Regulating Greenhouse Gas Emissions Under the Clean Air Act

Note: EPA no longer updates this information, but it may be useful as a reference or resource.
 
PDF Version (50 pp, 552K, About PDF)

[Federal Register: July 30, 2008 (Volume 73, Number 147)]
[Proposed Rules]
[Page 44453-44502]
From the Federal Register Online via GPO Access [wais.access.gpo.gov]
[DOCID:fr30jy08-38]

Regulating Greenhouse Gas Emissions Under the Clean Air Act

[[Continued from page 44452]]

[[Page 44453]]

    In conclusion, EPA seeks comment on whether a CO₂
emissions backstop is an appropriate complement to a footprint-based
regulatory approach under the CAA to ensure that the program would
achieve a minimum level of feasible carbon dioxide emissions
reductions. EPA invites comments on both the potential backstop
approaches discussed above, as well as suggestions for other
approaches.
iii. Potential Test Procedures for Light-Duty Vehicle Tailpipe
CO₂ Emissions
    For the program options EPA analyzed to date, EPA would expect
manufacturers and EPA to measure CO₂ for certification and
compliance purposes over the same test procedures currently used for
measuring fuel economy, except for A/C-related CO₂
emissions. This corresponds with the data used in our analysis of the
potential footprint-based CO₂ standards presented in section
VI.B.1.b of this advance notice, as the data on control technology
efficiency was also developed in reference to these test procedures.
These procedures are the Federal Test Procedure (FTP or ''city'' test)
and the Highway Fuel Economy Test (HFET or ''highway'' test). EPA
established the FTP for emissions measurement in the early 1970s. In
1976, in response to requirements in the Energy Policy and Conservation
Act (EPCA), EPA extended the use of the FTP to fuel economy measurement
and added the HFET. The provisions in the 1976 regulation, effective
with the 1977 model year, established procedures to calculate fuel
economy values both for labeling and for CAFE purposes. Under EPCA, EPA
is required to use these procedures (or procedures which yield
comparable results) for measuring fuel economy for cars for CAFE
purposes, but not for fuel economy labeling purposes. EPCA does not
impose this requirement on CAFE test procedures for light trucks, but
EPA does use the FTP and HFET for this purpose.
    On December 27, 2006, EPA established new ``5-cycle'' test
procedures for fuel economy labeling--the information provided to the
car-buying public to assist in making fuel economy comparisons from
vehicle to vehicle. These procedures were originally developed for
purposes of criteria emissions testing, not fuel economy labeling,
pursuant to section 206(h) of the Clean Air Act, which requires EPA to
review and revise as necessary test procedures for motor vehicles and
motor vehicle engines ``to insure that vehicles are tested under
circumstances which reflect the actual current driving conditions under
which motor vehicles are used.'' In updating the fuel economy labeling
regulations, EPA determined that these emissions test procedures take
into account several important factors that affect fuel economy in the
real world but are missing from the FTP and HFET tests. Key among these
factors are high speeds, aggressive accelerations and decelerations,
the use of air conditioning, and operation in cold temperatures.
Consistent with section 206 (h), EPA revised its procedures for
calculating the label estimates so that the miles per gallon (mpg)
estimates for passenger cars and light-duty trucks would better reflect
what consumers achieve in the real world. Under the new methods, the
city miles per gallon estimates for the manufacturers of most vehicles
have dropped by about 12% on average relative to the previous
estimates, with estimates for some vehicles dropping by as much as 30%.
The highway mpg estimates for most vehicles dropped on average by about
8%, with some estimates dropping by as much as 25% relative to the
previous estimates. The new test procedures only affect EPA's vehicle
fuel economy labeling program and do not affect fuel economy
measurements for the CAFE standards, which continue to be based on the
original 2-cycle test procedures (FTP/HFET).
    EPA continues to believe that the new 5-cycle test procedures more
accurately predict in-use fuel economy than the 2-cycle test
procedures. Although, as explained below, to date there has been
insufficient information to develop standards based on 5-cycle test
procedures, such information could be developed and there is no legal
constraint in the CAA to developing such standards. Indeed, section
206(h) provides support for such an approach. Now that automotive
manufacturers are using the 5-cycle test procedure for labeling
purposes, we anticipate significant amount of data regarding the impact
of the 5-cycle test on vehicle CO₂ emissions will be made
available to the Agency over the next several years.
    However, for the programs analyzed in the Light-duty Vehicle TSD,
EPA used the original 2-cycle test. Indeed, data were simply lacking
for the efficiencies of most fuel economy control measures as measured
by 5-cycle tests. Thus, existing feasibility studies and analyses, such
as the 2002 National Academy of Sciences (NAS) and the 2004 Northeast
States Center for a Clean Air Future (NESCCAF) studies that examined
technologies to reduce CO₂, were based on the 2-cycle test
procedures. However, as noted above, we expect that new data regarding
the 5-cycle test procedures will be made available and could be
considered in future analysis.
    It is important to note, however, that all of our benefits inputs,
modeling and environmental analyses underlying the potential programs
analyzed in the Light-duty Vehicle TSD accounted for the difference
between emissions levels as measured by the 2-cycle test and the levels
more likely to actually be achieved in real world performance. Thus,
EPA applied a 20% conversion factor (2-cycle emissions result divided
by 0.8) to convert industry-wide 2-cycle CO₂ emissions test
values to real world CO₂ emissions factors. EPA used this
industry-wide conversion factor for all of its emission reduction
estimates, and calculated such important values as overall emission
reductions, overall benefits, and overall cost-effectiveness using
these corrected values. In reality, this conversion factor is not
uniform across all vehicles. For example, the conversion factor is
greater than 20% for vehicles with higher fuel economy/lower
CO₂ values and is less than 20% for vehicles with lower fuel
economy/higher CO₂ values. But to simplify the technology
feasibility analysis, the analysis assumed a uniform conversion factor
of 20% for all vehicles. EPA does not believe the overall difference
would have a significant effect on the standards because the errors on
either side of 20% tend to offset one another.
    EPA thus analyzed CO₂ standards based on the 2-cycle
test procedures for our analysis to date. EPA would expect to continue
to gain additional experience and data on the 5-cycle test procedures
used in the labeling program. If EPA determined that analyzing
potential CO₂ standards based on these test procedures would
result in more robust control of those emissions, we would consider
this in future analyses. EPA requests comments on the above test
procedure issues, and the relative importance of using the 2-cycle
versus the 5-cycle test in any future EPA action to establish standards
for light-duty vehicle tailpipe CO₂ emissions.
2. Heavy-Duty Trucks
    Like light-duty vehicles, EPA's regulatory authority to address
pollution from heavy-duty trucks comes from section 202 of the CAA. The
Agency first exercised this responsibility for heavy-duty trucks in
1974. Since that time, heavy-duty truck and diesel engine technologies
have continued to improve, and the Agency has set increasingly
stringent emissions standards (today's diesel engines are 98% cleaner
than those from 1974). Over that same period, freight shipment

[[Page 44454]]

by heavy-duty trucks has more than doubled. Goods shipped solely by
truck account for 74% of the value of all commodities shipped within
the United States. Trucked freight is projected to double again over
the next two decades, growing from 11.5 billion tons in 2002 to over
22.8 billion tons in 2035.\139\ Total truck GHG emissions are expected
to grow with this increase in freight.
---------------------------------------------------------------------------

    \139\ Government Accountability Office. Freight Transportation:
National Policy and Strategies Can Help Improve Freight Mobility
GAO-08-287. Report to the Ranking Member, Committee on Environment
and Public Works, U.S. Senate. January 2008.
---------------------------------------------------------------------------

    Reflecting important distinctions between light and heavy-duty
vehicles, section 202 gives EPA additional guidelines for heavy-duty
vehicle regulations for certain pollutants, including defined
regulatory lead time criteria and authority to address heavy-duty
engine rebuild practices. The Agency has further used the discretion
provided in the CAA to develop regulatory programs for heavy-duty
vehicles that reflect their primary function. Key differences between
our light-duty and heavy-duty programs include vehicle standards for
cars versus engine standards for heavy-duty trucks, gram per distance
(mile) standards for cars versus gram per work (brake horsepower-hour)
for trucks, and vehicle test procedures for cars versus engine-based
tests for trucks. EPA has thus determined that in the heavy-duty
sector, the appropriate metric to evaluate performance is per unit of
work and that engine design plays a critical role in controlling
criteria pollutant emissions. EPA's rules also reflect the nature of
the heavy-duty industry with separate engine and truck manufacturers.
As EPA considers the best way to address GHG emissions from the heavy-
duty sector, we will again be considering the important ways that
heavy-duty vehicles differ from light-duty vehicles.
    In this section, we will characterize the heavy-duty GHG emissions
inventory, broadly discuss the technologies available in the near- and
long-term to reduce heavy-duty truck GHG emissions, and discuss
potential regulatory options to address these emissions. We invite
comment on the issues that are relevant to considering potential GHG
emission standards for heavy-duty trucks. In particular, we invite
commenters to compare and contrast potential heavy-duty solutions to
our earlier discussion of light-duty vehicles and our existing heavy-
duty criteria pollutant control program in light of the differences
between GHG emissions and traditional criteria air pollutants.
a. Heavy-Duty Truck GHG Emissions
    Heavy-duty on-road vehicles emitted 401 million metric tons of
CO₂ emissions in 2006, or approximately 19% of the mobile
source CO₂ emissions, the largest mobile source sub-category
after light-duty vehicles.\140\ CO₂ emissions from these
vehicles are expected to increase significantly in the future, by
approximately 29% between 2006 and 2030.\141\
---------------------------------------------------------------------------

    \140\ Emissions data in this section are from the United States
Environmental Protection Agency. Inventory of U.S. Greenhouse Gas
Emissions and Sinks: 1990-2006. EPA 430-R-08-005. April 2008.
    \141\ Growth data in this section is from United States
Department of Energy, Energy Information Administration. Annual
Energy Outlook 2008. #DOE/EIA-0383. April 2008.
---------------------------------------------------------------------------

    Diesel powered trucks comprise 91% of the heavy-duty CO₂
emissions, with the remaining 9% coming from gasoline and natural gas
engines. Heavy-duty GHG emissions come primarily from two types of
applications, combination and single unit trucks. Combination trucks
constitute 75% of the total heavy-duty GHG emissions--44% from long-
haul and 31% from short-haul operations. Short-haul single unit trucks
are the third largest source at 19%. The remaining 5% consists of long-
haul single unit trucks; intercity, school, and transit buses; refuse
trucks, and motor home emissions.\142\
---------------------------------------------------------------------------

    \142\ Breakdown of emissions data in this section is from United
States Environmental Protection Agency. MOVES model. April 8, 2008.
---------------------------------------------------------------------------

    GHG emissions from heavy-duty trucks are dominated by
CO₂ emissions, which comprise approximately 99% of the
total, while hydrofluorocarbon and N₂O emissions represent
0.5% and 0.3%, respectively, of the total emissions on a CO₂
equivalent basis.
b. Potential for GHG Emissions Reductions From Heavy-Duty Trucks
    Based on the work from EPA's SmartWay Transport Partnership and the
21st Century Truck Partnership, we see a potential for up to a 40%
reduction in GHG emissions from a typical heavy-duty truck in the 2015
timeframe, with greater reductions possible looking beyond 2015,
through improvements in truck and engine technologies.\143\ While
highly effective criteria pollutant control has been realized based on
engine system regulation alone, the following sections make clear that
GHG emissions improvements to truck technology provide a greater
potential for overall GHG emission reductions from this sector.
---------------------------------------------------------------------------

    \143\ 21st Century Truck Partnership. Technology Roadmap for the
21st Century Truck Program. 21CT-001. December 2000.
www.doe.gov/bridge.
---------------------------------------------------------------------------

    In this section, we will provide a brief summary of the potential
for GHG emission reductions in terms of engine technology, truck
technology and changes to fleet operations. The public docket for this
Advance Notice includes a technical memorandum from EPA staff
summarizing this potential in greater detail.\144\ In discussing the
potential for CO₂ emission reductions, it can be helpful to
think of work flow through a truck's system. The initial work input is
fuel. Each gallon of diesel fuel has the potential to produce some
amount of work and will produce a set amount of CO₂ (about
22 lbs. of CO₂ per gallon of diesel fuel). The engine
converts the chemical energy in the fuel to useable work to move the
truck. Any reductions in work demanded of the engine by the vehicle or
improvements in engine fuel conversion efficiency will lead directly to
CO₂ emission reductions. Current diesel engines are about
35% efficient over a range of operating conditions with peak efficiency
levels of a little over 40%. This means that approximately one-third of
the fuel's chemical energy is converted to useful work and two-thirds
is lost to waste heat in the coolant and exhaust. In turn, the truck
uses this work output from the engine to overcome vehicle aerodynamic
drag (53%), tire rolling resistance (32%), and friction in the vehicle
driveline (6%) and to provide auxiliary power for components such as
air conditioning and lights (9%).\145\ While it may be intuitive to
look first to the engine for CO₂ reductions given that only
about one-third of the fuel is converted to useable work, it is
important to realize that any improvement in vehicle efficiency reduces
both the work demanded and also the energy wasted in proportional amounts.
---------------------------------------------------------------------------

    \144\ Summary of GHG Emission Control Technologies for Heavy-
Duty Trucks, Memorandum to Docket XXX, May 2008.
    \145\ Approximate truck losses at 65 mph from 21st Century Truck
Partnership. 21st Century Truck Partnership Roadmap/Technical White
Papers: Engine Systems. 21CT-003. December 2006. www.doe.gov/bridge.
---------------------------------------------------------------------------

    In evaluating the potential to reduce GHG emissions from trucks and
operations as a whole, it will be important to develop an appropriate
metric to quantify GHG emission reductions. As discussed above, our
current heavy-duty regulatory programs measure emissions expressed on a
mass per work basis (g/bhp-hr). This approach has proven highly
effective at controlling criteria pollutant emissions while normalizing
the diverse range of

[[Page 44455]]

heavy-duty vehicle applications to a single engine-based test metric.
While such an approach could be applied to evaluate CO₂
emission reductions from heavy-duty engines, it would not readily
provide a mechanism to measure and compare reductions due to vehicle
improvements. Hence, we will need to consider other performance metrics
such as GHG emissions per ton-mile. We request comment on what types of
metrics EPA should consider to measure and express GHG emission rates
from heavy-duty trucks.
    We discuss below the wide range of engine, vehicle, and operational
technologies available to reduce GHG emissions from heavy-duty trucks.
Our discussion broadly assesses the availability of these technologies
and their GHG emissions reduction potential. We request comment on all
aspects of our current assessment summarized here and in more detail in
our technical memorandum, including supporting data with regard to
technology costs, GHG reduction effectiveness, the appropriate GHG
metric to evaluate the technology and the timeframe in which these
technologies could be brought into the truck market. More generally, we
request comment on the overall GHG emissions reductions that can be
achieved by heavy-duty trucks in the 2015 and 2030 timeframes.
i. Engine
    The majority of heavy-duty vehicles today utilize turbocharged
diesel engines. Diesel engines are more efficient compared to gasoline
engines due to the use of higher compression ratios, the ability to run
with lean air-fuel mixtures, and the ability to run without a throttle
for load control. Modern diesel engines have a peak thermal efficiency
of approximately 42%, compared to gasoline engines that have a peak
thermal efficiency of 30%. Turbochargers increase the engine's power-
to-weight ratio and recover some of the exhaust heat energy to improve
the net efficiency of the engine.
    Additional engine improvements could increase efficiency through
combustion improvements and reductions of parasitic and pumping losses.
Increased cylinder pressure, waste heat recovery, and low viscosity
lubricants could reduce CO₂ emissions, but are not widely
utilized in the heavy-duty industry. Individual improvements have a
small impact on engine efficiency, but a combination of approaches
could increase efficiency by 20% to achieve a peak engine efficiency of
approximately 50%.\146\
---------------------------------------------------------------------------

    \146\ 21st Century Truck Partnership. 21st Century Truck
Partnership Roadmap/Technical White Papers: Engine Systems. 21CT-
003. December 2006. www.doe.gov/bridge.
---------------------------------------------------------------------------

    Waste heat recovery technologies, such as Rankine bottoming cycle,
turbocompounding and thermoelectric materials, can recover and convert
engine waste heat to useful energy, leading to improvements in the
overall engine thermal efficiency and consequent reduction in
CO₂ emissions. We request comment on the potential of these
technologies to lower both GHG emissions and overall heavy-duty vehicle
operating costs.
    In section VI.D below, we discuss the Renewable Fuel Standard (RFS)
program and more broadly the overall role of fuel changes to reduce GHG
emissions. As we have previously noted, the Agency has addressed
vehicle emissions through a systems-based approach that integrates
consideration of fuel quality and vehicle or engine emission control
systems. For example, removing lead from gasoline and sulfur from
diesel fuel has enabled the introduction of very clean gasoline and
diesel engine emission control technologies. A systems approach may be
a means to address GHG emissions as well. Since 1989, European engine
maker Scania has offered an ethanol powered heavy-duty diesel cycle
engine with traditional diesel engine fuel efficiency (the current
version offers peak thermal efficiency of 43%).\147\ Depending on the
ethanol production pathway, such an approach could offer a significant
reduction in GHG emissions from a life cycle perspective when compared
to more traditional diesel fuels. We request comment on the potential
for a systems approach considering alternate fuel and engine
technologies to reduce GHG emission from heavy-duty trucks. We also
request comment on how EPA might structure a program to appropriately
reflect the potential for such GHG emission reductions.
ii. Vehicle systems
    An energy audit of heavy-duty trucks shows that vehicle efficiency
is strongly influenced by systems outside of the engine. As noted
above, aerodynamics, tire rolling resistance, drivetrain, and weight
are areas where technology improvements can significantly reduce GHG
emissions through reduced energy losses. The fuel savings benefits of
many of these technologies often offset the additional costs.
Opportunities for HFC and additional CO₂ reductions are
available through improved air conditioning systems.
---------------------------------------------------------------------------

    \147\ Green Car Congress. Scania Extending Heavy-Duty Ethanol
Engine Technology to Trucks. April 15, 2008. 
http://www.greencarcongress.com/2008/04/scania-extendin.html 
(April 30, 2008).
---------------------------------------------------------------------------

    For a typical combination tractor-trailer truck traveling at 65
mph, energy losses due to aerodynamic drag can total over 21% of the
total energy consumed.\148\ A recent study between industry and the
federal government demonstrated that reducing the tractor-trailer gap
and adding trailer side skirts, trailer boat tails, and aerodynamic
mirrors can reduce aerodynamic drag by as much as 23%. If aerodynamic
drag were reduced from 21% to 15% (a 23% reduction), GHG emissions at
65 mph would be reduced by almost 12%.\149\ The cost of aerodynamic
equipment installed on a new or existing trailer is generally paid back
within two years.\150\ As aerodynamic designs become more
sophisticated, more consistency in how aerodynamics is measured is
needed. There is no single, consistent approach used by industry to
measure the coefficient of aerodynamic drag of heavy trucks. As a
result, it is difficult for fleets to understand which truck
configurations have the lowest aerodynamic drag. We request comment on
the best approach to evaluate aerodynamic drag and the impact of
aerodynamic drag on truck GHG emissions.
---------------------------------------------------------------------------

    \148\ 21st Century Truck Partnership. Technology Roadmap for the
21st Century Truck Program. 21CT-001. December 2000. www.doe.gov/bridge.
    \149\ United States Department of Energy, Lawrence Livermore
National Laboratory. Working Group Meeting on Heavy Vehicle
Aerodynamic Drag: Presentation, Summary of Contents and Conclusion.
UCRL-TR-214683. May 2005.
    \150\ Bachman, L. Joseph,; Anthony Erb; Cheryl Bynum. Effect of
Single Wide Tires and Trailer Aerodynamics on Fuel Economy and
NO_x Emissions of Class 8 Line-Haul Tractor-Trailers. SAE
Paper 2005-01-3551. 2005.
---------------------------------------------------------------------------

