ozone and PM NAAQS, the
Agency identified those methods that were reasonably cost effective,
and showed that substantial progress toward attainment could be made.
However, we also concluded that methods beyond those that could be
identified as cost effective at the time were needed and we assumed
they would be identified in the future.
We believe that the Tier 2/gasoline sulfur proposal is one of those
methods. This proposal would reduce annual NO_X emissions by
about 2.2 million tons per year in 2020 and 2.8 million tons per year
in 2030 after the program is fully implemented. By way of comparison,
if all of the controls identified for the NAAQS analysis costing less
than $10,000/ton (the limit on cost effectiveness used in that
analysis) were implemented nationwide, they would produce
NO_X emission reductions of about 2.9 million tons per year.
That is, to achieve significant further reductions using control
approaches other than the proposed Tier 2/Gasoline Sulfur program could
mean adopting measures costing well beyond $10,000 per ton.
Further emission reductions are needed. Without Tier 2 and gasoline
sulfur controls, we project that in 2007 at least 8 metropolitan areas
and 2 rural counties with a combined population of 39 million will
exceed the 1-hour ozone NAAQS and 28 metropolitan areas and 4 rural
counties with a combined population of 80 million will exceed the 8-
hour ozone NAAQS. We project that cars and light trucks will contribute
17 percent of the nationwide NO_X inventory by 2007 and 20 to
40 percent in some cities with air quality problems. The NO_X
reductions from today's proposal range from 19 to 48 percent of the
reductions we estimate are needed for areas to achieve attainment. We
believe that the proposed program, as well as the technologies assumed
for the NAAQS analysis mentioned above, are clearly cost effective
approaches for attaining and maintaining the NAAQS.
The magnitude of emission reductions that can be achieved by a
comprehensive national Tier 2/gasoline sulfur program would be
difficult to achieve from any other source category. Given the
contribution that light-duty mobile source emissions make to the
national emissions inventory and the range of control programs ozone-
affected areas already have in place or would be expected to implement,
we believe it will be very difficult, if not impossible, to attain and
maintain the ozone NAAQS in a cost-effective manner without reducing
emissions from LDVs and LDTs. In addition, we project that the Tier 2/
gasoline sulfur program would reduce direct and secondary particulate
matter coming from LDVs and LDTs by over 70 percent, providing
reductions of almost 240,000 tons annually by 2010.
We believe, then, that today's proposal is a major and attractive
source of ozone and PM precursor emission reductions when compared to
other available options. It would represent a degree of emission
reduction beyond those programs identified in the NAAQS analysis that
we believe is currently unavailable from any other reasonable program.
We also believe that it would be a cost effective program, costing
approximately $2,000 per ton of NO_X plus hydrocarbon reduced
according to our estimates, which is quite attractive compared to other
alternatives. The discussion of cost and cost effectiveness later in
this preamble explains the derivation of these numbers and compares
them to other alternatives. That discussion indicates that today's
proposal would be as cost effective as both the Tier 1 and NLEV
standards and cost effective when compared to non-mobile source
programs as well.

III. Air Quality Need for and Impact of Today's Proposal

In the absence of significant new controls on emissions, tens of
millions of Americans would continue to be exposed to unhealthy levels
of air pollution. Emissions from passenger cars and light trucks are a
significant contributor to a number of air pollution problems. Today's
proposal would significantly reduce emissions from cars and light
trucks and hence would significantly reduce the health risks posed by
air pollution. This section summarizes the results of the analyses we
performed to arrive at our proposed determination that continuing air
quality problems are likely to exist, that these air quality problems
would be in part due to emissions from cars and light trucks, and that
the new standards being proposed today would improve air quality and
mitigate other environmental problems.

[[Page 26011]]

A. Americans Face Serious Air Quality Problems That Require Further
Emission Reductions

Air quality in the United States continues to improve. Nationally,
the 1997 air quality levels were the best on record for all six
criteria pollutants.⁹ In fact, the 1990s have shown a steady
trend of improvement, due to reductions in emissions from most sources
of air pollution, from factories to motor vehicles. Despite these
continued improvements in air quality, however, tens of millions of
Americans are still exposed to unhealthy levels of ozone and PM.
Moreover, unless there are reductions in overall emissions beyond those
that are scheduled to be achieved by already committed controls, many
of these Americans will continue to be so exposed.
---------------------------------------------------------------------------

\9\ National Air Quality and Emissions Trend Report, 1997, Air
Quality Trends Analysis Group, Office of Air Quality Planning and
Standards, U.S. Environmental Protection Agency, Research Triangle
Park, N.C., December 1998 (available on the World Wide Web at http:/
/www.epa.gov/oar/aqtrnd97/).
---------------------------------------------------------------------------

Ambient ozone is formed in the atmosphere through a complex
interaction of VOC and NO_X emissions. Cars and light trucks
emit a substantial fraction of these emissions. Ambient PM is emitted
directly from cars and light trucks; it also forms in the atmosphere
from NO_X, sulfur oxides (SO<INF>X</INF>) and VOC, all of
which are emitted by motor vehicles. When ozone exceeds the air quality
standards, otherwise healthy people often have reduced lung function
and chest pain, and hospital admissions for people with respiratory
ailments like asthma increase; for longer exposures, permanent lung
damage can occur. Similarly, particles can penetrate deep into the
lungs and are linked with premature death, increased hospital
admissions, increased respiratory symptoms, and changes in lung tissue.
When either ozone or PM air quality problems are present, those hardest
hit tend to be children, the elderly, and people who already have
health problems.
The health effects of high ozone and PM levels are not the only
reason for concern about continuing air pollution. Ozone and PM also
harm plants and damage materials. PM reduces visibility and contributes
to significant visibility impairment in our national parks and
monuments and in many urban areas. In addition, air pollution from
motor vehicles contributes to cancer and other health risks,
acidification of lakes and streams, eutrophication of coastal and
inland waters, and elevated drinking water nitrate levels. These
problems impose a substantial burden on public health, our economy, and
our ecosystems.
In recognition of this burden, Congress has passed and subsequently
amended the Clean Air Act. The Clean Air Act requires each state to
have an approved State Implementation Plan (SIP) that shows how an area
plans to meet its air quality obligations, including achieving and then
maintaining attainment of all of the National Ambient Air Quality
Standards (NAAQS), such as those for ozone and PM.
Under EPA's proposed policy for implementing the new 8-hour ozone,
revised PM₁₀, and new PM<INF>2.5</INF> ambient standards (63
FR 65593, November 27, 1998), states must prepare and submit SIP
revisions to demonstrate attainment of the 8-hour ozone standard
between 2000 and 2003, depending on ozone classification under the 8-
hour standard. The earlier submittal date applies to ``transitional''
areas, which are areas that are in attainment with the 1-hour standard
and can attain the 8-hour standard through local measures adopted prior
to classification (under the 8-hour standard) and the regional emission
reductions to be achieved under the Regional Ozone Transport Rule (63
FR 57356, October 27, 1998). In general, EPA expects these areas to
demonstrate attainment by 2007. Other 8-hour nonattainment areas will
be classified as ``traditional'' under the 8-hour standard, and we
believe that these areas will have attainment dates of 2007, 2009, or
2010 depending on their 1-hour classification status and 1-hour
attainment date.
Because it takes three ``clean'' years to qualify an area to be
redesignated as attainment for the ozone standard, the deadline for
each area to achieve the VOC and NO_X emission reductions
needed to meet the ozone standard generally should be two years earlier
than its attainment date. For example, 8-hour ozone nonattainment areas
for which we would establish an attainment date of 2009 would need to
implement emission reductions by the start of the 2007 ozone season in
order to have three ``clean'' years by their 8-hour attainment deadline
of 2009.
The SIP revisions to demonstrate attainment with the revised
PM₁₀ standard must be prepared by 2002, with attainment by
2006, unless this date is not practicable. As discussed below, EPA has
also finalized regulations that regions and states implement plans for
protecting and improving visibility in the 156 mandatory Federal Class
I areas as defined in section 162(a) of the Clean Air Act. These areas
are primarily national parks and wilderness areas.
To accomplish the goal of full attainment in all areas according to
the schedules for the various NAAQS and the visibility program, the
federal government must assist the states by reducing emissions from
sources that are not as practical to control at the state level as at
the federal level. Vehicles and fuels move freely among the states, and
they are produced by national or global scale industries. Most
individual states are not in a position to regulate these industries
effectively and efficiently. The Clean Air Act therefore gives EPA
primary authority to regulate emissions from the various types of
highway vehicles and their fuels. Our actions to reduce emissions from
these and other national sources are a crucial and essential complement
to actions by states to reduce emissions from more localized sources.
If we do not adopt new standards to reduce emissions from cars and
light trucks, emissions from these vehicles would remain a large
portion of the emissions burden that causes elevated ozone and
continued nonattainment with the ozone NAAQS, which in turn affects
tens of millions of Americans. Without new standards, steady annual
increases in fleet size and miles of travel will outstrip the benefits
of current emission controls, and will cause ozone-forming emissions
from cars and trucks to grow each year starting about 2014. The
contribution of these vehicles to PM exposure and PM nonattainment
would also remain significant, and could increase considerably if
diesel engines are used in more cars or light trucks. For ozone in
particular, the contribution of cars and light trucks--in terms of both
local emissions and transported pollution--will be so significant to
those areas expected to be in nonattainment in the 2007 to 2010 time
frame, and the expected emission reduction shortfall in these areas
will be so large, that further reductions from cars and light trucks
are an inescapable element of any attainment strategy.
The standards we are proposing would cut the contribution of ozone
and PM precursors from cars and light trucks greatly. Even with this
cut, many areas will likely still find it necessary to obtain
additional reductions from other sources in order to fully attain the
ozone and PM NAAQS. However, their task would be easier and the
economic impact on their industries and citizens would be lighter as a
result of the actions proposed today. This would be a critical benefit
of today's proposal. Following implementation of the Regional Ozone
Transport Rule, states

[[Page 26012]]

will have already adopted emission reduction requirements for nearly
all large sources of VOC and NO_X for which cost-effective
control technologies are known. Those that remain in nonattainment will
therefore have to consider their other alternatives. In fact, however,
many of the alternatives states will have to consider are very costly,
with a small impact from each additional category subjected to new
emission controls. The emission reductions from today's proposed
standards for gasoline, cars, and light trucks would ease the need for
states to find first-time reductions from the mostly smaller sources
that have not yet been controlled, including area sources that are
closely connected with individual and small business activities. They
would also reduce the need for states to seek even deeper reductions
from large and small sources already subject to emission controls.
In our meetings and correspondence with state and local officials,
they asked us to reduce the emissions from cars and trucks, so that
their charge of protecting the public against air pollution is one they
can accomplish on schedule and without adverse economic impacts. We
heard from the Northeast States for Coordinated Air Use Management, the
Ozone Transport Commission, the State and Territorial Air Program
Administrators, and the Association of Local Air Pollution Control
Officers. They consistently told us that it would be very difficult and
costly for the states to obtain comparable reductions from other
sources as substitutes for reductions from cars and light trucks,
especially on top of the additional reductions needed to reach ozone
attainment even with the reductions from today's proposal.
We project that today's proposal would also have important benefits
for regional visibility, acid rain, and coastal water quality.
For these and other reasons discussed in this document, we are
proposing to determine that significant emission reductions will still
be needed by the middle of the next decade and beyond to achieve and
maintain further improvements in air quality in many, geographically
dispersed areas. We also believe that a significant portion of these
emission reductions can be obtained by reducing emissions from cars and
light trucks. We believe that such reductions are in fact necessary
(since cars and light trucks are such large contributors to current and
projected ozone problems) and reasonable (since these reductions could
be achieved at a reasonable cost compared to other alternative
reductions).
The remainder of this section describes the health and
environmental problems that today's proposal would help mitigate and
the expected health and environmental benefits of this proposal. Ozone
is discussed first, followed by PM, other criteria pollutants,
visibility, air toxics, and other environmental impacts. The emission
inventories and air quality analyses are explained more fully in the
Draft Regulatory Impact Analysis for today's proposal.

B. Ozone

1. Ozone Levels Have Declined, but Unhealthy Levels of Ozone Persist
Ground-level ozone is the main harmful ingredient in
smog.¹⁰ It is produced by complex chemical reactions when
its precursors, VOC and NO_X, react in the presence of
sunlight. The chemical reactions that create ozone take place while the
wind is carrying the pollutants, which means that ozone can be more
severe many miles away from the source of ozone-forming emissions than
it is at the source. The movement of ozone and its precursors is called
``ozone transport'' and suggests two complementary approaches to reduce
ozone levels in areas affected by ozone transport:
---------------------------------------------------------------------------

\10\ Ozone also occurs naturally in the stratosphere and
provides a protective layer high above the earth.
---------------------------------------------------------------------------

(1) Reduce ozone precursor emissions in the area itself.
(2) Reduce ozone precursor emissions in upwind areas to reduce
incoming ozone and ozone precursor levels.
Within a nonattainment area itself, both VOC and NO_X
reductions are generally beneficial. Especially in the eastern portion
of the U.S., the second approach of controlling upwind emissions can
play an important part in efforts to reduce ozone levels in
nonattainment areas. Because individual states cannot control upwind
sources of air pollution that lie outside their borders, EPA has a
special role in managing transport impacts. Vehicle and fuel standards
should play a part in doing so.
Since NO_X affects downwind ozone levels in the eastern
U.S. over greater distances than VOC does, reductions in upwind
NO_X emissions are particularly important in reducing ozone
levels downwind. Modeling conducted by the Ozone Transport Assessment
Group, discussed below, indicates that VOC reductions substantially
upwind from nonattainment areas have little benefit in those
nonattainment areas across the eastern region of the U.S. By contrast,
VOC reductions in or near nonattainment areas do provide air quality
benefits. Since cars and light trucks meeting today's proposed
standards would operate everywhere, today's proposal would reduce VOC
and NO_X emissions in both nonattainment areas and in upwind
areas.
The new standards being proposed today would have their largest
effect on NO_X emissions. Sulfur in gasoline has been found
to increase NO_X emissions more than VOC emissions, and
reducing sulfur would therefore yield larger NO_X reductions
than VOC reductions. Similarly, the vehicle standards proposed today
represent a greater reduction from current NO_X standards
than is the case for VOC. We have taken this approach because air
quality modeling conducted for OTAG, and subsequent modeling we have
conducted, indicates that NO_X reductions would have larger
ozone benefits than would VOC reductions. In addition, we believe that
individual nonattainment areas have a wider range of alternative
control opportunities for VOC than they have for NO_X.
Ozone levels have decreased significantly over the past 20 years as
VOC and NO_X emissions have been reduced. However, ozone
levels in much of the country remain a major concern. Outside of
California, the 1990 census showed 72 million people living in areas
that were formally designated as non-attainment for the 1-hour standard
as of August 10, 1998. Measured ozone design values from 1995 to 1997
in the region analyzed by the Ozone Transport Assessment Group (OTAG)
¹¹ indicate that in this region alone, 26 metropolitan areas
and 8 rural counties together containing 75 million people experienced
ozone levels in excess of the 1-hour ozone standard.
---------------------------------------------------------------------------

\11\ OTAG evaluated a region that included all or part of the
easternmost 37 states.
---------------------------------------------------------------------------

The 8-hour ozone standard is more stringent and protective than the
1-hour standard, and more areas have exceeded it in the recent past. In
1995 to 1997, at least one county in each of 81 metropolitan areas and
an additional 30 rural counties together containing 110 million people
had ozone values in excess of the 8-hour ozone standard. Additional
areas in the OTAG region had ozone levels within 15 percent of the 8-
hour standard and hence faced potentially significant maintenance
challenges: 52 metropolitan areas and 44 rural counties together
containing 26 million people.
For several reasons, we expect to see substantial additional
progress in

[[Page 26013]]

reducing ozone levels over the next ten years despite continued growth
in electric power generation, industrial output, nonroad activity
levels, and vehicle miles traveled. NO_X and VOC emissions
from mobile sources will continue to decline as older, higher-emitting
vehicles and nonroad engines are retired from service and replaced with
newer vehicles and nonroad engines that must meet more stringent
federal emission standards. Other federal regulations that will reduce
ozone precursor emissions will take effect, such as regulations that
will reduce VOC emissions from paints and other architectural coatings.
Beginning in 2000, areas of the country participating in the federal
reformulated gasoline program will receive lower-emitting Phase 2
reformulated gasoline. States are expected to implement additional
measures to reduce NO_X and VOC emissions in 1-hour ozone
nonattainment areas. In addition, the final Regional Ozone Transport
Rule (ROTR) (63 FR 57356, October 27, 1998) requires the District of
Columbia and 22 states in the eastern U.S. to reduce their
NO_X emissions substantially by 2003 to reduce ozone levels
in downwind states.
Using the most recent improvements to the OTAG emission inventories
and the OTAG ozone model, we project that in the OTAG region, these
combined emission reductions will bring 18 of the aforementioned 26
metropolitan areas and 6 of the 8 rural counties, with 36 million
residents, into attainment with the 1-hour ozone standard by 2007. The
same emission reductions are projected to bring ozone design values
below the 8-hour standard in 53 out of 81 metropolitan areas and 26 out
of 30 rural counties, with a combined 1990 population of 30 million
people.¹²
---------------------------------------------------------------------------

\12\ The design value is the calculated ozone level, based on
ozone measurements in the area, that is compared to the NAAQS to
determine compliance with the standard.
---------------------------------------------------------------------------

However, we still project many areas in the OTAG region to have
ozone design values in 2007 in excess of the 1-hour and 8-hour
standards. Eight metropolitan areas and two counties with a combined
1990 population of 39 million are projected to experience ozone design
values in excess of the 1-hour ozone standard in 2007.¹³
Twenty-eight areas and 4 rural counties, with a combined 1990
population of 80 million, are projected to experience ozone design
values at levels in excess of the 8-hour standard in 2007.
---------------------------------------------------------------------------

\13\ Various states have submitted SIPs to meet a requirement
that they demonstrate attainment with the 1-hour ozone standard by
2005 or 2007 (the exact date is state-specific, depending on the
severity of their violation of the 1-hour standard). These plans
were submitted to EPA in the first half of 1998, and we are still
reviewing them for their completeness and approvability. We have not
fully evaluated the impact of the measures contained in these plans
on future ozone levels. As a result, they are not included in the
baseline emission inventory.
---------------------------------------------------------------------------

Additional areas outside the OTAG modeling region may also
experience high ozone levels, even with the additional emission
controls that will be implemented by 2007. The most recent assessment
for these areas was made in the Regulatory Impact Analysis for the
revised NAAQS (NAAQS RIA).¹⁴ That assessment predicted that
many areas in California will require substantial additional reductions
to attain the 1-hour and 8-hour ozone standards. Although the vehicle
and fuel standards being proposed today would not apply to vehicles and
fuel sold in California, we project that today's proposals would lead
to emission reductions within California. According to the State of
California, about 7 to 10 percent of all car and light truck travel in
California takes place in vehicles originally sold outside California.
These vehicles operate in California during visits and after relocation
of households from other states. Today's proposal would cause those
vehicles to be cleaner, assisting California's nonattainment areas to
meet the ozone standards. In addition, this proposal requires that
gasoline in all states (except California, which has its own low-sulfur
gasoline program) have a low sulfur content, in order to maintain
catalyst effectiveness. This would ensure that vehicles belonging to
California residents get clean gasoline when they travel outside of
California, so that they return to California with fully functioning
catalysts.
---------------------------------------------------------------------------

\14\ ``Regulatory Impact Analyses for the Particulate Matter and
Ozone National Ambient Air Quality Standards and Proposed Regional
Haze Rule,'' Innovative Strategies and Economics Group, Office of
Air Quality Planning and Standards, U.S. Environmental Protection
Agency, Research Triangle Park, NC, July 17, 1997.
---------------------------------------------------------------------------

Outside of California and the OTAG region, the NAAQS RIA modeling
indicated that all areas would attain the 1-hour standard by 2010. One
area (Phoenix, AZ) was projected not to attain the 8-hour standard.
Eleven other areas were projected to have ozone levels within 15
percent of the 8-hour standard and hence face potential challenges in
maintaining their attainment status.
Furthermore, even an area now in attainment or that reaches
attainment by 2007 can be at risk of becoming nonattainment in the face
of continued growth in its population, economy, vehicle traffic, and
nonroad equipment activity levels. Also, an area that we have estimated
will reach attainment in 2007 may fail to do so if growth is higher
than we project, if emission controls are less effective, or if the
modeling is otherwise in error. Our modeling for the OTAG region has
estimated that of the 1-hour nonattainment areas projected to reach
attainment by 2007 with the benefits of the Regional Ozone Transport
Rule (ROTR) and other already committed measures, 17 metropolitan areas
and 5 rural counties, with a combined 1990 population of 35 million
people, will remain within 15 percent of the 1-hour standard. These
areas would benefit from additional reductions to help ensure that they
will attain.
With respect to the 8-hour standard, we estimate that 80
metropolitan areas and 39 rural counties with a 1990 population of 49
million people will have design values within 15 percent of the 8-hour
standard. These areas have some risk of not actually being in
attainment in 2007, and will face potentially significant challenges
maintaining their attainment status in future years. Today's proposed
standards would help ensure these areas do attain, and help these areas
accommodate continued population and economic growth while staying in
attainment with the 8-hour ozone standard by further reducing levels of
ozone precursors.
EPA's best ozone projections at the current time for the OTAG
region are summarized in Tables III-1 and III-2, where ``ROTR'' refers
to the Regional Ozone Transport Rule. It should be noted that the
results for the OTAG regions discussed above and summarized in the
following tables apply to only a portion of the area that would benefit
from today's proposal.

Table III-1.--Extent of Potential 1-Hour Ozone Problem Areas in 2007 in
the OTAG Region.^a
------------------------------------------------------------------------
2007
projections
with ROTR
------------------------------------------------------------------------
Design values in excess of the 1-Hour NAAQS (<gr-thn-eq>125 ppb)
------------------------------------------------------------------------
Number of Metropolitan Areas............................ 8
Number of Rural Counties................................ 2
1990 Population of Metropolitan Areas and Rural Counties 39
(millions).............................................
------------------------------------------------------------------------
^a Additional potential problem areas in California.

[[Page 26014]]

Table III-2.--Extent of Potential 8-Hour Ozone Problem Areas in 2007 in
the OTAG Region ^a
------------------------------------------------------------------------
2007
projections
with ROTR
------------------------------------------------------------------------
Design values in excess of the 8-Hour NAAQS (<gr-thn-eq>85 ppb)
------------------------------------------------------------------------
Number of Metropolitan Areas............................ 28
Number of Rural Counties................................ 4
1990 Population of Metropolitan Areas and Rural Counties 80
(millions).............................................
------------------------------------------------------------------------
Design values within 15 percent of the 8-Hour NAAQS (72-84 ppb)
------------------------------------------------------------------------
Number of Metropolitan Areas............................ 80
Number of Rural Counties................................ 39
1990 Population of Metropolitan Areas and Rural Counties 49
(millions).............................................
------------------------------------------------------------------------
^a Phoenix, Arizona and multiple areas in California are also potential
problem areas.

It should be noted that the areas included in Table III-2 have not
been designated to be in nonattainment with the 8-hour ozone NAAQS.
Such designations will not be made by EPA until 2000, and these
designations will be based on the data that are most recently available
at that time.¹⁵ Instead, the areas included in Table III-2
have been projected to have design values that would place them in
nonattainment in 2007, using an approach described in the Draft
RIA.¹⁶ This approach enabled EPA to estimate the extent of
the 8-hour nonattainment problem after implementing the reductions set
forth in the Regional Ozone Transport Rule and the measures states have
adopted or are specifically required by the Clean Air Act to adopt for
their existing 1-hour nonattainment areas. (The modeling did not
consider the impact of additional measures that may appear in the SIP
revisions submitted by some states in the first half of 1998.)
---------------------------------------------------------------------------

\15\ It should also be noted that the number and 1990 population
of metropolitan areas projected to be near or above the 8-hour ozone
standard in Table III-2 are based on the boundaries of ozone
nonattainment areas as currently defined under the 1-hour ozone
standard. These boundaries will be reevaluated as 8-hour ozone
nonattainment areas are designated and may change from those used
above, affecting the count and population of the potential problem
areas.
\16\ The approach uses a combination of ambient monitoring data
and regional ozone photochemical grid modeling for specific ozone
episodes to develop statistical correlations between modeled ozone
levels and projected future monitoring results. The approach does
not reflect any further emission reductions that may have been
included in revisions to State Implementation Plans (SIPs) for ozone
that EPA received from some states in the first half of 1998. These
SIP revisions are still under review by EPA for completeness and
approvability.
---------------------------------------------------------------------------

We believe the large reductions called for in today's action would
substantially reduce ozone levels nationwide and would therefore reduce
ozone levels and design values in the areas projected to otherwise
exceed the 8-hour standard as well as in those areas facing potentially
significant maintenance challenges.
2. Cars and Light-Duty Trucks Are a Big Part of the NO_X and
VOC Inventory, and Today's Proposal Would Reduce This Contribution
Substantially
Emissions of VOCs and NO_X come from a variety of
sources, both natural and from human activity. Natural sources,
including emissions that have been traced to vegetation, account for a
substantial portion of total VOC emissions in rural areas. The
remainder of this section focuses on the contribution of motor vehicles
to emissions from human sources. Human-caused VOCs are released as
byproducts of incomplete combustion as well as evaporation of solvents
and fuels. For gasoline-fueled cars and light trucks, approximately
half of the VOC emissions come from the vehicle exhaust and half come
from the evaporation of gasoline from the fuel system. NO_X
emissions are dominated by human sources, most notably high-temperature
combustion processes such as those occurring in automobiles and power
plants. Emissions from cars and light trucks are currently, and will
remain, a major part of nationwide VOC and NO_X emissions. In
1996, cars and light trucks comprised 25 percent of the VOC emissions
and 21 percent of the NO_X emissions from human sources in
the U.S.¹⁷ The contribution in metropolitan areas was
generally larger.
---------------------------------------------------------------------------

\17\ Emission Trend Report, 1997.
---------------------------------------------------------------------------

Motor vehicle emission controls have led to significant
improvements in emission levels in the air (the ``emission inventory'')
and will continue to do so in the near term. As a result of the
introduction of cleaner reformulated gasoline in 2000, the introduction
of National Low Emission Vehicles (NLEVs) and vehicles complying with
the Enhanced Evaporative Test Procedure and Supplemental Federal Test
Procedures, and the continuing removal of older, higher-emitting
vehicles from the in-use vehicle fleet, total emissions from the car
and light truck fleet are projected to continue to decline through the
next decade, reaching a low point for NO_X in 2013 (Figure
III-1) and for VOC in 2015.¹⁸ On a per mile basis, average
VOC and NO_X emissions from cars and light trucks combined
will continue to decline well beyond 2015, reflecting the continuing
effect of existing emission control programs. However, projected
increases in vehicle miles traveled (VMT) will cause total emissions
from these vehicles to increase. With this increase in travel and
without additional controls, we project that combined NO_X
and VOC emissions for cars and light trucks will increase starting in
2013 and 2015, respectively, so that by 2030 they will have returned to
levels nearly the same as they will be in 2000. In cities experiencing
rapid growth, such as Charlotte, North Carolina, the near-term trend
toward lower emissions tends to reverse sooner.
---------------------------------------------------------------------------

\18\ The auto manufacturer and northeastern state commitments to
the NLEV program are scheduled to end in 2004 without further EPA
action on Tier 2 standards, although continued voluntary compliance
by automobile manufacturers and the affected states is a
possibility. Our analysis of emission trends and the emission
benefits expected from today's proposal assumes for the base
scenario a continuation of the NLEV program past 2004. It also
includes all other control measures assumed to be implemented for
the purposes of the proposed state-level NO_X budgets in
the Regional Ozone Transport Rule, such as reformulated gasoline in
all required and opt-in areas and enhanced I/M where required.
---------------------------------------------------------------------------

Figure III-1 illustrates this expected trend in car and light truck
NO_X emissions in the absence of today's proposed standards
for vehicles and gasoline. The figure also allows the contribution of
cars to be distinguished from that of light trucks. The figure clearly
shows the impact of steady growth in light truck sales and travel on
overall light-duty NO_X emissions; the decrease in overall
light-duty emission levels is due solely to reductions in LDV
emissions. In 2000, we project that trucks will produce about 50
percent of combined car and light truck NO_X emissions. We
project that truck emissions will actually increase after 2000, and
over the next 30 years, trucks will grow to dominate light-duty
NO_X emissions. By 2007, we project trucks will make up two-
thirds of light-duty NO_X emissions; by 2020, nearly three-
quarters of all light-duty NO_X emissions will be produced by
trucks.