    For a typical combination tractor-trailer truck traveling at 65
mph, energy losses due to tire rolling resistance can total nearly 13%
of the total energy consumed.\151\ Approximately 80-95% of the energy
losses from rolling resistance occur as the tire flexes and deforms
when it meets the road surface, due to viscoelastic heat dissipation in
the rubber. For heavy trucks, a 10% reduction in rolling resistance can
reduce GHG emissions by 1-3%.\152\ Improvements of this magnitude and
greater have already been demonstrated, and continued innovation in
tire design

[[Page 44456]]

has the potential to achieve even larger improvements in the future.
Specifying single wide tires on a new combination truck can have a
lower initial cost and lead to immediate fuel savings.\153\ Despite the
well-understood benefits of lower rolling resistance tires,
manufacturers differ in how they assess tire rolling resistance. We
seek comment on the potential for low rolling resistance tires to lower
GHG emissions, the need for consistent protocols to measure tire
rolling resistance, and the need for a common ranking or rating system
to provide tire rolling resistance information to the trucking industry.
---------------------------------------------------------------------------

    \151\ 21st Century Truck Partnership. Technology Roadmap for the
21st Century Truck Program. 21CT-001. December 2000. www.doe.gov/bridge.
    \152\ 21st Century Truck Partnership. Technology Roadmap for the
21st Century Truck Program. 21CT-001. December 2000. www.doe.gov/bridge.
    \153\ United States Environmental Protection Agency. A Glance at
Clean Freight Strategies: Single Wide-Based Tires. EPA420-F-04-004.
February 2004.
---------------------------------------------------------------------------

    Hybrid technologies, both electric and hydraulic, offer significant
GHG reduction potential. The hybrid powertrain is a combination of two
or more power sources: an internal combustion engine and a second power
source with an energy storage and recovery device. Trucks operating
under stop-and-go conditions, such as urban delivery trucks and refuse
trucks, lose a significant amount of energy during braking. In
addition, engines in most applications are designed to perform under a
wide range of requirements and are often oversized for the majority of
their requirements. Hybrid powertrain technologies offer opportunities
to capture braking losses and downsize the engine for more efficient
operation. We invite comment on the potential of GHG reductions from
hybrids in all types of heavy-duty applications.
    Currently most truck auxiliaries, such as the water pump, power
steering pump, air conditioning compressor, air compressor and cooling
fans, are mechanical systems typically driven by belts or gears off of
the engine driveshaft. The auxiliary systems are inefficient because
they produce power proportionate to the engine speed regardless of the
actual vehicle requirements and require conversion of fuel energy to
electrical or mechanical work. If systems were driven by electrical
systems they could be optimized for actual requirements and reduced
energy consumption. We request comment on the potential for these
auxiliary systems to lower GHG emissions from heavy-duty trucks.
    Air conditioning systems are responsible for GHG emissions from
refrigerant leakage and from the exhaust emissions generated by the
engine to produce the load required to run the air conditioning. The
emissions due to leakage can be reduced by the use of improved sealing
designs, low-permeation hoses, and refrigerant substitution. Replacing
today's refrigerant, HFC-134a, which has a high global warming
potential (GWP=1,300), with HFC-152a (GWP=120) or CO₂
(GWP=1) reduces the impact of the air conditioning leakage on the
environment.\154\ The load requirements of the air conditioning system
can be reduced through the use of improved condensers, evaporators, and
variable displacement compressors. We request comment on the impact of
air conditioning improvements on GHG reductions in heavy-duty trucks.
iii. Operational
---------------------------------------------------------------------------

    \154\ Frey, H. Christopher and Po-Yao Kuo. Best Practices
Guidebook for GHG Emissions Reductions in Freight Transportation.
Prepared for U.S. Department of Transportation via Center for
Transportation and the Environment. October 2007. Pages 26-27.
---------------------------------------------------------------------------

    The operation of the truck, including idle time and vehicle speed,
also has significant impact on the GHG emissions. Technologies that
improve truck operation exist and provide benefits to owners through
reduced fuel costs.
    Idling trucks emit a significant amount of CO₂ emissions
(as well as criteria pollutants). On average, a typical truck will emit
18 pounds of CO₂ per hour of idling.\155\ Long haul truck
idle reduction technologies can reduce main engine idling while still
meeting cab comfort needs. Some idle reduction technologies have no
upfront cost for the truck owner and hence represent an immediate
savings in operating costs with lower GHG emissions. Other idle
reduction technologies pay back within three years.\156\ In addition to
providing information about these systems, EPA seeks comment on whether
it should work with stakeholders to develop a formal evaluation
protocol for the effectiveness, cost, durability, and operability of
various idle-reduction technologies.
---------------------------------------------------------------------------

    \155\ United States Environmental Protection Agency. A Glance at
Clean Freight Strategies: Idle Reduction. EPA420-F-04-009. February 2004.
    \156\ EPA SmartWay Transport Partnership, Technology Package Savings 
Calculator, http://www.epa.gov/smartway/calculator/loancalc.htm.
---------------------------------------------------------------------------

    Vehicle speed is the single largest operational factor affecting
CO₂ emissions from large trucks. A general rule of thumb is
that every mph increase above 55 mph increases CO₂ emissions
by more than 1%. Speed limiters are generally available on new trucks
or as a low-cost retrofit, and assuming a five mph decrease in speed,
payback occurs within a few months.\157\
---------------------------------------------------------------------------

    \157\ American Trucking Associations Petition to National
Highway Traffic Safety Administration, (Docket NHTSA-2007-26851,
Document ID NHTSA-2007-26851-0005), October 20, 2006, and American
Trucking Associations Comment to Docket (Docket NHTSA-2007-26851,
Document ID NHTSA-2007-26851-3708), March 27, 2007.
---------------------------------------------------------------------------

    Automatic tire inflation systems maintain proper inflation
pressure, and thereby reduce tire rolling resistance. Studies indicate
that automatic tire inflation systems result in about 0.5 to 1%
reduction of CO₂ emissions for a typical truckload or less-
than-truckload over-the-road trucking fleet.\158\ Automatic tire
inflation systems can pay back in less than four years, assuming
typical underinflation rates.
---------------------------------------------------------------------------

    \158\ mission reduction and payback information from United
States Environmental Protection Agency. A Glance at Clean Freight
Strategies: Automatic Tire Inflation Systems. EPA420-F-04-010.
February 2004.
---------------------------------------------------------------------------

    All of the technologies summarized here can provide real GHG
reductions while providing value to the truck owner through reduced
fuel consumption. We request comment on the potential of these specific
technologies and on any other technologies that may allow vehicle
operators to reduce overall GHG emissions.
c. Regulatory Options for Reducing GHGs From Heavy-Duty Trucks
    In developing any GHG program for heavy-duty vehicles, we would
rely on our past experience addressing the multifaceted characteristics
of this sector. In the following sections, we discuss three potential
regulatory approaches for reducing GHG emissions from the heavy-duty
sector. We request comments on all aspects of these options. We also
encourage commenters to suggest other approaches that EPA should
consider to address GHG emissions from heavy-duty trucks, recognizing
that there are some important differences between criteria air
pollutants and GHG emissions.
    The heavy-duty engine manufacturers have made great strides in
reducing criteria pollutant emissions. We know these same manufacturers
have already achieved GHG emission reductions through the introduction
of more efficient engine technologies, and have the potential to
realize even greater reductions. We estimate that approximately 30% of
the overall GHG emission reduction potential from this sector comes
from engine improvements, 60% from truck improvements, and 10% from
operational improvements based on the technologies outlined in the 21st
Century Truck roadmap and Best Practices Guidebook for GHG Emissions
Reductions in Freight Transportation. We request comment on our assessment

[[Page 44457]]

of the relative contributions of engine, truck, and operational
technologies.
    The first approach we could consider would be a regulatory program
based on an engine CO₂ standard or weighted GHG standard
including N₂O and methane. One advantage to this option is
its simplicity because it preserves the current regulatory and market
structures. The heavy-duty engine manufacturers are familiar with
today's certification testing and procedures. They have facilities,
engine dynamometers, and test equipment to appropriately measure
emissions. The same equipment and test procedures can be, and already
are, used to measure CO₂ emissions. Measuring and reporting
N₂O and methane emissions would require relatively simple
additions to existing test cell instrumentation. We request comment
regarding issues that EPA should consider in evaluating this option and
the most appropriate means to address the issues raised. We recognize
that an engine-based regulatory structure would limit the potential GHG
emission reductions compared to programs that include vehicle
technologies and the crediting of fleets for operational improvements.
The other approaches considered below would have the potential to
provide greater GHG reductions by providing mechanisms to account for
vehicle and fleet operational changes.
    Recognizing that GHG emissions could be further reduced through
improvements to both engines and trucks, we request comment on an
alternative test procedure that would include vehicle aspects in an
engine-based standard. This option would still be based on an engine
standard. However, it would provide a mechanism to adjust the engine
test results to account for improvements in vehicle design. For
example, if through an alternate test procedure (e.g., a vehicle
chassis test) a hybrid truck were shown to reduce GHG emissions by 20%,
under this option an engine based GHG test result could be adjusted
downward by that same 20%. In this way, we could reflect a range of
vehicle or perhaps even operational changes into an engine based
regulatory program. In fact, we are already developing such an approach
for a vehicle based change to provide a better mechanism to evaluate
criteria emissions from hybrid vehicles.\159\ We are currently working
with the heavy-duty industry to develop these new alternate test
procedures and protocols. These new procedures could provide a
foundation for regulatory programs to address GHG emissions as well. We
request comment on the potential for alternate test procedures to
reflect vehicle technologies in an engine based GHG regulatory program.
---------------------------------------------------------------------------

    \159\ As discussed in section VI.C.2, we have also applied a
similar alternate test procedure approach in our new locomotive
standards (see 40 CFR 1033.530(h)).
---------------------------------------------------------------------------

    A second potential regulatory option for heavy-duty truck GHG
emissions would be to follow a model very similar to our current light-
duty vehicle test procedures. Each truck model could be required to
meet a GHG emissions standard based on a specified drive cycle. The
metric for the standard could be either a weighted GHG gram/mile with
prescribed test weight and payload or GHG gram/payload ton-mile to
recognize that heavy-duty trucks perform work. This option would
reflect an important change from our current regulatory approach for
most heavy-duty vehicles by direct regulation of trucks (and therefore
truck manufacturers) rather than engines.\160\ As discussed earlier in
this section, we have historically regulated heavy-duty engines rather
than vehicles reflecting in part the heavy-duty industry structure and
in part the preeminence of engine technology in controlling
NO_X and PM emissions. Clearly truck design plays a much more
important role in controlling GHG emissions due to significant energy
losses through aerodynamic drag and tire rolling resistance, and
therefore, this option directly considers the regulation of heavy-duty
trucks. We request comment on all aspects of this option including the
appropriate test metric, the need to develop new test procedures and
potential approaches for grouping heavy-duty vehicles into
subcategories for GHG regulatory purposes.
---------------------------------------------------------------------------

    \160\ For some years EPA has allowed gasoline and other non-
diesel vehicle manufactures to certify to and comply with a vehicle
based standard as compared to en engine based standard, at their
option. See, e.g., 40 CFR 86.005-10.
---------------------------------------------------------------------------

    As described earlier, there are a number of technologies and
operational changes that heavy-duty fleet operators can implement to
reduce both their overall operating costs and their GHG emissions.
Therefore, a third regulatory option that could be considered as a
complement to those discussed previously would be to allow heavy-duty
truck fleets to generate GHG emissions credits for applying
technologies to reduce GHG emissions, such as idle reduction, vehicle
speed limiters, air conditioning improvements, and improved aerodynamic
and tire rolling resistance. In order to credit the use of such
technologies, EPA would first need to develop procedures to evaluate
the potential for individual technologies to reduce GHGs. Such a
procedure could be based on absolute metrics (g/mile or g/ton-mile) or
relative metrics (percent reductions). We would further need to address
a wide range of complex potential issues including mechanisms to ensure
that the reductions are indeed realized in use and that appropriate
assurance of such future actions could be provided at the time of
certification, which occurs prior to the sale of the new truck. Such a
regulatory program could offer a significant opportunity to reward
trucking fleets for their good practices while providing regulatory
flexibility to help address the great diversity of the heavy-duty
vehicle sector. It would not lead to any additional GHG reductions,
however, as the credits generated by the fleet operators would be used
by the engine or vehicle makers to comply with their standards. We
welcome comments on the merits and issues surrounding potential
approaches to credit operational and technical changes from heavy-duty
fleets to reduce GHG emissions.
    In considering the regulatory options available, we are cognizant
of the significant burden that could result if these programs were to
require testing of every potential engine and vehicle configuration
related to its GHG emissions. Therefore, we have been following efforts
in Japan to control GHG emissions through a regulatory program that
relies in part on engine test data and in part on vehicle modeling
simulation. As currently constructed, Japan's heavy-duty fuel
efficiency regulation considers engine fuel consumption, transmission
type, and final drive ratio in estimating overall GHG emissions. Such a
modeling approach may be a worthwhile first step and may be further
improved by including techniques to recognize design differences in
vehicle aerodynamics, tire rolling resistance, weight, and other
factors. We request comment on the appropriateness of combining
emissions test data with vehicle modeling results to quantify and
regulate GHG emissions. In particular, we welcome comments addressing
issues including model precision, equality aspects of model based
regulation, and the ability to standardize modeling inputs.
    The regulatory approaches that we have laid out in this section
reflect incremental steps along a potential path to fully address GHG
emissions from this sector. These approaches should not be viewed as
discrete options but rather as potential building blocks that could be
mixed and matched in an

[[Page 44458]]

overall control program. Given the potential for significant burden,
EPA is also interested in considering how flexibilities such as
averaging, banking, and/or credit trading that may help to reduce costs
may be built into any of the regulatory options discussed above. We
request comment on all of the approaches described in this section and
the potential to implement one or more of these approaches in a phased
manner to capture the more straightforward approaches in the near-term
and the more complex approaches over a longer period.
3. Highway Motorcycles
    The U.S. motorcycle fleet encompasses a vast array of types and
styles, from small and light scooters with chainsaw-sized engines to
large and heavy models with engines as big as those found in many
family sedans. In 2006 approximately 850,000 highway motorcycles were
sold in the U.S., reflecting a near-quadrupling of sales in the last
ten years. Even as motorcycles gain in popularity, their overall GHG
emissions remain a relatively small fraction of all mobile source GHG
emissions. Most motorcycles are used recreationally and not for daily
commuting, and use is seasonally limited in much of the country. For
these reasons and the fact that the fleet itself is relatively small,
total annual vehicle miles traveled for highway motorcycles is about
9.5 billion miles (as compared to roughly 1.6 trillion miles for
passenger cars).\161\
---------------------------------------------------------------------------

    \161\ ``Highway Statistics 2003,'' U.S. Department of Transportation, 
Federal Highway Administration, Table VM-1, December 2004.
---------------------------------------------------------------------------

    The Federal Highway Administration reports that the average fuel
economy for motorcycles in 2003 was 50 mpg, almost twice that of
passenger cars in the same time frame. However, motorcycles are
generally designed and optimized to achieve maximum performance, not
maximum efficiency. As a result, many high-performance motorcycles have
fuel economy in the same range as many passenger cars despite the
smaller size and weight of motorcycles. Recent EPA emission regulations
are expected to reduce fuel use and hence GHG emissions from
motorcycles by: (1) Leading manufacturers to increase the use of
electronic fuel injection (replacing carburetors); (2) reducing
permeation from fuel lines and fuel tanks; and (3) eliminating the use
of two-stroke engines in the small scooter category.\162\
---------------------------------------------------------------------------

    \162\ See 69 FR 2398, January 15, 2004.
---------------------------------------------------------------------------

    There may be additional opportunities for further reductions in GHG
emissions. Options available to manufacturers may include incorporating
more precise feedback fuel controls; controlling enrichment on cold
starts and under load by electronically controlling choke operation;
allowing lower idle speeds when the opportunity exists; optimizing
spark for fuel and operating conditions through use of a knock sensor;
and, like light-duty vehicles, reducing the engine size and
incorporating a turbo-charger. The cost of these fuel saving and GHG
reducing technologies may be offset by the fuel savings realized over
the lifetime of the motorcycle.
    We request comment on information on what approaches EPA should
consider for potential further reductions in GHG emissions from
motorcycles. We also request comment and data regarding what technologies 
may be applicable to achieve further GHG reductions from motorcycles.

C. Nonroad Sector Sources

    As discussed previously, CAA section 213 provides broad authority
to regulate emissions from a wide array of nonroad engines and
vehicles,\163\ while CAA section 211 provides authority to regulate
fuels and fuel additives from both on-highway and nonroad sources and
CAA section 231 authorizes EPA to establish emissions standards for
aircraft. Collectively, the Title II nonroad and fuel regulation
programs developed by EPA over the past two decades provide a possible
model for how EPA could structure a long-term GHG reduction program for
nonroad engines and vehicles, fuels and aircraft.
---------------------------------------------------------------------------

    \163\ The Act does not define ``vehicle'', but we have
interpreted section 213 from its inception to include the broad
array of equipment, machines, and vessels powered by nonroad
engines, including those that are not self-propelled, such as
portable power generators. In keeping with common usage, we
typically use the generic terms ``equipment'', ``machine'', or
``application'', as well as the more application-specific terms
``vehicle'' and ``vessel'', to refer to these units, as appropriate.
---------------------------------------------------------------------------

    In this section, we first review and request comment on a number of
petitions received by EPA requesting action to regulate GHG emissions
from these sources and we highlight the similarities and key issues
raised in those petitions. We invite comment on all of the questions
and issues raised in these petitions. For each of three primary
groupings, nonroad, marine, and aircraft, we then discuss and seek
comment on the GHG emissions from these sources and the opportunities
to reduce GHG emissions through design and operational changes.
1. Petition Summaries
    Since the Massachusetts decision, EPA has received seven additional
petitions requesting that we make endangerment findings and undertake
rulemaking procedures using our authority under CAA sections 211, 213
and 231 to regulate GHG \164\ emissions from fuels, nonroad sources,
and aircraft. The petitioners represent states, local governments,
environmental groups, and nongovernmental organizations (NGO) including
the states of California, New Jersey, New Mexico, Friends of the Earth,
NRDC, OCEANA, International Center for Technology Assessment, City of
New York, and the South Coast Air Quality Management District. Copies
of these seven petitions can be found in the docket for this Advance
Notice. Following is a brief summary of these petitions. We request
comment on all issues raised by the petitioners.
---------------------------------------------------------------------------

    \164\ While petitioners vary somewhat in their definition of
GHGs, collectively they define carbon dioxide, methane, nitrous
oxide, hydrofluorocarbons, perfluorocarbons, water vapor, sulfur
hexaflouride, and soot or black carbon as GHGs.
---------------------------------------------------------------------------

a. Marine Engine and Vessel Petitions
    The Agency has received three petitions to reduce GHG emissions
from ocean-going vessels (OGVs). California submitted its petition on
October 3, 2007. A joint petition was filed on the same day by
EarthJustice on behalf of three environmental organizations: Oceana,
Friends of the Earth and the Center for Biological Diversity
(``Environmental Petitioners''). A third petition was received from the
South Coast Air Quality Management District (SCAQMD) on January 10, 2008.
    The California petition requests that EPA immediately begin the
process to regulate GHG emissions from Category 3 powered OGVs.\165\
According to the petition, the Governor of California has already
recognized that, ``California is particularly vulnerable to the impacts
of climate change,'' including the negative impact of increased
temperature on the Sierra snowpack, one of the State's primary sources
of water, and the further exacerbation of California's air quality
problems.\166\ The petition outlines the steps California has already
taken to reduce its own contributions to global warming and states that
it is petitioning the Administrator to take action to regulate GHG
emissions from

[[Page 44459]]

OGVs because it believes national controls will be most effective.
---------------------------------------------------------------------------

    \165\ A category 3 vessel is one where the main propulsion
engine(s) have a per-cylinder displacement of more than 30 liters.
    \166\ State of California, Petition for Rulemaking Seeking the
Regulation of Greenhouse Gas Emissions from Ocean--Going Vessels,
page3, October 3, 2007 (``California Petition'').
---------------------------------------------------------------------------

    California makes three key points in its petition. First,
California claims that EPA has clear authority to regulate OGV GHG
emissions under CAA section 213(a)(4). The State points out that the
``primary substantive difference'' between CAA section 202(a)(1), which
the Supreme Court found authorizes regulation of GHGs emissions from
new motor vehicles upon the Administrator making a positive
endangerment finding, and section 213 is that section 202(a)(1)
requires regulation if such an endangerment finding is made while
section 213(a)(4) authorizes, but does not require, EPA to regulate
upon making the requisite endangerment finding. But petitioner states
that EPA's discretion to decide whether to regulate OGVs under section
213(a)(4) is constrained in light of the overall structure and purpose
of the CAA. Citing the Massachusetts decision, California asserts that
the Supreme Court has ``set clear and narrow limits on the kinds of
reasons EPA may advance for declining to regulate significant sources
of GHGs''.
    The second claim California makes is that international law does
not bar regulation of GHG emissions from foreign-flagged vessels by the
U.S. California asserts that U.S. laws can operate beyond U.S. borders
(referred to as extra-territorial operation of laws) when the conduct
being regulated affects the U.S. and where Congress intended such
extra-territorial application.\167\ Petitioner believes that such
application of the CAA is both ``permissible and essential in this
case'' because to effectively control GHG emissions from shipping
vessels, the EPA must regulate foreign-flagged vessels since they
comprise 95% of the fleet calling on U.S. ports.\168\ Petitioner cites
two other instances where the U.S. has regulated foreign-flagged
vessels. First, in Specto v. Norwegian Cruiseline. 545 U.S. 119 (2005),
the Supreme Court held that the Americans with Disabilities Act (ADA)
could be applied to foreign-flagged cruise ships that sailed from U.S.
ports as long as the required accommodations for disabled passengers
did not require major, permanent modification to the ships involved.
Second, the National Park Service recently imposed air pollutant
emissions controls on cruise ships, including foreign-flagged cruise
ships that sail off the coast from Glacier Bay National Park, Alaska.
The petitioner points out that in this case they did so to protect and
preserve the natural resources of the Park, which is analogous to
California's reasons for why EPA must regulate GHG emissions from
foreign-flagged vessels.\169\
---------------------------------------------------------------------------

    \167\ Petitioners cite EEOC v. Arabian American Oil Co., 499
U.S. 244 (1991) (``Aramco'') as supporting this principle.
    \168\ California Petition, page 13.
    \169\ Petitioners cite regulations found at 36 CFR 13.65 (b)(4)
and 61 FR 27008, at 27011.
---------------------------------------------------------------------------

    The third claim raised in California's petition is that technology
is currently available to reduce GHG emissions from these vessels,
either through NO_X reductions or by reducing fuel
consumption. Options include, using marine diesel fuel oil instead of
bunker fuel, using selective catalytic reductions and exhaust gas
recirculation or by reducing speed. Petitioner states that the Clean
Air Act was intended to be a technology-forcing statute and that EPA
can and should consider OGV control measures that force the development
of new technology.
    California requests three forms of relief: (1) That EPA make a
finding that carbon dioxide emissions from new marine engines and
vessels significantly contribute to air pollution which may reasonably
be anticipated to endanger public health and welfare; (2) that EPA use
its CAA section 213(a)(4) authority to adopt regulations specifying
emissions standards for CO₂ emissions from these engines and
vessels; and (3) that EPA adopt regulations specifying fuel content or
type necessary to carry out the emission standards adopted for new
marine engines.
    The second group requesting EPA action on OGVs, Environmental
Petitioners, believes that climate change threatens public health and
welfare and that marine shipping vessels make a significant
contribution to GHG emissions, and that therefore EPA should quickly
promulgate regulations requiring OGVs to meet emissions standards by
``operating in a fuel-efficient manner, using cleaner fuels and/or
employing technical controls, so as to reduce emissions of carbon
dioxide, nitrous oxide, and black carbon.'' These petitioners further
state that EPA should also control ``the manufacture and sale of fuels
used in marine shipping vessels by imposing fuel standards'' to reduce
GHG emissions.\170\
---------------------------------------------------------------------------

    \170\ Environmental Petition, Petition for Rulemaking Under the
Clean Air Act to Reduce the Emissions of Air Pollutants from Marine
Shipping Vessels that Contribute to Global Climate Change, page 2,
October 3, 2007.
---------------------------------------------------------------------------

    The Environmental Petitioners focus their petition on four specific
arguments. First, like California, they assert that OGVs play a
significant role in global climate change. They focus on the emissions
of four pollutants: CO₂, NO_X, N₂0, and
black carbon (also known as soot). Petitioners cite numerous studies
that they assert document that the impact of these GHG emissions are
significant today and that industry trends indicate these emissions
will grow substantially in future decades. Second, petitioners lay out
a detailed legal argument asserting that EPA has clear authority to
regulate these four air pollutants from OGVs, and contending that the
Massachusetts decision must guide EPA's actions as it decides how to
regulate GHG emissions from OGVs. Third, petitioners discuss a number
of regulatory measures that can effectively reduce GHG emissions from
OGVs and which EPA could adopt using its regulatory authority under CAA
section 213(a)(4), including measures requiring restrictions on vessel
speed; requiring the use of cleaner fuels in ships and other technical
and operations measures petitioners believe are relatively easy and
cost-effective. Lastly, petitioners assert that the CAA section 213
provides EPA with clear authority to regulate GHG emissions from both
new and remanufactured OGV engines as well as from foreign-flagged
vessels.
    SCAQMD petition also requests Agency action under section 213 of
the CAA and states that it has a strong interest in the regulation of
GHG emissions from ships including emissions of NO_X, PM, and
CO₂. SCAQMD states that the net global warming effect of
NO_X emissions is potentially comparable to the climate
effect from ship CO₂ emissions and that PM emissions from
ships in the form of black carbon can also increase climate
change.\171\ Finally, because international shipping activity is
increasing yearly, SCAQMD asserts that if EPA dos not act quickly,
future ship pollution will become even worse, increasing both ozone and
GHG levels in the South Coast area of California. As with other
petitioners, SCAQMD states that there is a clear legal basis for EPA to
regulate ships GHG emissions under section 213(a)(4).
---------------------------------------------------------------------------

    \171\ SCAQMD, Petition for Rulemaking under the Clean Air Act to
Reduce Global Warming Pollutants from Ships, page 2, January 10, 2008.
---------------------------------------------------------------------------

    SCAQMD makes two additional assertions in its petition which mirror
the California and Environmental Petitions. First, EPA can avoid
regulation of ship GHG emissions only if it determines that
``endangerment'' can be avoided without regulation of ship
emissions.\172\ Second, SCAQMD believes that EPA has the authority to
regulate foreign-flagged vessels under at

[[Page 44460]]

least two circumstances: (1) For a foreign owned and operated vessel,
where the regulation(s) would not interfere with matters that ``involve
only the internal order and discipline of the vessel,'' Spector v.
Norwegian Cruise Lines, 545 U.S. 119, 131 (2005), and (2) where the
vessel is owned and operated by a U.S. corporation, even if it is
foreign-flagged.\173\
---------------------------------------------------------------------------

    \172\ SCAQMD Petition, page 9.
    \173\ SCAQMD Petition, page10.
---------------------------------------------------------------------------

    SCAQMD requests two types of relief: (1) That EPA, within six
months of receiving its petition, make a positive endangerment
determine for CO₂, NO_X, and black carbon
emissions from new marine engines and vessels ``because of their
contribution to climate change;'' and (2) that EPA promulgate
regulations under CAA section 213 (a)(4) to obtain the maximum feasible
reductions in emissions of these pollutants. We invite comment on all
elements of the petitioners' assertions and requests.
b. Aircraft Petitions
    The Agency has received two petitions to reduce GHG emissions from
aircraft.\174\ The first petition was submitted on December 4, 2007, by
California, Connecticut, New Jersey, New Mexico, Pennsylvania's
Department of Environmental Protection, the City of New York, the
District of Columbia, and the SCAQMD (``State Petitioners''). A second
petition was filed on December 31, 2007, by Earthjustice on behalf of
four environmental organizations: Friends of the Earth, Oceana, Center
for Biological Diversity and NRDC (``Environmental Petitioners'').
---------------------------------------------------------------------------

    \174\ While aircraft engines are not ``nonroad engines'' as
defined in CAA section 216(10) and aircraft are not ``nonroad
vehicles'' as defined in CAA section 216(11), such that aircraft
could be subject to regulation under CAA section 213, for
organizational efficiency we include aircraft in this ``Nonroad
Sector Sources'' section of today's notice.
---------------------------------------------------------------------------

    All petitioners request that EPA exercise its authority under
section 231(a) of the CAA to regulate GHG emissions from new and
existing aircraft and/or aircraft engine operations, after finding that
aircraft GHG emissions cause or contribute to air pollution which may
reasonably be anticipated to endanger public health or welfare.\175\
Petitioners suggest that these regulations could allow compliance
through technological controls, operational measures, emissions fees,
or a cap-and-trade system.
---------------------------------------------------------------------------

    \175\ Petitioners maintain that aircraft engine emissions of
CO₂, NO_X, water vapor, carbon monoxide, oxides
of sulfur, and other trace components including hydrocarbons such as
methane and soot contribute to global warming and that in 2005,
aircraft made up 3% of U.S. CO₂ emissions from all
sectors, and 12% of such emissions from the transportation sector.
States of California et al, Petition for Rulemaking Seeking the
Regulation of Greenhouse Gas Emissions from Aircraft, page 11,
December 4, 2007, and Friends of the Earth et al., Petition for
Rulemaking under the Clean Air Act to Reduce the Emissions of Air
Pollutants from Aircraft that Contribute to Global Climate Change,
pages 6-7, December 31, 2007.
---------------------------------------------------------------------------

    Both petitions discuss how aircraft engines emit GHG emissions
which they assert have a disproportionate impact on climate change.
Petitioners cite a range of scientific documents to support their
statements. They assert that ground-level aircraft NO_X, a
compound they identify as a GHG, contributes to the formation of ozone,
a relatively short-lived GHG. NO_X emissions in the upper
troposphere and tropopause, where most aircraft emissions occur, result
in greater concentrations of ozone in those regions of the atmosphere
compared to ground level ozone formed as a result of ground level
aircraft NO_X emissions. Petitioners contend that aircraft
emissions contribute to climate change also by modifying cloud cover
patterns. Aircraft engines emit water vapor, which petitioners identify
as a GHG that can form condensation trails, or ``contrails,'' when
released at high altitude. Contrails are visible line shaped clouds
composed of ice crystals that form in cold, humid atmospheres.
Persistent contrails often evolve and spread into extensive cirrus
cloud cover that is indistinguishable from naturally occurring cirrus
clouds. The petitioners state that over the long term this contributes
to climate change.
    State Petitioners highlight the effects climate change will have in
California and the City of New York as well as efforts underway in both
places to reduce GHG emissions. They argue that without federal
government regulation of GHG emissions from aircraft, their efforts at
mitigation and adaptation will be undermined. Both petitioners urge
quick action by EPA to regulate aircraft GHG emissions since these
emissions are anticipated to increase considerably in the coming
decades due to a projected growth in air transport both in the United
States and worldwide. They cite numerous reports to support this point,
including an FAA report, which indicates that by 2025 emissions of
CO₂ and NO_X from domestic aircraft are expected
to increase by 60%.\176\
---------------------------------------------------------------------------

    \176\ FAA, Office of Environment and Energy, Aviation and
Emission: A Primer, January 2005, page 10, available at 
http://www.faa.gov/regulations_policies/policy_guidance/envir_policy/
media/aeprimer.pdf.
---------------------------------------------------------------------------

    We request comment on all issues raised in the petitions,
particularly on two assertions made by Environmental Petitioners: (1)
That technology is available to reduce GHG emissions from aircraft
allowing EPA to take swift action, and (2) that EPA has a mandatory
duty to control GHG emissions from aircraft and can fulfill this duty
consistent with international law governing aircraft. In addition, we
invite comment on the petitioners' assessment of the impact of aircraft
GHG emissions on climate change, including the scientific understanding
of these impacts, and whether aircraft GHG emissions cause or
contribute to air pollution which may reasonably be anticipated to
endanger public health or welfare.
    With regard to technology, petitioners highlight existing and
developing aviation procedures and technologies which could reduce GHG
emissions from new and existing aircraft. For example, they point to
various aviation operations and procedures including minimizing engine
idling time on runways and employing single engine taxiing that could
be undertaken by aircraft to reduce GHG emissions. Petitioners also
discuss the availability of more efficient aircraft designs to reduce
GHG emissions, such as reducing their weight, and they suggest that
using alternative fuels could also reduce aviation GHG emissions.
    Environmental Petitioners contend that once EPA makes a positive
endangerment finding for aircraft GHG emissions, EPA has a mandatory
duty to act, but that the potential regulatory responses available to
EPA are quite broad and should be considered for all classes of
aircraft, including both new and in-use aircraft and aircraft engines.
In addition, petitioners argue that EPA's authority to address GHG
emissions from aircraft is consistent with international law-in
particular the Convention on International Civil Aviation (the
``Chicago Convention'')--and that the United States'' obligations under
the Convention do not constrain EPA's authority to adopt a program that
addresses aviation's climate change impacts, including those from
foreign aircraft.
    The State and Environmental Petitioners each request the following
relief: (1) That EPA make an explicit finding under CAA section
231(a)(2)(A) that GHG emissions from aircraft cause or contribute to
air pollution which may reasonably be anticipated to endanger public
health or welfare; (2) that EPA propose and adopt standards for GHG
emissions from both new and in-use aircraft as soon as possible; (3)
that EPA adopt regulations that allow a range of compliance approaches,
including emissions limits, operations practices and/or fees, a cap-
and-trade system, as well as measures that are more near-

[[Page 44461]]

term, such as reduced taxi time or use of ground-side electricity
measures. The Environmental Petitioners' also request that EPA issue
standards 90 days after proposal. We invite comment on all elements of
the petitioners' assertions and requests, as well as the scientific and
technical basis for their assertions and requests.
c. Nonroad Engine and Vehicle Petitions
    On January 29, 2008, EPA received two petitions to reduce GHG
emissions from nonroad engines and vehicles. The first petition was
submitted by California, Connecticut, Massachusetts, New Jersey and
Oregon and Pennsylvania's Department of Environmental Protection
(``State Petitioners''). The second petition was submitted by the
Western Environmental Law Center on behalf of three nongovernmental
organizations: the International Center for Technology Assessment,
Center for Food Safety, and Friends of the Earth (``NGO Petitioners'').
    Both petitions request that EPA exercise its authority under CAA
section 213(a)(4) to adopt emissions standards to control and limit GHG
emissions from new nonroad engines excluding aircraft and vessels. Both
petitions seek EPA regulatory action on a wide range of nonroad engines
and equipment, which the petitioners believe, contribute substantially
to GHG emissions, including outdoor power equipment, recreational
vehicles, farm and construction machinery, lawn and garden equipment,
logging equipment and marine vessels.\177\
---------------------------------------------------------------------------

    \177\ The two petitions request that EPA regulate slightly
different categories of nonroad engines and vehicles under CAA
section 213. State Petitioners exclude from their request aircraft,
locomotives and ocean-going vessels and do not include rebuilt
heavy-duty engines. The NGO Petitioners exclude only aircraft and
ocean-going vessels but also request that EPA use its CAA section
202 authority to regulate GHG emissions from rebuilt heavy-duty engines.
---------------------------------------------------------------------------

    The State Petitioners, mirroring the earlier State petitions on
ocean-going vessels and aircraft, describe the harms which they believe
will occur due to climate change, including reduced water supplies,
increased wildfires, and threats to agricultural outputs in California;
loss of coastal wetlands, beach erosion, saltwater intrusion of
drinking water in Massachusetts and Connecticut; and similar harms to
the Pennsylvania, New Jersey and Oregon. The petition highlights
actions that California has already taken to reduce its own
contributions to global warming but points out that only EPA has
authority to regulate emissions from new farm and construction
equipment under 175 horsepower, ``which constitutes a sizeable portion
of all engines in this category.* * * '' \178\
---------------------------------------------------------------------------

    \178\ States Petition for Nonroad, page 7-8.
---------------------------------------------------------------------------

    The State Petitioners present three claims which, they believe
compel EPA action to reduce GHG emissions from nonroad sources. First,
petitioners claim that GHG emissions from these sources are
significant.\179\ Petitioners cite various reports documenting national
GHG emissions from a broad range of nonroad categories which, they
contend, provide evidence that nonroad GHG emissions are already
substantial, and will continue to increase in the future. Petitioners,
also cite additional inventory reports that nonroad GHG emissions
already exceed total U.S. GHG emissions from aircraft as well as from
boats and ships, rail, and pipelines combined.\180\ Petitioner's
present California nonroad GHG emissions data which, they contend,
mirror national GHG emission trends for nonroad engines and bolster
their claim that GHG emissions from the nonroad sector, as a whole, are
significant and are substantial for three categories: Construction and
mining equipment, agricultural, and industrial equipment.
---------------------------------------------------------------------------

    \179\ Petitioners indicate that in 2007, non-transportation
mobile vehicles and equipment were responsible for approximately 220
million tons of CO₂ emissions (data derived from EPA's
Nonroad Emissions model for 2007). State of California et al,
Petition for Rulemaking Seeking the Regulation of Greenhouse Gas
Emissions from Nonroad Vehicles and Engines, page 8, January 29,
2008, and International Center for Technology Assessment et al,
Petition for Rulemaking Seeking the Regulation of Greenhouse Gas
Emissions from Nonroad Vehicles and Engines, page 5, January 29, 2008.
    \180\ State Petition for Nonroad, page 9.
---------------------------------------------------------------------------