BILLING CODE 6560-50-P

[[Page 26015]]

[GRAPHIC] [TIFF OMITTED] TP13MY99.000

BILLING CODE 6560-50-C

[[Page 26016]]

Today's action would significantly decrease NO_X and VOC
emissions from cars and light trucks, and would delay the date by which
NO_X and VOC emissions would begin to increase due to
continued VMT growth. With Tier 2/Sulfur control, light-duty vehicle
NO_X and VOC emissions are projected to continue their
downward trend past 2020. Table III-3 shows the annual tons of
NO_X that we project would be reduced if today's proposal
were adopted.¹⁹ These projections include the benefits of
low sulfur fuel and the introduction of Tier 2 car and light truck
standards.
---------------------------------------------------------------------------

\19\ Today's proposed standards for both vehicles and fuels
would apply in 49 states and the U.S. territories, excluding only
California. If today's proposal is adopted, there would also be
emissions reductions in California from vehicles that relocate or
visit from other states. However, much of the emissions inventory
analysis for this proposal was made for a 47-state region that
excludes California, Alaska, and Hawaii, since these states were not
included in the scope of ozone modeling.

Table III-3.--NO_X Emissions From Cars and Light Trucks as Percent of Total Emissions, and Reductions Due to Tier
2/Sulfur Control ^a
----------------------------------------------------------------------------------------------------------------
Light-duty
Light-duty percent of Light-duty
Year tons without total without tons reduced
tier 2 tier 2 by tier 2 ^b
(percent)
----------------------------------------------------------------------------------------------------------------
2007............................................................ 3,218,530 17 795,734
2010............................................................ 3,041,639 17 1,182,323
2015............................................................ 3,020,806 17 1,778,881
2020............................................................ 3,221,151 18 2,198,113
----------------------------------------------------------------------------------------------------------------
^a Estimates exclude California, Alaska, and Hawaii, although reductions would occur in all three. For all cases,
this table reflects implementation of ROTR and other measures assumed in the ROTR. For the ``Without Tier 2''
case, the estimates reflect continuation of NLEV beyond 2004.
^b Does not include emission reductions from heavy-duty gasoline vehicles.

The lower sulfur levels proposed today would produce large emission
reductions on pre-Tier 2 vehicles as soon as low-sulfur gasoline is
introduced, in addition to enabling Tier 2 vehicles to achieve lower
emission levels. Among the pre-Tier 2 vehicles, the largest per vehicle
emission reductions from lower sulfur in gasoline would be achieved
from vehicles that automobile manufacturers will have sold under the
voluntary National Low Emission Vehicle program. These vehicles are
capable of substantially lower emissions when operated on low sulfur
fuel. Older technology vehicles experience a smaller but significant
effect.
In 2007, when all gasoline would meet the new sulfur limit and when
large numbers of 2004 and newer vehicles meeting the proposed standards
would be in use, the combined NO_X emission reduction from
vehicles and fuels would be nearly 800,000 tons per year. After 2007,
emissions would be reduced further as the fleet turned over to Tier 2
vehicles operating on low sulfur fuel. By 2020, NO_X
emissions would be reduced by two-thirds from the levels that would
occur if today's proposal were not adopted. This reduction equals the
NO_X emissions from over 166 million pre-Tier 2/Sulfur cars
and light trucks. This reduction would represent a 12 percent
NO_X reduction in emissions from all manmade sources.
VOC emissions would also be reduced by today's proposal, with
reductions increasing as the fleet turns over. The reductions as a
percent of emissions from cars and light trucks would be 5 percent in
2007 and grow to 16 percent in 2020.
As discussed earlier, in California, smaller but still substantial
reductions in both NO_X and VOC would be achieved because
vehicles visiting and relocating to California would be designed to
meet today's proposed standards. Also, vehicles from California
visiting other states would not be exposed to high sulfur fuel.
These estimates of emission reductions reflect a mixture of urban,
suburban, and rural areas. As we noted in the Tier 2 Study, however,
cars and light trucks generally make up a larger fraction of the
emission inventory for urban and suburban areas, where human population
and personal vehicle travel is more concentrated than emissions from
other sources such as heavy-duty highway vehicles, power plants, and
industrial boilers. We have estimated emission inventories for three
cities using the same methods as were used to project the nationwide
inventories, and we present the results for 2007 below in Table III-4.
Inventory shares in 2010 are about the same.
These results confirm that light-duty vehicles make up a greater
share of the NO_X emission inventories in urban areas than
they do in the nationwide inventory. While these vehicles' share of
national NO_X emissions in 2007 is about 17 percent, it is
estimated to be about 38 percent in the Atlanta area. There is also a
range in VOC contributions, with Atlanta again being the area with the
largest car and light truck contribution at 33 percent. In metropolitan
areas with high car and light truck contributions, today's proposal
would represent a larger step toward attainment since it would have a
larger effect on total emissions.

Table III-4.--Proportion of the Total Urban Area NO_X and VOC Inventory
in 2007 Attributable to Light-Duty Vehicles ^a
------------------------------------------------------------------------
NO_X VOC
Region (percent) (percent)
------------------------------------------------------------------------
Nationwide ^a...................................... 17 18
New York urban area............................... 29 15
Atlanta urban area................................ 38 33
Charlotte urban area.............................. 18 15
------------------------------------------------------------------------
^a For all cases, this table reflects implementation of ROTR and other
measures assumed in the ROTR. The estimates reflect continuation of
NLEV beyond 2004.

Another useful perspective from which to view the magnitude of the
emission reductions from today's proposal is in terms of the additional
emission reductions from all human sources that areas will need to
attain the 8-hour ozone standard. For this analysis, we included the
implementation of the Regional Ozone Transport Rule but assumed that
today's proposal was not implemented. In the previously referenced
NAAQS RIA we estimated additional NO_X emission reductions
that, along with specific accompanying VOC reductions, would bring each
residual nonattainment area into attainment with the 8-hour ozone

[[Page 26017]]

standard by 2010. We have used these estimated reductions as the basis
for Table III-5, which shows the NO_X reductions needed to
reach attainment in 2007 for six metropolitan areas.²⁰ These
are areas for which both the NAAQS RIA and the ozone modeling for this
proposal forecasted continued 8-hour nonattainment in that year, even
with the emission reductions from the Regional Ozone Transport Rule.
---------------------------------------------------------------------------

\20\ We calculated the estimated reductions needed for
attainment in 2007 by adding the reductions due to NLEV vehicles
entering the fleet between 2007 and 2010 to the estimated reductions
needed for attainment in 2010.
---------------------------------------------------------------------------

Table III-5 also shows the NO_X emission reductions in
those same six areas that we project would result if today's proposal
were implemented. Although the two analyses differ in some emission
modeling estimates, the comparison is valid as a general indication of
the contribution today's proposal can make to attainment. Cars and
light trucks contribute about 20 to 40 percent of the NO_X
inventory in these six areas. The NO_X reductions estimated
for today's proposal range from 19 to 50 percent of the reductions that
are estimated to be needed for attainment. These figures show that
today's proposal would make a very substantial contribution to these
cities' attainment programs, but that there will still be a need for
additional reductions from other sources. The emission reductions from
today's proposal would clearly not exceed the reductions needed from an
air quality perspective for these areas; as described in the next
section, we project that about 20 other areas in the eastern U.S. would
also need reductions beyond those of today's proposed program to attain
the NAAQS for NO_X.

Table III-5.--Comparison of Tier 2/Sulfur NO_X Reductions to NO_X Reductions Estimated to Produce 8-Hour Ozone
Attainment in 2007
----------------------------------------------------------------------------------------------------------------
Tier 2/sulfur
NO_X reductions NO_X reductions NO_X reductions
estimated to from proposed as percent of
Metro area produce tier 2/sulfur reductions to
attainment standards produce
(tons/year) (tons/year) attainment
----------------------------------------------------------------------------------------------------------------
Atlanta......................................................... 69,802 17,271 25
Dallas.......................................................... 41,283 14,761 36
Memphis......................................................... 7,343 3,683 50
NY-NJ-CT........................................................ 186,880 35,906 19
Philadelphia.................................................... 63,456 19,942 31
Washington, DC-Baltimore........................................ 62,519 22,673 36
----------------------------------------------------------------------------------------------------------------

3. Tier 2/Sulfur Ozone Benefits and the Post-Tier 2/Sulfur Ozone
Problem
By reducing ozone precursor emissions from cars and light trucks in
areas where ozone levels are near or above the ozone standard, today's
proposal would reduce local ozone levels. And by reducing ozone
precursor emissions in upwind areas, today's proposal would reduce
ozone and ozone precursor levels in the air flowing into areas where
ozone levels are high. EPA's analysis of the ozone impact of today's
proposal suggests that it would yield large reductions in ozone,
particularly in areas where ozone transport plays a significant role in
local nonattainment problems. There are uncertainties associated with
the modeling we have used to estimate these reductions, but we are
certain that the emission reductions would be large.
Ozone levels in a few locations in the centers of large
metropolitan areas are VOC-limited; that is, the atmospheric chemistry
is such that ozone levels tend to respond to VOC reductions rather than
to NO_X reductions. Some of these areas may experience
essentially no change or a slight ozone increase on some days, if one
considers only the isolated effect of the emission reductions due to
today's proposal. However, it has long been recognized that
metropolitan areas containing such locations will need to implement
additional VOC reductions from local sources to reach attainment. If
these reductions and the reductions from today's proposal were
combined, the net effect would be a progressive drop in ozone levels
until attainment is reached.
To examine the impact of today's proposal on ozone levels, we
estimated the ozone effects of the emission reductions that would occur
in 2007 and 2010 for the area covered by the OTAG ozone model. The 1-
hour ozone reductions in 2007 are relevant to the several 1-hour
nonattainment areas required to reach attainment in that year. The 8-
hour reductions in 2007 and 2010 are of great relevance to the efforts
of states to achieve attainment with the 8-hour ozone standard, since
for many areas these dates bracket the three ``clean'' years required
to show attainment by their actual deadline.
The estimated emission reductions from our proposal in 2007 and
2010 would be substantial due to the effect of low sulfur fuel on the
entire in-use fleet of gasoline vehicles and trucks of all sizes,
especially those designed to meet NLEV standards, and due to the fact
that many cleaner 2004 and newer vehicles would be on the road. Table
III-6 provides a summary of the 1-hour ozone results for the OTAG
modeling area for 2007. Table III-7 provides a summary of the 2007 and
2010 results for the 8-hour standard. According to our best modeling,
the reductions in 2007 would make the difference between nonattainment
and attainment for four metropolitan areas with a combined 1990
population of 15 million people. In 2010, we estimate that the Tier2/
Sulfur reductions would be enough by themselves to bring eight
metropolitan areas with 13 million people into attainment with the 8-
hour standard.
Tables III-6 and III-7 indicate that we project that some areas
would not attain with only the emission reductions from the Tier 2/
Sulfur proposal. However, we do project that those areas would
experience reductions in ozone levels. These reductions would mean that
even the areas that are not brought all the way to attainment would not
need to reduce emissions from other sources as much as would be
required without today's proposal, as previously explained. Of the 18
areas that we projected would not be brought to attainment with the 8-
hour standard in 2010, we project that 10 areas would

[[Page 26018]]

have design values within 5 percent of the standard.
Today's proposal would also benefit ozone nonattainment areas
outside of the OTAG modeling region, including the one area (Phoenix,
Arizona) projected to be in nonattainment for ozone in 2010 in the
absence of Tier 2/Sulfur controls. The Tier 2/Sulfur controls being
proposed today would help Phoenix attain the ozone standard,
particularly since cars and light trucks are a relatively large part of
the Phoenix emission inventory. These controls also would help the 11
areas projected to face potential maintenance challenges stay in
attainment as their economies and populations grow. And as already
mentioned, because about 7 to 10 percent of travel in California is by
non-California vehicles, there would be a substantial benefit in that
state also.

Table III-6.--Projected Tier 2/Sulfur Impact on Potential 1-Hour Ozone Problem Areas in the OTAG Region in 2007
^a
----------------------------------------------------------------------------------------------------------------
2007 without 2007 with tier Change due to
tier 2/sulfur 2/sulfur tier 2/sulfur
----------------------------------------------------------------------------------------------------------------
Design values projected to be in excess of the 1-Hour NAAQS (<gr-thn-eq>125 ppb)
----------------------------------------------------------------------------------------------------------------
Number of Metropolitan Areas.................................... 8 4 -4
Number of Rural Counties........................................ 2 2 0
1990 Population of Metropolitan Areas and Rural Counties 39 24 -15
(millions).....................................................
----------------------------------------------------------------------------------------------------------------
^a For all cases, this table reflects implementation of ROTR and other measures assumed in the ROTR. For the
``Without Tier 2/Sulfur'' case, the estimates reflect continuation of NLEV beyond 2004.

Table III-7.--Projected Tier 2/Sulfur Impact on Potential 8-Hour Ozone Problem Areas in the OTAG Region in 2007
and 2010 ^a
----------------------------------------------------------------------------------------------------------------
Without tier 2/ With tier 2/ Change due to
sulfur sulfur tier 2/sulfur
----------------------------------------------------------------------------------------------------------------
Design values projected to be in excess of the 8-Hour NAAQS (<gr-thn-eq>85 ppb) in 2007
----------------------------------------------------------------------------------------------------------------
Number of Metropolitan Areas.................................... 28 25 -3
Number of Rural Counties........................................ 4 3 -1
1990 Population of Metropolitan Areas and Rural Counties 80 72 -8
(millions).....................................................
----------------------------------------------------------------------------------------------------------------
Design values projected to be in excess of the 8-Hour NAAQS (<gr-thn-eq>85 ppb) in 2010
----------------------------------------------------------------------------------------------------------------
Number of Metropolitan Areas.................................... 26 ^b 18 -8
Number of Rural Counties........................................ 3 3 0
1990 Population of Metropolitan Areas and Rural Counties 78 65 -13
(millions).....................................................
----------------------------------------------------------------------------------------------------------------
^a For all cases, this table reflects implementation of ROTR and other measures assumed in the ROTR. For the
``Without Tier 2/Sulfur'' case, the estimates reflect continuation of NLEV beyond 2004.
^b Of these 18 areas predicted to remain nonattainment, 10 would be within 5 percent of the 8-hour ozone
standard.

Much larger VOC and NO_X emission reductions would occur
in 2020, when the vehicle fleet would be almost fully turned over to
Tier 2 vehicles. The 2020 scenario is designed to help evaluate the
long-term impact of today's proposal on ozone levels, when the majority
of the vehicle fleet would consist of vehicles that meet the standards
being proposed today.
We present three indicators of the benefits of today's proposed
program in 2020. First, as shown in Table III-3, that today's proposal
would reduce NO_X emissions in 2020 by over 2,000,000 tons
per year, not counting reductions in California, Hawaii, and Alaska.
The reduction in each nonattainment area would also be very
substantial. Second, we have estimated how much design values in 2020
would change due to today's proposal. For all counties projected to
need emission reductions beyond the ROTR, the average reduction in 2020
design value was 6 ppb, or almost 8 percent of the 8-hour standard
itself. The range of design value reductions was 3 to 12 ppb. These
results included only the region covered by the OTAG ozone model.
Third, when we analyzed the 2020 scenario to take into account the
duration, severity, and geographic extent of high ozone levels, we
found that projected excessive 8-hour ozone levels, defined as grid
cell-days above 85 ppm ozone, were reduced by 43 percent.
The baseline scenario against which the ozone effects of today's
proposed standards in 2020 were compared assumes that no emission
control efforts beyond those assumed in the ROTR are implemented. We
believe this approximation is reasonable because our inventory modeling
shows that in 2020, total human-caused emissions in the absence of
today's proposed program change very little from their 2007 levels. We
subtracted the emission benefits of today's proposed program in 2020
from those baseline emissions to approximate the emissions that would
result in 2020.
We expect the requirement to achieve attainment with the 8-hour
standard will cause states with residual nonattainment areas to adopt
additional controls in pursuit of their attainment obligations. The
increasingly large emission reductions from today's proposal that would
occur over time would be of great value to those areas since these
areas would not need to implement as extensive or stringent additional
controls as would otherwise be the case. Furthermore, once an area
reaches attainment, it must adopt a SIP revision containing a strategy
to maintain the standard thereafter. The reductions from today's
proposal would help such areas overcome any loss of reductions due to
less-than-expected effectiveness from other controls, provide a safety
margin against the chance of new ozone violations, provide room for
population and economic growth to cause increases in emissions

[[Page 26019]]

from other sources with less need for the maintenance plan to increase
the stringency of controls on those other sources, and possibly even
allow selective relaxation of other control programs.
Because the ozone modeling for 2020 did not account for the
additional measures that states will adopt to attain and maintain the
ozone standard, an attainment vs. nonattainment distinction does not
apply in 2020. Instead, the changes that today's proposal would achieve
in 2020 precursor emissions and in predicted ozone concentrations are
more appropriate indicators of the benefits of the Tier2/Sulfur program
than would be a count of the areas that have design values move from
above to below the ozone standard.
These ozone results for 2007, 2010, and 2020 represent the best
modeling currently available to us, but should be considered
approximate. The Regulatory Impact Analysis documents all the methods
and assumptions used. The results presented are estimates of the future
that only apply to the OTAG region rather than the entire area that
would be subject to today's proposal. As previously mentioned, there
would also be ozone benefits outside this region, particularly for
nonattainment areas in California and for Phoenix, Arizona. We expect
to revise our ozone effects estimates prior to the final rule to
reflect further improvements in estimates of emissions from both mobile
and stationary sources.
In addition to the emission-reduction and ozone-reduction benefits
discussed above that we expect will result from the proposed rule, we
have done a separate analysis of economic benefits (and costs)
associated with the expected ozone reductions from today's proposed
program (see Section IV.D.5. below and the RIA).

C. Particulate Matter

1. Particulate Matter Presents Substantial Public Health Risks
Particulate matter (PM) is produced as a direct result of human
activity and natural processes, and it is also formed through chemical
and physical processes in the atmosphere. Natural sources include
windblown dust, salt from dried sea spray, fires, and volcanoes, as
well as so-called secondary particles formed from the transformation of
natural emissions of SO<INF>X</INF>, NO_X, and VOCs. Human
sources include industrial activities, agriculture, road dust, and
soot, as well as secondary particles produced from gases such as
SO<INF>X</INF>, NO_X, and VOCs that are emitted primarily
from combustion processes. PM includes fine particles with a diameter
smaller than 2.5 microns (also called PM<INF>2.5</INF>) and coarse
particles with larger diameters. Coarse particles are predominantly
from non-combustion sources and are dominated by soil dust and sea
salt. They remain in the atmosphere a relatively short period of time.
Fine particulate includes carbon-based particles emitted directly from
combustion processes but consists predominantly of secondary particles,
such as sulfate-based particles (produced from SO<INF>X</INF>),
nitrate-based particles (produced from NO_X), and carbon-
based particles created through transformation of VOC emissions. Mobile
sources can reasonably be estimated to contribute to ambient secondary
nitrate, sulfate and carbonaceous PM in proportion to their
contribution to total NO_X, SO<INF></INF>, and VOC emissions.
In 1997, 8 million Americans were living in 13 counties that
exceeded the recently revised PM₁₀ standard, and
PM₁₀ problems are projected to persist in the absence of
further actions to control PM₁₀ levels. Table III-8 presents
estimates of the extent of PM₁₀ and PM<INF>2.5</INF>
nonattainment in the future. In the NAAQS RIA, we projected that in
2010, eleven counties with a combined 1990 population of about 10
million people would not be in attainment with the revised
PM₁₀ standards.²¹ About half of the affected
population lives outside of California. In the same analysis, 102
counties were projected to violate the new PM<INF>2.5</INF> NAAQS, with
a combined 1990 population of about 55 million people. About 75 percent
of the affected population lives outside of California. (More
information about this analysis and its uncertainties may be found in
the NAAQS RIA and the Tier 2 Report to Congress.) Ambient PM reductions
from more stringent motor vehicle or fuel standards would primarily
affect areas outside of California, because California has its own
motor vehicle emission control program. California areas would also
benefit, however, through the temporary travel and permanent migration
of out-of-state vehicles into California, as discussed above.
---------------------------------------------------------------------------

\21\ The methods used to project PM concentrations in 2010 from
1990 emissions and ambient concentration data introduce several
sources of uncertainty. Also, the PM<INF>2.5</INF> values are
predicted from a regression model and hence are subject to the
uncertainty associated with this model. Other uncertainties exist
regarding emission inventory estimates from human and natural
sources, monitoring data, and the models used to account for
physical and chemical processes in the atmosphere. Even with the
anticipated delivery of more comprehensive modeling techniques, the
scarcity of speciated ambient PM data in both urban and rural areas
to evaluate model behavior will continue to compromise the certainty
of the best model-derived conclusions.

Table III-8.--Projected 2010 PM₁₀/PM<INF>2.5</INF> Nonattainment Counties and
Populations
------------------------------------------------------------------------
Outside
California California
------------------------------------------------------------------------
Violating Original PM₁₀ NAAQS
------------------------------------------------------------------------
Number of Counties............................ 33 12
1990 Population (millions).................... 11 7
------------------------------------------------------------------------
Violating Revised PM₁₀ NAAQS
------------------------------------------------------------------------
Number of Counties............................ 5 6
1990 Population (millions).................... 5 5
------------------------------------------------------------------------
Violating New PM<INF>2.5</INF> NAAQS
------------------------------------------------------------------------
Number of Counties............................ 92 10
1990 Population (millions).................... 42 13
------------------------------------------------------------------------

A significant number of areas are projected to exceed the
PM₁₀ NAAQS in 2010 with existing emission controls,
indicating that further PM and PM-precursor emission reductions will be
needed. Because the bulk of PM emissions from motor vehicles are fine
particles, any reduction in particulate emissions from motor vehicles
aimed at reducing PM₁₀ levels would also reduce ambient
levels of PM<INF>2.5</INF>. As mentioned above, the number of counties
projected to violate the new PM<INF>2.5</INF> NAAQS is much larger than
that for the revised PM₁₀ standards. Tier 2/Sulfur standards
that reduce particulate emissions for the purposes of facilitating
attainment with the PM₁₀ NAAQS could also benefit areas with
elevated PM<INF>2.5</INF> levels.
2. Reducing Emissions From Cars and Light Trucks Would Reduce Ambient
Levels
Today's proposal would reduce PM levels by reducing direct PM
emissions from cars and light trucks, and by reducing emissions of
sulfur and nitrogen oxides that are converted to PM in the atmosphere.
Direct PM emissions would be reduced in two ways. First, reductions in
gasoline sulfur levels would reduce PM emissions from gasoline
vehicles. Second, the more stringent PM standard included in today's
proposal would reduce PM emissions from cars and light trucks equipped
with diesel engines. Diesel engines are used in a small fraction of
current cars and light trucks, but this

[[Page 26020]]

fraction could grow as discussed in III.C.3. below.
With no growth in diesel sales, we project today's action would
reduce direct PM emissions from cars and light trucks mainly due to the
introduction of low-sulfur gasoline. Sulfur-based particles account for
a substantial portion of the particulate matter emitted by gasoline-
powered vehicles. More stringent PM emission standards are not
anticipated to alter PM emissions from gasoline vehicles but would
result in reductions in diesel PM emissions. The overall effect of
today's proposal under this assumption would be to reduce direct
exhaust PM emissions from cars and light trucks by 60 percent in 2007
and by 62-63 percent in 2015 and beyond. Tables III-9 and III-10 show
the contribution of cars and light trucks to total PM₁₀ and
PM<INF>2.5</INF> emissions, and the reductions that would be obtained
from today's proposal. The contribution of cars and light trucks to
either PM inventory will generally be higher in urban areas than on a
nationwide basis, and will vary from area to area. In 2007, for
example, cars and light trucks contribute 1.3 percent to the nationwide
PM₁₀ inventory (excluding natural sources and fugitive
dust). For comparison, this percentage is estimated to be 4.4 percent
in Atlanta and 1.9 percent in the New York City metropolitan area.
Later in this section we discuss the possibility that sales of
diesel-powered vehicles might increase from current levels, making the
effect of the more stringent PM standard in this proposal larger.

Table III-9.--Direct exhaust PM₁₀ Emissions From Cars and Light Trucks as Percent of Total Emissions, and
Reductions Due to Tier 2/Sulfur Control^a,b
----------------------------------------------------------------------------------------------------------------
Light-duty
Light-duty percent of Light-duty
Year tons without total without tons reduced
tier 2 tier 2 by tier 2
----------------------------------------------------------------------------------------------------------------
2007............................................................ 39,209 1.3 23,379
2010............................................................ 41,412 1.4 25,239
2015............................................................ 46,064 1.4 28,674
2020............................................................ 51,102 1.5 32,031
----------------------------------------------------------------------------------------------------------------
^a For all cases, this table reflects continuation of current diesel engine usage in the light truck fleet and
implementation of ROTR and other measures assumed in the ROTR.
^b The emission estimates shown exclude natural sources of PM and fugitive dust. They also do not include
California (which has its own vehicle and fuel standards), Alaska, or Hawaii. Today's proposal would have
additional emission benefits in these states.

Table III-10.--Direct exhaust PM<INF>2.5</INF> Emissions From Cars and Light Trucks As Percent of Total Emissions, and
Reductions Due to Tier 2/Sulfur Control ^a,b
----------------------------------------------------------------------------------------------------------------
Light-duty
Light-duty percent of Light-duty
Year tons without total without tons reduced
tier 2 tier 2 by tier 2
----------------------------------------------------------------------------------------------------------------
2007............................................................ 36,365 1.7 21,687
2010............................................................ 38,409 1.8 23,410
2015............................................................ 42,724 1.9 26,595
2020............................................................ 47,397 2.0 29,707
----------------------------------------------------------------------------------------------------------------
^a For all cases, this table reflects continuation of current diesel engine usage in the light truck fleet and
implementation of ROTR and other measures assumed in the ROTR.
^b The emission estimates shown exclude natural sources of PM and fugitive dust. They also do not include
California (which has its own vehicle and fuel standards), Alaska, or Hawaii. Today's proposal would have
additional emission benefits in these states.

Even larger PM reductions would result from the reductions in the
sulfur oxides (SO<INF>X</INF>), NO_X, and VOC emissions that
give rise to secondary PM that would result from today's proposal. The
reduction in ambient PM levels that would come from the proposed
reductions in these precursor emissions is about 6 to 7 times as large
as the reduction from lower emissions of direct PM. Essentially all
secondary PM is fine PM and hence is included in estimates of both
PM₁₀ and PM<INF>2.5</INF>.
We described the effect of today's proposal on VOC and
NO_X emissions above in Section III.B. Today's proposal also
would reduce SO<INF>X</INF> emissions from cars and light trucks by
dramatically lowering the level of sulfur in gasoline, since gaseous
SO<INF>X</INF> emissions are dependent entirely on fuel sulfur level.
In the absence of today's proposal, we project that SO<INF>X</INF>
emissions from cars and light trucks will increase steadily in
conjunction with VMT growth, from approximately 216,000 tons in 2005 to
300,000 tons in 2020--an increase of almost 40 percent (total
nationwide SO<INF>X</INF> emissions from all sources was 20,000,000
tons in 1997). Today's proposal would reduce SO<INF>X</INF> emissions
from all gasoline-powered engines, including cars, light trucks, heavy-
duty gasoline vehicles, and gasoline-powered nonroad engines, in any
year by 90 percent, once all gasoline meets the proposed sulfur limit.
The same percentage reductions in SO<INF>X</INF> emissions would occur
in subsequent years. The absolute emission reduction increases with
time, however, due to growth in VMT and nonroad engine use. Table III-
11 shows the impact of today's proposal on SO<INF>X</INF> emissions.

[[Page 26021]]

Table III-11.--SO<INF>x</INF> Emissions From Cars and Light Trucks as Percent of Total Emissions, and Reductions Due to
Tier 2/Sulfur control ^a
----------------------------------------------------------------------------------------------------------------
Light-duty
Light-duty percent of Light-duty
Year tons without total without tons reduced
tier 2 tier 2 by tier 2
----------------------------------------------------------------------------------------------------------------
2007............................................................ 225,673 1.2 202,748
2010............................................................ 240,694 1.3 216,437
2015............................................................ 270,174 1.4 242,964
2020............................................................ 299,959 1.6 269,756
----------------------------------------------------------------------------------------------------------------
^a The emission estimates shown do not include California (which has its own vehicle and fuel standards), Alaska,
or Hawaii. Today's proposal would have additional emission benefits in these states.