    State Petitioners' second claim is that EPA has the authority to
regulate GHG emissions from nonroad sources, although they acknowledge
that CAA section 213(a)(4) is discretionary. Petitioners contend this
discretion is not unlimited and that the structure of the CAA must
guide EPA's actions. Petitioners maintain that since the CAA prohibits
States from undertaking their traditional police power role in
regulating pollution from new construction or agricultural sources
under 175 horsepower, ``Congress has implicitly invested EPA with the
responsibility to act to prevent [these] harmful emissions.'' The third
and final claim raised by State Petitioners is that both physical and
operational controls are currently available to achieve fuel savings
and/or to limit GHG emissions. Such measures include idle reduction,
electrification of vehicles, the use of hybrid or hydraulic-hybrid
technology, as well as use of ``cool paints'' that reduce the need for
air conditioning.
    NGO petitioners make three similar claims in their petition. First,
petitioners argue that serious public health and environmental
consequences are projected for this century unless effective and timely
action is taken to mitigate climate change. Petitioners further contend
that GHG emissions from nonroad engines and vehicles are responsible
for a significant and growing amount of GHG emissions and, like the
State petitioners previously, they highlight three nonroad sectors
responsible for a large portion of these GHG emission--construction,
mining, and agriculture.
    Petitioners' second claim is that once EPA renders a positive
endangerment determination under CAA section 202 for motor vehicles and
engines, this finding should also satisfy the endangerment
determination required under CAA section 213(a)(4) for nonroad engines.
EPA's discretion under CAA section 213(a)(4) is limited, petitioners
assert, by the relevant statutory considerations, as held by the
Supreme Court in Massachusetts v. EPA, so that the Agency ``can decline
to regulate nonroad engine and vehicle emissions only if EPA determines
reasonably that such emissions do not endanger public health or
welfare, or else, taking into account factors such as cost, noise,
safety and energy, no such regulations would be appropriate.'' \181\
Like State petitioners, NGOs point out that because the CAA restricts
states' ability to regulate pollution from new construction or farm
vehicles and engines under 175 horsepower, Congress ``implicitly
invested EPA with unique responsibility to act in the states'' stead so
as to prevent such harmful emissions.'' Petitioners also argue that the
National Environment Policy Act (NEPA) section 101(b) compels EPA
action to fulfill its duty ``as a trustee of the environment for
succeeding generations.''
---------------------------------------------------------------------------

    \181\ NGO Petition, page 8.
---------------------------------------------------------------------------

    NGO Petitioners' third claim is that a wide range of technology is
currently available to reduce GHG emissions from nonroad engines and
vehicles and that, in addition, the CAA was intended to be a
technology-forcing statute so that EPA ``can and should'' establish
regulations that ``substantially limit GHG emissions.* * * even where
those regulations force the development of new technology.'' Regarding
technology availability, petitioners provide a list of technologies
that they believe are currently available to reduce GHG emissions from
nonroad vehicles and engines, including auxiliary power unit systems to
avoid engine use solely to

[[Page 44462]]

heat or cool the cab; tire inflation systems; anti-idling standards;
use of hybrid or hydraulic-hybrid technology; use of low carbon fuels;
and use of low viscosity lubricants.
    Both State and NGO Petitioners request three types of relief: (1)
That EPA make a positive endangerment determination for GHG emissions
from nonroad vehicles and engines; \182\ (2) that EPA adopt regulations
to reduce GHG emissions from this sector; and (3) that regulations
necessary to carry out the emissions standards also be adopted.\183\ We
invite comment on all of the petitioners' assertions and requests.
---------------------------------------------------------------------------

    \182\ In addition, NGO Petitioners also request that EPA make a
determination under CAA section 202 (a)(3)(D) that GHG emissions
from rebuilt heavy-duty engines also are significant contributors to
air pollution which may reasonably be anticipated to endanger public
health and welfare. NGO Petition, page 11.
    \183\ State Petitioners indicate that adopting regulations
specifying fuel type, for example, may be necessary to carry out the
emission limitations.
---------------------------------------------------------------------------

2. Nonroad Engines and Vehicles
    In this section, we discuss the GHG emissions and reduction
technologies that are or may be available for the various nonroad
engines and vehicles that are the subject of the petitioners described
above. Since section 213 was added to the CAA in 1990, the Agency has
completed a dozen major rulemakings which established programs that
reduce traditional air pollutants from nonroad sources by over 95%,
benefitting local, regional, and national air quality. EPA's approach
has been to set standards based on technology innovation, with
flexibility for the regulated industries to meet environmental goals
through continued innovation that can be integrated with marketing plans.
    With help from industry, environmental groups and state regulators,
EPA has designed nonroad regulatory programs that have resulted in
significant air quality gains with little sacrifice of products'
ability to serve their purpose. In fact, manufacturers have generally
added new features and performance improvements that are highly
desirable to users. Because GHG reductions from nonroad sources can be
derived from fuel use reductions that directly benefit the user's
bottom line, we expect that manufacturers' incentive to increase the
fuel efficiency of their products will be even stronger in the future.
This potential appears higher for nonroad engines compared to highway
engines because in the past energy consumption has been less of a focus
in the nonroad sector, so there may be more opportunity for improvement, 
while at the same time higher fuel prices are now beginning to make fuel 
expenses more important to potential equipment purchasers.
    The Agency and regulated industries have in the past grouped
nonroad engines in a number of ways. The first is by combustion cycle,
with two primary cycles in use: compression-ignition (CI) and spark-
ignition (SI). The combustion cycle is closely linked to grouping by
fuel type, because CI engines largely burn diesel fuel while SI engines
burn gasoline or, for forklifts and other indoor equipment, liquefied
petroleum gas (LPG). It has also been useful to group nonroad engines
by application category. Regulating nonroad engine application
categories separately has helped the Agency create effective control
programs, due to the nonroad sector's tremendous diversity in engine
types and sizes, equipment packaging constraints, affected industries,
and control technology opportunities. Although for the sake of
discussion we use these application groupings, we solicit comment on
what grouping engines and applications would make the most sense for
GHG regulation, especially if flexible emissions credit and averaging
concepts are pursued across diverse applications.
a. Nonroad Engine and Vehicle GHG Emissions
    Nonroad engines emitted 249 million metric tons of CO₂
in 2006, 12% of the total mobile source CO₂ emissions.\184\
CO₂ emissions from the nonroad sector are expected to
increase significantly in the future, approximately 46% between 2006
and 2030. Diesel engines emit 71% of the total nonroad CO₂
emissions. The other 29% comes from gasoline, LPG, and some natural
gas-fueled engines. CO₂ emissions from individual nonroad
application categories in decreasing order of prominence are: Nonroad
diesel (such as farm tractors, construction and mining equipment),
diesel locomotives, small SI (such as lawn mowers, string trimmers, and
portable power generators), large SI (such as forklifts and some
construction machines), recreational marine SI, and recreational
offroad SI (such as all terrain vehicles and snowmobiles).
---------------------------------------------------------------------------

    \184\ Emissions data in this section are from Inventory of U.S.
Greenhouse Gas Emissions and Sinks: 1990-2006. EPA 430-R-08-005.
April 2008, and EPA NONROAD2005a model.
---------------------------------------------------------------------------

    GHG emissions from nonroad applications are dominated by
CO₂ emissions which comprise approximately 97% of the total.
Approximately 3% of the GHG emissions (on a CO₂ equivalent
basis) from nonroad applications are due to hydrofluorocarbon
emissions, mainly from refrigerated rail transport. Methane and
N₂O make up less than 0.2% of the nonroad sector GHG
emissions on a CO₂ equivalent basis. Much of the following
discussion focuses on technology opportunities for CO₂
reduction, but we note that these technologies will generally reduce
N₂O and methane emissions as well, and we ask for comment on
measures and options for specifically addressing N₂O and
methane emissions.
b. Potential for GHG Reductions From Nonroad Engines and Vehicles
    The opportunity for GHG reductions from the nonroad sector closely
parallels the highway sector, especially for the heavy-duty highway and
nonroad engines that share many design characteristics. In addition,
there is potential for significant further GHG reductions from changes
to vehicle and equipment characteristics. A range of GHG reduction
opportunities is summarized in the following discussion. Comment is
requested on these opportunities and on additional suggestions for
reducing GHGs from nonroad sources.
    It should be noted that any means of reducing the energy
requirements necessary to power a nonroad application can yield the
desired proportional reductions of GHGs (and other pollutants as well).
Although in past programs, the Agency has typically focused on a new
engine's emissions per unit of work, such as gram/brake horsepower-hour
(g/bhp-hr), it may prove more effective to achieve GHG reductions by
redesigning the equipment or vehicle that the engine powers so that the
nonroad application accomplishes its task while expending less energy.
Improvements such as these do not show up in measured g/bhp-hr
emissions levels, but would be reflected in some other metric such as
grams emitted by a locomotive in moving a ton of freight one mile.
    EPA solicits comment on possible nonroad GHG emissions reduction
strategies for the various ``pathways'' by which GHGs can be impacted.
Although it is obvious that internal combustion engines emit GHGs via
the engine exhaust, it is helpful to take the analysis to another level
by putting it in the context of energy use and examining the pathways
by which energy is expended in a nonroad application, such as through
vehicle braking. Because of the diversity of nonroad applications, we
are taking a different approach here than in other sections of this
notice: first, we summarize some of the engine, equipment, and
operational pathways

[[Page 44463]]

and opportunities for GHG reductions that are common to all or at least
a large number of nonroad applications; next, we examine more closely
just one of the hundreds of nonroad applications, locomotives, to
illustrate the many additional application-specific pathways for GHG
reductions that are available. Our assessment is that, despite the
great diversity in nonroad applications, technology-based solutions
exist for every application to achieve cost-effective and substantial
GHG emissions reductions.
i. Common GHG Reduction Pathways
    To ensure that this advance notice initiates the widest possible
discussion of potential GHG control solutions, the following discussion
includes all three types of possible control measures: engine,
equipment, and operational.
(1) Engine Pathways
    To date, improving fuel usage in many nonroad applications has not
been of great concern to equipment users and therefore to designers.
There is potential for technologies now fairly commonplace in the
highway sector, such as advanced lubricants and greater use of
electronic controls, to become part of an overall strategy for GHG
emissions reduction in the nonroad sector. We welcome comment on the
opportunities and limitations of doing so.
    One engine technology in particular warrants further discussion.
Two-stroke gasoline engines have been popular especially in handheld
lawn care applications and recreational vehicles because they are
fairly light and inexpensive. However, they also produce more GHGs than
four-stroke engines. Much progress has been made in recent years in the
development of four-stroke engines that function well in these
applications. We ask for comment on the extent to which a shift to
four-stroke engines would be feasible and beneficial.
    Although today's nonroad gasoline and diesel engines produce
significantly less GHGs than earlier models, further improvements are
possible. Engine designers are continuing to work on new designs
incorporating technologies that produce less GHGs, such as homogeneous
charge CI, waste heat recovery through turbo compounding, and direct
fuel injection in SI engines. Most of this work has already been done
for the automotive sector where economies of scale can justify the
large investments. Much of this innovation can eventually be adapted to
nonroad applications, as has occurred in the past with such
technologies as electronic fuel injection and common rail fueling. We
therefore request comment on the feasibility and potential for these
advanced highway sector technologies, discussed in section VI.B, to be
introduced or accelerated in the nonroad sector.
(2) Equipment and Operational Pathways
    Technology solutions in both the equipment design and operations
can reach beyond the engine improvements to further reduce GHG
emissions. We broadly discuss the following technologies below:
Regenerative energy recovery and hybrid power trains, CVT
transmissions, air conditioning improvements, component design
improvements, new lighting technologies, reduced idling, and consumer
awareness.
    Locomotives, as an example, have significant potential to recover
energy otherwise dissipated as heat during braking. An 8,000-ton coal
train descending through 5,000 feet of elevation converts 30 MW-hrs of
potential energy to frictional and dynamic braking energy. Storing that
energy on board quickly enough to keep up with the energy generation
rate presents a challenge, but may provide a major viable GHG emissions
reduction strategy even if only partially effective. Another
regenerative opportunity relates to the specific, repetitive,
predictable work tasks that many nonroad machines perform. For example,
a forklift in a warehouse may lift a heavy load to a shelf and in doing
so expend work. Just as often, the forklift will lower such a load from
the shelf, and recover that load's potential energy, if a means is
provided to store that energy on board.
    There are, however, many nonroad applications that may not have
much potential for regenerative energy recovery (a road grader, for
example), but in those applications a hybrid diesel-electric or diesel-
hydraulic system without a regenerative component may still provide
some GHG benefits. A machine that today is made with a large engine to
handle occasional peak work loads could potentially be redesigned with
a smaller engine and battery combination sized to handle the occasional
peak loads.
    Besides pre-existing electrical or hydraulic systems, some nonroad
applications have one additional advantage over highway vehicles in
assessing hybrid prospects: They often have quite predictable load
patterns. A hybrid locomotive, for example, can be assigned to
particular routes, train sizes, and consist (multi-locomotive) teams,
to ensure it is used as close to full capacity as possible. The space
needs of large battery banks could potentially be accommodated on a
tender car, and the added weight would be offset somewhat by a smaller
diesel fuel load (typically 35,000 lbs today) and dynamic brake grid.
At least one locomotive manufacturer, General Electric, is already
developing a hybrid design, and battery energy storage has been
demonstrated for several years in rail yard switcher applications.
    We request comment on all aspects of the hybrid and regeneration
opportunity in the nonroad sector, including the extent to which the
electric and hydraulic systems already designed into many nonroad
machines and vehicles could provide some cost savings in implementing
this technology, and the extent to which plug-in technologies could be
used in applications that have very predictable downtime such as
overnight at construction sites, or that can use plug-in electric power
while working or while sitting idle between tasks.
    A Continuously Variable Transmission (CVT) has an advantage over
other conventional transmission designs by allowing the engine to
operate at its optimum speed over a range of vehicle speeds and
typically over a wider range of available ratios, which can provide GHG
emission reductions. It has been estimated that CVTs can provide a 3 to
8% decrease in fuel use over 4-speed automatic transmissions.\185\ They
are already in use some in nonroad vehicles such as snowmobiles and
all-terrain vehicles, and could possibly be used in other nonroad
applications as well. We request comment on the opportunities to apply
CVT to various nonroad applications.
---------------------------------------------------------------------------

    \185\ ``Effectiveness and Impact of Corporate Average Fuel
Economy (CAFE) Standards,'' National Research Council, National
Academy of Sciences, 2002.
---------------------------------------------------------------------------

    Some nonroad applications have air conditioning or refrigeration
equipment, including large farm tractors, highway truck transport
refrigeration units (TRUs), locomotives, and refrigerated rail cars.
Reducing refrigerant leakage in the field or reducing its release
during maintenance would work to reduce GHG emissions In addition, a
switch to refrigerants with lower GHG emissions than the currently-used
fluorinated gases can have a significant impact. We expect that the
measures used to reduce nonroad equipment refrigerant GHGs would most
likely involve the same strategies that have been or could be pursued
in the highway and stationary

[[Page 44464]]

source sectors, and the reader is referred to section VI.B.1 for
additional discussion. We request comment on the degree to which
nonroad applications emit fluorinated gases, and on measures that may
be taken to reduce these emissions.
    An extensive variety of energy-consuming electrical, mechanical,
and hydraulic accessories are designed into nonroad machines to help
them perform their tasks. Much of the energy output of a nonroad engine
passes through these components and systems in making the machine do
useful work, and all of them have associated energy losses through
bearing friction, component heating, and other pathways. Designing
equipment to use components with lower GHG impacts in these systems can
yield substantial overall reductions in GHG emissions.
    Some nonroad applications expend significant energy in providing
light, such as locomotive headlights and other train lighting.
Furthermore, diesel-powered portable light towers for highway
construction activities at night are increasingly being used to reduce
congestion from daytime lane closures. We request comment on the extent
to which a switch to less energy-intensive lighting could reduce GHG
emissions.
    Many nonroad diesel engines are left idling during periods when no
work is demanded of them, generally as a convenience to the operator,
though modern diesel engines are usually easy to restart. In some
applications this may occupy hours every day. Even though the hourly
fuel rate is fairly low during idle, in the past several years
railroads have saved considerable money by adding automatic engine stop
start (AESS) systems to locomotives. These monitor key parameters such
as state of battery charge, and restart the engine only as needed,
thereby largely eliminating unnecessary idling. They reduce GHG
emissions and typically pay for themselves in fuel savings within a
couple of years. Our recent locomotive rule mandated these systems for
all new locomotives as an emission control measure (40 CFR
1033.115(g)). AESS or similar measures may be feasible for other
nonroad applications with significant idling time as well. We request
comment on the availability and effectiveness of nonroad idle reduction
technologies.
ii. Application-Specific GHG Pathways
    As mentioned above, we discuss application-specific approach for
further reducting GHG emissions from one nonroad application,
locomotives, to illustrate application-specific opportunities for GHG
emission reductions beyond those discussed above that apply more
generally. We note that some of these application-specific
opportunities, though limited in breadth, may be among the most
important, because of their large GHG reduction potential.
    We have chosen locomotives for this illustration in part because
rail transportation has already been the focus of substantial efforts
to reduce its energy use, resulting in generally favorable GHG
emissions per ton-mile or per passenger-mile. The Association of
American Railroads calculates that railroads move a ton of freight 423
miles on one gallon of diesel fuel.\186\ Reasons for the advantage
provided by rail include the use of medium-speed diesel engines, lower
steel-on-steel rolling resistance, and relatively gradual roadway
grades. Rail therefore warrants attention in any discussion on mode-
shifting as a GHG strategy. Even if GHG emissions reduction were not at
issue, shippers and travelers already experience substantial mode-shift
pressure today from long-term high fuel prices. Growth in the rail
sector highlights the critical importance of locomotive GHG emissions
reduction.
---------------------------------------------------------------------------

    \186\ Comments of the Association of American Railroads on EPA's
locomotive and marine engine proposal, July 2, 2007. Available in
EPA docket EPA-HQ-OAR-2003-0190.
---------------------------------------------------------------------------

    We have listed some key locomotive-specific opportunities below. We
note that a number of these are aimed at addressing GHG pathways from
rail cars. Rail cars create very significant GHG reduction pathways for
locomotives, because all of the very large energy losses from railcar
components translate directly into locomotive fuel use. This is
especially important when one considers that an average train has
several dozen cars. We request comment on the feasibility of the ideas
on this list and on other possible ways to reduce GHG emissions.
Opportunities for Rail GHG Reduction
Locomotives
     Low-friction wheel bearings
     Aerodynamic improvements
     Idle emissions control beyond AESS (such as auxiliary power units)
     Electronically-controlled pneumatic (ECP) brakes
     High-adhesion trucks (wheel assemblies)
     Global positioning system (GPS)-based speed management (to minimize 
braking, over-accelerations, and run-out/run-in losses at couplings)
Railcars
     Low-torque rail car wheel bearings
     Tare weight reduction
     Aerodynamic design of rail cars and between-car gaps
     Better insulated refrigeration cars
Rail Infrastructure
     Application of lubricants or friction modifiers to
minimize wheel-to-track friction losses
     Higher-speed railroad crossings
     Targeted-route electrification
     Rail yard infrastructure improvements to eliminate
congestion and idling
Operational
     Consist manager (automated throttling of each locomotive
in a consist team for lowest overall GHG emissions)
     Optimized GPS-assisted dispatching/routing/tracking of
rail cars and locomotives
     Optimized matching of locomotives with train load for every route 
(including optimized placement of each locomotive along the train)
     Expanded resource sharing among railroads
     Reduction of empty-car trips
     Early scrappage of higher-GHG locomotives
c. Regulatory Options for Nonroad Engines and Vehicles
    There is a range of options that could be pursued under CAA section
213 to control nonroad sector GHGs. The large diversity in this sector
allows for a great number of technology solutions as discussed above,
while also presenting some unique challenges in developing a
comprehensive, balanced, and effective regulatory program, and
highlights the importance of considering multiple potential regulatory
strategies. We have met similar challenges in regulating traditional
air pollutants from this sector, and we request comment on the
regulatory approaches discussed below and whether they would address
the challenges of regulating GHGs from nonroad engines.
    As discussed in our earlier section on heavy-duty vehicles, the
potential regulatory approaches that we discuss here should be
considered not as discrete options but as a continuum of possible
approaches to address GHG emissions from this sector. Just as we have
in our technology discussion, these regulatory approaches begin with
the engine and then expand to included potential approaches to realize
reductions through vehicle and operational changes. In approaching the
discussion in this way, each step along such a path has the potential
to greater regulatory complexity but also has the

[[Page 44465]]

potential for greater regulatory flexibility, GHG reduction, and
program benefits. For large GHG reductions in the long term we expect
to give consideration to approaches that accomplish the largest
reductions, but we also note that, given the long time horizons for GHG
issues, we can consider a number of incremental regulatory steps along
a longer path. Also, given the absence of localized effects associated
with GHG emissions, EPA is interested in considering the incorporation
of banking, averaging, and/or credit trading into the regulatory
options discussed below.
    The first regulatory approach we consider is a relatively
straightforward extension of our existing criteria pollutant program
for nonroad engines. In its simplest form, this approach would be an
engine GHG standard that preserves the current regulatory structure for
nonroad engines. Nonroad engine manufacturers are already familiar with
today's certification testing and procedures. Just like the highway
engine manufacturers, they have facilities, engine dynamometers, and
test equipment to appropriately measure GHG emissions. Further,
technologies developed to reduce GHG emissions from heavy-duty engines
could be applied to the majority of diesel nonroad engines with
additional development to address differences in operating conditions
and engine applications in nonroad equipment. Hence, this approach
would benefit from both regulatory work done to develop a heavy-duty
engine GHG program and technology development for heavy-duty engines to
comply with a GHG program. While we do not expect that new test cycles
would be needed to effect meaningful GHG emissions control, we request
comment on whether new test cycles would allow for improved control,
and especially on whether there are worthwhile GHG control technologies
that would not be adequately exercised and measured under the current
engine test cycles and test procedures.
    A second approach that would extend control opportunities beyond
engine design improvements involves developing nonroad vehicle and
equipment GHG standards. Changes to nonroad vehicles and equipment can
offer significant opportunity for GHG emission reductions, and
therefore any nonroad GHG program considered by EPA would need to
evaluate the potential for reductions not just from engine changes but
from vehicle and equipment changes as well. In section VI.B.2 we
discussed a potential heavy-duty truck GHG standard (e.g., a gram per
mile or gram per ton-mile standard). A similar option could be
considered for at least some portion of nonroad vehicles and equipment.
For example, a freight locomotive GHG standard could be considered on a
similar mass per ton mile basis. This would be a change from our
current mass per unit work approach to locomotive regulation, but
section 213 of the Clean Air Act does authorize the Agency to set
vehicle-based and equipment-based nonroad standards as well.
    However, we are concerned that there may be significant drawbacks
to widespread adoption of this application-specific standards-setting
approach. For the freight locomotive example given above, a gram per
ton-mile emissions standard measured over a designated track route
might be a suitable way to express a GHG standard, but such a metric
would not necessarily be appropriate for other applications. Instead
each application could require a different unit of measure tied to the
machine's mission or output-- such as grams per kilogram of cuttings
from a ``standard'' lawn for lawnmowers and grams per kilogram-meter of
load lift for forklifts. Such application-specific standards would
provide the clearest metric for GHG emission reductions. The standards
would directly reflect the intended use of the equipment and would help
drive equipment and engine designs that most effectively meet that need
while reducing overall GHG emissions. However, the diversity of tasks
performed by the hundreds of nonroad applications would lead to a
diverse array of standard work units and measurement techniques in such
a nonroad GHG program built on equipment-based standards. We request
comments on this second regulatory approach, and in particular comments
that identify specific nonroad applications that would be best served
by such a nonroad vehicle-based regulatory approach.
    A variation on the above-described approaches would be to maintain
the relative simplicity of an engine-based standard while crediting the
GHG emission reduction potential of new equipment designs. Under this
option, the new technology would be evaluated by measuring GHG
emissions from a piece of equipment that has the new technology while
performing a standard set of typical tasks. The results would then be
compared with data from the same or an identical piece of equipment,
without the new technology, performing the same tasks. This approach
could be carried out for a range of equipment models to help improve
the statistical case for the resulting reductions. The percentage
reduction in GHG emissions with and without the new equipment
technology could then be applied to the GHG emissions measured in
certification testing of engines used in the equipment in helping to
demonstrate compliance with an engine-based GHG standard. Thus if a new
technology were shown to reduce the GHG emissions of a typical piece of
equipment by 20%, that 20% reduction could be applied at certification
to the GHG emission results from a more traditional engine-based test
procedure and engine-based standard.
    In fact, a very similar approach has been adopted in EPA's recently
established locomotive program (see 73 FR 25155, May 6, 2008). In this
provision, credit is given to energy-saving measures based on the fact
that they provide proportional reductions in the criteria pollutants.
This credit takes the form of an adjustment to criteria pollutant
emissions measured under the prescribed test procedure for assessing
compliance with engine-based standards.
    A more flexible extension of this approach would be to de-link the
equipment-based GHG reduction from the compliance demonstration for the
particular engine used in the same equipment. Instead the GHG
difference would provide fungible credits for each piece of equipment
sold with the new technology, credits that then could be used in a
credit averaging and trading program. Under this concept it would be
important to collect and properly weight data over an adequate range of
equipment and engine models, tasks performed, and operating conditions,
to ensure the credits are deserved. We request comments on the option
of applying the results of equipment testing to an engine-based GHG
standard and the more general concept of generating GHG emission
credits from such an approach. We also request comment on whether such
credit-based approaches to accounting for the many promising equipment
measures are likely to obtain similar GHG reductions as the setting of
equipment based standards, and on whether some combined approach
involving both standards and credits may be appropriate.
    There are also a number of ways to reduce GHG emissions in the
nonroad sector that do not involve engine or equipment redesign.
Rather, reductions can be achieved by altering the way in which the
equipment is used. For example, intermodal shipping moving freight from
trucks and onto lower GHG rail or marine services, provides a means of
reducing these emissions for

[[Page 44466]]

freight shipments that can accommodate the logistical constraints of
intermodal shipping. Many of the operational measures with GHG-reducing
potential do involve a significant technology component, perhaps even
hardware changes, but they can also involve actions on the part of the
equipment operator or owner that go beyond simply maintaining and not
tampering with the emission controls. For example, a railroad may make
the capital and operational investment in sophisticated computer
technology to dispatch and schedule locomotive resources, using onboard
GPS-based tracking hardware. The GHG reduction benefit, though enabled
in part by the onboard hardware, is not realized without the people and
equipment assigned to the dispatch center.
    Credit for such operational measures could conceivably be part of a
nonroad GHG control program and could be calculated and assigned using
the same ``with and without'' approach to credit generation described
above for equipment-based changes. However, some important
implementation problems arise from the greater human element involved.
This human element becomes increasingly significant as the scope of
creditable measures moves further away from automatic technology-based
solutions. Assigning credits to such measures must involve good
correlation between the credits generated and the GHG reductions
achieved in real world applications. It therefore may make sense to
award these credits only after an operational measure has been
implemented and verified as effective. This might necessitate that such
credits have value for equipment or sources other than the equipment
associated with the earning of the credit, such as in a broader credit
market. This is because nonroad equipment and engines must demonstrate
compliance with EPA standards before they are put into service. They
therefore cannot benefit from credits created in the future unless
through some sort of credit borrowing mechanism.
    Once verified, however, we would expect credits reflecting these
operational reductions could be banked, averaged and traded, just as
much as credits derived from equipment- or engine-based measures.
Verifiable GHG reductions, regardless of how generated, have equal
value in addressing climate change. We also note, however, that an
effective credit program, especially one with cross-sector utility,
should account for the degree to which a credit-generating measure
would have happened anyway, or would have happened eventually, had no
EPA program existed; this is likely to be challenging. We request
comment on the appropriateness of a much broader GHG credit-based
program as described here.
    In this section, we have laid out a range of regulatory approaches
for nonroad equipment that takes us from a relatively simple extension
of our existing engine-based regulatory program through equipment based
standards and finally to a fairly wide open credit scheme that would in
concept at least have the potential to pull in all aspects of nonroad
equipment design and operation. In describing these approaches, we have
noted the increasing complexity and the greater need for new mechanisms
to ensure the emission reductions anticipated are real and verifiable.
We seek comment on the relative merits of each of these approaches but
also on the potential for each approach along the continuum to build
upon the others.
3. Marine Vessels
    Marine diesel engines range from very small engines used to propel
sailboats, or used for auxiliary power, to large propulsion engines on
ocean-going vessels. Our current marine diesel engine emission control
programs distinguish between five kinds of marine diesel engines,
defined in terms of displacement per cylinder. These five types include
small (<=37 kW), recreational, and commercial marine engines.
Commercial marine engines are divided into three categories based on
per cylinder displacement: Category 1 engines are less than 5 l/cyl,
Category 2 engines are from 5 l/cyl up to 30 l/cyl, and Category 3
engines are at or above 30 l/cyl. Category 3 engines are 2- or 4-stroke
propulsion engines that typically use residual fuel; this fuel has high
energy content but also has very high fuel sulfur levels that result in
high PM emissions. Most of the other engine types are 4-stroke and can
be used to provide propulsion or auxiliary power. These operate on
distillate fuel although some may operate on a blend of distillate and
residual fuel or even on residual fuel (for example, fuels commonly
known as DMB, DMC, RMA, and RMB).
    There are also a wide variety of vessels that use marine diesel
engines and they can be distinguished based on where they are used.
Vessels used on inland waterways and coastal routes include fishing
vessels that may be used either seasonally or throughout the year,
river and harbor tug boats, towboats, short- and long-distance ferries,
and offshore supply and crew boats. These vessels often have Category 2
or smaller engines and operate in distillate fuels. Ocean-going vessels
(OGVs) include container ships, bulk carriers, tankers, and passenger
vessels and have Category 3 propulsion engines as well as some smaller
auxiliary engines. As EPA deliberates on how to potentially address GHG
emissions from marine vessels, we will consider the significance of the
different engine, vessel, and fuel types. We invite comment on the
marine specific issues that EPA should consider; in particular, we
invite commenters to compare and contrast potential marine vessel
solutions to our earlier discussions of highway and nonroad mobile
sources and our existing marine engine criteria pollutant control programs.
a. Marine Vessel GHG Emissions
    Marine engines and vessels emitted 84.2 million metric tons of
CO₂ in 2006, or 3.9 percent of the total mobile source
CO₂ emissions. CO₂ emissions from marine vessels
are expected to increase significantly in the future, more than
doubling between 2006 and 2030. The emissions inventory from marine
vessels comes from operation in ports, inland waterways, and offshore.
The CO₂ inventory estimates presented here refer to
emissions from marine engine operation with fuel purchased in the
United States.\187\ OGVs departing U.S. ports with international
destinations take on fuel that emits 66 percent of the marine vessel
CO₂ emissions; the other 34 percent comes from smaller
commercial and recreational vessels.
---------------------------------------------------------------------------

    \187\ U.S. EPA, ``Inventory of U.S. Greenhouse Gas Emissions and
Sinks: 1990-2006,'' April 15, 2008.
---------------------------------------------------------------------------

    GHG emissions from marine vessels are dominated by CO₂
emissions which comprise approximately 94 percent of the total.
Approximately 5.5 percent of the GHG emissions from marine vessels are
due to HFC emissions, mainly from reefer vessels (vessels which carry
refrigerated containers). Methane and nitrous oxide make up less than 1
percent of the marine vessel sector GHG emissions on a CO₂
equivalent basis. Comment is requested on the contribution of marine
vessels to GHG emissions and on projections for growth in this sector.
b. Potential for GHG Reductions From Marine Vessels
    There are significant opportunities to reduce GHG emissions from
marine vessels through both traditional and innovative strategies.
These strategies include technological improvements to engine and
vessel design as well as changes in vessel operation. This

[[Page 44467]]

section provides an overview of these strategies, and a more detailed
description is available in the public docket.\188\ EPA requests
comment on the advantages and drawbacks of each of the strategies
described below, as well as on additional approaches for reducing
greenhouse gases from marine vessels.
---------------------------------------------------------------------------

    \188\ ``Potential Technologies for GHG Reductions from
Commercial Marine Vessels'', memorandum from Michael J. Samulski,
U.S. EPA, to docket xx, DATE.
---------------------------------------------------------------------------

i. Reducing GHG Emissions Through Marine Engine Changes
    GHG emissions may be reduced by increasing the efficiency of the
marine engine. As discussed earlier for heavy-duty trucks, there are a
number of improvements for CI engines that may be used to lower GHGs.
These improvements include higher compression ratios, higher injection
pressure, shorter injection periods, improved turbocharging, and
electronic fuel and air management. Much of the energy produced in a CI
engine is lost to the exhaust. Some of this energy can be reclaimed
through the use of heat recovery systems. We request comment on the
feasibility of reducing GHG emissions through better engine designs and
on additional technology which could be used to achieve GHG reductions.
    As discussed above, marine engines are already subject to exhaust
emission standards. Many of the noxious emissions emitted by internal
combustion engines may also be GHGs. These pollutants include
NO_X, methane, and black carbon soot. Additionally, some
strategies used to mitigate NO_X and PM emissions can also
indirectly impact GHGs through their impact on fuel use--for example,
use of aftertreatment rather than injection timing retard to reduce
NO_X emissions. We request comment on the GHG reductions
associated with HC+NO_X and PM emissions standards for these
engines.
    The majority of OGVs operate primarily on residual fuel, while
smaller coastal vessels operate primarily on distillate fuel. Shifting
more shipping operation away from residual fuel would reduce GHG
emissions from the ship due to the lower carbon/hydrogen ratio in
distillate fuel. Marine engines have been developed that operate on
other lower carbon fuels such as natural gas and biodiesel. Because
biodiesel is a renewable fuel, lifecycle GHG emissions are much lower
than for operation on petroleum diesel. We request comment on these and
other fuels that may be used to power marine vessels and the impact
these fuels would have on lifecycle GHG emissions.
    A number of innovative alternatives are under development for
providing power on marine vessels. These alternative power sources
include fuel cells, solar power, wind power, and even wave power. While
none of these technologies are currently able to supply the total power
demands of larger, ocean-going vessels, they may prove to be capable of
reducing GHG emissions through auxiliary power or power-assist
applications. Hybrid engine designs are used in some vessels where a
bank of engines is used to drive electric motors for power generation.
The advantage of this approach is that the same engines may be used
both for propulsion and auxiliary needs. Another advantage is that
alternative power sources could be used with a hybrid system to provide
supplemental power. We request comment on the extent to which
alternative power sources and hybrid designs may be applied to marine
vessels to reduce greenhouse gases.
ii. Reducing GHG Emissions Through Vessel Changes
    GHG emissions may be reduced by minimizing the power needed by the
vessels to perform its functions. The largest power demand is generally
for overcoming resistance as the vessel moves through the water but is
also affected by propeller efficiency and auxiliary power needs.
    Water resistance is made up of the effort to displace water and
drag due to friction on the hull. The geometry of the vessel may be
optimized in many ways to reduce water resistance. Ship designers have
used technologies such as bulbous bows and stern flaps to help reduce
water resistance from the hull of the vessel. Marine vessels typically
use surface coatings to inhibit the growth of barnacles or other sea
life that would increase drag on the hull. Innovative strategies for
reducing hull friction include coatings with textures similar to marine
animals and reducing water/hull contact by enveloping the hull with
small air bubbles released from the sides and bottom of the ship.
    Both the wetted surface area and amount of water displaced by the
hull may be reduced by lowering the weight of the vessel. This may be
accomplished through the use of lower weight materials such as aluminum
or fiberglass composites or by simply using less ballast in the ship
when not carrying cargo. Other options include ballast-free ship
designs such as constantly flowing water through a series of pipes
below the waterline or a pentamaran hull design in which the ship is
constructed with a narrow hull and four sponsons which provide
stability and eliminate the need for ballast water. We request comment
to the extent that these approaches may be used to reduce GHGs by
reducing fuel consumption from marine vessels in the future. We also
request comment on other design changes that may reduce the power
demand due to resistance on the vessel.
    In conventional propeller designs, a number of factors must be
considered including load, speed, pitch, diameter, pressure pulses, and
cavitation (formation of bubbles which may damage propeller and reduce
thrust). Proper maintenance of the propeller can minimize energy losses
due to friction. In addition, propeller coatings are available that
reduce friction on the propeller and lead to energy savings. Because of
the impact of the propeller on the operation of the vessel, a number of
innovative technologies have been developed to increase the efficiency
of the propeller. These technologies include contra-rotating
propellers, azimuth thrusters, ducted propellers, and grim vane wheels.
We request comment on the GHG reductions that may be achieved through
improvements in vessel propulsion efficiency, either through the
approaches listed here or through other approaches.
    Power is also needed to provide electricity to the ship and to
operate auxiliary equipment. Power demand may be reduced through the
use of less energy intensive lighting, improved electrical equipment,
improved reefer systems, crew education campaigns, and automated air-
conditioning systems. We request comment on the opportunities to
provide auxiliary power with reduced GHG emissions.
    In addition, GHG emissions may be released from leaks in air
conditioning or refrigeration systems. There is a large amount of
fluorinated and chlorinated hydrocarbons used in refrigeration and air-
conditioning systems on ships. We request comment on the degree to
which marine vessels emit fluorinated and chlorinated hydrocarbons to
the atmosphere, and on measures that may be taken to mitigate these
emissions.
iii. Reducing GHG Emissions Through Vessel Operational Changes
    In addition to improving the design of the engine and vessel, GHG
emissions may be reduced through operational measures. These
operational measures include reduced speeds, improved routing and fleet
planning, and shore-side power.