3. Today's Proposal Would Limit the Potential Health Risks From
Increased Diesel Engine Use in Cars and Light Trucks
Of particular concern from a PM perspective is the possibility that
diesels will become more prevalent in the light-duty truck fleet. This
development is a reasonable possibility since vehicle and engine
manufacturers have indicated their intent to sell more diesel-powered
light-duty trucks and in some cases have made capital investments to
implement these plans. The Partnership for a New Generation of Vehicles
(PNGV), a public-private research and development effort that has been
pursuing several promising technologies for greatly improved vehicle
fuel economy combined with low emissions, has identified improved
diesel engines as a technology likely to be able to deliver large fuel
economy improvements in the near future, by about 2004. In order to
assess the potential impact of increased diesel sales penetration on
PM<INF>2.5</INF> emissions, we analyzed benefits from our proposed Tier
2 PM standards under a scenario in which the use of diesel engines in
light trucks increases rapidly, by five percentage points per year from
2001 through 2010, when diesels would account for 50 percent of light-
duty truck sales; beyond 2010, diesel sales were assumed to be stable
at 50 percent of the light-truck market. Table III-12 presents the
results of our analysis of this scenario.
This scenario of increased diesels would result in dramatic
increases in direct PM<INF>2.5</INF> emissions from cars and light
trucks, if there is no change in the PM standards for light trucks. The
increase in diesel exhaust PM<INF>2.5</INF> emissions would more than
overcome the reduction in direct PM<INF>2.5</INF> attributable to the
sulfur reduction in gasoline. Assuming no change in the existing PM
standards for light trucks, our analysis of this scenario shows that
direct PM<INF>2.5</INF> emissions in 2020 would be approximately
140,000 tons, nearly three times the 47,000 tons projected in the base
diesel sales case from Table III-10. The portion of the
PM<INF>2.5</INF> inventory attributable to cars and light trucks would
climb steadily, reaching almost 6 percent in 2020 instead of the 2
percent shown in Table III-10 for a scenario where diesel engines do
not increase their presence in the light truck fleet. In some cities
with relatively high vehicle use and lower industrial emissions, the
car and truck contribution would be even higher.
This increase would be accompanied by increases in the mortality
and morbidity associated with PM<INF>2.5</INF> exposure. Fortunately,
the standards being proposed today would result in a steady decrease in
total direct PM<INF>2.5</INF> from cars and light trucks despite a
possible increase in diesel engines in light trucks. Direct PM
emissions in 2020 with today's proposal would be about 25,000 tons per
year, less than at present.
If this scenario for increased diesel engines in light trucks were
to occur, today's proposal would reduce diesel PM<INF>2.5</INF> by over
90 percent in 2020. Stated differently, by 2020 today's proposal would
reduce over 113,000 tons of the potential increase in PM emissions from
passenger cars and light trucks. The result would be less direct
PM<INF>2.5</INF> than is emitted today, because the increase in diesel
PM would be more than offset by the reduction in gasoline PM.

Table III-12.--Direct Exhaust PM<INF>2.5</INF> Emissions From Light Duty Vehicles and Reductions Due to Tier 2/Sulfur
Control, With Greater Diesel Engine Sales ^a,b
----------------------------------------------------------------------------------------------------------------
Light-duty Light-duty
Year exhaust tons exhaust tons Light-duty
without tier 2 with tier 2 tons reduced
----------------------------------------------------------------------------------------------------------------
2007............................................................ 52,907 22,478 30,429
2010............................................................ 72,626 22,542 50,084
2015............................................................ 109,622 23,275 86,347
2020............................................................ 138,177 24,754 113,424
----------------------------------------------------------------------------------------------------------------
^a For all cases, this table reflects implementation of ROTR and other measures assumed in the ROTR and an
increase in diesel-powered light truck market share from 5 percent of light truck sales in 2001 to 50 percent
in 2010 and beyond.
^b The emission estimates shown exclude natural sources of PM and fugitive dust. They also do not include
California (which has its own vehicle and fuel standards), Alaska, or Hawaii. Today's proposal would have
additional emission benefits in these states.

4. Today's Proposal Would Have Substantial PM Benefits
In general, we project that today's proposal would reduce both
direct and secondary PM from cars and light trucks substantially,
regardless of the future market share for diesel engines in the light-
duty fleet. The larger part of the reduction is due to large reductions
in VOC, NO_X, and SO<INF>X</INF> emissions, with
corresponding reductions in secondary PM formation.
Low sulfur fuel would greatly reduce direct PM emissions and
sulfate-based secondary PM formation from SO<INF>X</INF> emissions from
gasoline vehicles, while tailpipe PM standards are projected to
mitigate excess PM emissions from diesel vehicles, even at very
aggressive rates of diesel vehicle sales growth. Substantial reductions
in NO_X

[[Page 26022]]

emissions would carry over to reductions in indirect PM. These
reductions would help reduce the number of areas with PM₁₀
and PM<INF>2.5</INF> levels in excess of national standards, reduce the
severity of PM nonattainment in other areas, and help areas facing PM
maintenance challenges stay in attainment.
The magnitude of the PM reductions from today's proposal in a given
area depends on conditions such as the contribution of light-duty
vehicles to the local PM, SO<INF>X</INF>, NO_X, and VOC
inventory; the contribution of light-duty vehicles to the PM,
SO<INF>X</INF>, NO_X, and VOC inventories in upwind areas;
local and upwind ammonia inventories (involved in secondary PM
formation); control measures being implemented on both local and upwind
sources of PM and its precursors, and local meteorology. We have
incorporated these factors into the air quality modeling used to
develop the benefit/cost analysis presented in Section IV.D.5., which
includes the economic benefits of the direct and secondary PM
reductions expected to result from today's proposal.
The PM modeling results from that analysis suggest that if all cars
and trucks used in 2010 met the emission standards being proposed
today, significant PM reductions would result in urban and substantial
PM reductions would result in much of the continental U.S. The annual
average level of both PM₁₀ and PM<INF>2.5</INF> was
projected to decline by 0.25 to 0.64 micrograms per cubic meter
(<greek-m>/m³) in many cities; average levels were projected
to decline by 0.1 to 0.25 <greek-m>/m³ throughout most of
the country east of the Great Plains, Nebraska, and parts of Colorado,
Arizona, and other western states. Similarly, daily maximum PM levels
²² were projected to decline substantially, with many cities
projected to see declines of 0.75 to 4.5 <greek-m>/m³ and
over half the continental U.S. projected to experience declines of 0.25
to 0.75 <greek-m>/m³. Note that this analysis assumed no
growth in sales of diesel-powered light trucks. It also did not account
for the direct PM reductions that would be achieved when the small
number of diesel-powered trucks already being sold now will reduce
their PM emissions to meet the lower proposed PM standard.
---------------------------------------------------------------------------

\22\ Daily maximum PM levels are the PM levels (averaged over 24
hours) for days that are projected to be in the 98th or 99th
percentile when ranked by their PM<INF>2.5</INF> and PM₁₀
levels, respectively.
---------------------------------------------------------------------------

D. Other Criteria Pollutants: Carbon Monoxide, Nitrogen Dioxide, Sulfur
Dioxide

This proposal would help reduce levels of three other pollutants
for which NAAQS have been established: carbon monoxide (CO), nitrogen
dioxide (NO<INF>2</INF>), and sulfur dioxide (SO<INF>2</INF>). The
extent of nonattainment for these three pollutants is small, so the
primary effect of today's proposal would be to provide areas concerned
with maintaining their attainment status a greater margin of safety. As
of 1998, every area in the United States has been designated to be in
attainment with the NO<INF>2</INF> NAAQS. As of 1997, only one area
(Buchanan County, Missouri) did not meet the primary SO<INF>2</INF>
short-term standard, due to emissions from the local power plant. In
1997, only 6 of 537 monitoring sites reported ambient CO levels in
excess of the CO NAAQS; all six sites were located in California, which
has established its own vehicle and fuel emission standards.
The reductions in SO<INF>2</INF> precursor emissions from today's
proposal are essentially equal to the SO<INF>X</INF> reductions
described in Section III.B. and III.C., respectively. The impact of
today's proposal on NO<INF>2</INF> emissions depends on the specific
emission control technologies used to meet the standards being proposed
today. However, essentially all of the NO_X emitted by cars
and light trucks converts to NO<INF>2</INF> in the atmosphere;
therefore, it is reasonable to assume that today's proposal would
substantially reduce ambient NO<INF>2</INF> levels by the same
proportion. Today's proposal also would require light trucks to meet
more stringent CO standards; we will evaluate the impact of these
standards more fully before publishing our final rule. The analysis of
economic benefits and costs found in Section IV.D.-5. does not account
for the economic benefits of the CO reductions expected to result from
today's proposal.

E. Visibility

Visibility impairment occurs as a result of the scattering and
absorption of light by particles and gases in the atmosphere. It is
most simply described as the haze that obscures the clarity, color,
texture, and form of what we see. The principal cause of visibility
reduction is fine particles between 0.1 and 1 <greek-m>m in size. Of
the pollutant gases, only NO<INF>2</INF> absorbs significant amounts of
light; it is partly responsible for the brownish cast of polluted
skies. While the contribution of NO<INF>2</INF> to visibility
impairment varies from area to area, it is generally responsible for
less than ten percent of visibility reduction.
The CAA requires EPA to protect visibility, or visual air quality,
through a number of programs. These programs include the national
visibility program under Sections 169a and 169b of the Act, the
Prevention of Significant Deterioration program for the review of
potential impacts from new and modified sources, and the secondary
NAAQS for PM₁₀ and PM<INF>2.5</INF>. The national visibility
program established in 1980 requires the protection of visibility in
156 mandatory federal Class I areas across the country (primarily
national parks and wilderness areas). More than 65 million visitors
travel each year to these parks and wilderness areas. The CAA
established as a national visibility goal, ``the prevention of any
future, and the remedying of any existing, impairment of visibility in
mandatory federal Class I areas in which impairment results from
manmade air pollution.'' The Act also calls for state programs to make
``reasonable progress'' toward the national goal. In addition, a recent
national opinion poll on the state of the national parks found that
more than 80 percent of Americans believe air pollution affecting these
parks should be cleaned up for the benefit of future
generations.²³
---------------------------------------------------------------------------

\23\ ``National Parks and the American Public: A National Pubic
Opinion Survey on the National Park System,'' Summary Report,
National Parks and Conservation Association, June 1998.
---------------------------------------------------------------------------

There has been improvement in visibility in the western part of the
country over the last ten years. However, visibility impairment remains
a serious problem in Class I areas. Visibility in the East does not
seem to have improved. As one part of addressing this national problem,
EPA has proposed that states be required to adopt and implement
effective plans for protecting and improving visibility in Class I
federal areas (including 156 major national parks and wilderness
areas), integrated with plans to achieve the revised ozone and PM
standards.
Today's proposal should result in visibility improvements due to
the reduction in local and upwind PM and PM precursor emissions. Since
mobile source emissions contribute to the formation of visibility-
reducing PM, control programs that reduce the mobile source emissions
of direct and secondary PM would have the effect of improving
visibility. The Grand Canyon Visibility Transport Commission's final
recommendations report ²⁴ found that

[[Page 26023]]

reducing total mobile source emissions is an essential part of any
program to protect visibility in the Western U.S. The Commission found
that motor vehicle exhaust is responsible for about 14 percent of
human-caused visibility reduction (excluding road dust). A substantial
portion of motor vehicle exhaust comes from cars and light trucks. In
light of that impact, the Commission's recommendations in 1996
supported federal Tier 2/Sulfur standards, as EPA is proposing today.
More recently, a number of Western Governors noted the importance of
controlling mobile sources as part of efforts to improve visibility in
their comments on the Regional Haze Rule and on the need to protect the
16 Class I areas on the Colorado Plateau. In their joint letter dated
June 29, 1998, they stated that, ``* * * the federal government must do
its part in regulating emissions from mobile sources that contribute to
regional haze in these areas. * * *'' and called on EPA to make a
``binding commitment * * * to fully consider the Commission's
recommendations related to the * * * federal national mobile source
emission control strategies.'' These recommendations included Tier 2
vehicle standards and reductions in gasoline sulfur levels.
---------------------------------------------------------------------------

\24\ ``Recommendations for Improving Western Vistas,'' Report of
the Grand Canyon Visibility Transport Commission to the United
States Environmental Protection Agency, June 10, 1996.
---------------------------------------------------------------------------

As an indication of how important car and light truck emissions can
be to fine PM and visibility, the recent Northern Front Range Air
Quality Study has reported findings that indicate that cars and light
trucks are responsible for 39 percent of fine PM at a site within the
metropolitan Denver area, and for 40 percent at a downwind rural site.
This contribution includes both direct PM and indirect PM formed from
sulfur dioxide and NO_X from these vehicles.
The analysis of economic benefits and costs found in Section
IV.D.5. accounts for the economic benefits of the visibility
improvements expected to result from today's proposal.

F. Air Toxics

Emissions from cars and light trucks include a number of air
pollutants that are known or suspected human or animal carcinogens such
as benzene, formaldehyde, acetaldehyde, 1,3-butadiene, and diesel
particulate matter, or that are known or suspected to have other, non-
cancer health impacts. For several of these pollutants, motor vehicle
emissions are believed to account for a significant proportion of total
nation-wide emissions. All of these compounds are present in exhaust
emissions; benzene is also found in evaporative emissions from
gasoline-fueled vehicles.
The health effects of diesel particulate are of particular
relevance to this rulemaking, because of the possibility for increased
diesel-powered truck sales and our proposal for a more stringent PM
standard that would apply to these trucks. While we have not finalized
our decision about the carcinogenicity of diesel exhaust particulate,
we are in the process of addressing this question. Several other
agencies and international organizations have already made such a
determination, including the California Air Resources Board (ARB). Our
own quantitative risk assessment for diesel particulate is still in
draft form,²⁵ and is presently being revised to address the
comments of a peer review panel of the Clean Air Science Advisory
Committee.
---------------------------------------------------------------------------

\25\ EPA's diesel health assessment (Health Assessment Document
for Diesel Emissions, SAB Review Draft, U.S. Environmental
Protection Agency, Washington, DC. EPA/600/8-90/057C, February
1998.) can be found at the following EPA website: http://
www.epa.gov/ncea/diesel.htm. The Clean Air Science Advisory
Committee's review of that assessment (CASAC Review of the Draft
Diesel Health Assessment Document, U.S. Environmental Protection
Agency Science Advisory Board, Washington, DC EPA-SAB-CASC-99-001.)
can be found at the following SAB website: http://www.epa.gov/sab/.
---------------------------------------------------------------------------

Because our assessment for diesel particulate is not complete, we
are not presenting absolute estimates of how potential cancer risks
from diesel particular could be affected by today's proposal. However,
we can give a qualitative or relative discussion. Diesel engines are
used in a very small portion of the cars and light-duty trucks in
service today. By far, heavy duty highway and nonroad diesel engines
are the larger source of diesel PM. Engine and vehicle manufacturers
have projected that diesel engines are likely to be used in an
increasing share of light trucks, and some manufacturers have announced
capital investments to build such engines.
If these projections are valid and the proportion of light-duty
trucks powered by diesel engines increases, the potential health risks
from diesel PM could increase substantially. Light trucks could become
a larger source of diesel PM than heavy-duty diesel trucks. We estimate
that if the percentage of light duty diesel truck sales were to
increase to 50 percent of light-duty truck sales by 2010, the increased
presence of light duty diesel trucks on the nation's roads could
increase the potential cancer risks associated with PM emissions from
all diesel-powered highway vehicles (including heavy-duty diesel
trucks, diesel buses, and light-duty diesel vehicles) by approximately
130 percent as of 2020, under the current light-duty diesel PM
standards. Though the actual levels of diesel engine use may be
considerably different than the projections used in both analyses, the
analyses are useful in illustrating the potential impact of increased
diesel engine use in light trucks.
Today's proposal would limit the increase in the potential cancer
risks from cars and light trucks associated with any potential increase
in light-duty diesel sales. We have estimated that in 2020, today's
proposal would limit the increase in total highway diesel PM emissions
due to growth in light truck diesels to 24 percent, in contrast to the
more than doubling that would occur without our proposal for a tighter
PM standard for light trucks. The comparison in terms of potential
cancer risk from car and light truck diesel PM likely would closely
follow this emissions comparison.
The VOC emission reductions resulting from today's proposal would
further reduce the potential cancer risk posed by air pollutants other
than diesel PM emitted by cars and light trucks, since many of these
pollutants are themselves VOCs. The analysis of economic benefits and
costs found in Section IV.D.5. does not account for the economic
benefits of the reduction in cancer risk from air toxics that could
result from today's proposal, because we have not yet completed our
study of this issue or engaged in a peer-reviewed assessment of the
baseline air toxics risks (including a final quantitative risk
assessment of the diesel particulate risks) or of the reductions that
would be achieved by today's proposal. Therefore, the estimates
included in the Draft RIA should be considered preliminary. A peer-
reviewed assessment is planned and may be completed in time to be
available for incorporation into the impact analysis for the final
rule. EPA will place this document in the docket as soon as it is
available for public review.
Section 202(l)(2) of the Clean Air Act requires EPA to establish
regulations for the control of hazardous air pollutants, or air toxics,
from motor vehicles. The regulations may address vehicle emissions or
fuel properties that influence emissions, or both. We will issue a
proposal to address this requirement in September of this year, and a
final rule in July 2000.

G. Acid Deposition ²⁶

Acid deposition, or acid rain as it is commonly known, occurs when
SO<INF>2</INF>

[[Page 26024]]

and NO_X react in the atmosphere with water, oxygen, and
oxidants to form various acidic compounds that later fall to earth in
the form of precipitation or dry deposition of acidic particles. It
contributes to damage of trees at high elevations and in extreme cases
may cause lakes and streams to become so acidic that they cannot
support aquatic life. In addition, acid deposition accelerates the
decay of building materials and paints, including irreplaceable
buildings, statues, and sculptures that are part of our nation's
cultural heritage. To reduce damage to automotive paint caused by acid
rain and acidic dry deposition, some manufacturers use acid-resistant
paints, at an average cost of $5 per vehicle--a total of $61 million
per year if applied to all new cars and trucks sold in the U.S. The
general economic and environmental effects of acid rain are discussed
at length in the Draft RIA.
---------------------------------------------------------------------------

\26\ Much of the information in this section was excerpted from
the EPA document, Human Health Benefits from Sulfate Reduction,
written under Title IV of the 1990 Clean Air Act. Amendments, U.S.
EPA, Office of Air and Radiation, Acid Rain Division, Washington, DC
20460, November 1995.
---------------------------------------------------------------------------

Acid deposition primarily affects bodies of water that rest atop
soil with a limited ability to neutralize acidic compounds. The
National Surface Water Survey (NSWS) investigated the effects of acidic
deposition in over 1,000 lakes larger than 10 acres and in thousands of
miles of streams. It found that acid deposition was the primary cause
of acidity in 75 percent of the acidic lakes and about 50 percent of
the acidic streams, and that the areas most sensitive to acid rain were
the Adirondacks, the mid-Appalachian highlands, the upper Midwest and
the high elevation West. The NSWS found that approximately 580 streams
in the Mid-Atlantic Coastal Plain are acidic primarily due to acidic
deposition. Hundreds of the lakes in the Adirondacks surveyed in the
NSWS have acidity levels incompatible with the survival of sensitive
fish species. Many of the over 1,350 acidic streams in the Mid-Atlantic
Highlands (mid-Appalachia) region have already experienced trout losses
due to increased stream acidity. Emissions from U.S. sources contribute
to acidic deposition in eastern Canada, where the Canadian government
has estimated that 14,000 lakes are acidic. Acid deposition also has
been implicated in contributing to degradation of high-elevation spruce
forests that populate the ridges of the Appalachian Mountains from
Maine to Georgia. This area includes national parks such as the
Shenandoah and Great Smoky Mountain National Parks.
The SO<INF>X</INF> and NO_X reductions from today's
proposal would help reduce acid rain and acid deposition, thereby
helping to reduce acidity levels in lakes and streams throughout the
U.S. These reductions would help accelerate the recovery of acidified
lakes and streams and the revival of ecosystems adversely affected by
acid deposition. Reduced acid deposition levels would also help reduce
stress on forests, thereby accelerating reforestation efforts and
improving timber production. Deterioration of our historic buildings
and monuments, and of buildings, vehicles, and other structures exposed
to acid rain and dry acid deposition, also would be reduced, and the
costs borne to prevent acid-related damage may also decline.
While the reduction in sulfur and nitrogen acid deposition would be
roughly proportional to the reduction in SO<INF>X</INF> and
NO_X emissions, respectively, the precise impact of today's
proposal would differ across different areas. Each area is affected by
emissions from different source regions, and the mobile source
contribution to the total SO<INF>X</INF> and NO_X emission
inventory will differ across different source regions. Nonetheless, the
projected impact of today's proposal on SO<INF>X</INF> and
NO_X emission inventories provides a rough indicator of the
likely effect of today's proposal on acid deposition. As discussed in
Section III.D. above, today's proposal would reduce SOx emissions by
1.6 percent and NO_X emissions by 12.5 percent in 2020.
The analysis of economic benefits and costs found in Section
IV.D.5. was not able to account for the economic benefits of the
reduction in acid deposition expected to result from today's proposal.

H. Eutrophication/Nitrification

Nitrogen deposition into bodies of water can cause problems beyond
those associated with acid rain. Elevated levels of nitrate in drinking
water pose significant health risks, especially to infants. The
Ecological Society of America has included discussion of the
contribution of air emissions to increasing nitrogen levels in surface
waters in a recent major review of causes and consequences of human
alteration of the global nitrogen cycle in its Issues in Ecology
series.²⁷ Long-term monitoring in the United States, Europe,
and other developed regions of the world shows a substantial rise of
nitrogen levels in surface waters, which are highly correlated with
human-generated inputs of nitrogen to their watersheds. These nitrogen
inputs are dominated by fertilizers and atmospheric deposition.
---------------------------------------------------------------------------

\27\ Vitousek, Peter M., John Aber, Robert W. Howarth, Gene E.
Likens, et al. 1997. Human Alteration of Global Nitrogen Cycle:
Causes and Consequences. Issues in Ecology. Published by Ecological
Society of America, Number 1, Spring 1997.
---------------------------------------------------------------------------

Human activity can increase the flow of nutrients into those waters
and result in excess algae and plant growth. This increased growth can
cause numerous adverse ecological effects and economic impacts,
including nuisance algal blooms, dieback of underwater plants due to
reduced light penetration, and toxic plankton blooms. Algal and
plankton blooms can also reduce the level of dissolved oxygen, which
can also adversely affect fish and shellfish populations. This problem
is of particular concern in coastal areas with poor or stratified
circulation patterns, such as the Chesapeake Bay, Long Island Sound, or
the Gulf of Mexico. In such areas, the ``overproduced'' algae tends to
sink to the bottom and decay, using all or most of the available oxygen
and thereby reducing or eliminating populations of bottom-feeder fish
and shellfish, distorting the normal population balance between
different aquatic organisms, and in extreme cases causing dramatic fish
kills.
Collectively, these effects are referred to as eutrophication,
which the National Research Council recently identified as the most
serious pollution problem facing the estuarine waters of the United
States (NRC, 1993). Nitrogen is the primary cause of eutrophication in
most coastal waters and estuaries.²⁸ On the New England
coast, for example, the number of red and brown tides and shellfish
problems from nuisance and toxic plankton blooms have increased over
the past two decades, a development thought to be linked to increased
nitrogen loadings in coastal waters. Airborne NO_X
contributes from 12 to 44 percent of the total nitrogen loadings to
United States coastal water bodies. For example, approximately one-
quarter of the nitrogen in the Chesapeake Bay comes from atmospheric
deposition.
---------------------------------------------------------------------------

\28\ Much of this information was taken from the following EPA
document: Deposition of Air Pollutants to the Great Waters-Second
Report to Congress, Office of Air Quality Planning and Standards,
June 1997, EPA-453/R-97-011.
---------------------------------------------------------------------------

Excessive fertilization with nitrogen-containing compounds can also
affect terrestrial ecosystems. ²⁹ Research suggests that
nitrogen fertilization can alter growth patterns and change the balance
of species in an ecosystem. In

[[Page 26025]]

extreme cases, this process can result in nitrogen saturation when
additions of nitrogen to soil over time exceed the capacity of the
plants and microorganisms to utilize and retain the nitrogen. This
phenomenon has already occurred in some areas of the U.S.
---------------------------------------------------------------------------

\29\ Terrestrial nitrogen deposition can act as a fertilizer. In
some agricultural area, this effect can be beneficial.
---------------------------------------------------------------------------

Deposition of nitrogen from cars and light trucks contributes to
these problems. As discussed in Section III.B. above, today's proposal
would reduce total NO_X emissions by 12.5 percent in 2020.
These reductions should reduce drinking water nitrate levels by
reducing the amount of nitrate deposited from the atmosphere onto
drinking water sources or onto the watersheds of drinking water sources
by similar amounts. The NO_X reductions would also reduce the
eutrophication problems associated with atmospheric deposition of
nitrogen into watersheds and onto bodies of water, particularly in
aquatic systems where atmospheric deposition of nitrogen represents a
significant portion of total nitrogen loadings. Since air deposition
accounts for 12-44 percent of total nitrogen loadings in coastal
waters, the 12.5 percent reduction in NO_X from today's
proposal are projected to reduce nitrogen loadings by 1.5-5.5 percent.
To put these reductions in perspective, the reductions expected in the
Chesapeake Bay area would amount to about 6 percent of the total
reduction in nitrogen loading needed to maintain the reduction in
nutrient loads agreed to by the signatory states in the Chesapeake Bay
Agreement (40 percent of ``controllable by the year 2000).
The analysis of economic benefits and costs found in Section
IV.D.5. does not account for the economic benefits of reduced drinking
water nitrate levels and reduced terrestrial nitrogen deposition
expected to result from today's proposal, if implemented. The analysis
does, however, account for the economic benefits of reduced
eutrophication.

I. Conclusion: Cleaner Cars and Light Trucks Are Critically Important
to Improving Air Quality

Despite continued progress in reducing emissions from cars and
light trucks, these vehicles will continue to contribute a substantial
share of the ozone and PM precursors in current and projected
nonattainment areas, and in upwind areas whose emissions contribute to
downwind nonattainment, unless additional measures are taken to reduce
their emissions. These vehicles will also continue to contribute to the
ambient PM that affects visibility in Class I federal areas and some
urban areas. Emissions from cars and light trucks also play a
significant role in a wide range of health and environmental problems,
including known and potential cancer risks from inhalation of air
pollutants (a problem that could become more significant if sales of
diesel-powered cars and light trucks were to increase), health risks
from elevated drinking water nitrate levels, acidification of lakes and
streams, and eutrophication of inland and coastal waters.
Today's proposal would reduce NO_X, VOC, CO, PM, and
SO<INF>X</INF> emissions from these vehicles substantially. These
reductions would help reduce ozone levels nationwide and reduce the
extent and severity of violations of both the 1-hour and 8-hour ozone
standards. These reductions would also help reduce PM levels, both by
reducing direct PM emissions and by reducing emissions that give rise
to secondary PM. The NO_X and SO<INF>X</INF> reductions would
help reduce acidification problems, and the NO_X reductions
would help reduce eutrophication problems and drinking water nitrate
levels. The PM standards proposed today would help improve visibility
and would help mitigate the adverse health effects due to possible
increases in light-duty diesel engine sales.
Section IV.D.5. of this preamble describes the comprehensive
analysis EPA has made of the net economic benefit of the requirements
we are proposing today. In that analysis, we have quantified many of
the public health and environmental benefits of the actions on an
annual, national scale. Estimates of the economic value of these
effects have been made for as many of the effects as possible, and
compared to the cost of compliance. This rulemaking is the first
instance in which EPA has conducted such a cost-benefit analysis for a
set of proposed vehicle emission standards.

IV. What Are We Proposing and Why?

In the previous section, we showed why many states need as much
emission reduction as is reasonably possible from LDVs and LDTs--plus
reductions from other sources--if they are to reach and maintain
compliance with the 1-hour and 8-hour ozone NAAQS. We also pointed out
that these reductions would also be important in addressing PM and
other air quality and environmental problems in every major region of
the country.
In this section, we describe the comprehensive vehicle/fuel program
we are proposing to respond to these serious air quality needs.
Specifically, we discuss:

Our reasons for proposing a comprehensive vehicle and fuel
program, including why stringent LDV and LDT standards are feasible in
conjunction with low sulfur gasoline.

Our proposed vehicle-related requirements and our
rationales for proposing them.

Our proposed fuel-related requirements and our rationales.

Our projections of the economic impacts, cost
effectiveness, and monetized environmental and health benefits of the
proposed program.

Other program design options we have considered.

A. Why Are We Proposing Vehicle and Fuel Standards Together?

1. Feasibility of Stringent Standards for Light-Duty Vehicles and
Light-Duty Trucks.
a. Gasoline Fueled Vehicles. We believe that the standards being
proposed today for gasoline-fueled vehicles are well within the reach
of existing control technology. Our proposed determination of
feasibility is based on the use of catalyst-based strategies that are
already in use and are well proven on the existing fleet of vehicles.
In fact, as you will see below, many current engine families are
already certified to levels at or below the proposed new Tier 2
requirements. All of the certification and research testing discussed
below was performed on low-sulfur test fuel (nominally 30 ppm).
Certainly, larger vehicles and trucks, which are heavier and have
larger frontal areas, will face the biggest challenges. However,
conventional technology will be sufficient for even these vehicles,
especially in light of the extra leadtime we have provided before LDT3s
and LDT4s have to meet Tier 2 levels. We are also proposing to change
the test conditions for these trucks from ``adjusted loaded vehicle
weight'' to ``loaded vehicle weight.'' Adjusted loaded vehicle weight,
suitable for commercial truck operation, loads the truck to half of its
full payload. Loaded vehicle weight, on the other hand, represents curb
weight plus 300 pounds. The proposed change more accurately reflects
how these vehicles are used and makes heavy LDT testing consistent with
passenger car and light LDT testing. This change will make it
substantially easier for the heavier trucks to meet our proposed
standards.
Emission control technology has evolved rapidly in recent years.
Emission standards applicable to 1990 model year vehicles required
roughly 90% reductions in exhaust HC and CO

[[Page 26026]]

emissions and a 75% reduction in NO_X emissions compared to
uncontrolled emissions. Today, some vehicles currently in production
are well below these levels, showing overall emissions reductions of
all three of these pollutants. These vehicles' emissions are well below
those necessary to meet the current federal Tier 1 and even California
Low-Emission Vehicle (LEV) standards. The reductions have been brought
about by ongoing improvements in engine air-fuel management hardware
and software plus improvements in catalyst designs, all of which are
described fully in the Draft RIA.
The types of changes being seen on current vehicles have not yet
reached their technological limits and continuing improvement will
allow both LDVs and LDTs to meet the proposed standards. The Draft RIA
describes a range of specific techniques that we believe could be used.
These range from improved computer software and engine air-fuel
controls to increases in precious metal loading and other exhaust
system/catalyst system improvements. All of these technologies are
currently used on one or more production vehicle models. There is no
need to invent new approaches or technologies. The focus of the effort
is primarily development, application, and optimization of these
existing technologies.
We can gain significant insight into the difficulty of meeting the
proposed new standards by looking at current full-life certification
data. There are at least 48 engine family-control systems combinations
certified in 1999 at levels below the Tier 2 NO_X standard of
0.07 g/mi. Of these, 35 also have hydrocarbon levels of 0.09 g/mi or
below. Looking at a somewhat higher threshold to identify vehicles
certified near the proposed standard, there are an additional 113 car
and light truck families certified at levels between 0.07 g/mi and 0.10
g/mi NO_X.
All of the above vehicles are already able, or close to being able,
to certify to our proposed standards. The further reductions needed are
those to provide an ample safety margin, or cushion, between the
certified level and the emission standard. The degree of compliance
margin required is a function of a variety of factors designed to
provide the manufacturer a high confidence that production vehicles
will meet the standards in-use over their useful life. Historically,
these determinations are manufacturer specific, with cushions generally
growing smaller as standards decline (reflecting more precision and
repeatability in vehicle performance as more sophisticated controls are
developed). The 1999 certification data reflects compliance cushions
from as little as 20 percent below the standard to as high as 80
percent below the standard.
The cushion to be expected for Tier 2 vehicles is difficult to
establish, although some manufacturers claim a cushion of 50 percent
below the standard would be needed. We believe that manufacturers would
strive to use the smallest cushions possible in order to minimize the
impacts of the standards on their vehicles. Looking at 1999
certification data from this perspective and using a threshold of 0.04
g/mi NO_X, there are fully 22 engine family-control system
configurations at or below the 0.04 g/mi level (one of which is a
LDT4). Thus, even at such low levels, current technology is already
demonstrating the performance that would be necessary to meet the
proposed standards.
Since the most difficult compliance effort would be faced by the
larger LDTs, we have undertaken a technology demonstration program
aimed at lowering the emissions of a large 1999 LDT3 vehicle. This
vehicle has a high horsepower engine, four wheel drive, and a curb
weight of 4,500 pounds (GVWR ³⁰ of 6,100 lbs). The exhaust
system of the vehicle was modified to incorporate two close-coupled and
two underfloor catalytic converters. The catalytic converters were aged
to full useful life conditions using the accelerated aging methods
described by Theiss.³¹ For further details of the
modifications to this vehicle, please refer to the draft RIA.
---------------------------------------------------------------------------

\30\ Gross Vehicle Weight Rating. The curb weight of the vehicle
plus its maximum recommended load of passengers and cargo.
\31\ Theiss, J.R., ``Catalytic Converter Diagnosis Using the
Catalyst Exotherm,'' SAE Technical Paper Series, Paper No. 942058,
SAE Fuels and Lubricants Meeting and Exposition, Baltimore, MD,
October 17-20, 1994.
---------------------------------------------------------------------------

In our initial work we made no attempts to alter the calibration of
the electronic engine controls. In this configuration, the vehicle
achieved emissions levels of 0.060 <plus-minus> 0.002 g/mi
NO_X and 0.09 <plus-minus> 0.01 g/mi NMHC. Thus, by these
straightforward modifications to the catalyst system based upon
existing catalyst hardware, this vehicle was able to reach the proposed
Tier 2 levels. In order to achieve additional reductions in the test
vehicle's emissions, we are planning further work consisting largely of
elimination of fuel cut-offs during decelerations, slight increases in
EGR, and a minor degree of air injection during cold-start. However,
given the amount of leadtime before any of the proposed Tier 2
standards would begin, we believe that the work already done clearly
shows the feasibility of our proposal, even for large light-duty
trucks.
Figure IV.A.-1 shows the results of our testing in comparison to
the California LEV-1 standards applicable to this vehicle, and the
proposed Tier 2 standards.

BILLING CODE 6560-50-P

[[Page 26027]]

[GRAPHIC] [TIFF OMITTED] TP13MY99.001

BILLING CODE 6560-50-C

[[Page 26028]]

One of the challenges facing larger truck catalyst systems is
overtemperature protection. Because of this, our work on this vehicle
included temperature evaluation of the catalyst under maximum load
conditions. We found that the original fuel calibration for the truck
provided sufficient enrichment under wide-open-throttle conditions to
prevent exceeding the catalyst bed temperature limits (<difference>950
to 1000 deg.C) specified by the manufacturer of the catalytic
converters. We conducted chassis dynamometer testing over the
aggressive US06 cycle with the dynamometer inertia greatly increased to
simulate full GVWR load conditions (6,100 lbs) for the pickup. Catalyst
bed temperatures did not exceed 850 deg.C at any point during the
testing.
In addition to the EPA work, others have conducted several test
programs recently that help demonstrate the feasibility of our proposed
levels. The Coordinating Research Council (CRC), automobile
manufacturers, and the American Petroleum Institute (API) all tested a
number of light-duty vehicles capable of complying with the California
LEV or ULEV standards as part of an evaluation of the effects of sulfur
levels on emissions. Of the vehicles tested, seven met or nearly met
the Tier 2 design targets, and all were below the proposed 0.07 g/mi
NO_X and 0.09 g/mi NMOG standards.
Another program sponsored by MECA took two LDVs (a Crown Victoria
and a Buick LeSabre) and one LDT2 (a Toyota T100) certified to the
federal Tier 1 standards and replaced the original catalytic converters
with more advanced catalytic converters, thermally aged to roughly
50,000 miles. With these systems and some related emission control
modifications, all three vehicles' emissions were well below our
proposed 50,000 mile standards (0.05 g/mi NO_X, 0.075 g/mi
NMOG), and the Buick and the Toyota LDT2 met our estimated design
targets for those standards.
Finally, the California Air Resources Board (ARB) tested six
different production LEV light-duty vehicle models. Two of the six
models met the proposed Tier 2 design targets for NMOG and
NO_X. After installing low mileage advanced catalytic
converters and making some minor adjustments, all of the vehicles had
emission levels well below the proposed Tier 2 NMOG and NO_X
design targets. ARB also tested several Ford Expeditions (LDT4)
equipped with advanced catalytic converters. By adjusting several
parameters, they were able to reduce NO_X emissions to 0.06
g/mi and NMOG to 0.07 g/mi with a catalyst aged to 50,000 miles of use.
Neither the MECA nor the ARB test programs modified the basic
engine calibrations of the vehicles tested. It is very likely that such
recalibration could reduce emissions even further. Therefore, we
consider these actual test results to be a conservative estimate of the
capability of these advanced catalytic converters. This is especially
true for the Ford Expedition testing by ARB, where the engine software
appeared to modify its own calibration with the new catalyst,
counteracting some of the advantages of the new catalyst.
A more expanded analysis of the feasibility of the proposed
standards for gasoline fueled vehicles can be found in the Draft RIA,
considering the types of changes that will allow manufacturers to
extend effective new controls to the entire fleet of affected vehicles.
That analysis includes discussion of gasoline direct-injection engines,
as well as the feasibility of the proposed CO, formaldehyde and
evaporative emission standards. The conclusion of all of our analyses
is that the proposed standards would be feasible for gasoline-fueled
vehicles operated on low-sulfur gasoline. As gasoline-fueled vehicles
represent the overwhelming majority of the light-duty vehicle and truck
population, EPA proposes to find that the proposed standards would be
feasible overall for LDVs and LDTs.
b. Diesel Vehicles. As outlined above, we have decided to propose
standards that are intended to be ``fuel neutral.'' In today's
document, we propose to find that the Tier 2 standards are
technologically feasible and cost-effective for light-duty vehicles and
light-duty trucks overall, based on the discussion in Section IV.A.1.a.
above. Under the principal of fuel neutrality, all cars and light
trucks, including those using diesel engines, would be required to meet
the proposed Tier 2 standards. EPA believes that the proposed program,
including the phase-in periods, would facilitate the advancement of
clean diesel engine technologies. EPA further believes that in the long
term the standards would be within reach for diesel-fueled vehicles in
combination with appropriate changes to diesel fuel to facilitate
aftertreatment technologies.
As with gasoline engines, manufacturers of diesels have made
abundant progress over the past 10 years in reducing engine-out
emissions from diesel engines. In heavy trucks and buses, PM emission
standards, which were projected to require the use of exhaust
aftertreatment devices, were actually met with only engine
modifications. NO_X emissions from heavy trucks and buses
sold starting in 2002 will also reflect deep reductions from emission
levels typical of engines produced in the mid-1980's. Indeed, emissions
and performance of lighter diesel engines are rapidly approaching the
characteristics of gasoline engines, while retaining the durability and
fuel economy advantages that diesels enjoy. Against this background of
continuing progress, we believe that the technological improvements
that would be needed could be made in the time that would be available
before diesels would have to meet the new Tier 2
standards.³²
---------------------------------------------------------------------------

\32\ We generally expect that manufacturers would take advantage
of the flexibilities in today's proposal to delay the need for
diesel vehicles to meet the final Tier 2 levels until late in the
phase-in period. Because diesel vehicles represent a very small
percentage of the LDV/LDT market, diesels would not fall under the
final Tier 2 standards until 2009, giving manufacturers a relatively
large amount of leadtime. As discussed below, we are issuing an
Advance Notice of Proposed Rulemaking intended to solicit comment on
the need for reduced sulfur in diesel fuel in order to meet these
standards. We also believe that the proposed interim standards would
be feasible for diesels by 2004, with or without the fuel change,
given the flexibilities associated with those standards.
---------------------------------------------------------------------------

While reductions in ``engine-out'' emissions, including
incorporation of EGR strategies, will continue to be made, increasing
emphasis is being placed on various aftertreatment devices for diesels.
This is because further reductions in engine-out emissions will be
unlikely, by themselves, to allow diesels to comply with the proposed
Tier 2 standards for NO_X and PM. Rather, diesels would
require the use of highly effective aftertreatment devices.
For NO_X emissions, potential aftertreatment technologies
include lean NO_X catalysts, NO_X adsorbers and
selective catalytic reduction (SCR). Lean NO_X catalysts are
still under development, but generally appear capable of reducing
NO_X emissions by about 15-30%. This efficiency is not likely
to be sufficient to enable compliance with the proposed Tier 2
standards, but it could be used to meet the interim standards that
would begin in 2004.
NO_X adsorbers appear to be up to 90% efficient at
removing NO_X from the exhaust. Efficiency in this range is
likely to be sufficient to enable compliance with the proposed Tier 2
standards. NO_X adsorbers temporarily store the
NO_X and thus the engine must be run periodically for a brief
time with excess fuel, so that the stored NO_X can be
released and converted to nitrogen and oxygen using a conventional
three-way

[[Page 26029]]

catalyst, like that used on current gasoline vehicles.
There is currently a substantial amount of development work being
directed at NO_X adsorber technology. While there are
technical hurdles to be overcome, progress is continuing and it is our
judgement that the technology should still be available by the time it
would be needed for the proposed Tier 2 standards.
One serious concern with current NO_X adsorbers is that
they are quickly poisoned by sulfur in the fuel. Some manufacturers
have strongly emphasized their belief that, in order to meet the Tier 2
levels, low sulfur diesel fuel would also be required to mitigate or
prevent this poisoning problem. One solution would be to reduce sulfur
to very low levels. Another solution would be to reduce sulfur
somewhere below current levels and develop a way to periodically remove
the sulfur from the adsorber. In any event, this technique, if used,
would also require low sulfur diesel fuel.
SCR has been demonstrated commercially on stationary diesel engines
and can reduce NO_X emissions by 80-90%. This efficiency
would be sufficient to enable compliance with the proposed Tier 2
standards. However, SCR requires that the chemical urea be injected
into the exhaust before the catalyst to assist in the destruction of
NO_X. The urea must be injected at very precise rates, which
is difficult to achieve with an on-highway engine, because of widely
varying engine operating conditions. Otherwise, emissions of ammonia,
which have a very objectionable odor, can occur. Substantial amounts of
urea are required, meaning that vehicle owners would have to replenish
their vehicles' supply of urea frequently. As the engine and vehicle
will operate satisfactorily without the urea (only NO_X
emissions would be affected), some mechanism would be needed to ensure
that vehicle owners maintained their supply of urea. Otherwise, little
NO_X emission reduction would be expected in-use.
Regarding PM, applicable aftertreatment devices tend to fall into
two categories: oxidation catalysts and traps. Diesel oxidation
catalysts can reduce total PM emissions by roughly 15-30%. They would
need to be used in conjunction with further reductions in PM engine-out
emissions in order to meet the proposed Tier 2 standards. Diesel
particulate traps, on the other hand, can eliminate up to 90% of diesel
PM emissions. However, some of the means of accomplishing the
regeneration of particulate traps involve catalytic processes that also
convert sulfur dioxide in the exhaust to sulfate. These techniques, if
used, would also require a low sulfur fuel.
Since we have noted that some of the options for diesel
aftertreatment may require lower sulfur diesel fuel than is currently
available, the question of diesel fuel quality improvement arises.
Manufacturers have argued that low sulfur diesel fuel will be required
to permit diesels to meet the proposed new standards. While we believe
that low sulfur diesel fuel would likely be required to enable diesel
engines to meet the proposed Tier 2 standards, this proposal does not
include provisions for such fuel. We need additional information about
the specific aftertreatment solutions that could be used to meet the
standards, the effectiveness of these approaches in reducing PM and
NO_X emissions and their sensitivity to diesel sulfur, and
improvements or alternatives that might reduce the impacts of fuel
sulfur.
To deal more thoroughly with this matter, we are issuing an
Advanced Notice of Proposed Rulemaking on a parallel path with today's
Tier 2 proposal. As a part of that process, EPA will assess the effect
of low-sulfur fuel on the ability of diesels to meet Tier 2 standards
for LDVs and LDTs. It will also consider the issue of the relation of
diesel fuel quality to future standards for heavy-duty on-highway
diesel engines and nonroad diesel engines. Our plans for this Advanced
Notice are discussed further in section IV.C. below. In any case, we
believe that the standards proposed today are appropriate and feasible
overall for LDVs and LDTs.
2. Gasoline Sulfur Control Is Needed To Support the Proposed Vehicle
Standards
As we discussed in the previous section, we believe that the
stringent standards we propose are needed to meet air quality goals are
feasible for LDVs and LDTs. At the same time, we believe that for these
standards to be feasible for gasoline LDVs and LDTs, low sulfur
gasoline must be made available. The following paragraphs explain why
we think gasoline sulfur control must accompany Tier 2 vehicle
standards.
Catalyst manufacturers generally use low sulfur gasoline in the
development of their catalyst designs. Vehicle manufacturers then equip
their vehicles with these catalysts and EPA certifies them to the
exhaust emission standards, usually based on testing the manufacturer
does using low sulfur gasoline. However, fundamental chemical and
physical characteristics of exhaust catalytic converter technology
generally result in a significant degradation of emission performance
when these vehicles use gasoline with sulfur levels common in most of
the country today. This sensitivity of catalytic converters to gasoline
sulfur varies somewhat depending on a number of factors, some better
understood than others. Clearly, however, as we discuss in the
following paragraphs, gasoline sulfur's impact is large, especially in
vehicles designed to meet very low emission standards like those
proposed today.
This is the reason EPA has decided to propose a comprehensive
approach to addressing emissions from cars and light trucks, including
provisions to get low sulfur gasoline into the field in the same time
frame needed for Tier 2 vehicles. (We discuss the related fact that the
sulfur impact on catalyst performance is not fully reversible in
Section IV.C. below, in the context of EPA's preference for a
nationwide versus a regional gasoline sulfur control program, and in
the Draft RIA.)
a. How Does Gasoline Sulfur Affect Vehicle Emission Performance? We
know that gasoline sulfur has a negative impact on vehicle emission
controls. Vehicles depend on the catalytic converter to reduce
emissions of HC, CO, and NO_X. Sulfur and sulfur compounds
attach or ``adsorb'' to the precious metal catalysts that are required
to convert these emissions. Sulfur also blocks sites on the catalyst
designed to store oxygen that are necessary to optimize NO_X
emissions conversions. While the amount of sulfur contamination can
vary depending on the metals used in the catalyst and other aspects of
the design and operation of the vehicle, some level of sulfur
contamination will occur in any catalyst.
Sulfur sensitivity is impacted not only by the catalyst formulation
(the types and amounts of precious metals used in the catalyst) but
also by factors including the following:

the materials used to provide oxygen storage capacity in
the catalyst, as well as the general design of the catalyst,

the location of the catalyst relative to the engine, which
impacts the temperatures inside the catalyst,

the mix of air and fuel entering the engine over the
course of operation, which is varied by the engine's computer in
response to the driving situation and affects the mix of gases entering
the catalyst from the engine, and

the speeds the car is driven at and the load the vehicle
is carrying, which

[[Page 26030]]

also impact the temperatures experienced by the catalyst.
Since these factors vary for every vehicle, the sulfur impact
varies for every vehicle to some degree. There is no single factor that
guarantees that a vehicle will be very sensitive or very insensitive to
sulfur. We now believe that there are not (and will not be in the
foreseeable future) emission control devices available for gasoline-
powered vehicles that can meet the proposed Tier 2 emission standards
that would not be significantly impaired by gasoline with sulfur levels
common today.
b. How Large Is Gasoline Sulfur's Effect on Emissions? High sulfur
levels have been shown to significantly impair the emission control
systems of cleaner, later technology vehicles. The California LEV
standards and Federal NLEV standards, as well as California's new LEV-
II standards and our proposed Tier 2 standards, require catalysts to be
extremely efficient to adequately reduce emissions over the full useful
life of the vehicle. Recent test programs conducted by the automotive
and oil industries show that LEV and ULEV vehicles can experience, on
average, a 40% increase in NMHC and 134% increase in NO_X
emissions when operated on 330 ppm sulfur fuel (approximately the
current national average sulfur level) compared to 30 ppm sulfur fuel.
This level of emissions increase is significant enough on its own
to potentially cause a vehicle to exceed the proposed full useful life
emission standards when operated on sulfur levels that are
substantially higher than the levels proposed today, even with the
margin of safety that auto manufacturers generally include. Average
sulfur levels in the U.S. are currently high enough to significantly
impair the emissions control systems in new technology vehicles, and to
potentially cause these vehicles to fail emission standards required
for vehicles up through 100,000 miles (or more) of operation.
For older vehicles designed to meet Tier 0 and Tier 1 emission
standards, the effect of sulfur contamination is somewhat less. Still,
testing shows that gasoline sulfur increases emissions of NMHC and
NO_X by almost 17% when one of these vehicles is operated on
gasoline containing 330 ppm sulfur compared to operation on gasoline
with 30 ppm sulfur. Thus, Tier 0 and Tier 1 vehicles can also have
higher emissions when they are exposed to sulfur levels substantially
higher than the proposed sulfur standard. This increase is generally
not enough to cause a vehicle to exceed the full useful life emission
standards in practice, but it can result in in-use emissions increases
since the vehicle could emit at levels higher than it would if it
operated consistently on 30 ppm sulfur gasoline.
Gasoline sulfur control to 30 ppm would achieve about 700,000 tons
of NO_X reductions per year from LDVs and LDTs by 2020. This
represents about a third of the national NO_X emission
reductions otherwise available from these vehicles. Without these
potential emission reductions, many states would face the potentially
unmeetable challenge of finding enough other cost-effective sources of
NO_X emission reductions to address their ozone nonattainment
and maintenance problems.
Other implications of continued use of high-sulfur gasoline include
the following:

Other important potential air quality benefits would not
be realized throughout the country, including reduction in direct
emissions of sulfur dioxide, secondary formation of nitrate PM from
NO_X emissions, reductions in regional haze, reductions in
air toxics emissions and other pollution problems described in Section
III above.

The immediate and very significant improvements that lower
sulfur gasoline would bring in the emissions performance of vehicles
already on the road would not occur.

Advanced emission control technologies now being
developed, all of which appear equally or even more sensitive to
gasoline sulfur levels than current technologies, would not be
available to the U.S. vehicle market (for example, very fuel efficient
technologies like gasoline direct injection technology and fuel cells).

Finally, any interference with onboard emission control
system diagnostic (OBD) systems that high-sulfur gasoline causes would
remain in the absence of a low-sulfur gasoline program.
3. A Comprehensive Vehicle/Fuel Approach Is Therefore Necessary
Based on this information, we have concluded that sulfur levels in
gasoline must be reduced to enable these catalysts to operate properly
and for the needed air quality benefits of this program to be achieved.
In today's action, therefore, we are proposing a comprehensive,
integrated program of stringent vehicle emission standards in
combination with stringent gasoline sulfur standards. The proposal is
carefully designed to address the need for refiners to make low-sulfur
gasoline available at very nearly the same time as auto makers begin
selling large numbers of Tier 2 vehicles. We have tried to take into
account all potential areas of interaction between the vehicle and
gasoline sulfur parts of the proposal, and as a result we believe that
the overall proposed program would achieve the expected environmental
goals while minimizing the economic and administrative burdens on the
affected industries. We encourage all commenters to consider and
discuss the interrelationships among the elements of the program when
they comment on individual provisions.

B. Our Proposed Program for Vehicles

We have held a series of meetings with the various stakeholders
impacted by this action. We have seriously considered their input in
developing our proposal and believe the program laid out below and the
areas upon which we are seeking comment are responsive to their
concerns. One part of this input was provided by a broad representation
of the LDV/LDT manufacturing industry, represented by the Alliance of
Automobile Manufacturers, and offered constructive recommendations on a
number of elements of a vehicle emission control program. We have
considered many of their ideas and issues in the design of the proposed
program and we are seeking comment on a number of others. The
``Alliance'' proposal is documented in the docket in a letter to EPA
dated March 26, 1999.
The next sections of the preamble describe our proposal in detail.
1. Overview of the Proposed Vehicle Program
The vehicle-related part of today's proposal covers a wide range of
standards, concepts, and provisions that affect how vehicle
manufacturers would develop, certify, produce, and market Tier 2
vehicles. This Overview subsection provides readers with a broad
summary of the major vehicle-related aspects of the proposal. Readers
for whom this Overview is sufficient may want to move on to the
discussion of the key gasoline sulfur control provisions (Section
IV.C.). Readers wishing a more detailed understanding of the proposed
vehicle provisions can continue beyond the Overview to deeper
discussions of key issues and provisions (Sections IV.B.-2, 3, and 4)
as well as discussions of additional provisions (Section V.A.). Readers
should refer to the regulatory language found at the end of this
preamble for a complete compilation of the proposed requirements.
a. Introduction. Today's proposal for Tier 2 vehicle standards
incorporates concepts from the federal NLEV program. The program takes
the

[[Page 26031]]

corporate averaging concept and other provisions from NLEV but changes
the focus from NMOG to NO_X. The emission standard ``bins''
used for this average calculation are different in several respects
from those of the California LEV II program, yet we have designed them
to allow harmonization of federal and California vehicle technology. As
discussed below, the Tier 2 corporate average NO_X level to
be met through these requirements ultimately applies to all of a
manufacturer's LDVs and LDTs (subject to two different phase-in
schedules) regardless of what fuel is used.
In the discussions below, we propose different Tier 2 phase-in
schedules for two different groups of vehicles as well as two different
sets of interim standards for 2004 and later model year vehicles not
yet phased-in to the Tier 2 standards. To understand how the program
would work, it is necessary first to understand EPA's classification
system for light-duty vehicles and trucks.
The light duty category of motor vehicles includes all vehicles and
trucks under 8500 pounds gross vehicle weight rating, or GVWR (i.e.,
vehicle weight plus rated cargo capacity). Table IV.B.-1 shows the
various light duty categories. In the discussion below, we make
frequent reference to two separate groups of light vehicles: (1) LDV/
LLDTs, which include all LDVs and all LDT1s and LDT2s; and (2) HLDTs,
which include LDT3s and LDT4s.

Table IV.B.-1.--Light Duty Vehicles and Trucks; Category Characteristics
------------------------------------------------------------------------
Characteristics
------------------------------------------------------------------------
LDV.................................... A passenger car or passenger
car derivative seating 12
passengers or less.
Light LDT (LLDT)....................... Any LDT rated at up through
6,000 lbs GVWR. Includes LDT1
and LDT2.
Heavy LDT (HLDT)....................... Any LDT rated at greater than
6,000 lbs GVWR, but not more
than 8,500 lbs GVWR. Includes
LDT3 and LDT4.
------------------------------------------------------------------------

As discussed below, the Tier 2 program would take effect in 2004,
with full phase in occurring by 2007 for LDV/LLDTs and 2009 for HLDTs.
During the phase-in years of 2004-2008, vehicles not certified to Tier
2 requirements would meet interim requirements that would also employ a
bins system, but with less stringent corporate average NO_X
standards.