[[Page 44468]]

    In general, the power demand of a vessel increases with at least
the square of the speed; therefore, a 10 percent reduction in speed
could result in more than a 20 percent reduction in fuel consumption,
and therefore in GHG emissions. An increased number of vessels
operating at slower speeds may be able to transport the same amount of
cargo while producing less GHGs. In some cases, vessels operate at
higher speeds than necessary simply due to inefficiencies in route
planning or congestion at ports. Ship operators may need to speed up to
correct for these inefficiencies. GHG reductions could be achieved
through improved route planning, coordination between ports, and
weather routing systems. GHG reductions may also be achieved by using
larger vessels and through better fleet planning to minimize the time
ships operate at less than full capacity. We request comment on the
extent to which greenhouse gas emissions may be practically reduced
through vessel speed reductions and improved route and fleet planning.
    Many ports have shore-side power available for ships as an
alternative to using onboard engines at berth. To the extent that the
power sources on land are able to produce energy with lower GHG
emissions than the auxiliary engines on the vessel, shore-side power
may be an effective strategy for GHG reduction. In addition to more
traditional power generation units, shore-side power may come from
renewable fuels, nuclear power, fuel cells, windmills, hydro-power, or
geothermal power. We request comment on GHG reductions that could be
achieved through the use of shore-side power.
c. Regulatory Options for Marine Vessels
    EPA could address GHG emissions from marine vessels using
strategies from a continuum of different regulatory tools, including
emission standards, vessel design standards, and strategies that
incorporate a broader range of operational controls. These potential
regulatory strategies are briefly described below. As is the case with
other source categories, EPA is also interested in exploring the
potential applicability of flexible mechanisms such as banking and
credit trading. With regard to ocean-going vessels, we are also
exploring the potential to address GHG emissions through the
International Maritime Organization under a program that could be
adopted as a new Annex to the International Convention for the
Prevention of Pollution from Ships (MARPOL). Those efforts are also
described below. EPA requests comment on the advantages and drawbacks
of each of these regulatory approaches.
    As with trucks and land-based nonroad equipment, the first
regulatory approach we could consider entails setting GHG emission
limits for new marine diesel engines. For engines with per cylinder
displacement up to 30 liters (i.e., Category 1 and Category 2), EPA has
already adopted stringent emission limits for several air pollutants
that may be GHGs, including NO_X, methane (through
hydrocarbon standards) and black carbon soot (through PM standards).
This emission control program could be augmented by setting standards
for GHG emissions that could be met through the application of the
technologies described above (e.g., improved engine designs, hybrid
power). We request comment regarding issues that EPA should consider in
evaluating this approach and the most appropriate means to address the
issues raised. We recognize that an engine-based regulatory structure
would limit the potential GHG emission reductions compared to programs
that include vessel technologies and crediting operational
improvements. In the remainder of this section, we consider other
options that would have the potential to provide greater GHG reductions
by providing mechanisms to account for vessel and operational changes.
    A second regulatory approach to address GHG emissions from marine
vessels is to set equipment standards. As described above, these could
take the form of standards that require reduced air and/or water
resistance, improved propeller design, and auxiliary power
optimization. Equipment standards could also address various equipment
onboard vessels, such as refrigeration units. While Annex VI currently
contains standards for ozone depleting substances, this type of control
could be applied more broadly to U.S. vessels that are not subject to
the Annex VI certification requirements.
    A critical characteristic of marine vessels that must be taken into
account when considering equipment standards is that not all marine
vessels are designed alike for the same purpose. A particular hull
design change that would lower GHGs for a tugboat may not be
appropriate for a lobster vessel or an ocean-going vessel. These
differences will have an impact on how an equipment standard would be
expressed. We request comment on how to express equipment standards in
terms of an enforceable limit, and on whether it is possible to set a
general standard or if separate standards would be necessary for
discrete vessel types/sizes. We also request comment on the critical
components of a compliance program for an equipment standard, how it
can be enforced, and at what point in the vessel construction process
it should be applied.
    In addition to the above, the spectrum of regulatory approaches we
outline in section VI.C.2.c for nonroad engines and vehicles could
potentially be applied to the marine sector as well, with corresponding
GHG reductions. These would include: (1) Setting mission-based vessel
standards (such as GHG gram per ton-mile shipping standards) for at
least some marine applications where this can be reliably measured and
administered, (2) allowing vessel changes such as lower resistance hull
designs to generate credits against marine engine-based standards, (3)
granting similar credits for operational measures such as vessel speed
reductions, and (4) further allowing such credits to be used in wider
GHG credit exchange programs. We note too that the implementation
complexities for these approaches discussed in section VI.C.2.c apply
in the marine sector as well, and these complexities increase as
regulatory approaches move further along the continuum away from
engine-based standards.
    Separate from the Annex VI negotiations for more stringent
NO_X and PM standards discussed above, the United States is
working with the Marine Environment Protection Committee of the IMO to
explore appropriate ways to reduce CO₂ emissions from ships
for several years. At the most recent meeting of the Committee, in
April 2008, the Member States continued their work of assessing short-
and long-term GHG control strategies. A variety of options are under
consideration, including all of those mentioned above. The advantage of
an IMO-based program is that it could provide harmonized international
standards. This is important given the global nature of vessel traffic
and given that this traffic is expected to increase in the future.
4. Aircraft
    In this section we discuss and seek comment on the impact of
aircraft operations on GHG emissions and the potential for reductions
in GHG emissions from these operations. Aircraft emissions are
generated from aircraft used for public, private, and national defense
purposes including air carrier commercial aircraft, air taxis, general
aviation, and military aircraft.

[[Page 44469]]

Commercial aircraft include those used for scheduled service
transporting passengers, freight, or both. Air taxis fly scheduled and
for-hire service carrying passengers, freight or both, but they usually
are smaller aircraft than those operated by commercial air carriers.
General aviation includes most other aircraft (fixed and rotary wing)
used for recreational flying, business, and personal transportation
(including piston-engine aircraft fueled by aviation gasoline).
Military aircraft cover a wide range of airframe designs, uses, and
operating missions.
    As explained previously, section 231 of the CAA directs EPA to set
emission standards, test procedures, and related requirements for
aircraft, if EPA finds that the relevant emissions cause or contribute
to air pollution which may reasonably be anticipated to endanger public
health or welfare. In setting standards, EPA is to consult with FAA,
particularly regarding whether changes in standards would significantly
increase noise and adversely affect safety. CAA section 232 directs FAA
to enforce EPA's aircraft engine emission standards, and 49 U.S.C.
section 44714 directs FAA to regulate fuels used by aircraft.
Historically, EPA has worked with FAA and the International Civil
Aviation Organization (ICAO) in setting emission standards and related
requirements. Under this approach international standards have first
been adopted by ICAO, and subsequently EPA has initiated CAA
rulemakings to establish domestic standards that are at least as
stringent as ICAO's standards. In exercising EPA's own standard-setting
authority under the CAA, we would expect to continue to work with FAA
and ICAO on potential GHG emission standards, if we found that aircraft
GHG emissions cause or contribute to air pollution which may reasonably
be anticipated to endanger public health or welfare.
    Over the past 25-30 years, EPA has established aircraft emission
standards covering certain criteria pollutants or their precursors and
smoke; these standards do not currently regulate emissions of
CO₂ and other GHGs.\189\ However, provisions addressing test
procedures for engine exhaust gas emissions state that the test is
designed to measure various types of emissions, including
CO₂, and to determine mass emissions through calculations
for a simulated aircraft landing and takeoff cycle (LTO). Currently,
CO₂ emission data over the LTO cycle is collected and
reported.\190\ Emission standards apply to engines used by essentially
all commercial aircraft involved in scheduled and freight airline
activity.\191\
---------------------------------------------------------------------------

    \189\ Our existing standards include hydrocarbon emissions and
CH₄ is a hydrocarbon. If CH₄ is present in the
engine exhaust, it would be measured as part of the LTO test
procedure. There is not a separate CH₄ emission standard
for aircraft engines.
    \190\ Certification information includes fuel flow rates over
the different modes (and there are specified times in modes) of the
LTO cycle. Utilizing this information, the ICAO Engine Emissions
Databank reports kilograms of fuel used during the entire LTO cycle
(see www.caa.co.uk/default.aspx?catid=702&pagetype=90).
    \191\ Regulated aircraft engines are used on commercial aircraft
including small regional jets, single-aisle aircraft, twin-aisle
aircraft, and 747s and larger aircraft.
---------------------------------------------------------------------------

a. GHG Emissions From Aircraft Operations
    Aircraft engine emissions are composed of about 70 percent
CO₂, a little less than 30 percent water vapor, and less
than one percent each of NO_X, CO, sulfur oxides
(SO_X), non-methane volatile organic carbons (NMVOC),
particulate matter (PM), and other trace components including hazardous
air pollutants (HAPs). Little or no nitrous oxide (N₂O)
emissions occur from modern gas turbines. Methane (CH₄) may
be emitted by gas turbines during idle and by relatively older
technology engines, but recent data suggest that little or no
CH₄ is emitted by more recently designed and manufactured
engines.\192\ By mass, CO₂ and water vapor are the major
compounds emitted from aircraft operations that relate to climate change.
---------------------------------------------------------------------------

    \192\ IPCC, Aviation and the Global Atmosphere, 1999, at 
http://www.grida.no/climate/ipcc/aviation/index.htm. 
---------------------------------------------------------------------------

    In 2006, EPA estimated that among U.S. transportation sources,
aircraft emissions constituted about 12 percent of CO₂
emissions, and more broadly, about 12 percent of the combined emissions
of CO₂, CH₄, and N₂O. Together
CH₄ and N₂O aircraft emissions constituted only
about 0.1 percent of the combined CO₂, CH₄, and
N₂O emissions from U.S. transportation sources, and they
make up about one percent of the total aircraft emissions of
CO₂, CH₄, and N₂O.\193\ Aircraft
emissions were responsible for about 4 percent of CO₂
emissions from all U.S. sources, and about 3 percent of CO₂,
CH₄, and N₂O emissions collectively. While
aircraft CO₂ emissions have declined by about 6 percent
between 2000 and 2006, from 2006 to 2030, the U.S. Department of Energy
projects that the energy use of aircraft will increase by about 60
percent (excluding military aircraft operations).\194\ Commercial
aircraft make up about 83 percent of both CO₂ emissions and
the combined emissions of CO₂, CH₄, and
N₂O for U.S. domestic aircraft operations. In addition, U.S.
domestic commercial aircraft activity represents about 24 percent of
worldwide commercial aircraft CO₂ emissions. With
international aircraft departures, the total U.S. CO₂
emissions from commercial aircraft are about 35 percent of the total
global commercial aircraft CO₂ emissions.195 196
Globally, 93 percent of the fuel burn (a surrogate for CO₂)
and 92 percent of NO_X emissions from commercial aircraft
occur outside of the basic LTO cycle (i.e., operations nominally above
3,000 feet).\197\
---------------------------------------------------------------------------

    \193\ U.S. EPA, Inventory of U.S. Greenhouse Gas Emissions and
Sinks: 1990-2006, April 2008, USEPA #430-R-08-005, available
at http://www.epa.gov/climatechange/emissions/
usinventoryreport.html.
    \194\ Energy Information Administration, Annual Energy Outlook
2008, Report No.: DOE/EIA-0383 (2008), March 2008, available at
http://www.eia.doe.gov/oiaf/aeo/. These Department of Energy
projections are similar to FAA estimates (FAA, Office of Environment
and Energy, Aviation and Emission: A Primer, January 2005, at pages
10 and 23, available at http://www.faa.gov/regulations_policies/
policy_guidance/envir_policy/media/aeprimer.pdf ). The FAA
projections were based on FAA long-range activity forecasts that
assume a constant rate of emissions from aircraft engines in
conjunction with an increase in aviation operations. It does not
take into account projected improvements in aircraft, aircraft
engines, and operational efficiencies.
    \195\ FAA, System for Assessing Aviation's Global Emissions,
Version 1.5, Global Aviation Emissions Inventories for 2000 through
2004, FAA-EE-2005-02, September 2005, available at http://
www.faa.gov/about/office_org/headquarters_offices/aep/models/sage/.
    \196\ International flights are those that depart from the U.S.
and arrive in a different country.
    \197\ FAA, System for Assessing Aviation's Global Emissions,
Version 1.5, Global Aviation Emissions Inventories for 2000 through
2004, FAA-EE-2005-02, September 2005, at page 10, at Table 3,
available at http://www.faa.gov/about/office_org/headquarters_
offices/aep/models/sage/.
---------------------------------------------------------------------------

    The compounds emitted from aircraft that directly relate to climate
change are CO₂, CH₄, N₂O and, in
highly specialized applications, SF₆.\198\ Aircraft also
emit other compounds that are indirectly related to climate change such
as NO_X, water vapor, and PM. NO_X is a precursor
to cruise-altitude ozone, which is a GHG. An increase in ozone also
results in increased tropospheric hydroxyl radicals (OH) which reduces
ambient CH₄, thus potentially at least partially offsetting
the warming effect from the increase in ozone. Water vapor and PM
modify or create cloud cover, which in turn can either amplify or

[[Page 44470]]

dampen climate change.\199\ Contrails are unique to aviation
operations, and persistent contrails are of interest because they
increase cloudiness.\200\ The IPCC Fourth Assessment Report (2007) has
characterized the level of scientific understanding as low to very low
regarding the radiative forcing of contrails and aviation induced
cirrus clouds.\201\ EPA requests information on the climate change
compounds emitted by aircraft and the scientific understanding of their
climate effects, including contrail formation and persistence.
---------------------------------------------------------------------------

    \198\ SF₆ is used as an insulating medium in the
radar systems of some military reconnaissance planes. 2006 IPCC
Guidelines for National Greenhouse Gas Inventories, Volume 3,
Industrial Processes and Product Use, Chapter 8, Other Product
Manufacture and Use, Section 8.3, Use of SF₆ and HFCs in
Other Products; http://www.ipcc-nggip.iges.or.jp/public/2006gl/
index.htm. 
    \199\ IPCC, Climate Change 2007--The Physical Science Basis,
Contribution of Working Group I to the Fourth Assessment Report of
the IPCC, Chapter 2, Changes in Atmospheric Constituents and in
Radiative Forcing.
    \200\ EPA, Aircraft Contrails Factsheet, EPA430-F-00-005,
September 2000, developed in conjunction with NASA, the National
Oceanic and Atmospheric Administration (NOAA), and FAA, available at
http://www.epa.gov/otaq/aviation.htm.
    \201\ IPCC, Climate Change 2007--The Physical Science Basis,
Contribution of Working Group I to the Fourth Assessment Report of
the IPCC, Chapter 2, Changes in Atmospheric Constituents and in
Radiative Forcing, (page 202).
---------------------------------------------------------------------------

b. Potential for GHG Reductions From Aircraft Operations
    There are both technological controls and operational measures
potentially available to reduce GHG emissions from aircraft and
aircraft operations. These are discussed below.
i. Reducing GHG Emissions Through Aircraft Engine Changes
    Fuel efficiency and therefore GHG emission rates are closely linked
to jet aircraft engine type (e.g., high bypass ratio) and choice of
engine thermodynamic cycles (e.g., pressure and temperature ratios),
but modifications in the design of the engine's combustion system can
also have a substantial effect on the composition of the exhaust.\202\
Turbofan engines, with their high bypass ratios and increased
temperatures, introduced in the 1970s and 1980s reduced CO₂,
HC, and CO emissions, but in many cases put upward pressure on
NO_X emission rates. Also, a moderate increase in the engine
bypass ratio (high bypass turbofan) decreases fuel burn (and
CO₂) by enhancing propulsive efficiency and reduces noise by
decreasing exhaust velocity, but it may lead to increased engine
pressure ratio and potentially higher NO_X. \203\ There is no
single relationship between NO_X and CO₂ that
holds for all engine types. As the temperatures and pressures in the
combustors are increased to obtain better efficiency, emissions of
NO_X increase, unless there is also a change in combustor
technology.\204\ There are interrelationships among the different
emissions and noise to be considered in engine design.
---------------------------------------------------------------------------

    \202\ IPCC, Aviation and the Global Atmosphere, 1999, at
Aircraft Technology and Its Relation to Emissions, at page 221, at
section 7.1, available at http://www.grida.no/climate/ipcc/aviation/
index.htm. 
    \203\ ICCIA, Technical Design Interrelationships, Presentation
by Dan Allyn, ICCAIA Chair, at Aviation and the Environment
Conference, March 19, 2008, available at http://www.airlines.org/
government/environment/
Aviation+and+the+Environment+Conference+Presentations.htm. 
    \204\ IPCC, Aviation and the Global Atmosphere, 1999, at
Aircraft Technology and Its Relation to Emissions, at page 237, at
section 7.5.6, available at http://www.grida.no/climate/ipcc/
aviation/index.htm. 
---------------------------------------------------------------------------

    The three major jet engine manufacturers in the world are General
Electric (GE), Pratt and Whitney, and Rolls-Royce. All of these
manufacturers supply engines to both U.S. and non-U.S. aircraft
manufacturers, and their engines are installed on aircraft that operate
worldwide. These three manufacturers are now (or will be in the future)
producing more fuel efficient (lower GHG) engines with improved
NO_X. The General Electric GEnx jet engine is being developed
for the new Boeing 787, and GE's goal is to have the GEnx engine meet
NO_X levels 50 percent lower than the ICAO standards approved
in 2005.\205\ The combustor technology GE is employing is called the
Twin Annular, Pre-mixing Swirler (TAPS) combustor. In addition, the
GEnx is expected to improve specific fuel consumption by 15 percent
compared to the previous generation of engine technology (GE's CF6
engine).\206\
---------------------------------------------------------------------------

    \205\ The NO_X standards adopted at the sixth meeting
of ICAO's Committee on Aviation Environmental Protection (CAEP) in
February 2004 were approved by ICAO in 2005.
    \206\ General Electric, Press Release, Driving GE Ecomagination
with the Low-Emission GEnx Jet Engine, July 20, 2005, available at http://
www.geae.com/aboutgeae/presscenter/genx/genx_20050720.html. 
---------------------------------------------------------------------------

    Pratt and Whitney has developed the geared turbofan technology that
is expected to deliver 12 percent reduction in fuel burn while emitting
half of the NO_X emissions compared to today's engines. In
addition to an advanced gear system, the new engine design includes the
next generation technology for advanced low NO_X (TALON). The
rich-quench-lean TALON combustor utilizes advanced fuel/air atomizers
and mixers, metallic liners, and advanced cooling management to
decrease NO_X emissions during the LTO and high-altitude
cruise operations. Flight testing of the engine is expected this year,
and introduction into service is expected in 2012.\207\ Mitsubishi
Heavy Industries has chosen the engine for its regional jet.208 209
---------------------------------------------------------------------------

    \207\ Engine Yearbook, Pratt & Whitney changing the game with
geared turbofan engine, 2008, at page 96.
    \208\ Aviation, Japanese Airliner to Introduce PW's New Engine
Technology, by Chris Kjelgaard, October 9, 2007, available at 
http://www.aviation.com/technology/071009-pw-geared-turbofan-powering-
mrj.html. 
    \209\ The New York Times, A Cleaner, Leaner Jet Age Has Arrived,
by Matthew L. Wald, April 9, 2008, available at http://www.nytimes.com/
2008/04/09/technology/techspecial/09jets.html?_r=1&ex=1208491200&
en=6307ad7d1372acdf&ei=5070&emc=eta1&oref=slogin. 
---------------------------------------------------------------------------

    Rolls-Royce's Trent 1000 jet engine will power the Boeing 787s on
order for Virgin Atlantic airlines. The Trent 1000 powered 787 is
expected to improve fuel consumption by up to 15 percent compared to
the previous generation of engines (Rolls-Royce's Trent 800
engine).\210\ The technology in the Trent 1000 improves the operability
of the compressors, and enables the engine to run more efficiently at
lower speeds. This contributes to better fuel burn, especially in
descent.\211\
---------------------------------------------------------------------------

    \210\ Rolls-Royce, Trent and the environment, available at
http://www.rolls-royce.com/community/downloads/trent_env.pdf  
and the Rolls-Royce environmental report, Powering a better world:
Rolls-Royce and the environment, 2007, available at 
http://www.rolls-royce.com/community/environment/default.jsp. 
    \211\ Green Car Congress, Rolls-Royce Wins $2.6B Trent 1000
Order from Virgin Atlantic; The Two Launch Joint Environmental
Initiative, March 3, 2008, available at 
http://www.greencarcongress.com/2008/03/rolls-royce-win.html. 
---------------------------------------------------------------------------

ii. Reducing GHG Emissions Through Aircraft Changes
    Aircraft (or airframe) efficiency gains are mainly achieved through
aerodynamic drag and weight reduction.\212\ Most of the fuel used by
aircraft is needed to overcome aerodynamic drag, since they fly at very
high speeds. Reduction of aerodynamic drag can substantially improve
the fuel efficiency of aircraft thus reducing GHG emissions.
Aerodynamic drag can be decreased by installing add-on devices, such as
film surface grooves, hybrid laminar flow technology, blended winglets,
and spiroid tips, and GHG emissions can be reduced by each of these
measures from 1.6 to 6 percent.

[[Page 44471]]

Further discussion of these devices is provided below.