References to California LEV II Program

Throughout this preamble, we make reference to California's LEV II
program and its requirements. The LEV II program was approved by the
California ARB at a hearing of November 5, 1998. Numerous draft
documents were prepared by ARB staff in advance of that hearing and
made available to the public. Some of those documents have now been
modified as a result of changes to the proposed program made at the
hearing and due to comments received after the hearing.
However, when this NPRM was assembled for signature, the documents
related to the LEV II program had still not been finalized. In fact, a
15 day public review of the program was scheduled for April 15-30,
1999. After that review, ARB expected to be able to formally adopt the
program and issue final documents without significant change.
We have placed copies of the latest available documents, some of
which we used in the preparation of this NPRM, in the docket. You may
also obtain these documents and other information about California's
LEV II program from ARB's web site: (www.arb.ca.gov/regact/levii/
levii.htm).
In the regulatory text that follows this preamble, we propose to
incorporate by reference a number of documents related to LEVII and
California test procedures under LEVII. ARB expects to finalize the LEV
II program without significant changes before we issue a final rule. We
will review any changes to the final version of the LEV II program and
its supporting documents and consider them for inclusion in the federal
program when we prepare our final rule.
b. Corporate Average NO_X Standard. The program we are
proposing today would ultimately require each manufacturer's average
NO_X emissions over all of its Tier 2 vehicles each model
year to meet a NO_X standard of 0.07 g/mi. Manufacturers
would have the flexibility to certify Tier 2 vehicles to different sets
of exhaust standards that we refer to as ``bins,'' but would have to
choose the bins so that their corporate sales weighted average
NO_X level for their Tier 2 vehicles was no more than the
0.07 g/mi. (We discuss the bins in the next subsection.)
The value of a corporate average standard is that the program's air
quality goals would be met while allowing manufacturers the flexibility
to certify some models above and some models below the standard. Each
manufacturer would determine its year-end corporate average
NO_X level by computing a sales-weighted average of the
NO_X standards from the various bins to which it certified
any Tier 2 vehicles. The manufacturer would be in compliance with the
standard if its corporate average NO_X emissions for its Tier
2 vehicles met the 0.07 g/mi level.
c. Tier 2 Emission Standard ``Bins''. We are proposing seven
emission standard bins, each one a set of standards to which
manufacturers could certify their vehicles. (Table IV.B.-2. in Section
IV.B.-4.a. below shows all the standards associated with each bin.)
Several bins have the same values as the California LEV II program.
Further, we added three bins that are not a part of the California
program to increase the flexibility of the program for manufacturers.
As further discussed in Section IV.B.4. below, we believe these extra
bins would help provide incentives for manufacturers to produce
vehicles with emissions below 0.07 g/mi NO_X.
The corporate average concept using the seven bins would provide a
program that gets the same emission reductions we would expect from a
straight 0.07 g/mi standard for all vehicles because all NO_X
emissions from Tier 2 vehicles in bins above 0.07 g/mi would need to be
offset by NO_X emissions from Tier 2 vehicles in bins below
0.07 g/mile. This focus on NO_X allows NMOG emissions to
``float'' in that the fleet NMOG emission rate depends on the mix of
bins used to meet the NO_X standard. However, you can see by
examining the bins we are proposing, that any combination of vehicles
meeting the 0.07 g/mi average NO_X standard would have
average NMOG levels at or below 0.09 g/mi. In addition, there will be
overall improvements in NMOG since Tier 2 incorporates HLDTs, which are
not covered by the NLEV program.
d. Schedules for Implementation. We recognize that the Tier 2
standards pose greater technological challenges for larger light duty
trucks than for LDVs and smaller trucks. We believe that additional
leadtime is appropriate for HLDTs. HLDTs have historically been subject
to the least stringent vehicle-based standards. Also, HLDTs were not
subject to the voluntary emission reductions implemented for LDVs,
LDT1s and LDT2s in the NLEV program. Consequently we have designed
separate phase-in programs for the two groups. Our phase-in approach
would provide HLDTs with extra time before they would need to begin
phase-in to the Tier 2 standards and also provide two additional years
for them to fully comply. Figure IV.B-1 provides a graphical
representation of how the phase-in of the Tier 2 program would work for
all vehicles. This figure shows several aspects of the proposed
program:

Phase-in/phase-out requirements of the interim programs;

[[Page 26032]]

Phase-in requirements of new evaporative standards;

Years that could be included in alternative phase-in
schedules;

Years in which manufacturers could bank NO_X
credits through ``early banking''; and

``Boundaries'' on averaging sets in the Tier 2 and interim
programs.
We discuss each of these topics in detail below and make numerous
references to Figure IV.B-1.

BILLING CODE 6560-50-P

[[Page 26033]]

[GRAPHIC] [TIFF OMITTED] TP13MY99.002

BILLING CODE 6560-50-C

[[Page 26034]]

i. Implementation Schedule for LDVs and LLDTs

We are proposing that the Tier 2 standards take effect beginning
with the 2004 model year for light duty vehicles and trucks at or below
6000 pounds GVWR (LDV/LLDTs). We are proposing that manufacturers would
phase their vehicles into the Tier 2 program beginning with 25 percent
of LDV/LLDT sales that year, 50 percent in 2005, 75 percent in 2006,
and 100 percent in 2007. Manufacturers would be free to choose which
vehicles were phased-in each year. However, in each year during (and
after) the phase-in, the manufacturer's average NO_X for its
Tier 2 vehicles would have to meet the 0.07 g/mi corporate average
standard. This phase-in schedule would provide between five and eight
years of leadtime for the manufacturers to bring all of their LDV/LLDT
production into compliance. These vehicles constitute nearly 90 percent
of the light duty fleet.
To increase manufacturer flexibility and provide incentives for
early introduction of Tier 2 vehicles, we are proposing that
manufacturers could use alternative phase-in schedules that would
require 100 percent phase-in by 2007, but would recognize the benefits
of early introduction of Tier 2 vehicles, and allow manufacturers to
adjust their phase-in to better fit their own production plans.

ii. Implementation Schedule for HLDTs

To provide greater leadtime for HLDTs we are proposing that the
Tier 2 phase-in schedule would start later and end later than that for
LDVs and LLDTs. In our proposal 50 percent of each manufacturer's HLDTs
would be required to meet Tier 2 standards in 2008, and 100 percent
would have to meet Tier 2 standards in 2009. As with the LDV/LLDTs, the
Tier 2 HLDTs would have to meet a corporate average NO_X
standard of 0.07 g/mi. This delayed phase-in schedule would provide
manufacturers with nine years of lead time before they would need to
bring any HLDTs into compliance with Tier 2 standards. As for the LDV/
LLDTs above, to encourage early introduction of Tier 2 HLDTs and to
provide manufacturers with greater flexibility, we are proposing that
manufacturers could use alternative phase-in schedules that would still
result in 100% phase-in by 2009.
We request comment on the appropriateness of this separate schedule
for HLDTs.
e. LDVs and LDTs Not Covered by Tier 2. The two groups of vehicles
(LDV/LLDTs and HLDTs) will be approaching the Tier 2 standards from
quite different emission ``backgrounds.'' LDV/LLDTs will be at NLEV
levels, which require NO_X emissions of either 0.3 or 0.5g/mi
on average ³³, while HLDTs will be at Tier 1 levels facing
NO_X standards of either 0.98 or 1.53 g/mi, depending on
truck size. These Tier 1 NO_X levels for HLDTs are very high
relative to our 0.07 g/mi Tier 2 NO_X average. To address the
disparity in emission ``backgrounds'' while gaining air quality
benefits from vehicles during the phase-in period, we are proposing
separate sets of interim standards for the two vehicle groups during
the phase-in period. The provisions described below would apply in 2004
for all LDVs and LDTs not certified to Tier 2 standards. The
relationship of the interim programs to the final Tier 2 standards is
shown in Figure IV.B-1.
---------------------------------------------------------------------------

\33\ The NLEV program imposes NMOG average standards that would
lead to full useful life NO_X levels of about 0.3 g/mi for
LDV/LDT1s and 0.5 g/mi for LDT2s.
---------------------------------------------------------------------------

i. Interim Standards for LDV/LLDTs

Beginning with the 2004 model year, all new LDVs and LLDTs not
incorporated under the Tier 2 phase-in would be subject to an interim
corporate average NO_X standard of 0.30 g/mi. This is the
nominal LEV NO_X emission standard for LDVs and LDT1s under
the NLEV program. This interim program would hold LDVs and LLDTs not
covered by the Tier 2 standards during the phase-in to NLEV levels and
bring about NO_X emission reductions from LDT2s . By
implementing these interim standards for LDVs and LLDTs we will ensure
that the accomplishments of the NLEV programs are continued. Because
the Tier 2 standards are phased-in beginning in the 2004 model year,
the interim standards for LDVs and LLDTs apply to fewer vehicles each
year, i.e., they are ``phase-out'' standards. Figure IV.B-1 shows the
maximum percentage of LDVs and LLDTs that would normally be subject to
the interim standards each year.
As the interim program for LDV/LLDTs is designed to hold these
vehicles to NLEV levels, it employs bins derived from the NLEV program.
These bins are shown in Tables IV.B.-6 and -7.

ii. Interim Standards for HLDTs.

Our interim standards for HLDTs would begin in 2004. The Interim
Program for HLDTs would set a corporate average NO_X standard
of 0.20 g/mi that would be phased in between 2004 and 2007. The interim
HLDT standards, like those for LDV/LLDTs would be built around a set of
bins (See Tables IV.B.-8 and -9).
As shown in Figure IV.B.-1, the phase-in would be 25 percent in the
2004 model year, 50 percent in 2005, 75 percent in 2006, and 100
percent in 2007. The program would remain in effect through 2008 to
cover those HLDTs not yet phased into the Tier 2 standards (a maximum
of 50%). Vehicles not subject to the interim corporate average
NO_X standard during the 2004-2006 phase-in years would be
subject to the least stringent bin (Bin 5) so their NO_X
emissions would be effectively capped at 0.60 g/mi. These vehicles
would be excluded from the calculation to determine compliance with the
interim 0.20 g/mi average NO_X standard.
This proposed approach would implement standards significantly
lower than the Tier 1 NO_X standards currently applicable to
these vehicles. While manufacturers already certify many HLDTs at or
below these levels, we believe these interim standards represent a
reasonable step toward the Tier 2 standards and would provide
meaningful control in the near term relative to current levels and Tier
1. This approach would allow more time for manufacturers to bring the
more difficult HLDTs to Tier 2 levels while achieving real reductions
from those HLDTs that may present less of a challenge.

iii. Interim Programs Would Provide Reductions over Previous Standards

As was the case with the primary Tier 2 bin structure, the bin
structure for the interim programs would focus on NO_X and
yet should provide further reductions in NMOG beyond the NLEV program
(See Tables IV.B.-6,7,8 and 9). This is because the interim programs
would reduce emissions from LDT2s and HLDTs compared to their previous
standards. Without the interim standards, HLDTs could be certified as
high as 0.46 g/mi or 0.56 g/mi, the Tier 1 NMHC levels. With the
interim standards, however, exhaust NMOG should average approximately
0.09 g/mi for all non-Tier 2 LDV/LLDTs. and 0.25 g/mi or less for
HLDTs.

iv. Alternative Approach for Interim Standards

An alternative flexible approach for reducing the emissions from
vehicles and trucks prior to their phase-in to Tier 2 standards would
be to employ a declining NO_X average, or perhaps separate
declining NO_X averages for LDV/LLDTs and HLDTs. In this
approach, manufacturers would certify vehicles to their choice of bins,
but

[[Page 26035]]

would have to meet an average NO_X standard (or standards)
that became lower each year. Manufacturers could bank NO_X
credits in early years of such a program for use in later years when
the standard tightened. We request comment on the benefits,
implications and drawbacks of such an approach. Commenters should
address the issues of (1) what added flexibility does this approach
provide beyond that provided by the bins and phase-in approach proposed
above, (2) how to handle potential windfall credits that could arise in
the early years under such an approach, (3) how a standard that changes
each year would impact technology phase-in and phase-out, and (4)
whether such an approach would require the implementation of declining
average standards for the other exhaust pollutants.
f. Generating, Banking, and Trading NO_X Credits. As
described above, we are proposing that manufacturers average the
NO_X emissions of their Tier 2 vehicles and comply with a
corporate average NO_X standard. In addition, we are
proposing that when a manufacturer's average NO_X emissions
fall below the corporate average NO_X standard, it could
generate NO_X credits that it could save for later use
(banking) or sell to another manufacturer (trading). NO_X
credits would be available under the Tier 2 standards, the interim
standards for LDVs and LLDTs, and the interim standards for HLDTs.
These NO_X credit provisions would facilitate compliance with
the fleet average NO_X standards and would be very similar to
those currently in place for NMOG emissions under California and
federal NLEV regulations.
A manufacturer with an average NO_X level for its Tier 2
vehicles in a given model year below the 0.07 gram per mile corporate
average standard would generate Tier 2 NO_X credits that it
could use in a future model year when its average NO_X might
exceed the 0.07 standard. Manufacturers would calculate their corporate
average NO_X emissions and then compute credits based on how
far below 0.07 g/mi the corporate average fell.
Manufacturers would be free to retain any credits they generate for
future use or to trade (sell) those credits to other manufacturers.
Credits retained or purchased could be used by manufacturers with
corporate average Tier 2 NO_X levels above 0.07 g/mi.
Manufacturers could certify LDVs and LLDTs to Tier 2 standards as early
as the 2001 model year and receive NO_X credits for their
efforts. They could use credits generated under these ``early banking''
provisions after the Tier 2 phase-in begins in 2004 (2008 for HLDTs).
Banking and trading of NO_X credits under the interim
non-Tier 2 standards would be similar, except that a manufacturer would
determine its credits based upon the 0.30 or 0.20 gram per mile
corporate average NO_X standard applicable to vehicles in the
interim programs. There would be no provisions for early banking under
the interim standards and manufacturers would not be allowed to use
interim credits to address the Tier 2 NO_X average standard.
Interim credits from LDVs/LLDTs and interim credits from HLDTs could
not be used interchangeably due to the differences in the interim
corporate average NO_X standards. We seek comment on allowing
exchanges of credits between the LDV/LLDT interim program and the HLDT
interim program.
Banking and trading of NO_X credits and related issues
are discussed in greater detail in Section IV.B.-4.d. below.
2. Why Are We Proposing the Same Set of Standards for Tier 2 LDVs and
LDTs?
Before we provide a more detailed description of the proposed
vehicle program, two overarching principles of today's proposal are
worth explaining in some detail. The first of these is our proposal to
bring all LDVs and LDTs under the same set of emission standards.
Historically, LDTs--and especially the heavier trucks in the LDT3 and
LDT4 categories--have been subject to less stringent emission standards
than LDVs (passenger cars). In recent years the proportion of light
truck sales has grown to approximately 50 percent. Many of these LDTs
are minivans, passenger vans, sport utility vehicles and pick-up trucks
that are used primarily or solely for personal transportation; i.e.,
they are used like passenger cars and there are more annual vehicle
miles of travel as a result.
As vehicle preferences have increasingly shifted from passenger
cars to light trucks there has been an accompanying increase in
emissions over what otherwise would have occurred, because of the
increase in miles traveled and the less stringent standards for LDTs as
compared to LDVs. As Section III. above makes clear, reductions in
these excess emissions (and in other mobile and stationary source
emissions) are seriously needed. Since both LDVs and LDTs are within
technological reach of the standards in the proposed Tier 2 bin
structure, we are proposing to equalize the regulatory useful life
periods for LDVs and LDTs and to apply the same Tier 2 exhaust emission
standard bins to all of them.
Once the phase in periods end for all vehicles in 2009,
manufacturers would include all LDVs and LDTs together in calculating
their corporate average NO_X levels.³⁴ As
mentioned above and described in more detail in Section IV.B.-4. below,
manufacturers could choose the emission bin for any test group of
vehicles provided that on a sales weighted average basis, the
manufacturer met the average NO_X standard of 0.07 g/mi for
its Tier 2 vehicles that year.
---------------------------------------------------------------------------

\34\ Because of the different phase-in percentages and phase in
schedules for the two groups, we are proposing that during the
duration of the phase-in (through 2008) manufacturers would average
Tier 2 LDV/LLDTs separately from HLDTs.
---------------------------------------------------------------------------

Some have suggested that a program with different requirements
would be needed for heavy LDTs. Recognizing that compliance will be
most challenging for HLDTs, the delay in the start of the phase-in and
the additional phase-in years for those vehicles would allow
manufacturers to delay the initial impact of the Tier 2 standards until
the 2008 model year. This represents four additional model years of
leadtime beyond the time when passenger cars and LDT1s and LDT2s would
have achieved Tier 2 standards in substantial numbers. We believe this
phase-in and other provisions of this proposal respond to these
concerns. However, we request comments on the need for different
standards for these vehicles. Specifically, we request comment on
different levels for NMOG standards for these vehicles, including how
NMOG standards less stringent than our proposed standards might affect
the technological challenges presented by the proposed NO_X
standards.

Considerations for a 2004 Technology Review

EPA is seeking comment on whether it should conduct a technology
review of the Tier 2 standards in the future. As part of the input
received from stakeholders while developing this proposal, the Alliance
of Automobile Manufacturers suggested that the proposal include
consideration of a technology review, principally designed to assess
the status of Tier 2 technology development. As discussed above, we
recognize that HLDTs will face the greatest technological challenge in
complying with our proposed standards. Some manufacturers have
suggested that the approach of applying the same standard to cars and
light-duty trucks presents sufficient challenge as to raise serious
uncertainty about compliance for the larger vehicles, even in the 2008

[[Page 26036]]

time frame. In addition to the concerns expressed regarding the time
frame for implementation of the more stringent standards for HLDTs in
2008, manufacturers have indicated that there are questions of
feasibility for introduction of advanced technologies for improved fuel
economy, such as lean burn, fuel cell, and hybrid electric technology.
The review could assess the feasibility of the standards relative
to the state of technology development for HLDTs. Further, the review
could consider gasoline and diesel fuel quality and its impact on the
effectiveness of aftertreatment, and whether lower sulfur levels are
necessary for HLDTs to meet the Tier 2 standards. We may also examine
the feasibility of the standards for vehicles using technologies to
advance fuel economy. In addition, the review could consider whether
additional air quality improvements are necessary and the feasibility
of additional reductions of vehicle emissions to achieve such air
quality improvements. EPA believes that serious consideration of this
concept is warranted and if it determines such a review to be
appropriate, the best time to conduct such a review may be in the 2004
time frame, before the final Tier 2 standards go into effect for HLDTs.
EPA could conduct such a review to assess the feasibility, timing
and stringency of the standards relative to the state of technology
development. In doing so, EPA would determine whether or not there was
a need to formally consider a change in the final Tier 2 standards. If
such a change were determined to be necessary, EPA would conduct a
formal rulemaking, including conducting public hearings.
As part of the technology review, EPA would seek advice from all
appropriate stakeholders and could engage a peer review process. In
addition, such a process, if undertaken, could include public notice
and opportunity for comment on the review, including the holding of
public hearings by EPA. One way to structure the process would include
the establishment of an advisory panel under the Clean Air Act Advisory
Committee to provide assessment of the state of technology and the
feasibility of the standards. The Committee could recommend appropriate
action for the Administrator based on their findings. The Administrator
would then determine if any changes were needed to adjust the Tier 2
standards for HLDTs, advanced technologies, or the fuel parameters. We
request comment on the need for a technology review, scope of the
review and on the design of the process and its timing.
3. Why Are We Proposing the Same Standards for Both Gasoline and Diesel
Vehicles?
The second overarching principle of our vehicle proposal is to
apply the same Tier 2 standards to all light vehicles, regardless of
the fuel they are designed to use. The same exhaust emission standards
and useful life periods we are proposing today would apply whether the
vehicle is built to operate on gasoline or diesel fuel or on an
alternative fuel such as methanol or natural gas. Diesel engines used
in LDVs and LDTs tend to be used in the same applications as their
gasoline counterparts, and thus we believe they should meet the same or
very similar standards.
Manufacturers have expressed concerns that diesel-fueled vehicles
would have difficulty meeting NO_X and particulate matter
levels like those contained in today's proposal. Clearly, these
standards would be challenging. As discussed in Section IV.A.-1. above,
we expect that the proposed Tier 2 NO_X and NMOG standards
would be challenging for gasoline vehicles, but that major
technological innovations would not be required. For diesels, however,
the proposed NO_X and PM standards would likely require
applications of new types of aftertreatment with, perhaps, changes in
diesel fuel. We anticipate that manufacturers that chose to build
diesel vehicles would adopt aftertreatment technologies such as
NO_X storage catalysts and continuously regenerating
particulate traps to meet Tier 2 requirements.
Today, diesels comprise less than one-half of one percent of all
LDV/LDT sales. While this is a small fraction, the potential exists for
diesels to gain a considerable market share in the future. All one need
do is review the dramatic increase in recent years of diesel engine use
in the lightest category of heavy duty vehicles (8500-10,000 pounds
GVWR) to see the potential for significant diesel engine use in LDTs,
and perhaps LDVs, in the future. Just ten ago years diesels made up
less than 10 percent of this class of vehicles. In 1998, this fraction
approached 50 percent.
The potential impact of large-scale diesel use in the light-duty
fleet underscores the need for the same standards to apply to diesels
as for other vehicles. Given the health concerns associated with diesel
PM emissions (see Section III. above), we believe that it is prudent to
address PM emissions from diesel LDVs and LDTs while their numbers are
relatively small. In this way the program can minimize the PM impact
that would accompany significant growth in this market segment while
allowing manufacturers to incorporate low-emission technology into new
light-duty diesel engine designs.
4. Key Elements of the Proposed Vehicle Program
The previous subsections IV.B.-1., 2., and 3. provided an overview
of today's proposed vehicle program and the two overarching principles
that it is built on. This subsection elaborates on the major vehicle-
related elements of today's proposal. Later in this preamble, Section
V.A. discusses the rest of the proposed vehicle provisions.
a. Basic Exhaust Emission Standards and ``Bin'' Structure. The
program we are proposing today contains a basic requirement that each
manufacturer meet, on average, a full useful life NO_X
standard of 0.07 g/mi for all its Tier 2 LDVs and LDTs. Manufacturers
would have the flexibility to choose the set of standards that a
particular test group ³⁵ of vehicles must meet. For a given
test group of LDVs or LDTs, manufacturers would select a set of full
useful life ³⁶ standards from the same row (``emission bin''
or simply ``bin'') in Table IV.B.-1. below. Each bin contains a set of
individual NMOG, CO, HCHO, NO_X, and PM standards. The
vehicles would have to comply with each of those standards and would
also be subject to the corresponding bin of intermediate useful life
standards, if applicable, found in Table IV.B-2. For technology
harmonization purposes, our proposed

[[Page 26037]]

emission bins include all of those adopted in California's LEV II
program.³⁷
---------------------------------------------------------------------------

\35\ A ``test group'' is the basic classification unit proposed
for certification of light-duty vehicles and trucks under EPA
certification procedures for the CAP2000 program. This preamble
assumes that manufacturers will be certifying under the provisions
of the CAP2000 program. ``Test group'' is a broader classification
unit than ``engine family'' used prior to the implementation of the
CAP2000 program. We discuss the CAP2000 program in more detail in
section V.A.9. of this preamble.
\36\ The regulatory ``useful life'' value for Tier 2 vehicles is
specifically addressed in Section V.A.2. of this preamble. Full
useful life is proposed to be 10 years or 120,000 miles for all
vehicles except LDT3s and LDT4s, for which it is 11 years or 120,000
miles. Intermediate useful life, where standards are applicable, is
5 years or 50,000 miles.
\37\ EPA's current standards for Clean Fuel Vehicles are less
stringent than the proposed Tier 2 standards. See 40 CFR 88.104-94.
The Tier 2 standards would supercede the current CFV standards, and,
if EPA adopts the standards proposed today, the Agency intends to
undertake a rulemaking to revise the CFV standards accordingly.

Table IV.B.-2.--Tier 2 Light-Duty Full Useful Life (120,000 mile) Exhaust Emission Standards
[Grams per mile]
----------------------------------------------------------------------------------------------------------------
Bin No. NO_X NMOG CO HCHO PM
----------------------------------------------------------------------------------------------------------------
7............................... 0.20 0.125 4.2 0.018 0.02
6............................... 0.15 0.090 4.2 0.018 0.02
5............................... 0.07 0.090 4.2 0.018 0.01
4............................... 0.07 0.055 2.1 0.011 0.01
3............................... 0.04 0.070 2.1 0.011 0.01
2............................... 0.02 0.010 2.1 0.004 0.01
1............................... 0.00 0.000 0.0 0.000 0.00
----------------------------------------------------------------------------------------------------------------

Table IV.B.-3.--Light-Duty Intermediate Useful Life (50,000 mile) Exhaust Emission Standards
[Grams per mile]
----------------------------------------------------------------------------------------------------------------
Bin No. NO_X NMOG CO HCHO PM
----------------------------------------------------------------------------------------------------------------
7.............................. 0.14 0.100 3.4 0.015 ..............
6.............................. 0.11 0.075 3.4 0.015 ..............
5.............................. 0.05 0.075 3.4 0.015 ..............
4.............................. 0.05 0.040 1.7 0.008
----------------------------------------------------------------------------------------------------------------

Under a ``bins'' approach, a manufacturer may select a set of
emission standards (a bin) to comply with, and a test group must meet
all standards within that bin. Ultimately, the manufacturer must also
ensure that the emissions of a targeted pollutant-- NO_X in
this case--from all of its vehicles taken together meet a ``corporate
average'' emission standard. This corporate average emission standard
ensures that a manufacturer's production yields the required overall
emission reductions. (See Section IV.B.-4.c. below for more discussion
of the corporate average NO_X standard.)
In addition to the Tier 2 standards described above, we are also
proposing interim standards derived from the LDV/LDT1 NLEV standards to
cover all non-Tier 2 LDVs and LLDTs during the Tier 2 phase-in. We are
proposing separate interim standards for HLDTs. (We describe the
interim standards in detail in Section IV.B.4.e. below.)

i. Why Are We Proposing Extra Bins?

Compared to the CalLEV II program, our Tier 2 proposal includes
additional bins. The California program contains no bins that would
allow NO_X levels above the 0.07 g/mi level of LEVs.
Therefore, under the California program, no engine family can be
certified above LEV levels, even with the application of offsetting
credits. We propose to add two bins above the LEV bin (Bins 6 and 7)
and another below the LEV bin (Bin 3) to provide manufacturers with
additional flexibility to reduce costs and to account for greater
technological challenges faced in getting certain vehicles to levels of
0.07 g/mi NO_X or less.
During the Tier 2 phase-in years (through 2006 for LDV/LLDTs and
2008 for HLDTs), we are also proposing that the bins from the
applicable interim program would be available. Vehicles certified to
these levels could, at the manufacturer's option, be included in
calculating the Tier 2 corporate average NO_X level. This
would enhance the flexibility of the program by providing manufacturers
with three additional bins having NO_X standards above 0.07
g/mi. Since a manufacturer could elect these bins under the interim
program anyway, there would be no impact on air quality. The interim
program and the interim bins for non-Tier 2 vehicles are described in
detail in section IV.B.4.e.
The additional bins would also provide an incentive for
manufacturers to produce vehicles below 0.07 g/mi of NO_X. We
believe this incentive would exist because manufacturers would have
some vehicles (especially larger LDTs) that they might find more cost
effective to certify to levels above the 0.07 g/mi average standard.
However, to do this they would have to offset those vehicles in our
NO_X averaging system with vehicles certified below 0.07 g/
mi, and the 0.04 g/mi bin would provide greater opportunity to do this.
Thus, the extra bins would serve two purposes; they would provide
additional flexibility to manufacturers to address technological
differences and costs, and they would provide those manufacturers with
incentives to produce cleaner vehicles and thus advance emission
control technology.
We are proposing a bins approach and the proposed bins because we
believe they would provide adequate and appropriate emission reductions
and manufacturer flexibility. In addition, this structure will help to
accelerate technological innovation. We request comment on the
appropriateness of the proposed bin structure and whether the levels
proposed are appropriate. Also, we request comment on whether we should
include up to two additional bins between bin 5 (NO_X = 0.07)
and bin 6 (NO_X = 0.15). Our proposed bin structure is
intended to assure that nearly all vehicles comply with a
NO_X standard of 0.07 g/mi. These additional bins would
provide greater flexibility for manufacturers who may find it more
cost-effective to produce some vehicles slightly above 0.07 but would
have difficulties meeting a 0.07 g/mi average NO_X standard
if they had to certify them to a NO_X level of 0.15 g/mi. We
request specific comment on whether we should

[[Page 26038]]

establish these bins and if so what standards for each pollutant we
should include. As we indicated above, we believe that the existence of
bins above 0.07 g/mi NO_X provide an incentive for
technological advancement. We request comment as to whether these
additional bins would limit this incentive in any way.
On the other hand, Bin 7 is intended primarily to aid manufacturers
during the transition to Tier 2 standards. We request comment on
whether this bin should be eliminated when the Tier 2 phase-in is
completed (after 2007 for LDV/LLDTs and after 2009 for HLDTs).
b. The Proposed Program Would Phase in the Tier 2 Vehicle Standards
over Several Years

i. Primary Phase-In Schedule

We are proposing to phase in the Tier 2 standards for LDVs/LLDTs
over a four year period beginning in 2004 and we are proposing a
delayed two year phase-in beginning in 2008 for HLDTs. These phase-in
schedules are shown in Tables IV.B.-2 and are also shown separately in
Tables IV.B.-4 and 5. We believe the flexibility of this dual phase-in
approach is appropriate because the proposed Tier 2 program would
encompass all light-duty vehicles and trucks and would result in
widespread applications of upgraded and improved technology across the
fleet. The program would require research, development, proveout, and
certification of all light-duty models, and manufacturers would need
longer lead time for some vehicles, especially HLDTs. Also,
manufacturers might wish to time compliance with the Tier 2 standards
to coincide with other changes such as the roll out of new engines or
new models. In order to begin the introduction of very clean vehicles
as soon as possible while avoiding imposing unnecessary inefficiencies
on vehicle manufacturers, we believe a practical but aggressive phase-
in schedule like the one we are proposing effectively balances air
quality, technology, and cost considerations.
In each year, manufacturers would have to ensure that the specified
fraction of their U.S. sales ³⁸ met Tier 2 standards for
evaporative emissions (discussed in Section IV.B.-4.f. below) and
exhaust emissions, including Supplemental Federal Test Procedure (SFTP)
standards (discussed in Section V.A.-3. below), as well as the
corporate average Tier 2 NO_X standard. Manufacturers would
have to meet the Tier 2 exhaust requirements (i.e., all the standards
of a particular bin plus the SFTP standards) using the same vehicles.
Vehicles not covered by the Tier 2 standards during the phase-in years
(2004-2008) would have to meet interim standards described in Section
IV.B.-.4.e. below and the existing evaporative emission as well as the
applicable SFTP standards.
---------------------------------------------------------------------------

\38\ For Tier 2 vehicles (and for interim vehicles), the term
``U.S. sales'' means, for a given model year, those sales in states
other than California and any states that have adopted the
California program.
---------------------------------------------------------------------------

Manufacturers could elect to meet the percentage phase-in
requirements for evaporative and exhaust emissions using two different
sets of vehicles. We believe that because of interactions between
evaporative and exhaust control strategies, manufacturers would
generally address the Tier 2 evaporative phase-in with the same
vehicles that they used to meet the exhaust phase-in. However, the
primary focus of today's proposal is on exhaust emissions, and the
flexibility for manufacturers to use different sets of vehicles in
complying with the phase-in schedule for evaporative standards and for
the exhaust standards would have no environmental down side that we are
aware of. It is possible that some exhaust emission improvements might
even occur sooner than they otherwise would if a manufacturer were able
to move ahead with the roll-out of a model with cleaner exhaust
emissions without having to wait for the development of suitable
evaporative controls to be completed for that model.