    \212\ U.S. Department of Transportation, Best Practices
Guidebook for Greenhouse Gas Reductions in Freight Transportation--
Final Report, Prepared for U.S. Department of Transportation via
Center for Transportation and the Environment, Prepared by H.
Christopher Frey and Po-Yao Kuo, Department of Civil, Construction,
and Environmental Engineering, North Carolina State University, October 4, 
2007, available at http://www4.ncsu.edu/~frey/Frey_Kuo_071004.pdf.
---------------------------------------------------------------------------

    --Film surface grooves: This technology is undergoing testing, and
it is an adhesive-backed film with micro-grooves placed on the outer
surfaces of the wings and the fuselage of the aircraft. Film surface
grooves are estimated to reduce total aerodynamic drag and GHG
emissions by up to 1.6 percent.
    --Hybrid laminar flow technology: Contamination on the airframe
surface, such as the accumulation of ice, insects or other debris,
degrades laminar flow. A newly developed concept, hybrid laminar flow
technology (replace turbulent air flow), integrates approaches to
maintain laminar flow. This technology can reduce fuel use by 6 to 10
percent and potentially GHG emissions by 6 percent.
    --Blended winglets: A blended winglet is a commercially available
wing-tip device that can decrease lift-induced drag. This technology is
an extension mounted at the tip of a wing. The potential decreases in
both GHG emissions and fuel use are estimated to be 2 percent.
    --Spiroid tip: A spiroid tip has been pilot tested and, similar to
blended winglets, it is intended to reduce lift-induced drag. This
technology is a spiral loop formed by joining vertical and horizontal
winglets. Greenhouse gas emissions and fuel use are both potentially
estimated to be decreased by 1.7 percent.

    Reductions in the weight of an aircraft by utilizing light-weight
materials and weight reduction of non-essential components could lead
to substantial decreases in fuel use. The weight of an airframe is
about 50 percent of an aircraft's gross weight. The use of advanced
lighter and stronger materials in the structural components of the
airframe, such as aluminum alloy, titanium alloy, and composite
materials for non-load-bearing structures, can decrease airframe
weight. These materials can reduce structural weight by 4 percent. The
potential reduction in greenhouse gas emissions and fuel use are
estimated to both be 2 percent.
iii. Reducing GHG Emissions Through Operational Changes
    Rising jet fuel prices tend to drive the aviation industry to
implement practices to decrease fuel usage and lower fuel usage reduces
GHG emissions.\213\ Indeed this has occurred in the recent past where
several airlines have reduced flights and announced plans to retire
older aircraft. However, such practices are voluntary, and there is no
assurance that such practices would continue or not be reversed in the
future. Technology developments for lighter and more aerodynamic
aircraft and more efficient engines which reduce aircraft fuel
consumption and thus GHG emissions are expected to improve in the
future. However, technology changes take time to find their way into
the fleet. Aircraft and aircraft engines operate for about 25 to 30 years.
---------------------------------------------------------------------------

    \213\ According to the Energy Information Administration, jet
fuel prices increased by about 140 percent from 2000 to 2007 (see
http://tonto.eia.doe.gov/dnav/pet/hist/rjetnyhA.htm.).
---------------------------------------------------------------------------

    Air traffic management and operational changes are governed by FAA.
The FAA, in collaboration with other agencies, is in the process of
developing the next generation air transportation system (NextGen), a
key environmental goal of which is to decrease aviation's contribution
to GHG emissions by reducing aviation system-induced congestion and
delay and accelerating air traffic management improvements and
efficiencies. As will be discussed below, measures of this type
implemented together with technology changes may be a way to reduce GHG
emissions in the near term. A few examples of the advanced systems/
procedures and operational measures are provided below.
    Reduced Vertical Separation Minimum (RSVM) allows air traffic
controllers and pilots to reduce the standard required vertical
separation from 2,000 feet to 1,000 feet for aircraft flying at
altitudes between 29,000 and 41,000 feet. This increases the number of
flight altitudes at which aircraft maximize fuel and time efficiency.
RSVM has led to about a 2 percent decrease in fuel burn.\214\
Continuous Descent Approach is a procedure that enables continuous
descent of the aircraft on a constant slope toward landing, as opposed
to a staggered or staged approach, thus allowing for a more efficient
speed requiring less fuel and reducing GHG emissions. Aircraft
auxiliary power units (APUs) are engine-driven generators that supply
electricity and pre-conditioned cabin air for use aboard the aircraft
while at the gate. Ground-based electricity sources or electrified
gates combined with preconditioned air supplies can reduce APU fuel use
and thus CO₂ emissions substantially. Single-engine taxiing,
a practice already used by some airlines, could be utilized more
broadly to reduce CO₂ emissions.\215\ Fuel consumption, and
thus GHG emissions, could be reduced by decreasing the aircraft weight
by reducing the amount of excess fuel carried. More efficient routes
and aircraft speeds would be directly beneficial to reducing full
flight GHG emissions. Operational safety must be considered in the
application of all of these measures.
---------------------------------------------------------------------------

    \214\ PARTNER, Assessment of the impact of reduced vertical
separation on aircraft-related fuel burn and emissions for the
domestic United States, PARTNER-COE-2007-002, November 2007,
available at web.mit.edu/aeroastro/partner/reports/rsvm-caep8.pdf.
    \215\ ICAO, Operational Opportunities to Minimize Fuel Use and
Reduce Emissions, Circular 303 AN/176, February 2004, available at
http://www.icao.int/icao/en/m_publications.html. 
---------------------------------------------------------------------------

    In regard to the above three sections, we request information on
potentially available technological controls (technologies for
airframes, main engines, and auxiliary power units) and operational
measures to reduce GHG emissions from aircraft operations. Since FAA
currently administers and implements air traffic management and operational 
procedures, EPA would share information on these items with FAA.
    Efforts are underway to potentially develop alternative fuels for
aircraft in the future. Industry (manufacturers, operators and
airports) and FAA established the Commercial Aviation Alternative Fuels
Initiative (CAAFI) in 2006 to explore the potential use of alternative
fuels for aircraft for energy security and possible environmental
improvements. CAAFI's goals are to have available for certification in
2008 a 50 percent Fischer-Tropsch synthetic kerosene fuel, 2010 for 100
percent synthetic fuel, and as early as 2013 for other biofuels.
However, any alternative fuel would need to be compatible with current
jet fuel for commercial aircraft to prevent the need for tank and
system flushing on re-fueling and to meet comprehensive performance and
safety specifications. In February 2008, Boeing, General Electric, and
Virgin Atlantic airlines tested a Boeing 747 that was partly powered by
a biofuel made from babassu nuts and coconut oil, a first for a
commercial aircraft.
    EPA requests information on decreasing aircraft emissions related
to climate change through the use of alternative fuels, including what
is feasible in the near-term and long-term and information regarding
safety, distribution and storage of fuels at airports, life-cycle
impacts, and cost information. Given the Agency's work to develop a
lifecycle methodology for fuels as required by the Energy Independence
and Security Act, EPA also is interested in information on the
lifecycle impacts of alternative fuels.

[[Page 44472]]

c. Options To Address GHG Emissions From the Aviation Sector
    In the preceding nonroad sections, we have described a continuum of
regulatory approaches that take us from traditional engine standards
through a range of potential approaches for vehicle standards and even
potential mechanisms to credit operational changes. For commercial
aircraft, although the reasons to consider such continuum are just as
valid, the means to accomplish these could be simpler. We see at least
two potential basic approaches for regulating aircraft GHG emissions
under the CAA, engine emission standards or a fleet average standard.
These approaches are discussed further below.
    The first approach we can consider is setting emission standards as
an extension of our current program. Under this approach we would
establish, for example, CO₂ exhaust emission standards and
related requirements for all newly and previously certified engines
applicable in some future year and later years. These standards could
potentially cover all phases of flight. Depending on timing, this first
set of standards could effectively be used to either establish baseline
values and/or to require reductions.
    As described earlier, ICAO and EPA currently require measurement
and reporting of CO₂ emissions during engine exhaust gaseous
emissions testing for the current certification cycle (although the
current absence of this information for other GHGs does not rule out a
similar approach for those GHGs).\216\ Although test procedures for
measuring CO₂ are in place already and LTO cycle
CO₂ data exists, test requirements to simulate full-flight
emissions are a significant consideration. Further work is needed to
determine how CO₂ and other GHG emissions measured over the
various modes of LTO cycle might be used to as a means to estimate or
simulate cruise or full-flight emissions. A method has been developed
by ICAO for determining NO_X for climb/cruise operations
(outside the LTO) based on LTO data, and this could be a good starting
point.217 218 For CO₂, and potentially
NO_X and other GHGs as well, the climb/cruise methods could
then be codified as test procedures, and we could then establish
emission standards for these GHGs. We request comments on the need to
develop a new test procedure for aircraft engines and the best approach
to developing such a procedure, including the viability and need for
altitude simulation tests for emissions certification.
---------------------------------------------------------------------------

    \216\ EPA's regulations at 40 CFR 87.62 require testing at each
of the following operating modes in order to determine mass emission
rates: taxi/idle, takeoff, climbout, descent and approach.
    \217\ ICAO, CAEP/7 Report, Working Paper 68, CAEP/7-WP/68,
February 2007, see http://www.icao.int. 
    \218\ ICAO has deferred work on using the NO_X climb/
cruise method for a certification procedure and standards since
future engines (potential new technologies) may behave in a
different way. There may need to be future work to consider the
aircraft mission, taking into account all phases of flight and the
performance of the whole aircraft.
---------------------------------------------------------------------------

    Furthermore, to drive the development of engine technology, we
could pursue near- and long-term GHG exhaust emission standards. Near-
term standards, which could for example apply 5 years from their
promulgation, would encourage engine manufacturers to use the best
currently available technology. Long-term standards could require more
significant reductions in emissions beyond the near-term values. In
both cases, new standards could potentially apply to both newly and
previously certified engines, but possibly at different levels and
implementation dates based on lead time considerations. Under this
approach, we would expect that no engines would be able to be produced
indefinitely if they did not meet the new standards, except possibly
based on the inclusion of an emissions averaging program for GHG as
discussed below.
    For emission standards applied to other mobile sources, EPA has
often incorporated emission averaging, banking and trading (ABT)
programs to provide manufacturers more flexibility in phasing-in and
phasing-out engine models as they seek to comply with emission
standards. In these types of programs, the average emissions within a
manufacturer's current year product line are required to meet the
applicable standard, which allows a manufacturer to produce some
engines with emission levels above the standard provided they are
offset with some below the standard. The calculation for average
compliance is usually sales, activity, and power weighted. In addition,
emissions credits and debits may be generated, banked and traded with
other engine manufacturers. We request comment on the approaches to
engine standards for reducing GHG emissions and an engine ABT program
for new GHG emission standards, including whether certain GHGs, such as
CO2, are more amenable than are other GHGs to being addressed by such a
program.
    As part of this option, we could pursue new standards and test
procedures for PM that would encompass LTO and climb/cruise operations
(ICAO and EPA currently do not have test procedures or emission
standards for PM from aircraft), if we find that aircraft PM emissions
cause or contribute to air pollution which may reasonably be
anticipated to endanger public health or welfare.\219\ Work has been
underway for several years under the auspices of the Society of
Automotive Engineers E-31 Committee, and EPA/FAA are working actively
with this committee to bring forth a draft recommended test procedure.
In addition, requirements could potentially be proposed and adopted
using the same approach as discussed above for GHGs for near- and long-
term standards and newly and already certified engines.
---------------------------------------------------------------------------

    \219\ As mentioned earlier, PM modifies or creates cloud cover,
which in turn can either amplify or dampen climate change. Aircraft
are also a source of PM emissions that contribute to local air
quality near the ground, and the public health and welfare effects
from these emissions are an important consideration.
---------------------------------------------------------------------------

    In the preceding nonroad sections, we have discussed several
approaches or variations on approaches to include vehicle and
operational controls within a GHG emission control program for nonroad
equipment. In doing so, we have not discussed direct regulation of
equipment or fleet operators. Instead, we have focused on approaches
that would credit fleet operators for improvements in operational
controls within a vehicle or engine GHG standards program. Those
approaches described in section VI.C.2 could apply to aircraft GHG
emissions as well, and we request comments on the potential to apply
those approaches to aircraft.
    As a second approach, in the case of aircraft, it may be more
practical and flexible to directly regulate airline fleet average GHG
emissions. Under such an approach we would set a declining fleet
average GHG emission standard for each airline, based on the GHG
emission characteristics of its entire fleet. This would require GHG
certification emission information for all engines in the fleet from
the aircraft engine manufacturers and information on hours flown and
average power (e.g., thrust). Airlines would have GHG emission
baselines for a given year based on the engine emission characteristics
of their fleet, and beginning in a subsequent year, airlines would be
required to reduce their emissions at some annual rate, at some rolling
average rate, or perhaps to some prescribed lower level in a future
year. This could be done as a fleet average GHG emission standard for
each airline or through a surrogate measure of GHGs such as airline
total fuel consumption, perhaps adjusted for flight activity in some
way. This could

[[Page 44473]]

cover all domestic operations and international departures of domestic
airlines. The fleet average program could potentially be implemented in
the near term since it is not as reliant on lead times for technology
change.
    Although we might develop such a declining fleet average emissions
program based on engine emissions, an operational declining fleet
average program could potentially be designed to consider the whole
range of engine, aircraft and operational GHG control opportunities
discussed above. Under this approach compliance with a declining fleet
average standard would be based not only on parameters such as engine
emission rates and activity, but could also consider efficiencies
gained by use of improved operational controls. It is important to note
that as part of this approach, a recordkeeping and reporting system
would need to be established for airlines to measure and track their
annual GHG emissions. Perhaps this could be accomplished through a
surrogate measure of GHGs such as airline total fuel consumption. Today
each airline reports its annual fuel consumption to the Department of
Transportation. We request comment on the operational fleet average GHG
emission standard concept, how it could be designed and implemented,
what are important program design considerations, and what are
potential metrics for establishing standards and determining
compliance. While we have discussed two basic concepts above, we invite
comment and information on any other approaches for regulating aircraft
GHG emissions.
d. Other Considerations
    We are aware that the European Commission (EC) has proposed a
program to cap aviation-related CO₂ emissions (cap is 100%
of sector's emissions during 2004-2006). They would by 2012 include
CO₂ emissions from all flights arriving at and departing
from European airports, including U.S.-certified aircraft, in the
European Union Emissions Trading Scheme (ETS).220, 221 If
the proposal is adopted, airlines from all countries (EU and non-EU)
will be required to submit allowances to cover emissions from all such
aircraft flights over the compliance period (e.g., 5 years). The EU has
expressed some interest in developing a program to waive this
requirement for foreign-flagged carriers (non-EU carriers) whose
nations develop ``equivalent'' measures. The petitioners discussed this
program, and we invite comments on it.
---------------------------------------------------------------------------

    \220\ Commission Proposal for a Directive of the European
Parliament and of the Council amending Directive 2003/87/EC so as to
include aviation activities in the scheme for greenhouse gas
emission allowance trading within the Community, 2006/0304 (COD),
COM(2006) 818 final, December 20, 2006, available at 
eur-lex.europa.eu/smartapi/cgi/sga_
doc?smartapi!celexplus!prod!DocNumber&1g=en&type_doc=COMfinal&an_
doc=2006&nu_doc=818.
    \221\ Proposal for a Directive of the European Parliament and of
the Council amending Directive 2003/87/EC so as to include aviation
activities in the scheme for greenhouse gas emission allowance trading 
within the Community--Political agreement, December 21, 2007 available at 
http://register.consilium.europa.eu/pdf/en/07/st16/st16855.en07.pdf.

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

    The 36th Session of ICAO's Assembly met in September 2007 to focus
on aviation emissions related to climate change, including the use of
emissions trading.\222\ In response to the EC's proposed aviation
program, the Assembly agreed to establish a high-level group through
ICAO to develop a framework of action that nations could use to address
these emissions. A report with recommendations is due to be completed
before the next Assembly Session in 2010. In addition, the Assembly
urged all countries to not apply an emissions trading system to other
nations' air carriers except on the basis of mutual consent between
those nations.\223\
---------------------------------------------------------------------------

    \222\ ICAO, Assembly--36th Session, Report of the Executive
Committee on Agenda Item 17, A36-WP/355, September 27, 2007.
    \223\ ICAO, Assembly--36th Session, Report of the Executive
Committee on Agenda Item 17, A36-WP/355, September 27, 2007.
---------------------------------------------------------------------------

    To address greenhouse gas emissions, ICAO's focus currently appears
to be on the continued development of guidance for market-based
measures.\224\ These measures include emissions trading (for
CO₂), environmental levies, and voluntary measures.
Emissions trading is when an overall target or cap is established and a
market for carbon is set. This approach allows participants to buy and
sell allowances, the price of which is established by the market.
Environmental levies include taxes and charges with the objective of
generating an economic incentive to decrease emissions. Voluntary
measures are unilateral actions by industry or in an agreement between
industry and government to decrease emissions beyond the base case.
Note, for ICAO's efforts on CO₂ emission charges, it
evaluated an aircraft efficiency parameter, and in early 2004 ICAO
decided that there was not enough information available at the time to
create a parameter that correlated properly with aircraft/engine
performance.\225\ However, it is important to note, that unlike EPA,
ICAO has not been petitioned under applicable law to determine whether
GHG emissions from aircraft may reasonably be anticipated to endanger
public health or welfare or to take any action if such a finding is
made. We invite information on reducing overall emissions that relate
to climate change from aircraft through a cap-and-trade system or other
market-based system.
---------------------------------------------------------------------------

    \224\ ICAO, ICAO Environmental Report 2007, available at 
http://www.icao.int/env/. 
    \225\ ICAO, CAEP/6 Report, February 2004, available at 
http:/www.icao.int. 
---------------------------------------------------------------------------

    Another consideration in the GHG program is the regulation of
emissions from engines commonly used in general aviation aircraft. As
indicated earlier, our current aircraft engine requirements apply to
gas turbine engines that are mainly used by commercial aircraft, except
in cases where general aviation aircraft sometimes use commercial
engines. Our requirements do not currently apply to many engines used
in business jets or to piston-engines used in aircraft that fall under
the general aviation category, although our authority under the Clean
Air Act extends to any aircraft emissions for which we make the
prerequisite finding that those emissions cause or contribute to air
pollution which may reasonably be anticipated to endanger public health
or welfare.\226\ In 2006, general aviation made up about one percent of
the CO₂ emissions from U.S. domestic transportation sources,
and about 8 percent of CO₂ emissions from U.S. domestic
aircraft operations.\227\ Regulating GHG emissions from this sector of
aviation would require the development of test procedures and emission
standards. EPA requests comment on this matter and on any elements we
should consider in potentially establishing test procedures and
emission standards for these currently unregulated engines.
---------------------------------------------------------------------------

    \226\ As specified in 40 CFR 87.10, our emission standards apply
to different classes of aircraft gas turbine engines, which have a
particular minimum rated output. The engine class and rated output
specifications correspond to certain engine operational or use
practices, but we do not, by the terms of the rule, exempt general
aviation aircraft or engines as such.
    \227\ U.S. EPA, Inventory of U.S. Greenhouse Gas Emissions and
Sinks: 1990-2006, April 2008, USEPA #430-R-08-005, available at 
http://www.epa.gov/climatechange/emissions/usinventoryreport.html.
---------------------------------------------------------------------------

5. Nonroad Sector Summary
    There are a number of potential approaches for reducing GHG
emissions from the nonroad sector within the regulatory structure of
the CAA. In considering our next steps to address GHG emissions from
this sector, we seek comment on all of the issues raised in this notice
along with recommendations

[[Page 44474]]

on the most appropriate means to address the issues.