Table IV.B.-4.--Primary Phase-in Schedule for Sales of Tier 2 LDVs and
LLDTs
------------------------------------------------------------------------
Required
percentage of
light-duty
Model year vehicles and
light light-
duty trucks
------------------------------------------------------------------------
2004.................................................... 25
2005.................................................... 50
2006.................................................... 75
2007.................................................... 100
------------------------------------------------------------------------

Table IV.B.-5.--Primary Phase-in Schedule for Sales of Tier 2 HLDTs
------------------------------------------------------------------------
Required
percentage of
Model year heavy light-
duty trucks
------------------------------------------------------------------------
2008.................................................... 50
2009.................................................... 100
------------------------------------------------------------------------

According to the proposed phase-in approach, vehicle sales would be
determined according to the ``point of first sale'' method outlined in
the NLEV rule. Vehicles with points of first sale in California or a
state that had adopted the California LEV II program would be excluded
from the calculation. The ``point of first sale'' method recognizes
that most vehicle sales will be to dealers and that the dealers' sales
will generally be to customers in the same geographic area. While some
sales to California residents (or residents of states that adopt
California standards) may occur from other states and vice-versa, we
believe these sales will be far too small to have any significant
impact on the air quality benefits of the Tier 2 program.

ii. Alternative Phase-In Schedule

While our primary proposal is based upon a phase-in of 25%, 50%,
75% and 100% of sales over the 2004, 2005, 2006 and 2007 model years,
respectively (or 50% and 100% in 2008 and 2009 for HLDTs), we are
proposing to permit alternative phase-in schedules as an option to
provide additional flexibility to manufacturers. The alternative phase-
in schedule provisions are structured to provide incentive to
manufacturers to introduce Tier 2 vehicles before 2004 (or 2008 for
HLDTs).
Under this alternative, manufacturers that introduced vehicles
earlier than required could earn the flexibility to make offsetting
adjustments, on a one-for-one basis, to the phase-in percentages in
later years. However, they would still need to reach 100% of sales in
the 2007 model year (2009 for HLDTs). Manufacturers would have the
option to use this alternative to meet phase-in requirements for LDV/
LLDTs and/or HLDTs. They could use separate alternative phase-in
schedules for exhaust and evaporative emissions, or an alternative
phase-in schedule for one set of standards and the primary (25/50/75/
100%) schedule for the other.
An alternative phase-in schedule would be acceptable if it passed a
specific mathematical test. We have designed the test to provide
manufacturers benefit from certifying to the Tier 2 standards early
while ensuring that significant numbers of Tier 2 vehicles would be
introduced during each year of the alternative phase-in schedule. To
test an alternative schedule, a manufacturer would sum its yearly
percentages of Tier 2 vehicles beginning with model year 2001 and
compare the resulting sum to the sum that results from the primary
phase-in schedule. If an alternative schedule scored as high or higher
than the base

[[Page 26039]]

option, then the alternative schedule would be acceptable.
For LDV/LLDTs, the final sum of percentages would have to equal or
exceed 250--the sum that results from a 25/50/75/100 percent phase-in.
For example, a 10/25/50/65/100 percent phase-in that began in 2003
would have a sum of 250 percent and would be acceptable. In this
example, each Tier 2 vehicle sold early (i.e. in 2003) would permit the
manufacturer to sell one less Tier 2 vehicle in the last phase-in year
(2006). A 10/20/40/70/100 percent phase-in that began the same year
would have a sum of 240 percent and would not be acceptable. For HLDTs,
the sum would have to equal or exceed 150 percent.
To ensure that significant numbers of Tier 2 vehicles are
introduced in the 2004 time frame, manufacturers would not be permitted
to use alternative phase-in schedules that delayed the implementation
of the Tier 2 LDV/LLDT requirements, even if the sum of the phase-in
percentages met or exceeded 250. Such a situation could occur if a
manufacturer delayed implementation of its Tier 2 production until 2005
and began a 75/85/100 percent phase-in that year. To protect against
this possibility, we are proposing that in any alternate phase-in
schedule, a manufacturer's phase-in percentages from the 2004 and
earlier model years sum to at least 25%.
The mathematical technique to evaluate alternative phase-in schemes
is somewhat similar to that used in our NLEV rule and in California
rules. We request comment on its appropriateness for this application.
We also request comment on other approaches that might serve to provide
incentive to manufacturers to introduce Tier 2 vehicles early, and to
provide additional flexibility, while at the same time assuring that
environmental gains equivalent to or greater than those of the primary
phase-in option are produced. We have considered whether it would be
appropriate to provide a ``multiplier'' that would serve to increase
the value of the percentage of vehicles introduced before 2004 (2008
for HLDTs) in the mathematical test described above. Such a multiplier
might start at 1 for 2004-2007 vehicles and increase for each year
prior to 2004 (2008 for HLDTs). We request comment as to whether such a
multiplier would be appropriate and whether it would produce real
environmental gains by speeding the introduction of Tier 2 vehicles
into the fleet.
All of the discussion on alternative phase-in schedules to this
point has been premised on 100% compliance in 2007 (2009 for HLDTs). We
request comment as to whether alternative phase-in schedules should be
structured in such a way that, if a manufacturer introduced Tier 2
vehicles in excess of the minimum required during the phase-in years,
that manufacturer could extend its phase-in beyond 2007 or 2009.
Commenters should address the time period beyond 2007 or 2009 that
would be appropriate as well as how EPA would determine the fraction of
vehicles that could be delayed until that time.
Phase-in schedules, in general, add little flexibility for
manufacturers with limited product offerings. A manufacturer with only
one or two test groups can not take full advantage of a 25/50/75/100
percent or similar phase-in. However for manufacturers that meet EPA's
definition of ``small volume manufacturer,'' we are proposing elsewhere
in this preamble that those manufacturers be exempt from the phase-in
schedules and would simply have to comply with the final 100%
compliance requirement. Still, we request comment on how alternative
phase-in schedules might be structured to provide flexibility and
incentive for early introduction to smaller manufacturers.
Later in this preamble (in Section V) we request specific comment
on whether we should include a scheme to provide extra NO_X
credits for manufacturers that introduce Tier 2 vehicles early.
Commenters to the above discussion on alternate phase-in schedules
should address whether a provision for extra NO_X credits
might be a more appropriate way to provide inducements to smaller
manufacturers to introduce Tier 2 vehicles early. Commenters should
consider the interactions such extra credits might have with alternate
phase-in schedules, particularly in situations where a ``multiplier,''
as described above, might be applied.
c. Manufacturers Would Meet a ``Corporate Average'' NO_X
Standard. While the manufacturer would be free to certify a test group
to any bin of standards in Table IV.B.-2, it would have to ensure that
the sales-weighted average of NO_X standards from all of its
test groups of Tier 2 vehicles met a full useful life standard of 0.07
g/mi. Using a calculation similar to that for the NMOG corporate
average standard in the California and NLEV programs, manufacturers
would determine their compliance with the corporate average
NO_X standard at the end of the model year by computing a
sales weighted average of the full useful life NO_X standards
from each bin. Manufacturers would use the following formula:

Corporate Average NO_X = <greek-S>(Tier 2 NO_X
std for each bin) x (sales for each bin) total Tier 2 sales

Manufacturers would exclude vehicles sold in California or states
adopting California LEV II standards from the calculation. As indicated
above, manufacturers would compute separate NO_X averages for
LDV/LLDTs and HLDTs through the year 2008.
The corporate average NO_X standards of the primary Tier
2 program and the interim programs for LDVs/LLDTs and HLDTs would
ensure that expected fleet-wide emission reductions are achieved. At
the same time, the corporate average standards allow us to permit the
sale of some vehicles above the levels of the average standards to
address the greater technological challenges some vehicles face and to
reduce the overall costs of the program. We discuss how manufacturers
could generate, use, and buy or sell NO_X credits under the
proposed program in the next subsection.
Given the corporate average NO_X standards, we do not
believe a corporate average NMOG standard as used by California is
essential because meeting the corporate average NO_X standard
would automatically bring the NMOG fleet average to approximately LEV
levels. However, we request comment on the need for such a corporate
average NMOG standard, as well as suggestions and rationales for what
that standard, if any, should be. Commenters are encouraged to address
any interactions with the bin structure, if appropriate.
d. Manufacturers Could Generate, Bank, and Trade NO_X
Credits.

i. General Provisions

As mentioned in the Overview above, we are proposing that
manufacturers with year-end corporate average NO_X emissions
for their Tier 2 vehicles below 0.07 g/mi could generate Tier 2
NO_X credits. Credits could be saved (banked) for use in a
future model year or for trading (sale) to another manufacturer.
Manufacturers would consume credits if their corporate average
NO_X emissions were above 0.07 g/mi.
We are proposing the Tier 2 standards to apply regardless of the
fuel the vehicle is designed for, and there would be no restrictions on
averaging, banking or trading of credits across vehicles of different
fuel types. Consequently, a gasoline fueled LDV might help a
manufacturer generate NO_X credits in one year that could be
banked for the next year when they could be used to average against
NO_X emissions of a diesel fueled LDT.
Because of the split phase-in and the different interim programs we
are

[[Page 26040]]

proposing for the two different groups of vehicles (LDV/LLDTs and
HLDTs), we are also proposing to require that manufacturers compute
their corporate Tier 2 NO_X averages separately for LDV/LLDTs
and HLDTs through 2008. Credit exchanges between LDVs/LLDTs and HLDTs
would not be allowed nor would credit exchanges across the interim
program and Tier 2 program be allowed. These restrictions would end
with the 2009 model year at which time both phase-ins and all interim
standards will have ended and the program would permit free averaging
across all Tier 2 vehicles. In the context of the whole program we are
proposing, we are concerned that allowing cross-trading between interim
and Tier 2 vehicles would reduce the expected benefits of the program
and delay fleet turnover to Tier 2 emission levels. For this reason we
are not proposing to allow such exchanges. We seek comment on this
issue.

ii. Averaging, Banking, and Trading of NO_X Credits Would
Fulfill Several Goals

There are several reasons why we believe the proposed provisions
for averaging, banking, and trading of NO_X credits (ABT)
would be valuable.

ABT allows us to consider a more stringent emission
standard than might otherwise be appropriate under the CAA, since ABT
reduces the cost and improves the technological feasibility of
achieving the standard.

ABT enhances the technological feasibility and cost
effectiveness of the proposed standard, helping to ensure that the
standard would be attainable earlier than would otherwise be possible.

ABT would provide manufacturers with additional product
planning flexibility and the opportunity for a more cost effective
introduction of product lines meeting the new standard.

ABT would create an incentive for early introduction of
new technology, allowing certain engine families to act as trail
blazers for new technology. This could help provide valuable
information to manufacturers on the technology prior to manufacturers
needing to apply the technology throughout their product line. The
early introduction of new technology would also further improve the
feasibility of achieving the standard and could also provide valuable
information for use in other regulatory programs that may benefit from
similar technologies (e.g., heavy-duty vehicle standards).
EPA views the proposed ABT provisions as environmentally neutral
because the use of credits by some vehicles would be offset by the
generation of an equal number of credits generated by other vehicles.
However, when coupled with the new standards, ABT could have
environmental benefits because it could allow the new standards to be
implemented earlier than would otherwise be appropriate under the Act.

iii. How Manufacturers Would Generate and Use NO_X Credits

As described in the previous subsection, and subject to the phase-
in restrictions described in that subsection, manufacturers would
determine their year-end corporate average NO_X emission
level by computing a sales-weighted average of the NO_X
standard from each bin to which the manufacturer certified any LDVs or
LDTs. The manufacturer would round this average to one more decimal
place than in the corporate average NO_X standard. Tier 2
NO_X credits would be generated when a manufacturer's average
was below the 0.07 gram per mile corporate average NO_X
standard, according to this formula:

NO_X Credits = (0.07 g/mi--Corporate Average
NO_X) x Sales

The manufacturer could then use these NO_X credits in
future years when its corporate NO_X average was above 0.07,
or it could trade (sell) the credits to other manufacturers. The use of
NO_X credits would not be permitted to address Selective
Enforcement Auditing or in-use testing failures.
The enforcement of the NO_X averaging standard would
occur through the vehicle's certificate of conformity. A manufacturer's
certificate of conformity would be conditioned upon compliance with the
averaging provisions. The certificate would be void ab initio if a
manufacturer failed to meet the corporate average NO_X
standard and did not obtain appropriate credits to cover their
shortfalls in that model year or in the subsequent model year (see
proposed deficit carryforward provision below). Manufacturers would
need to track their certification levels and sales unless they produced
only vehicles certified to bins containing NO_X levels of
0.07 g/mi or below and did not plan to bank NO_X credits.

iv. Manufacturers Could Earn and Bank Credits for Early NO_X
Reductions

To provide manufacturers with greater flexibility and with
incentives to certify, produce and sell Tier 2 vehicles as early as
possible, we are proposing that manufacturers could utilize alternative
phase in schedules. (See IV.B.4.b.ii above.) Under such schedules, a
manufacturer could certify vehicles to bins having NO_X
standards of 0.07 g/mi or below in years prior to the first required
phase-in year and then phase its remaining vehicles in over a more
gradual phase-in schedule that would still lead to 100% compliance by
2007 (2009 for HLDTs). To the extent that a manufacturer's corporate
average NO_X level of its ``early Tier 2'' vehicles was below
0.07 g/mi, the manufacturer could bank NO_X credits for later
use. Manufacturers would compute these early credits by calculating a
sales-weighted corporate average NO_X emission level of their
Tier 2 vehicles, as in the basic Tier 2 program described above.
These credits would have all the same properties as credits
generated by vehicles subject to the primary phase-in schedule. These
credits could not be used in the NLEV, Tier 1 or interim program for
non-Tier 2 vehicles in any way. However, the NMOG emissions of these
vehicles (LDVs and LLDTs only) could be used in the calculation of the
manufacturer's corporate average NMOG emissions under NLEV through
2003.
To provide manufacturers with maximum flexibility in the period
prior to 2004, when LDV/LLDT useful lives will still be at 100,000
miles, we are proposing that manufacturers could choose between the
Tier 2 120,000 mile useful life or the current 100,000 mile useful life
requirement for early Tier 2 LDV/LLDTs. (HLDTs already have a 120,000
mile useful life.) Early LDV/LLDT NO_X credits for 100,000
mile useful life vehicles would have to be prorated by 100,000/120,000
(5/6) so that they could be properly applied to 120,000 mile Tier 2
vehicles in 2004 or later.
We are proposing that early banking of HLDT NO_X credits
could not begin until the 2004 model year. This provides a four year
period during which early credits could be generated for use in the
2008/2009 HLDT Tier 2 phase-in. We are concerned that allowing
generation of early HLDT credits in years prior to 2004 could result in
credits that are largely windfall credits. Still, we recognize that
vehicles that meet the Tier 2 standards early represent an
environmental benefit and we request comment on the need for and
appropriateness of allowing early banking of HLDT credits before the
2004 model year.
We recognize that vehicles generating early NO_X credits
may be doing so without the emissions benefit of low sulfur fuel, and
thus these vehicles may not achieve the full in-use emission reduction
for which they received credit. When these credits are used to

[[Page 26041]]

permit the sale of higher-emitting vehicles, there may be a net
increase in emissions. We believe that the benefits of early
introduction of Tier 2 technology described above are significant
enough that they are worth the risk of some emission losses that might
occur if and when the early credits are used. Also, we believe that
some fuel sulfur reductions will occur prior to 2004 as refiners
upgrade their refineries or bring new refining capacity on stream in
anticipation of the 2004 requirements and take advantage of the phase-
in proposed in the gasoline sulfur ABT program (described in Section
IV.C. below). We request comment on all aspects of early introduction
of Tier 2 vehicles and the proposed provisions for early NO_X
credits.

v. NO_X Credits Would Have Unlimited Life

We are not proposing to apply the California schedule of
discounting unused credits that was adopted for NMOG credits in the
NLEV program. This schedule serves to limit credit life throughout the
program by reducing unused credits to 50, 25 and 0 percent of their
original number at the end of the second, third and fourth year,
respectively, following the year in which they were generated. Because
of the declining corporate average NMOG standards in that program,
California has decided, and we agree, that it is prudent to limit the
lives of credits to prevent manufacturers from being able to accumulate
credits and then apply them in such a way as to delay the impact of
declining standards. But in this proposed federal program, once the
proposed phase-in period ends in model year 2009, all light duty
vehicles and trucks would comply on average with a fixed Tier 2
NO_X standard.
Credits would allow manufacturers a way to address unexpected
shifts in their sales mix and yet would prevent the program from being
abused to allow emission increases by design, since emissions would be
capped by the levels in the least stringent bin. The NO_X
emission standards in the Tier 2 and interim programs are quite
stringent and do not present easy opportunities to generate credits.
The degree to which manufacturers invest the resources to achieve extra
NO_X reductions provides true value to the manufacturer and
the environment. We do not want to take measures to reduce the
incentive for manufacturers to bank credits nor do we want to take
measures to encourage unnecessary credit use. Consequently we are
proposing that Tier 2 NO_X credits would have unlimited
lives. We request comment on the need for discounting of credits or
limits on credit life and what those discount rates or limits, if any,
should be.

vi. NO_X Deficits Could Be Carried Forward

When a manufacturer has a NO_X deficit at the end of a
model year--that is, its corporate average NO_X level is
above the required corporate average NO_X standard--we are
proposing that the manufacturer be allowed to carry that deficit
forward into the next model year. Such a carry-forward could only occur
after the manufacturer used any banked credits. If the deficit still
existed and the manufacturer chose not to or was unable to purchase
credits, the deficit could be carried over. At the end of that next
model year, the deficit would need to be covered with an appropriate
number of NO_X credits that the manufacturer generated or
purchased. Any remaining deficit would be subject to an enforcement
action.
To prevent deficits from being carried forward indefinitely, the
manufacturer would not be permitted to run a deficit for two years in a
row.³⁹ We believe that it is reasonable to provide this
flexibility to carry a deficit for one year given the uncertainties
that light duty vehicle and truck manufacturers face with changing
market forces and consumer preferences, especially during the
introduction of new technologies. These uncertainties can make it hard
for manufacturers to accurately predict sales trends of different
vehicle models. We request comment on this provision.
---------------------------------------------------------------------------

\39\ Because of the limited duration of the interim programs, we
are proposing that a manufacturer could carry a credit deficit in
the interim program forward until the 2006 model year (2008 for
HLDTs). The interim program, in its entirety, lasts only five years
and therefore we see little risk of prolonged deficits.
---------------------------------------------------------------------------

e. Interim Standards.

i. Interim Standards for LDV/LLDTs

The NLEV program referenced throughout this discussion is a
voluntary program in which all major manufacturers have opted to
produce LDVs and LLDTs to tighter standards than those required by
EPA's Tier 1 regulations. Under the NLEV program, manufacturers must
meet an NMOG average outside of California that is equivalent to
California's current intermediate-life LEV requirement--0.075 g/mi for
LDVs and LDT1s (0.10 g/mi for LDT2s). Currently, NLEV requirements
apply only to LDVs and LLDTs, not to HLDTs.
The NLEV program is effective beginning in the northeastern states
in 1999 and in the remaining states in 2001, except that the program
does not apply to vehicles sold in California or in states that adopted
California's LEV program. The program runs at least through 2003 and
can run through model year 2005.
Given the Tier 2 phase-in we are proposing, not all LDV/LLDTs
covered under NLEV will be subject to Tier 2 standards in the 2004 to
2006 period. Unless EPA adopts a program for full Tier 2 compliance in
2004 (i.e., without a phase-in), these vehicles could revert to Tier 1
standards. The NLEV program, moreover, is a voluntary program that
contains several provisions that restrict EPA's flexibility and that
could lead to a manufacturer or a covered Northeastern state leaving
the program in or prior to 2004. To resolve these concerns we are
proposing interim standards for all non-Tier 2 LDV/LLDTs for the 2004-
2006 model years. Our interim standards would replace the NLEV program,
which would then terminate at the end of 2003. The transition from NLEV
to Tier 2 should be smooth because the interim standards are derived
from the NLEV standards for LDVs and LDT1s and would ensure that all
LDVs, LDT1s and LDT2s that are not certified to Tier 2 levels during
the 2004-2006 phase-in period remain at levels at least as stringent as
NLEV levels. The standards would also arguments prebring the emission
standards for LDT2s into line with those for the LDVs and LDT1s. We
propose to align the useful life periods for interim standards with
those of the Tier 2 standards (full useful life of 120,000 miles,
intermediate useful life of 50,000 miles, as discussed in Section V.A.-
2 below)
Tables IV.B.-6 and IV.B.-7 below present interim standards we are
proposing for LDVs and LLDTs not covered by Tier 2 standards during the
phase in period.

[[Page 26042]]

Table IV.B.-6.--Full Useful Life (120,000 mile) Interim Exhaust Emission Standards for LDV/LLDTs
[Grams per mile]
----------------------------------------------------------------------------------------------------------------
Bin No. NO_X NMOG CO HCHO PM
----------------------------------------------------------------------------------------------------------------
5............................... 0.60 0.156 4.2 0.018 0.06
4............................... 0.30 0.090 4.2 0.018 0.06
3............................... 0.30 0.055 2.1 0.011 0.04
2............................... 0.07 0.090 4.2 0.018 0.01
1............................... 0.00 0.00 0.0 0.000 0.0
----------------------------------------------------------------------------------------------------------------

Table IV.B.-7.--Intermediate Useful Life (50,000 mile) Interim Exhaust Emission Standards for LDV/LLDTs
[Grams per mile]
----------------------------------------------------------------------------------------------------------------
Bin No. NO_X NMOG CO HCHO PM
----------------------------------------------------------------------------------------------------------------
5............................... 0.40 0.125 3.4 0.015 ..............
4............................... 0.20 0.075 3.4 0.015 ..............
3............................... 0.20 0.040 1.7 0.008 ..............
2............................... 0.05 0.075 3.4 0.015 ..............
----------------------------------------------------------------------------------------------------------------

We are proposing a corporate average full useful life
NO_X standard of 0.30 g/mi for this interim program. LDV/
LLDTs, which will already be at NLEV levels, should readily be able to
meet this average NO_X standard. Although we have not shown
it in the tables of interim standards above, we are also proposing that
all of the bins shown for the Tier 2 program (see Tables IV.B.-2 and -
3) could be used in the interim program. Thus if a manufacturer had
vehicles certified to Tier 2 bins that it did not need to comply with
the Tier 2 NO_X average standard and phase in percentage, it
would have the additional option to use them in the interim program. We
request comment as to whether the number of bins provided in the
interim program and their emission levels are appropriate.
The 0.30 g/mi corporate average NO_X standard (and the
bins of standards in the above two tables) would apply only to non-Tier
2 LDV/LLDTs and only for the 2004-2006 model years. Manufacturers would
compute, bank, average, trade, account for, and report NO_X
credits via the same processes and equations described in this preamble
for Tier 2 vehicles, substituting the 0.30 g/mi corporate average
standard for the 0.07 g/mi corporate average standard in the basic
program. Also, EPA would condition the certificates of conformity on
compliance with the corporate average standard, as described for Tier 2
vehicles. These NO_X credits would be good only for the 2004-
2006 model years and would only apply to the interim non-Tier 2 LDV/
LLDTs. Credits would not be subject to any discounts, and credit
deficits from the 2004 and 2005 model year could be carried forward,
provided they were covered with appropriate credits by the end of the
2006 model year. NMOG credits from the NLEV program could not be used
in this interim program in any way. Credits generated under this
interim program would not be applicable to the Tier 2 NO_X
average standard of 0.07 g/mi because of our concern that a windfall
credit situation could occur. This could happen because credits are
relatively easy to generate under a 0.30 g/mi standard compared to
generating credits under a 0.07 g/mi standard. The application of
credits earned under the interim standard to the Tier 2 standards could
significantly delay the fleet turnover to Tier 2 vehicles. The
requirements of the interim program would be monitored and enforced in
the same fashion as for Tier 2 vehicles.
For the reasons cited above, we believe it is appropriate to extend
interim, NLEV-like standards beyond 2003 as a mandatory program and to
bring all LDVs and LLDTs within its scope. Manufacturers have already
demonstrated their ability to make LDVs and LLDTs that comply at levels
well below these standards, and, as the interim standards for LDV/LLDTs
are essentially ``phase-out'' standards, we are not proposing any
alternative phase-in schedules or early banking provisions for
NO_X credits from the interim LDV/LLDTs.
We request comment on all aspects of the interim standards for LDVs
and LLDTs.
ii. Interim Standards for HLDTs.
We are also proposing interim standards to begin in 2004 for HLDTs.
These vehicles are not included in the NLEV program and will be subject
only to the Tier 1 standards prior to model year 2004. Tier 1 standards
permit NO_X emissions of 0.98 g/mi for LDT3s and 1.53 g/mi
for LDT4s.
The interim standards for HLDTs would apply beginning in the 2004
model year and would phase-in through the 2007 model year, as shown in
Figure IV.B.-1. The proposed interim program is based on a corporate
average full-life NO_X standard of 0.20 g/mi. Manufacturers
would comply with the corporate average HLDT NO_X standard by
certifying their interim HLDTs to any of the full useful life bins
shown in Table IV.B.-8. Where applicable, manufacturers would also
comply with the intermediate useful life standards shown in Table
IV.B.-9. Interim HLDTs not needed to meet the phase-in percentages
during model years 2004-2006 would have to be certified to the
standards of one of the bins in Table IV.B.-8 (and -9), but would not
be included in the calculation to demonstrate compliance with the 0.20
g/mi average. Thus, the emissions of all interim HLDTs would be capped
at a NO_X value of 0.60 g/mi.
As with LDV/LLDTs, manufacturers would also have the flexibility to
use any of the Tier 2 bins shown in Tables IV.B.-2 and IV.B.-3 as
additional bins for interim HLDTs. At the end of each model year,
manufacturers would determine their compliance with the 0.20
NO_X standard by calculating a sales weighted average of all
the bins to which they certified any interim HLDTs, excluding those not
needed to meet the phase-in requirements during 2004-2006.
We believe these interim standards are necessary and reasonable for
HLDTs. While these trucks make up a fairly small portion of the light-
duty fleet (about 11%), their current standards under Tier 1 are far
less stringent than the NLEV standards that apply to current model year
LDVs and LLDTs.

[[Page 26043]]

Given the delayed phase-in we are proposing for HLDTs, we believe it is
appropriate to bring about some interim reductions from these vehicles.
Further, manufacturers have already demonstrated their ability to meet
these interim standards with HLDTs. These standards are a reasonable
first step toward the Tier 2 program and would provide meaningful
reductions in the near term relative to current certification levels
under the Tier 1 emission standards.

Table IV.B.-8.--Full Useful Life (120,000 mile) Interim Exhaust Emission Standards for HLDTs
[Grams per mile]
--------------------------------------------------------------------------------------------------------------------------------------------------------
Bin No. NO_X NMOG CO HCHO PM
--------------------------------------------------------------------------------------------------------------------------------------------------------
5................................................... 0.60 0.230 4.2 0.018 0.06
4................................................... 0.30 0.180 4.2 0.018 0.06
3................................................... 0.20 0.156 4.2 0.018 0.02
2................................................... 0.07 0.090 4.2 0.018 0.01
1................................................... 0.0 0.0 00.0 0.000 0.0
--------------------------------------------------------------------------------------------------------------------------------------------------------

Table IV.B.-9.--Intermediate Useful Life (50,000 mile) Interim Exhaust Emission Standards for HLDTs
[Grams per mile]
----------------------------------------------------------------------------------------------------------------
Bin No. NO_X NMOG CO HCHO PM
----------------------------------------------------------------------------------------------------------------
5............................... 0.40 0.160 3.4 0.015 ..............
4............................... 0.20 0.140 3.4 0.015 ..............
3............................... 0.14 0.125 3.4 0.015 ..............
2............................... 0.05 0.075 3.4 0.015 ..............
----------------------------------------------------------------------------------------------------------------

Given that the interim HLDT standards are ``phase-in'' standards
through 2007 (as opposed to the interim LDV/LLDT standards, which are
``phase-out'' standards), we are proposing that manufacturers could
employ alternative phase-in schedules as proposed for the Tier 2
standards and described in detail in section IV.B.4.b.ii. of this
preamble. These schedules provide manufacturers with greater
flexibility and we believe they also provide incentive for
manufacturers to introduce advanced emission control technology at an
earlier date. Alternative phase-in schedules would have to provide 100%
phase-in by the same year as the primary phase-in schedule (2007).
Because we are concerned about the possibility of windfall credits from
some vehicles that might easily meet the 0.20 corporate average
NO_X standard, we are not proposing to permit the generation
of credits from interim HLDTs prior to the 2004 model year, although we
request comment on this issue.
f. More Stringent Proposed Light-Duty Evaporative Emission
Standards. We are proposing to adopt a set of more stringent
evaporative emission standards for all Tier 2 light-duty vehicles and
light-duty trucks. The standards we are proposing in Table IV.B.-10
represent, for most vehicles, more than a 50% reduction in diurnal plus
hot soak standards from those that will be in effect in the years
immediately preceding Tier 2 implementation. The higher standards for
HLDTs provide allowance for greater non-fuel emissions related to
larger vehicle size.

Table IV.B.-10.--Proposed Evaporative Emission Standards [Grams per
test]
------------------------------------------------------------------------
Supplemental
3 day 2 day
Vehicle class diurnal + diurnal +
hot soak hot soak
------------------------------------------------------------------------
LDVs and LLDTs............................... 0.95 1.2
HLDTs........................................ 1.2 1.5
------------------------------------------------------------------------

Evaporative emissions from light-duty vehicles and trucks represent
nearly half of the light duty VOC inventory projected for the 2007-2010
time frame, according to MOBILE5 projections. We are proposing today to
reduce the light-duty evaporative emission standards applicable to
diurnal and hot soak emissions by more than 50 percent for most
vehicles. Manufacturers are currently certifying to levels that are, on
average, about half of the current standards, and in many cases, much
less than half the standards. Thus, meeting these proposed standards
appears readily feasible. Even though manufacturers are already
certifying at levels much below the current standard, we believe that
reducing the standards will result in emission reductions as all
manufacturers seek to certify with adequate margins to allow for in-use
deterioration. Further, we believe that tighter standards will prevent
``backsliding'' toward the current standards as manufacturers pursue
cost reductions.
As mentioned in section IV.B.-4.b above, we are proposing to phase
in the Tier 2 evaporative standards by the same mechanism as the Tier 2
exhaust standards; e.g., 25/50/75/100 percent beginning in 2004 for
LDV/LLDTs and 50/100 percent beginning in 2008 for HLDTs. (as shown in
Figure IV.B.-1) As for the proposed exhaust standards, alternative
phase-in plans would also be available.
The evaporative emissions standards we are proposing are the same
as those that manufacturers' associations proposed during the
development of California's LEV II proposal; California ultimately did
not adopt these standards. We request comment on all aspects of these
proposed evaporative standards and their likely impact on in-use
evaporative emission levels. We also request comment on adopting the
evaporative emissions standards and phase-in schedule that California
adopted (representing about a 75 percent reduction from the standards
that will otherwise be in place).

C. Our Proposed Program for Controlling Gasoline Sulfur

When we discussed gasoline sulfur control with the American
Petroleum Institute, the National Petrochemical and Refiners
Association, and other representatives of the oil industry, they laid
out several major points for us to consider in development of our
proposal:

[[Page 26044]]

A regional approach to gasoline sulfur control would be
more appropriate than a nationwide program. Gasoline sulfur control
should be targeted primarily at the areas of greatest environmental
need.

Within the regions, gasoline sulfur standards should be
uniform. State fuel initiatives different from any federal regional
standards could result in supply disruption and price volatility and
should be avoided.

Adequate lead time would be critical to a successful
implementation. Implementing gasoline sulfur control over the next few
years involves a number of demands and uncertainties. For example, the
technology that is the lowest cost and more cost effective requires
sufficient time to develop.

Permitting and construction of all of these refineries in
just four years would be a major challenge. Therefore, streamlining of
the permitting process could help address lead time concerns.

If sulfur levels in diesel fuel were also going to be
reduced (or any other changes to gasoline or diesel fuel required)
industry would need to know soon so investment discussions could be
coordinated.
We have seriously considered the oil industry's input in developing
our proposal. While we are not proposing a regionally-based program, as
discussed below, we believe the nationwide program we are proposing
would provide flexibility in response to many of these concerns about
uncertainty and would provide uniformity on a national basis.
The next section of the preamble describes in more detail the
industry proposal and our response to their approach, including the
concepts of national versus regional scope and the level of the
standard. We recognize that refineries face many uncertainties and
constraints, including potential future regulation of diesel sulfur
that would affect the timing of their ability to meet the proposed
gasoline sulfur levels. Consequently, also in this section we propose
and request comment on two provisions, a sulfur averaging, banking and
trading program and permit streamlining, designed to provide
flexibility, to increase lead time, and to ease concern about how other
uncertainties would affect decision making concerning gasoline sulfur
control.
1. Oil Industry Proposal
During the development of this proposal, a large part of the oil
refining industry, represented by the American Petroleum Institute
(API) and the National Petrochemical and Refiners Association (NPRA),
offered a series of constructive recommendations for the design of a
gasoline sulfur control program. These proposals, which have
progressively addressed more and more of the concerns we had raised
about such a program, have a key element in common--the suggestion that
different levels of gasoline sulfur control be applied to different
regions of the country. These industry representatives observe that
some areas of the country need the emission reductions to be achieved
from Tier 2 LDVs and LDTs more than others, and that the gasoline
distribution system can supply different gasolines to different
geographical regions.
The most recent proposal from these members of the oil industry
would provide gasoline meeting an average sulfur level of 150 ppm
(capped at 300 ppm) to a large region of the U.S. This proposal would
cover all states east of the Mississippi river, plus Missouri,
Louisiana, and the eastern half of Texas (and any RFG areas in the
West), and would begin in 2004.⁴⁰, ⁴¹ The
remainder of the country (excluding California) would receive gasoline
meeting a 300 ppm average (450 ppm cap). Further reductions in sulfur
levels in eastern states, to a 30 ppm average/80 ppm cap, would be
required starting in 2010, unless a study performed in 2004-06
demonstrated no air quality need for further sulfur reductions. If this
study found an air quality need for additional reductions, EPA would
make recommendations about the appropriate sulfur levels (if different
from the proposed 30/80 ppm levels) and the area to receive this lower
sulfur gasoline (if different from the region receiving the 150 ppm
average in 2004). The industry representatives thus characterized the
2010 standards as ``rebuttable,'' standards because EPA could have to
initiate additional regulatory actions to implement the final 2010
standards.
---------------------------------------------------------------------------

\40\ The industry representatives offered to meet these
standards earlier if Tier 2 vehicles were introduced before 2004.
\41\ While a majority of oil companies have approved this
proposal, not every U.S. refiner supports all of the provisions
summarized here.
---------------------------------------------------------------------------

The arguments presented by the members of the oil industry for why
this regional program would be reasonable include a consideration of
the technical needs of the vehicles and the ability of refining
industry to meet the requirements. Based on testing and analyses
performed by oil companies and their trade associations, they
concluded:

Automakers can select from a range of design factors to
reduce sulfur sensitivity, including engine design, catalyst size,
catalyst location, control of air/fuel mixtures, the types and amounts
of precious metals used in the catalyst;

Vehicles can be designed to fully reverse the sulfur
effect while meeting both Tier 2 and SFTP emission standards, even if
operated for a long time (1,000 miles) on high sulfur fuel;

This division of the country into two sulfur regions
``matches cost to consumers with benefits,'' since the areas with the
greatest air quality need would get the lower sulfur gasoline, while
consumers and refiners located in areas without substantial air quality
need would not have to pay the higher costs resulting from the lower
levels; and

The regions, as defined, would optimize gasoline
distribution based on the existing distribution system, thus reducing
the potential for supply shortfalls or other difficulties.
Following the same methodology we used to estimate the future
emissions and emissions reductions that would result from our combined
Tier 2/gasoline sulfur proposal (presented above in Section III), we
estimated the emissions that would occur from a program that combined
our proposed Tier 2 vehicle standards with the gasoline sulfur program
proposed by the oil industry.⁴² As explained below, we
believe vehicles meeting the proposed Tier 2 standards that
consistently use the higher sulfur gasoline would emit at higher levels
than those that consistently use 30 ppm sulfur gasoline, and that
vehicles that travel between the East and West (as defined by the oil
industry proposal) would experience an irreversible (permanent) loss in
as much as 50 percent of the emissions performance after being exposed
to high sulfur levels. As a result, our analysis shows somewhat higher
total emissions for the program incorporating the oil industry's
proposal than would occur if this sulfur effect did not occur. Since
the ``rebuttable standard'' leaves open the possibility that the
eastern region will not receive 30 ppm sulfur levels in 2010 and beyond
(upon a finding of no air quality need for further reductions), we
analyzed that scenario as well. Table IV.C.-1 shows the NO_X
emissions we

[[Page 26045]]

calculated for select years for these two scenarios, compared to our
proposal.
---------------------------------------------------------------------------

\42\ As explained in this section, because of sulfur's effect on
emissions, we do not believe we could finalize the proposed Tier 2
vehicle standards with sulfur levels averaging significantly above
30 ppm. However, for the purposes of this analysis we did not change
the modeled Tier 2 vehicle standards.

Table IV.C.-1.--Nationwide NO_X Emissions from Tier 2 Standards and Oil Industry Proposed Gasoline Sulfur Program
----------------------------------------------------------------------------------------------------------------
Total NO_X tons
-----------------------------------------------
Oil industry
Oil industry proposal, 2010
Year proposal 2004: standard
EPA proposal 150/300 ^a rebutted 2004:
2010: 30/300 150/300 ^a
2010: 150/300
----------------------------------------------------------------------------------------------------------------
2007............................................................ 2,423,000 2,821,000 2,821,000
2010............................................................ 1,859,000 2,021,000 2,292,000
2015............................................................ 1,242,000 1,424,000 1,701,000
2020............................................................ 1,023,000 1,221,000 1,508,000
----------------------------------------------------------------------------------------------------------------
^a Sulfur average in East/sulfur average in West.

The industry's proposals have been valuable in helping EPA and all
the major stakeholders focus on key issues of the design of gasoline
sulfur control options. We have seriously considered these proposals as
well as the responses of others to the proposals. We have paid
particularly close attention to the issue of the reversibility of
gasoline sulfur's emissions impacts, since the environmental benefits
to be gained from a regional sulfur program in combination with
national Tier 2 vehicle standards hinge on the degree to which the
negative impact of high sulfur levels can be reversed when a vehicle is
operated later on low sulfur gasoline. We encourage comments on the
appropriateness and feasibility of a regional gasoline sulfur program
such as the one recommended by the oil industry (in combination with
national Tier 2 vehicle standards as proposed today). We are
particularly interested in analyses of the environmental and economic
consequences of such a proposal.
In addition, others have raised the idea of an alternative
temporary regional gasoline sulfur control program. Under this program,
which would last from 2004 through 2008, gasoline refined in PADD IV
(generally covering the Rocky Mountain states and representing about 5
percent of U.S. gasoline production) would meet an average sulfur
standard of 150 ppm with a 300 ppm cap while the remainder of the
country would meet a 30 ppm average beginning in 2004. Gasoline refined
in PADD IV would have to comply with the 30 ppm average/80 ppm cap
beginning in 2009.
This approach would provide the smaller refineries in this region
with additional time to make the significant capital investments to
desulfurize gasoline. In part because of the smaller scale of the PADD
IV refineries, we estimate that the cost of desulfurization would be
larger for these refineries than the estimated average cost of meeting
a 30 ppm standard.
While the Rocky Mountain region's air quality problems are
generally less severe than those in many other parts of the country, we
believe that the emission reductions provided by today's proposed
program would still be important, for several reasons.

The Denver and Salt Lake City areas will have ozone levels
in the 2007 time frame within 15 percent of the national ambient
standards and would benefit from the lowest possible gasoline sulfur
levels to assist their efforts to maintain their ozone attainment
status.

Other benefits of the proposed program would also be
forgone during the interim period, as discussed above, including the
lower secondary PM emissions, improved visibility, and reduced toxic
emissions.

Irreversible damage to vehicle emission control systems in
those vehicles that have been fueled in this region at any time during
their life would occur.

PADD IV gasoline is marketed outside the borders of PADD
IV.

The vehicle emission standards would be more difficult to
enforce if there were an extended period when vehicles were exposed to
gasolines of more than one sulfur level.

We seek comment on the appropriateness of this approach, including
consideration of the cost, air quality, and public health impacts as
compared to our proposal.
As discussed below, however, we are not proposing a gasoline sulfur
control program that incorporates a regional element. We have not been
able to satisfy our concerns with the irreversibility of the sulfur
effect, since it is not clear that vehicle or catalyst design changes
will solve the problem and since we do not believe that the effect is
negligible. Without a national low sulfur gasoline program, the air
quality benefits of our program would be reduced, particularly in the
initial years when the emissions reductions will be most required to
help many states achieve attainment with the National Ambient Air
Quality Standards. A national program providing low sulfur gasoline
everywhere could ensure that the vehicles designed to meet the proposed
Tier 2 standards achieved the desired emissions performance, that the
investments made by car buyers in cleaner technology would be
justified, and that the needed emissions reductions occurred beginning
as early as 2004.
2. Why EPA Believes Gasoline Sulfur Program Must be Nationwide
As explained in Section IV.C.3. below, we are proposing that our
gasoline sulfur control program apply throughout the country, rather
than in a more limited geographic area along the lines of what the oil
industry has proposed. In determining the appropriate geographic scope
for our proposed program, we considered the implications for the
emission control hardware of Tier 2 vehicles, based on the degree to
which the sulfur impact on catalysts may be reversible. We considered
the degree to which sulfur will impact advanced technology engines and
aftertreatment systems. We weighed the impact that sulfur has on
onboard diagnostic systems, and what that may mean for state inspection
and maintenance programs. We evaluated the environmental implications
beyond the ozone benefits to be realized. We also considered the
ability of the entire refining industry to control gasoline

[[Page 26046]]

sulfur at essentially the same time. After review of all of these
issues, it is our judgement that a national program is appropriate and
reasonable. The following sections explore these issues in more detail.
a. Sulfur's Negative Impact on Tier 2 Catalysts Is Irreversible. We
have reviewed data from several test programs designed to characterize
both the effect of high sulfur levels on vehicle emissions and the ease
and completeness with which this effect was eliminated or ``reversed''
once the vehicle was operated on low sulfur gasoline. These test
programs were performed by auto manufacturers, oil companies, emission
control equipment manufacturers and their various associations. All of
the vehicles included in these test programs met either EPA Tier 1 or
California LEV emission standards and were not designed to comply with
either EPA or California supplemental federal test procedure (SFTP)
standards. The SFTP standards are intended to better address and
control emissions under driving conditions not captured when compliance
with our FTP-based exhaust emissions standards is demonstrated, such as
operation with the air conditioning turned on or driving at very high
rates of acceleration and vehicle speeds (hereafter referred to simply
as aggressive driving). This is an important factor in assessing sulfur
reversibility, because in contrast to the vehicles that have been
tested to date, Tier 2 vehicles would have to meet more stringent
exhaust emission standards and would have to meet these standards over
the wider variety of operating conditions included in the SFTP
provisions. Hence, they would have to be designed to meet the emission
standards under all such operating conditions; these design changes may
influence the ease with which the sulfur effect is reversed, as
explained below.
The vehicles tested exhibited a wide range of reversibility, for
reasons that are not fully understood. The LEVs tested in these
programs showed, on average, that the effect of operation on high
sulfur fuel was reversed after operation on low sulfur fuel if
aggressive driving conditions occurred once the vehicle was switched to
low sulfur fuel. Roughly 85% of the increase in NMOG and NO_X
emissions resulting from high sulfur levels was reversible after
operation on low sulfur fuel coupled with more moderate urban driving.
(CO emissions were somewhat less reversible under these conditions.)
Individual vehicles showed a wide range of responses, however. For
example, many vehicles showed substantial irreversibility for one
pollutant (NO_X or NMOG) while very high reversibility for
the other. In some cases, only half of the initial emission increase
due to high sulfur could be removed by driving on low sulfur fuel.
Catalyst temperature, the mixture of air and fuel in the engine and the
design of the catalyst are all believed to be important factors that
affect the reversibility of the sulfur impact. However, to date, no one
has been able to demonstrate the specific contributions of these
various factors. Also, no one has been able to design a catalyst with
both high conversion efficiencies and no or very low sensitivity to
sulfur.
These data indicate that the effect of high sulfur levels on
emissions from current LEV models driven over a wide variety of
operating conditions appears to be partially reversible, particularly
if the vehicle is periodically driven aggressively. However, were these
vehicles required to meet the SFTP standards, we believe that the
degree of reversibility would have been substantially worse.
Studies of the adsorption and removal of sulfur on catalysts have
demonstrated that wide variations in the mixture of air and fuel
entering the engine (alternating between having a shortage to having an
excess of oxygen) directionally help to remove sulfur from the
catalytic surface. When driven aggressively, the mixture of air and
fuel in the engines of most current vehicles (those not certified to
SFTP standards) is quite variable, because precise control of the
mixture of air and fuel is primarily done to control emissions. Meeting
the SFTP standards will ensure that manufacturers carefully control the
mixture of air and fuel over essentially all in-use driving conditions.
This absence of widely varying mixtures of air and fuel could therefore
inhibit the removal of sulfur from the catalyst once operation on high
sulfur fuel ceased. Thus, we project that the sulfur effect on vehicles
meeting both the LEV and SFTP standards (vehicles sold after 2000) and
vehicles meeting the Tier 2 standards (which will include low exhaust
emissions and low SFTP emission standards, too) will be less reversible
than the effect shown on the vehicles included in the test programs
discussed here.
Another factor that may substantially influence sulfur
reversibility is the amount of time the catalyst is exposed to high
sulfur fuel. With only a few exceptions, the vehicles in the test
programs mentioned above were only driven on high sulfur fuel for a few
miles (well under 100) before low sulfur fuel was reintroduced. This
appears to limit the extent to which sulfur could permanently disable
the effectiveness of the catalyst. However, one vehicle was tested with
an aged catalyst system (to simulate a vehicle near the end of the
useful life of 100,000+ miles) and driven for extended mileage (more
than 1,000 miles) on high sulfur fuel before being retested on low
sulfur fuel. (As with the other vehicles, this test vehicle was not
designed to be SFTP-compliant; SFTP compliance could further complicate
the ability of a vehicle to reverse the sulfur effect.) For this
vehicle, only 50% of the NO_X emission effect of high sulfur
fuel was reversed upon operation on low sulfur fuel. This is much less
than the 85-100% reversibility found with short term exposure to
sulfur. Thus, we project that in-use emissions performance of Tier 2
vehicles operated for some time on high sulfur fuel (as would occur if
a regional sulfur control program permitted high sulfur levels in a
large geographic area) might be substantially compromised. For example,
in-use emissions of passenger cars designed to meet the 0.07 g/mi
NO_X standard and operate on 30 ppm gasoline would actually
be increased by about 50 percent if they were operated on 300 ppm
gasoline at any point in their life. Such vehicles might only recover
half of the emissions performance otherwise expected, perhaps even less
once SFTP compliant designs are incorporated. Furthermore, we believe
this effect would be essentially permanent; continued operation with
low sulfur gasoline would be unlikely to improve the emissions
performance.
The Draft RIA presents our complete evaluation of sulfur
irreversibility, based on the data we have obtained to date. We
encourage comments on this analysis. Furthermore, we are seeking
comment on and will be considering the studies described in Appendix B
of the Draft RIA, plus any new information developed or received before
a final decision. We welcome any additional data characterizing the
irreversibility of the sulfur effect, including what vehicle or
catalyst design factors may make exposure to sulfur more or less
reversible.
The preceding discussion focused on the irreversibility of the
sulfur impact on emissions from current gasoline engine technologies.
There are new technologies under development, which could be sold in
the U.S. in the middle of the next decade (the same time that Tier 2
vehicles are being introduced), which also appear to be very sensitive
to sulfur and largely unable to reverse this sulfur impact. One of
these

[[Page 26047]]

technologies is the direct injection gasoline (GDI) engine. These
engines utilize much more air than is needed to burn the fuel, unlike
conventional gasoline engines that operate under conditions where only
just enough air to completely burn the fuel is introduced into the
engine. This GDI technology allows these engines to be up to 25% more
fuel efficient than current gasoline engines and to emit up to 20% less
carbon dioxide. GDI engines are currently being introduced in both
Japan and Europe (which have or will soon require low sulfur
gasolines). Because of the significant operating differences with GDI
engines, these vehicles will likely require emission control technology
substantially different from that used on conventional gasoline
engines. For example, a GDI engine may require a NO_X
adsorber to meet the proposed Tier 2 NO_X standard. High fuel
sulfur levels quickly and permanently degrade the performance of these
NO_X adsorbers. Thus, to enable the sale of advanced, high
efficiency GDI engines in the U.S. under the Tier 2 standards, it
appears that low sulfur gasoline would have to be available nationwide
by the time this technology becomes available.
The fuel cell is another promising propulsion system that is being
developed for possible introduction to consumers early in the next
century. Fuel cells are being designed to operate on a variety of
fuels, including gasoline and diesel fuel. The basic fuel cell
technology is highly sensitive to sulfur. Almost any level of sulfur in
the fuel will disable the fuel cell. One possible solution is to
install a technology that essentially filters out the sulfur before it
enters the fuel cell. However, such sulfur ``guards'' are costly and
could not practically be used like a disposable filter (requiring the
vehicle owner to change the sulfur guard frequently, much like changing
an oil filter) in situations where constant exposure to high sulfur
levels occurs. (Even exposure to relatively low sulfur levels will
likely require periodic replacement of the sulfur guard to ensure
adequate protection for the fuel cell.) Therefore, the amount of sulfur
in the fuel must be limited to that which can be removed by one or at
most two sulfur guards over the life of the vehicle. Thus, in order for
fuel cells operating on gasoline to be feasible in the U.S., low sulfur
fuels would have to be available nationwide by the time this technology
becomes available.
b. Sulfur Has Negative Impacts on OBD Systems and I/M Programs. As
discussed in more detail in the RIA, EPA believes that sulfur in
gasoline can adversely impact the onboard diagnostic (OBD) systems of
current vehicles as well as vehicles meeting the proposed Tier 2
standards. This is an important factor supporting the need for a
national sulfur control program. EPA's onboard diagnostics (OBD)
regulations require that all vehicles be equipped with a system that
monitors, among other things, the performance of the catalyst and warns
the owner if the catalyst is not functioning properly. The OBD catalyst
monitor is designed to identify those catalysts with pollutant
conversion efficiencies that have been reduced to the extent that
tailpipe emissions would exceed a specified multiple of the applicable
hydrocarbon emissions standard. For California LEV and federal NLEV
vehicles, that multiple is 1.75 times the applicable hydrocarbon
emissions standard; for federal Tier 1 vehicles, that multiple is 1.5
times the applicable hydrocarbon standard added to the 4,000 mile
emission level.
We want to ensure that OBD systems operate correctly, and thus the
possibility that gasoline sulfur may interfere with these systems was
another consideration when evaluating the need for a national sulfur
program. Our evaluation of sulfur's effect on OBD systems was
summarized in a staff paper in 1997.⁴³ We concluded that
sulfur can affect the decisions made by the OBD systems. Sulfur appears
to affect the oxygen sensor downstream of the catalyst, which is used
in the OBD systems, and it is not clear that the conditions that seem
to reverse sulfur's effect on the catalyst will also reverse any sulfur
impact on the downstream oxygen sensors. Indirectly, sulfur impacts OBD
systems because it can impair a catalyst that would otherwise be
operating satisfactorily, thereby triggering the OBD warning lights.
While this would be indicate a properly operating OBD system, auto
manufacturers have expressed the concern that consumers using high
sulfur fuel may experience OBD warnings much more frequently than they
would if operating on low sulfur gasoline, and that this could lead to
a loss of consumer confidence in or support for OBD systems. Consumers
may then ignore the OBD warning system and drive a potentially high
emitting vehicle (which may have nothing to do with exposure to
sulfur), contributing even more to air quality problems. Another
possible scenario is that the OBD system may be impaired by sulfur in
such a way that it does not register an improperly functioning
catalyst, even if the catalyst is impaired for reasons unrelated to
exposure to sulfur. This would defeat the purpose of OBD systems.
---------------------------------------------------------------------------

\43\ U.S. EPA, ``OBD & Sulfur Status Report: Sulfur's Effect on
the OBD Catalyst Monitor on Low Emission Vehicles,'' March 1997,
updated September 1997.
---------------------------------------------------------------------------

The NLEV program provides manufacturers the opportunity to request
extra preconditioning of vehicles that they believe may be negatively
impacted by high sulfur levels, when such vehicles may be included in
in-use testing by EPA. We consider such requests on a case-by-case
basis. One manufacturer has already requested, and received approval
for, a special preconditioning cycle to remove any sulfur from the
catalyst of a specific vehicle model, should that vehicle model be
included in any in-use testing. We are concerned that a regional
gasoline sulfur program would increase the likelihood that
manufacturers would be compelled to request special preconditioning
cycles for test programs, and believe that the one request we have
granted already is indicative of the potential problems that would
arise under a regional gasoline sulfur program. While the use of a
special preconditioning cycle can protect the manufacturer from
liability for high in-use emissions resulting purely from exposure to
high sulfur, the in-use emissions from these vehicles would still be
higher than expected based on the certified design.
To the extent that future catalysts are more sensitive to sulfur as
emission standards become more stringent, the impact of sulfur on
catalysts and catalyst monitors becomes proportionately more critical.
The more stringent the Tier 2 vehicle emission standards are, the more
stringent the OBD malfunction thresholds will be, because those
thresholds are expressed as multiples of the applicable hydrocarbon
emission standard. Therefore, even if the sulfur effect on future
technology vehicles were equivalent in absolute terms to the effect on
current technology vehicles, would become more significant in relative
terms on those future technology vehicles. Because of this (and our
concern about how reversible the effect of sulfur may be), we are
concerned that a regional sulfur program could create widespread
problems with OBD catalyst monitors for vehicles traveling outside of
the low sulfur region. A regional sulfur program would likely result in
higher emissions from Tier 2 vehicles in high sulfur regions, and may
also result in more OBD-identified catalyst failures in those areas. We
are not aware of a technical solution to this problem.

[[Page 26048]]

The geographic scope of a sulfur control program also has
implications for inspection and maintenance (I/M) programs. A regional
sulfur control program could affect I/M programs located outside of the
sulfur control region. The emissions measured in these I/M programs
would likely be higher than those measured in the low sulfur region,
possibly necessitating the use of unique emission cut points for Tier 2
vehicles registered in the higher sulfur region. I/M programs located
outside of the sulfur control area would need to consider the
possibility that the presence of OBD failure codes may be caused
primarily by the use of high sulfur fuels, and may have to provide for
a catalyst regeneration procedure to try to reverse the sulfur buildup
to get a reading of how the catalyst is operating. This could lead to
unequal treatment of vehicles located in different regions of the
country based solely on their exposure to sulfur, unnecessarily
complicating I/M programs. Furthermore, many I/M areas intend to rely
heavily on OBD checks rather than emission checks in the future, making
the correlation of OBD checks to the emissions from the vehicles very
important. Therefore, the potential scenario of increased emissions
without OBD detections (due to sulfur-fouled catalyst monitors) would
make OBD a less attractive I/M tool in areas with high sulfur fuel. A
national program, even one providing limited, temporary exemptions for
small refiners, would avoid many of these concerns.
c. Sulfur Reductions Would Ensure Lower Emissions of Many
Pollutants. One of the major arguments supporting a regional program is
that such a program could be targeted at the majority of areas needing
ozone controls by getting the NO_X and VOC reductions in the
areas with the greatest ozone pollution problems. However, as our
estimates of the total emission reductions to be achieved through the
combined Tier 2/gasoline sulfur program show (presented above in
Section III), there are substantial NO_X and VOC reductions
to be attained nationwide with our proposal. In Table IV.C.-1 above, we
estimated that our national sulfur control proposal would result in 9-
22% fewer NO_X emissions compared to the regional sulfur
program proposed by the oil industry, presuming that we implemented
Tier 2 vehicle standards consistent with today's proposal and depending
on the year in which the emissions reduction is evaluated. The higher
emissions from a regional program would be due to the reduced emissions
performance of vehicles (Tier 2 and others) located in the West where
higher sulfur levels would be permitted and the loss of emissions
performance for vehicles located in the East that travel to the West
(or are relocated from the West) and are expected to suffer
irreversible catalyst damage due to the higher sulfur levels in the
West. Even in 2010 and beyond, when the oil industry's proposed program
would result in sulfur levels consistent with our proposal in the East,
Tier 2 vehicles located in the West or traveling from West to East
would see substantial reductions in emissions performance. Furthermore,
if the oil industry's proposed 2010 standard were not implemented (on
the basis of the findings of the study they propose for 2004-06), the
difference in emissions reductions between our proposal and the oil
industry proposal climbs to 16-47% fewer NO_X emissions.
Hence, the ozone benefits of this proposal would be somewhat smaller if
a regional gasoline sulfur program were adopted.⁴⁴
---------------------------------------------------------------------------

\44\ See the Draft RIA for information on the evaluation of this
and other alternatives.
---------------------------------------------------------------------------

While the benefits of reducing ozone precursors through gasoline
sulfur reductions are generally limited to a nonattainment area (as
well as areas trying to maintain their attainment status, including
those within 15% of the NAAQS standard and upwind locations that
contribute transported ozone precursors into those areas), reductions
in emissions of other pollutants have broader geographic benefits, as
discussed in Section III. For example, sulfur reductions would help
reduce emissions of particulate matter, providing some benefit to PM
nonattainment areas (which may or may not coincide with ozone
nonattainment areas) as well as areas with visibility problems. Sulfur
reductions will also have benefits for areas across the country with
acid deposition problems. Furthermore, sulfur reduction, by enabling
tighter Tier 2 standards and by improving the emissions performance of
the vehicles already on the road, will lead to fewer NMOG emissions,
since, as explained in the Draft RIA, NMOG emissions are also impacted
by gasoline sulfur (although to a lesser extent than NO_X
emissions). Some of the NMOG emissions reduced are air toxics. As
described in Section III above, air toxics, also known as hazardous air
pollutants, or HAPs, contribute to a variety of human health problems.
Thus, a national sulfur reduction program would achieve larger benefits
than a regional program, and people living in the region with higher-
sulfur gasoline would not get the full benefits of reduced air toxics
emissions and could suffer adverse health consequences.
d. The Refining Industry Can Control Gasoline Sulfur. While
evaluating the merits of a national gasoline sulfur program, in
addition to considering the technical requirements for vehicles to meet
the proposed Tier 2 standards and the potential air quality benefits
that could be realized, we also considered the ability of refiners to
reduce gasoline sulfur in essentially every gallon of gasoline by 2004.
Based on this evaluation, we believe it is technically feasible for
refiners to meet the proposed standards and that it is possible for
them to do so in the proposed time frame. A summary of our analysis is
presented here; we refer the reader to the Draft RIA for more details.
Technologies that enable refiners to significantly reduce the level
of sulfur in gasoline have been available for many years. California
began requiring low sulfur gasoline (30 ppm average/80 ppm cap) in
1996.⁴⁵ Refiners in California are currently producing
gasoline that averages around 20 ppm sulfur. In addition, low sulfur
gasoline standards similar to our proposal are, or soon will be,
implemented by countries in Asia and Europe, and by Canada. These
programs provide additional evidence that desulfurization technologies
are available to meet a low sulfur gasoline standard, and that the
majority of refiners in the industry can reasonably be expected to
install and operate these technologies if given a reasonable amount of
lead time.
---------------------------------------------------------------------------

\45\ Prior to that date, gasoline in California was capped at
300 ppm sulfur.
---------------------------------------------------------------------------

When considering the implications of a sulfur standard, U.S.
refiners can be grouped into two major groups: those already producing
gasoline that meets, or nearly meets, the proposed requirements, and
those that would have to make processing changes to comply. The
majority of refiners currently producing relatively low sulfur gasoline
today (roughly 15 percent of domestic production) could meet the
proposed gasoline sulfur standard with no or very little additional
capital investment, and at most a small increase in operating cost.
These refiners have achieved their current sulfur levels using
traditional sulfur removal technologies, or, in some cases, with
refinery configurations that can accommodate very low sulfur crude
oils.
Two examples of these traditional technologies are hydrotreating or
hydrocracking the feed to the fluidized catalytic cracker unit (FCC),
the unit in

[[Page 26049]]

the refinery that produces the largest fraction of gasoline blendstock.
These processes are capital intensive and demand large amounts of
hydrogen and other utilities, resulting in high operating expenses.
Another example is desulfurization of the gasoline stream coming from
the FCC unit. Treating the FCC gasoline stream has the advantage of
lower capital and operating costs than treating the FCC feed. The major
concern with this approach is that the octane value of this gasoline
blendstock is reduced at the same time that sulfur is reduced,
particularly when the sulfur is being reduced to low levels. This lost
octane must be made up by increasing the production of high-octane
blendstocks from other units of the refinery, or by the addition of
oxygenates. Making up this octane loss adds significantly to the cost
of desulfurizing FCC gasoline. We seek comment on any implications of
this proposal of recent activities in California relating to the
oxygenate MTBE, and of refiners' possible use of oxygenates other than
MTBE to make up any octane loss.
Based on current sulfur levels, we believe the majority of U.S.
refiners would have to install at least one desulfurization processing
unit to lower gasoline sulfur to the proposed levels. Since
installation of traditional desulfurization technologies could be quite
costly for most refiners, we have been very encouraged to see the
recent development of several improved desulfurization processes that
are now available at reduced capital investment and operating costs
(and which avoid the octane loss that increases the costs of
traditional technologies). Examples of these technologies are CDHydro
and CDHDS (licensed by the company CDTECH) and OCTGAIN 220 (licensed by
Mobil Oil).⁴⁶ These technologies use conventional refining
processes combined in new ways, with improved catalysts and other
design changes that minimize the undesirable impacts (such as the
substantial loss in octane) and maximize the effectiveness of the
desulfurization approach. Since these processes provide less costly
ways to reduce gasoline sulfur, we presume that they would be used by
most refiners to meet the proposed gasoline sulfur standard, and have
based our economic assessment (summarized in Section IV.D. below) on
that presumption.
---------------------------------------------------------------------------

\46\ In addition to these technologies, other companies have
told us that they are working on developing their own
desulfurization technologies. Furthermore, there have been recent
advances in an approach called biodesulfurization, which employs
bacteria that selectively desulfurizes petroleum. We believe
refiners will have an increasing number of technology options to
meet our proposed standards.
---------------------------------------------------------------------------

Some in the refining industry have told us that since there have
not been long-term commercial demonstrations of these newer
technologies, they would not consider these technologies to be viable
and, if faced with our proposed requirements in 2004, they might select
the more traditional sulfur reduction processes, resulting in a higher
cost to produce low sulfur gasoline. While we understand the hesitation
on the part of some in the oil industry to invest in these improved
sulfur reduction technologies, we believe many, if not all, of their
concerns would be addressed in the next few years. The industry would
have four years to prepare to meet our proposed gasoline sulfur
requirements. Refiners have been provided a similar amount of time to
comply with fuel programs in the past (highway diesel fuel sulfur
control, reformulated gasoline under the complex model) and some have
told us that three to four years is adequate to allow them to meet
gasoline sulfur standards similar to those proposed today. Refiners
would have time to grow more comfortable with the improved processes
after they have obtained additional data and information from the
vendors that license these technologies. Refiners would be able to have
their FCC gasolines tested in vendors' pilot plant facilities, which
would provide each refiner with more specific information on how the
process would function in their particular refineries. Furthermore, we
have been informed that there will soon be demonstrations of at least
two of the improved desulfurization technologies in existing
refineries; the entire industry will benefit from these efforts.
We have heard concerns that small refiners, particularly those in
the Rocky Mountain region, would bear proportionately higher economic
burdens if they were required to produce gasoline meeting the same
sulfur levels as larger refineries located in the Gulf Coast and East.
The severity of these economic impacts could result in unreasonably
high gasoline prices, potential refinery closures, and supply
shortages, according to those raising the concerns. Our analysis,
presented here and in the Draft RIA, leads us to conclude that these
severe events would not occur. Furthermore, we have recently received a
study that suggests that, in fact, small refiners in the Rocky Mountain
region will incur costs only slightly higher than the national
average.⁴⁷ This study concludes that the potential for
refinery closures in this region in response to a gasoline sulfur
regulation is small, and that even if ten percent of gasoline were
negatively impacted there would not be a significant supply shortfall
in the region. We have not yet reviewed this study in detail, and we
encourage comments on the analysis presented in it. However, having
considered the concerns raised about small refiners in general,
including those in the Rocky Mountain region, we are proposing special
provisions for small refiners to address their unique challenges.
---------------------------------------------------------------------------

\47\ MathPro, Inc., ``Likely Effects on Gasoline Supply in PADD
4 of a National Standard for Gasoline Sulfur Content,'' Prepared for
Association of International Automobile Manufacturers,
DaimlerChrysler Corporation, Ford Motor Company, and General Motors
Corporation, March 19, 1999.
---------------------------------------------------------------------------

The advent of the improved desulfurization technologies creates an
opportunity for a stringent, nationwide, and yet relatively low-cost,
sulfur control program. Such a program would still likely be
challenging for many if not most refiners. In the program proposed
today, we have built in a number of flexibilities that would ease the
task of compliance for refiners while maintaining the level of air
quality improvements of a less flexible program. In particular, Section
IV.C.-3 below presents a sulfur averaging, banking, and trading program
that effectively extends the final compliance date by two years. In
consideration of all these factors, we believe that under the proposed
program, all refiners nationwide should be able to produce very low
sulfur gasoline without suffering severe financial consequences.
e. Other Stakeholders Support National Gasoline Sulfur Control. In
addition to our technical arguments for concluding that gasoline sulfur
should be controlled nationwide, we have considered the positions of
other parties. Many stakeholders to our decision have expressed to us
their support for a national sulfur control program. Automakers,
represented by the American Automobile Manufacturers Association (AAMA)
and the Association of International Automotive Manufacturers (AIAM),
have petitioned the Agency to implement a national, low sulfur gasoline
program ``as soon as possible.'' State organizations such as STAPPA/
ALAPCO and the Ozone Transport Commission (OTC) have made similar
resolutions, and many individual states have also voiced support for a
national program. Environmental organizations, such as the American
Lung Association

[[Page 26050]]

and the American Council for an Energy Efficient Economy, favor a
national sulfur control program, as well. The arguments for a national
program presented by these parties include:

High sulfur levels significantly impair the performance of
today's emission control technologies, reducing the emissions benefits
of current and advanced vehicles,

Gasoline sulfur contributes to air quality problems not
directly benefitted by vehicle emission standards (PM, SO<INF>X</INF>,
hazardous air pollutants),

The sulfur impact on emission controls is largely
irreversible, and

If sulfur levels are not controlled, new, more fuel-
efficient vehicle technologies that are as or more sulfur-sensitive
than today's vehicles will not be introduced in the U.S.
3. Proposed Gasoline Sulfur Standards
We are proposing to require substantial reductions in gasoline
sulfur levels nationwide. Not only would these standards enable the
stringent tailpipe emission standards we're proposing for Tier 2
vehicles and ensure that these low emission levels would be realized
throughout the life of the vehicle, but they would also help to reduce
emissions of pollutants that endanger public health and welfare from
vehicles already on the road, including NLEV vehicles. The following
sections summarize the proposed requirements for gasoline refiners and
importers, special provisions for small refiners, and possible changes
to construction permitting requirements that would enable refiners to
install gasoline desulfurization technology in a timely manner. We also
raise the potential need for changes to diesel fuel to enable diesel
technologies to meet the proposed Tier 2 standards. Section VI.
provides additional information about the compliance and enforcement
provisions that would accompany these proposed requirements. More
detailed information in support of the conclusions presented in this
section of the proposal is found in the draft Regulatory Impact
Analysis.
a. Standards for Refiners and Importers. Our proposed gasoline
sulfur program balances the goal of enabling Tier 2 emission control
technologies with the goal of lowering sulfur as early as the refining
industry can practically achieve the required levels. To accomplish
both of these goals, we are proposing a set of standards combined with
a sulfur averaging, banking, and trading (ABT) program. This proposed
overall program would achieve the desired sulfur levels, on average,
beginning in 2004--the first year Tier 2 vehicles will be sold--while
proposing to allow the use of credits towards compliance with refinery
average standards indefinitely (within the limits of per-gallon caps).
These requirements would apply to all gasoline sold in the
U.S.,⁴⁸ based on our belief that emissions must be reduced
nationwide to adequately protect public health and the environment and
that Tier 2 vehicles operated everywhere in the U.S. require protection
from the harmful impacts of gasoline sulfur.
---------------------------------------------------------------------------

\48\ Gasoline sold in California that meets California's
standards would be exempt from meeting the proposed standards, due
to our belief tht California gasoline already meets or exceeds these
requirements. See Section VI.B.
---------------------------------------------------------------------------

Table IV.C.-2. presents the proposed standards for gasoline
refiners and importers. The proposal would require all gasoline
refiners and importers to produce gasoline that meets an average
standard of 30 ppm sulfur at the refinery gate on an annual basis,
beginning in 2004. These requirements would apply to all gasoline,
reformulated as well as conventional. In 2004 and beyond this standard
could be met through the use of credits generated as early as 2000 by
refiners who substantially reduce sulfur levels from current (1997-
1998) levels, under the provisions of the proposed sulfur ABT program
discussed below in Section IV.C.3.c. Hence, the actual average sulfur
levels for gasoline in use could be somewhat higher than 30 ppm.
However, to ensure that sulfur levels are being reduced significantly
(for the benefit of Tier 2 vehicles and to achieve the other emissions
benefits of reducing gasoline sulfur), these in-use sulfur levels would
be constrained by maximum corporate pool average standards of 120 ppm
in 2004 and 90 ppm in 2005. These standards would represent the maximum
allowable average sulfur levels for each refiner, measured across all
refineries owned and operated by that refiner, rather than at each
refinery. In 2006 and beyond, there would be no corporate pool average
standard. Every refinery would have to meet the 30 ppm average refinery
gate standard, although refiners could use any banked/purchased credits
to meet this standard (as explained in the ABT discussion below). Thus,
in 2006 and beyond, the majority of gasoline would average 30 ppm,
although some individual refineries could average slightly more or
less.

Table IV.C.-2.--Proposed Gasoline Sulfur Standards for Refiners and Importers [Excluding small refiners]
----------------------------------------------------------------------------------------------------------------
January 1, January 1,
Compliance as of: 2004 2005 January 1, 2006+
----------------------------------------------------------------------------------------------------------------
Refinery Average, ppm...................... ^a30 ^a30 ^a30
Corporate Pool Average, ppm................ 120 90 not applicable
Per-Gallon Cap, ppm........................ ^b300 180 80
----------------------------------------------------------------------------------------------------------------
a This standard can be met through the use of credits as long as the applicable corporate pool average and per-
gallon caps are not exceeded, as explained in the text.
b This initial per-gallon cap standard begins October 1, 2003.

To ensure that, even as average sulfur levels are reduced in 2004-
2006, gasoline sulfur levels do not exceed a maximum level that we
believe is particularly harmful to Tier 2 vehicles, we are also
proposing ``caps'' on the sulfur content of every batch of gasoline
produced or imported into the country. As shown in Table IV.C.-2, these
caps decline over time, ultimately resulting in a per-gallon limit of
80 ppm in 2006 and beyond. Since Tier 2 vehicles would be sold prior to
the start of calendar year 2004, the actual date when the initial
sulfur cap standard would take effect at the refinery is October 1,
2003. We are also proposing caps on the sulfur content of gasoline sold
at the retail level or otherwise distributed downstream of the
refinery, as explained in Section VI.B.
For purposes of compliance, we propose that a joint venture, in
which two or more refiners own and operate one or more refineries, be
treated as separate refining corporations under the proposed gasoline
sulfur requirements.

[[Page 26051]]

Hence, a refinery owned by a joint venture would be included in the
corporate pool calculations of the joint venture, and would not be
allowed to be included with other refineries owned by one of the
parties to the joint venture in the corporate pool calculations for
that party. Given the large number of joint ventures that have been
announced recently in the oil industry, we believe this would be an
equitable way to handle compliance for joint venture refineries.
Furthermore, this approach would increase the number of companies that
can generate and trade sulfur credits; a more limited number of multi-
refinery companies would tend to bank and trade credits within rather
than across corporations. We welcome comments on alternatives to this
approach, such as requiring the majority owner in a joint venture to
include the jointly owned refinery in his compliance calculations. If
you recommend such an approach, please discuss how joint ventures that
have (nearly) equal ownership among the parties should be treated for
compliance and aggregation purposes.
i. Why Begin the Program In 2004?
The primary reason for our proposal to begin the gasoline sulfur
standards in 2004 is that this is the first year that Tier 2 vehicles
would be required to be sold, and these lower sulfur levels would be
needed to avoid significant impairment of the Tier 2 emissions control
technology. Furthermore, vehicles already in the fleet would benefit
and we would like to maximize that benefit by starting the program as
soon as is reasonable. States need the emission reductions that sulfur
control would bring as soon as possible due to their SIP requirements
in 2007 and 2010. This is reinforced by the fact that several states
have already taken the initiative to develop state gasoline sulfur
standards. In fact, since model year 2004 vehicles will likely be on
the market in the fall of 2003, we are proposing to implement the caps
on sulfur levels beginning October 1, 2003. This would help to ensure
that sulfur levels are reduced coincidentally with the sale of Tier 2
vehicles, and would also ensure that sulfur levels throughout the
gasoline distribution system have been reduced by the start of 2004.
We request comment on the feasibility of the compliance dates
summarized in Table IV.C.-2. If these dates are not feasible, what
date(s) would be more appropriate, given that Tier 2 vehicles will be
introduced no later than model year 2004 and our conclusion that
gasoline sulfur reductions must coincide with the introduction of these
vehicles? For example, we request comment on the implications of
implementing the 30 ppm average standard beginning later than 2004,
including potential implication on cost, air quality, and
implementation of the proposed Tier 2 vehicle standards. What other
factors should we consider if you believe that the proposed
implementation dates are not feasible and should be postponed?
We also seek comment on the implications of implementing an average
sulfur standard different than the proposed 30 ppm average standard,
including levels higher and lower than 30 ppm. Specifically, commenters
should address the feasibility of different standards they support, the
time frame in which different average standards could be implemented
(i.e., in 2003, 2004, or 2005), the potential air quality impacts of
such standards, and how such standards would affect the implementation
of the proposed Tier 2 vehicle standards.
ii. How Did We Arrive At the 80 ppm Cap and 30 ppm Average
Standards?
We believe a 30 ppm averaging standard is important and necessary
to enable the emission reductions needed from Tier 2 vehicles. The test
data we have reviewed, referenced in previous sections of this notice
and in the Draft RIA, show that even very low levels of sulfur have
some negative impact on catalyst performance. Most of the data
available to us were generated through testing with minimum sulfur
levels near 30 ppm. We have used this data to conclude that sulfur
levels need to be reduced, and to assess, as part of our analysis, the
technical feasibility of the proposed Tier 2 vehicle standards. The
non-linear relationship between sulfur level and emissions impact (the
lower the sulfur level, the greater the incremental increase in
emissions) suggests that emission reductions would be ensured by sulfur
levels at or near 30 ppm. We believe that requiring the 30 ppm average
standard would be necessary to ensure that vehicles regularly use
gasoline containing very low amounts of sulfur, regardless of where the
vehicles were driven, what time of year it was, or how gasoline
production varied from batch-to-batch in a given refinery.
We also believe that an 80 ppm cap standard would be required to
provide appropriate insurance for maintaining Tier 2 standards in use
and to give automakers an indication of the maximum sulfur levels for
which they would need to design their vehicles. The test data we have
reviewed show that the greatest increase in emissions comes as the
sulfur level is increased from the lowest levels (i.e., 30 ppm). At
higher sulfur levels (i.e., above 100 ppm), the catalyst performance is
impaired to the extent that an additional increase in sulfur content
has a smaller additional impact on emissions. Since the factors that
influence sulfur sensitivity vary from vehicle to vehicle, different
vehicles will experience different impacts from exposure to specific
sulfur levels. None of the data that we have reviewed indicates that a
vehicle can be designed to be completely insensitive to sulfur for all
types of emissions. Furthermore, as discussed in Section IV.C.2., our
concern that roughly half of the sulfur impact on the catalyst would be
irreversible for Tier 2 vehicles (with other vehicles being negatively
affected as well) provides additional arguments for trying to keep the
sulfur cap as close to the average as possible. Hence, to ensure that
Tier 2 vehicles maintain the designed emission performance over the
life of the vehicle, we believe a cap on gasoline sulfur levels would
be necessary, and that 80 ppm would be the appropriate level for this
cap.
Setting a cap also would enhance enforcement of sulfur standards by
setting a maximum level of sulfur that could be checked at all points
in the gasoline distribution process. A sulfur cap significantly lower
than 80 ppm could have the unintended consequence of forcing a sulfur
average lower than the 30 ppm standard, increasing the overall costs of
the program. The proposed level of 80 ppm sulfur for the cap reflects
our balancing of several factors, including the potential air quality
benefits, economic impacts, compliance flexibility, and the
irreversibility of the effects of gasoline sulfur on vehicle emission
controls.
As explained in Section IV.D. below, we believe that the
combination of our proposed gasoline sulfur standards and the proposed
Tier 2 standards would be cost-effective. This judgement about cost-
effectiveness reflects what we believe would be an appropriate balance
between the costs to be borne by the affected industries and the
emissions reductions to be gained. Even though few refiners currently
produce gasoline at or near these levels, as explained in Section
IV.C.2 above there appear to be no significant obstacles to refiners
achieving this level of sulfur control by 2004 (or 2006 if they were to
take advantage of the sulfur ABT program). Unless a substantially
higher average sulfur standard were set or a substantially smaller
fraction of gasoline were affected by our regulations, refiners would
have to make a significant investment in technology to desulfurize
gasoline. Hence the cost to

[[Page 26052]]

refiners would not be substantially reduced if we selected a less
stringent average standard. Furthermore, we believe that a lesser
reduction in gasoline sulfur levels could require us to reduce the
stringency of the proposed Tier 2 standards. A higher average sulfur
level would require less stringent standards or more vehicle hardware
costs; either would reduce the effectiveness of our proposed combined
program.
At the same time, we recognize the need to provide some
flexibilities to refiners in meeting our proposed standards, to ensure
that the program is implemented in an orderly manner, without severe
consequences in the initial months (for example, supply shortages or
substantial spikes). Hence, we have proposed to allow less stringent
caps in 2004 and 2005 (through 2007 under the small refiner provisions
discussed below) to balance the needs of the technology with the
regulatory burden, economic impact, and ability of the refining
industry to reduce sulfur levels in this time frame. Given that Tier 2
vehicles would be phased in over several years and that the vast
majority of gasoline would be capped at 80 ppm by 2006 (when 75% of new
LDV, LDT1, and LDT2 sales would be required to meet the proposed Tier 2
standards), we believe that the potential damage to Tier 2 catalysts
would be minimized. Furthermore, since the gasoline distribution system
is fungible (i.e., gasoline from multiple refiners may be mixed
together, and gasoline produced at one company's refinery may be sold
at another company's retail station), any gasoline that approached the
higher caps in 2004 and 2005 would be highly likely to be diluted by
lower sulfur gasoline, further limiting the potential negative impact
on Tier 2 vehicles.
We have also proposed to permit compliance with the 30 ppm refinery
average with the use of credits indefinitely, not just in the years
during which the corporate average is reduced, as long as the
applicable per-gallon caps are not exceeded. We would like comments on
whether this provision should end, and if so, what date would be
appropriate to require every refinery to meet the 30 ppm standard with
actual production. We also encourage comments on whether corporate
averaging (aggregation of refineries owned by a single entity) should
be allowed for compliance with the 30 ppm standard, in 2004 and 2005
(in addition to corporate averaging to the pool standard) and/or
beginning in 2006.
In light of our technical conclusions about the need for these
standards, and our concerns about the irreversibility of the sulfur
effect, we believe the 30 ppm average/80 ppm cap is the appropriate
sulfur level to enable vehicles to meet the proposed Tier 2 standards
and to maximize the emissions reductions to be achieved from this
program in a cost-effective way. We welcome comments on these
conclusions. We are also interested in any information on the
reversibility of the sulfur impact on NLEV and Tier 2 catalysts that
may supplement our understanding of how reversibility may differ with
exposure to different sulfur levels and how this difference would
impact our selection of the 30/80 standards. We also solicit
information about what, if any engine or catalyst design modifications
could minimize the irreversibility of the sulfur impact and about how
compliance with the SFTP standards could impact irreversibility (for
either NLEV or Tier 2 vehicles).
iii. Should a Near-Zero Gasoline Sulfur Standard Be Considered?
The auto industry, represented by the Alliance of Automobile
Manufacturers, have supported a gasoline sulfur control program that
would require 30 ppm gasoline in 2004 with a further reduction to
``near-zero'' levels (less than 5 ppm) by 2007. They believe that near-
zero sulfur levels would enable the emission control technology that
would ultimately be necessary to meet standards similar to those we are
proposing today. They also believe that very low sulfur gasoline would
significantly increase the emission reductions of the program as
compared to a 30 ppm sulfur program.
We are also aware of concerns that advanced emission control and
fuel efficient technologies, such as gasoline direct injection engines
and automotive fuel cells, may require zero or near-zero sulfur levels
to achieve Tier 2 emission levels over their full useful life (or in
some cases, even to operate for a significant length of time). At the
same time, we're aware that there may be technological solutions to
these problems that may allow these technologies to operate on gasoline
averaging 30 ppm sulfur. For example, it may be possible to regenerate
(remove the sulfur from) the emission control technologies used by
gasoline direct injection engines on an ongoing basis. Similarly, it
may be possible to prevent sulfur from entering a fuel cell through the
use of a sulfur ``guard'' made, for example, of zinc oxide, that might
need to be replaced periodically.
We believe at this time that our proposed Tier 2 standards could be
met with conventional technology if gasoline averaging 30 ppm is
available. Nonetheless, for the reasons put forward by the auto
industry and others, we also believe that it may be desirable in the
long term for all gasoline in the U.S. to average substantially below
30 ppm sulfur. We encourage you to comment on the question of requiring
gasoline sulfur levels under 5 ppm in the 2007 and later time frame. If
you are commenting on this issue, we encourage you to take a broad view
and to discuss all of the following questions in your comments:

What technological options would be opened to
manufacturers of vehicles and emission control hardware if near-zero
sulfur fuel were available?

What additional air quality benefits would be achieved?

What changes in vehicle engines and emission control
technology would be needed to achieve these emission benefits, absent
reductions in gasoline sulfur levels beyond our proposed 30 ppm
standard? What would these changes cost?

What is the maximum sulfur level that advanced
technologies, including gasoline direct injection and automotive fuel
cells, could be designed to withstand if they are to be commercialized
under the proposed Tier 2 standards? In what time frame might
substantial commercialization of these technologies occur?

How feasible is production of near-zero sulfur gasoline
for the refining industry? What technologies would be required? How
would this vary from refinery to refinery? What additional costs,
beyond those expected for a 30 ppm sulfur program, would be incurred?
How would the timing of a near-zero sulfur requirement affect refining
costs?

Would equipment used to make 30 ppm have to be modified or
replaced to make near-zero sulfur gasoline? If so, how would this
affect the time frame in which a near-zero sulfur level in gasoline
could be achieved? Would the time frame for achieving these levels be
different if refiners were not required to meet a 30 ppm standard? Is
there another sulfur concentration that could be easily achieved as an
intermediate level before achieving near-zero levels?

What other issues should we consider in evaluation of
further reductions in gasoline sulfur levels?
iv. Why Are We Proposing Less Stringent Standards for 2004 and
2005?
We are proposing to permit corporate average sulfur levels to be
somewhat higher than 30 ppm, and maximum sulfur levels to be higher
than 80 ppm, under the ABT program in 2004 and 2005. This proposal is
meant to provide greater flexibility for refiners to meet

[[Continued on page 26053]]

Notices For
2009
2008
2007
2006
2005
2004
2003
2002
2001
2000
1999
1998
1997
1996
1995
1994

Control of Air Pollution From New Motor Vehicles: Proposed Tier 2 Motor Vehicle Emissions Standards and Gasoline Sulfur Control Requirements

Related Material

Local Navigation