D. Fuels

1. Recent Actions Which Reduce GHG Impacts of Transportation Fuels
    Historically under Title II of the CAA, EPA has treated vehicles,
engines and fuels as a system. The interactions between the designs of
vehicles and the fuels they use must be considered to assure optimum
emission performance at minimum cost. While EPA continues to view its
treatment of vehicles, engines and fuels as a system as appropriate, we
request comment on whether it would continue to be advantageous to take
this approach for the purpose of controlling GHG emissions from the
transportation sector. This section describes existing authorities
under the CAA for regulating the GHG emissions contribution of fuels.
In this discussion, we ask for comment on the combination of
authorities that would suit the goal of GHG emission reductions from
transportation fuel use.
    In response to CAA section 211(o) adopted as part of the Energy
Policy Act of 2005 (Energy Act of 2005), EPA issued regulations
implementing a Renewable Fuels Standard (RFS) program (72 FR 23900, May
1, 2007). These regulations were designed to ensure that 4.0 billion
gallons of renewable fuel were used in motor vehicles beginning in
2006, gradually increasing to 7.5 billion gallons in 2012. While the
primary purpose of this provision of the Energy Act of 2005 was to
reduce U.S. dependence on petroleum-based fuel and promote domestic
sources of energy, EPA analyzed the extent to which reductions in GHG
emissions would also result from the new RFS program. Therefore, for
the first time in a major rule, EPA presented estimates of the GHG
impacts of replacing petroleum-based transportation fuel with fuel made
from renewable feedstocks.
    In December 2007, EISA revised section 211(o) to set three specific
volume standards for biomass-based diesel, cellulosic biofuel, and
advanced biofuel as well as a total renewable fuel standard of 36
billion gallons annually by 2022. Certain eligible fuels must also meet
specific GHG performance thresholds based upon a lifecycle GHG
assessment. In addition to being limited to renewable fuels, EISA puts
constraints on what land sources can be used to produce the renewable
fuel feedstock, requires assessment of both primary and significant
secondary land use impacts as part of the required lifecycle GHG
emissions assessment, and has a number of other specific provisions
that affect both the design of the rule and the required analyses. EISA
requires that EPA adopt rules implementing these provisions by January
2009.
    The U.S. federal government is not alone in considering or pursuing
fuel changes which can result in reductions of GHG emissions from the
transportation sector California is moving toward adopting a low carbon
fuel standard that it anticipates will result in significant reductions
in GHG emissions through such actions as increasing the use of
renewable fuel and requiring refiners to offset any emission increases
that might result from changes in crude oil supply. Canada, the
countries of the European Union, and a number of other nations are
considering or in the process of requiring fuel changes as part of
their strategy to reduce GHG emissions from the transportation sector.
2. GHG Reductions Under CAA Section 211(o)
    The two principal CAA authorities available to EPA to regulate
fuels are sections 211(c) and 211(o). As explained in previously,
section 211(o), added by the Energy Act of 2005 and amended by EISA,
requires refiners and other obligated parties to assure that the
mandated volumes of renewable fuel are used in the transportation
sector. Section 211(o) only addresses renewable fuels; other
alternative fuels such as natural gas are not included nor are any
requirements imposed on the petroleum-based portion of our
transportation fuel pool. EPA is authorized to waive or reduce required
renewable fuel volumes specified in EISA under certain circumstances,
and is also authorized to establish required renewable fuel volumes
after the years for which volumes are specified in the Act (2012 for
biomass-based diesel and 2022 for total renewable fuel, cellulosic
biofuel and advanced biofuel). One of the factors EPA is to consider in
setting standards is the impact of production and use of renewable
fuels on climate change. In sum, EPA has limited discretion under
211(o) to improve GHG performance of fuels.
    Changes in fuel feedstock sources (for example, petroleum versus
biomass) and processing technologies can have a significant impact on
GHG emissions when assessed on a lifecycle basis. As analyzed in
support of the RFS rules, a lifecycle approach considers the GHG
emissions associated with producing a fuel and bringing it to market
and then attributes those emissions to the use of that fuel. In the
case of petroleum, the lifecycle would account for emissions resulting
from extraction of crude oil, shipping the oil to a refiner, refining
the oil into a fuel, distributing the fuel to retail markets and
finally the burning the gasoline or diesel fuel in an engine. This
assessment is sometimes referred to as a ``well-to-wheels'' assessment.
A comparable assessment for renewable fuel would include the process of
growing a feedstock such as corn, harvesting the feedstock,
transferring it to a fuel production facility, turning the feedstock
into a fuel, getting the renewable fuel to market and then assessing
its impact on vehicle emissions. EPA presented estimates of GHG impacts
as part of the assessment for the Energy Act of 2005 RFS rulemaking
that increasing renewable fuel use from approximately 4 billion gallons
to 7.5 billion gallons by 2012. However, as noted below, the
methodology used in that RFS rulemaking did not consider a number of
relevant issues.
    The 7.5 billion gallons of renewable fuel required by the Energy
Act of 2005 program represents a relatively small portion of the total
transportation fuel pool projected to be used in 2012 (add figure as %
of energy). The much larger 36 billion gallons of renewable fuel
required by EISA for 2022 would be expected to displace a much larger
portion of the petroleum-based fuel used in transportation and would
similarly be expected to have a greater impact on GHG emissions.
Comments on the RFS proposal suggested improvements to the lifecycle
assessment used in that rule. For instance, the RFS analysis did not
fully consider the impact of land use changes both domestically and
abroad that would likely result from increased demand for corn and
soybeans as feedstock for ethanol and biodiesel production in the U.S.
EPA largely agreed with these comments but was not able to incorporate
a more thorough assessment of land use impacts and other enhancements
in its lifecycle emissions modeling in time. We are undertaking such a
lifecycle assessment as we develop the proposal to implement EISA fuel
mandates. Because this updated lifecycle assessment will incorporate
more factors and the latest data, it will undoubtedly change the
estimates of GHG reductions included in the Energy Act 2005 RFS package.
    EISA recognizes the importance of distinguishing between renewable
fuels on the basis of their impact on lifecycle GHG emissions.
Nevertheless, EISA stops short of directly comparing and crediting each
fuel on the basis of its

[[Page 44475]]

estimated impact on GHG emissions. For example, while requiring a
minimum of 60% GHG emission reduction for cellulosic biomass fuel
compared to the petroleum-based fuel displaced, EISA does not
distinguish among the multiple pathways for producing cellulosic
biofuel even though these pathways might differ significantly in their
lifecycle GHG emission performance. It may be that the least costly
fuels meeting the cellulosic biofuel GHG performance threshold will be
produced which may not be the fuels with the greatest GHG benefit or
even the greatest GHG benefit when considering cost (e.g., GHG
reduction per dollar cost). The same consideration applies to other
fuels and pathways. Without further delineating fuels on the basis of
their lifecycle GHG impact, no incentive is provided for production of
particular fuels which would minimize lifecycle GHG emissions within
the EISA fuel categories.
    We request comment on the importance of distinguishing fuels beyond
the categories established in EISA and how an alternative program might
further encourage the development and use of low GHG fuels. We also
request comment on the ability (including considerations of uncertainty
and the measurement of both direct and indirect emissions associated
with the production of fuels) of lifecycle analysis to estimate the GHG
emissions of a particular fuel produced and used for transportation and
how EPA should delineate fuels (e.g., on the basis of feedstock,
production technology, etc.). EPA notes that a certain level of
aggregation in the delineation of fuels may be necessary, but that the
greater the aggregation in the categories of fuels, the fewer
incentives exist for changes in behavior that would result in
reductions of GHG emissions. EPA asks for comment on this idea as well
as how and whether methods for estimating lifecycle values for use in a
regulatory program can take into account the dynamic nature of the
market. EPA also requests comment on the relative efficacy of a
lifecycle-based regulatory approach versus a price-based (e.g., carbon
tax or cap and trade) approach to incentivize the multitude of actors
whose decisions collectively determine the GHG emissions associated
with the production, distribution and use of transportation fuels.
Finally, we request comment on the ability to determine lifecycle GHG
performance for fuels and fuel feedstocks that are produced outside the U.S.
    EISA addresses impacts of renewable fuels other than GHG impacts.
Section 203 of EISA directs that the National Academy of Sciences be
asked to consider the impacts on producers of feed grains, livestock,
and food and food products, energy producers, individuals and entities
interested in issues relating to conservation, the environment and
nutrition, users and consumers of renewable fuels, and others
potentially impacted. Section 204 directs EPA to lead a study on
environmental issues, including air and water quality, resource
conservation and the growth and use of cultivated invasive or noxious
plants. We request comment on what impacts other than GHG impacts
should be considered as part of a potential fuels GHG regulation and
how such other impacts should be reflected in any policy decisions
associated with the rule. These impacts could include the potential
impacts on food prices and supplies.
    Programs under section 211(o) are subject to further limitations.
Limited to renewable fuels, these programs do not consider other
alternative fuels such as coal-to-liquids fuel that could be part of
the transportation fuel pool and could impact the lifecycle GHG
performance of the fuel pool. Additionally, EISA's GHG performance
requirements are focused on the renewable fuels, not the petroleum-
based fuel being replaced. Under EISA, the GHG performance of renewable
fuels is tied to a 2005 baseline for petroleum fuel. No provision is
included for considering how the GHG impacts of the petroleum-based
fuel pool might change over time, either for the purpose of determining
the comparative performance for threshold compliance of renewable fuels
or for assessing the impact of the petroleum fuel itself on
transportation fuel GHG emissions. Thus, for example, there is no
opportunity under EISA to recognize and credit improvements in refinery
operation which might improve the lifecycle GHG performance of the
petroleum-based portion of the transportation fuel pool. Comments are
requested on the importance of lowering GHG emissions from
transportation fuels via the inclusion of alternative, non-renewable
fuels in a GHG regulatory program as well as the petroleum portion of
the fuel pool, thus providing opportunity to reflect improvements in
refinery practices.
    Finally while the current RFS and anticipated EISA programs will
tend to improve the GHG performance of the transportation fuel pool
compared to a business as usual case, they would not in any way cap the
GHG emissions due to the use of fuels. In fact, under both programs,
the total amount of fuel consumed and thus the total amount of GHG
emissions from those fuels can both increase. We note that other
lifecycle fuel standard programs being developed such as those in
California, Canada, and Europe, while also taking into account the GHG
emissions reduction potential from petroleum fuels, do not cap the
emissions from the total fuel pool; the GHG per gallon of
transportation fuel consumed may decrease but the total gallons
consumed are not constrained such that the total GHG emissions from
fuel may continue to grow. We request comment on setting a GHG control
program covering all transportation fuels used in the United States
which would also cap the total emissions from these transportation fuels.
    Elsewhere in this notice, comments are solicited on the potential
for regulating GHG emissions from stationary sources which could
include petroleum refineries and renewable and alternative fuel
production facilities. EPA recognizes the potential for overlapping
incentives to control emissions at fuel production facilities. We
request comment on the implications of using a lifecycle approach in
the regulation of GHG emissions from fuels which would include refinery
and other fuel production facilities while potentially also directly
regulating such stationary source emission under an additional control
program. Recognizing that the use of biomass could also be a control
option for stationary sources seeking to reduce their lifecycle GHG
impacts, EPA requests comment on the implications of using biomass for
transportation fuel in potential competition as an energy source in
stationary source applications.
3. Option for Considering GHG Fuel Regulation Under CAA Section 211(c)
    Section 211(c)(1) of the CAA has historically been the primary
authority used by EPA to regulate fuels. It provides EPA with authority
to ``control or prohibit the manufacture, introduction into commerce,
offering for sale, or sale of any fuel or fuel additive for use in a
motor vehicle, motor vehicle engine, or nonroad engine of nonroad
vehicle [(A)] if in the judgment of the Administrator any emission
product of such fuel or fuel additive causes or contributes to air
pollution or water pollution (including any degradation in the quality
of groundwater) which may reasonably be anticipated to endanger public
health or welfare.'' Section 211(c)(2) specifies that EPA must consider
all available relevant medical and scientific information, including
consideration of other technologically or economically feasible means of

[[Page 44476]]

achieving vehicle emission standards under CAA section 202 before
controlling a fuel under section 211(c)(1)(A). A prerequisite to action
under 211(c)(1) is an EPA finding that a fuel or fuel additive, or
emission product of a fuel or fuel additive, causes or contributes to
air or water pollution that may reasonably be anticipated to endanger
public health or welfare. Issues related to an endangerment finding are
discussed in section V of this advance notice.
    EPA asks for comment on whether section 211(c) could be read as
providing EPA a broader scope of authority to establish a new GHG fuel
program than section 211(o). Specifically, EPA asks for comment on
whether section 211(c)(1)(A) could allow EPA to start the program as
soon as appropriate in light of our analysis and similarly cover the
time period most appropriate; whether it could allow a program that
would encourage the use of both renewable and alternative fuels with
beneficial GHG emissions impacts and discourage those fuels with
relatively detrimental GHG impacts; and whether it could allow EPA to
establish requirements for all fuels (gasoline, diesel, renewables,
alternative and synthetic fuel, etc.) used in both highway and nonroad
vehicles and engines. EPA requests comment on whether the flexibilities
under section 211(c) allow it to consider a broad set of options for
controlling GHG emissions through fuels, including those that solely
regulate the final point of emissions such as tailpipe emissions rather
than also controlling the emissions at the fuel production facility
through a lifecycle approach.
    Typically EPA has acted through CAA section 211(c) to prohibit the
use of certain additives (e.g., lead) in fuel, to control the level of
a component of fuel to reduce harmful vehicle emissions (e.g., sulfur,
benzene), or to place a limit on tailpipe emissions of a pollutant
(e.g., the reformulated gasoline standards for volatile organic
compounds and toxics emissions performance). While multiple approaches
may be available to regulate GHG emissions under section 211(c), one
option could require refiners and importers of gasoline and diesel meet
a GHG performance standard based on reducing their lifecycle GHG
emissions of the fuel they import or produce. They would comply with
this performance standard by ensuring the use of alternative and/or
renewable fuels that have lower lifecycle GHG emissions than the
gasoline and diesel they displace and through selection of lower
petroleum sources that also reduce the lifecycle GHG performance of
petroleum-based fuel. EPA asks comment on whether section 211(c) could
authorize such an approach because it would be a control on the sale or
manufacture of a fuel that addresses the emissions of GHGs from the
transportation fuels that would be the subject the endangerment finding
discussed in section V. Comments are requested on this interpretation
of 211(c) authority.
    As pointed out above, neither the Energy Act of 2005 RFS program
nor the forthcoming program under EISA directly addresses the varying
GHG emission reduction potential of each fuel type and production
pathway. EPA asks comment on whether it could have the authority under
CAA section 211(c) to design and implement a program that includes not
only renewable fuels but other alternative fuels, considers the GHG
emissions from the petroleum portion of the fuel pool and reflects
differences in fuel production not captured by the GHG thresholds
established under EISA, including differences in technology at the fuel
production facility. We request comment on the factors EPA should
consider in developing a GHG fuel control program under section 211(c)
and how including such factors could serve to encourage the use of low
GHG-emitting practices and technology.
    We note that the RFS and the forthcoming EISA programs require
refiners and other obligated parties to meet specified volume standards
and that these programs are anticipated to continue. We request comment
on the impacts and opportunities of implementing both a GHG program
under 211(c) and volume mandates under 211(o).
    EPA seeks comment on the potential for reducing GHG emissions from
transportation fuel over and above those reductions that could be
achieved by RFS and the anticipated EISA requirements. Although EPA has
not completed its analysis of the GHG emission reductions expected
under the combined RFS and EISA programs, EPA seeks comment on how it
might structure a program that could reduce GHG emissions from
transportation fuel over and above those reductions that could be
achieved by the RFS and anticipated EISA requirements.

VII. Stationary Source Authorities and Potential Options for Regulating
Greenhouse Gases Under the Clean Air Act

    In this section, we explore three major pathways that the CAA
provides for regulating stationary sources, as well as other stationary
source authorities of the Act, and their potential applicability to
GHGs. The three pathways include NAAQS and implementation plans
(sections 107-110 and related provisions); performance standards for
new and existing stationary sources (section 111); and hazardous air
pollutant standards for stationary sources (section 112).\228\ Special
provisions for regulating solid waste incinerators are contained in
section 129.
---------------------------------------------------------------------------

    \228\ As explained in this section, the NAAQS pathway is not solely 
a stationary source regulatory authority; plans for implementating the 
NAAQS can involve regulation of stationary and mobile sources.
---------------------------------------------------------------------------

    We also review the implications of regulating GHGs under Act's
programs for preconstruction permitting of new emissions sources, with
emphasis on the PSD program under Part C of the Act. These programs
require permits and emission controls for major new sources and
modifications of existing major sources. The permitting discussion
closes by examining the implications of requiring operating permits
under Title V for major sources of GHGs. Finally, we describe four
different types of market-oriented regulatory designs that (in addition
to other forms of regulation) could be considered for programs to
reduce GHG emissions from stationary sources to the extent permissible
under the CAA: cap-and-trade, rate-based emissions trading, emissions
fees, and a hybrid approach.
    For each potential pathway of stationary source regulation, this
notice discusses the following basic questions:
     What does the section require?
     What sources would be affected if GHGs were regulated
under this authority?
     What would be the key milestones and implementation timeline?
     What are key considerations regarding use of this
authority for GHGs and how could potential issues be addressed?
     What possible implications would use of this authority for
GHGs have for other CAA programs?
    In discussing these questions, EPA considers the President's core
principles and other policy design principles enumerated in Section
III.F.1. EPA seeks comment on the advantages and disadvantages of
alternative regulatory authorities in light of those policy design
principles. EPA further invites comments on the following aspects of
each CAA stationary source authority:
     How much flexibility does the CAA section provide for
implementing its requirements? For example, can EPA set compliance
dates that reflect the global

[[Page 44477]]

and long-lived nature of GHGs and that allow time for technological
advances and new technology deployment?
     To what extent would the section allow for consideration
of the costs and economic impacts of regulating GHGs? For example,
would the section provide opportunities for sending a price signal,
such as through cap and trade programs (with or without cost
containment mechanisms) and emission fees.
     To what extent can each section account for the
international aspects of GHG emissions, atmospheric concentrations, and
emission impacts, including ways for potentially addressing
international pollutant transport and emission leakage?
     How does each section address the assessment of available
technologies, and to what extent could the section promote or require
the advancement of technology?
     To what extent does the section allow for the ability to
prioritize regulation of significant emitting sectors and sources?
     To what extent could each authority be adapted to GHG
regulation without compromising the Act's effectiveness in regulating
traditional air pollutants?
    Finally, for each regulatory authority, EPA requests comment on a
range of program-specific issues identified in the discussion below.
EPA also requests comment on whether there are specific statutory
limitations that would best be addressed by new legislation. Additional
information concerning potential CAA regulation of stationary source
GHGs may be found in the Stationary Source Technical Support Document
(Stationary Source TSD) placed in the docket for this notice.

A. National Ambient Air Quality Standards (NAAQS)

1. What Are the Requirements for Setting and Implementing NAAQS?
a. Section 108: Listing Pollutant(s) and Issuing Air Quality Criteria
    Section 108(a)(1) establishes three criteria for listing air
pollutants to be regulated through NAAQS. Specifically, section
108(a)(1) states that: EPA ``shall from time to time * * * list * * *
each air pollutant--
    (A) emissions of which, in [the Administrator's] judgment, cause or
contribute to air pollution which may reasonably be anticipated to
endanger public health or welfare;
    (B) the presence of which in the ambient air results from numerous
or diverse mobile or stationary sources; and
    (C) for which air quality criteria had not been issued before the
date of enactment of the Clean Air Amendments of 1970, but for which
[the Administrator] plans to issue air quality criteria under this
section.''
    In determining whether a pollutant meets these criteria, EPA must
consider a number of issues, including many of those discussed in
section IV above regarding an endangerment finding. As discussed there,
in the context of the ICTA petition remand, EPA is considering defining
the ``air pollution'' as the elevated current and future concentration
of six GHGs (CO₂, CH₄, N₂O, HFCs,
PFCs, and SF₆). Also in that context, EPA is considering
alternative definitions of ``air pollutant'' as the group of GHGs or
each individual GHG for purposes of the ``cause or contribute''
determination.
    In considering the potential listing of GHGs under section 108, EPA
solicits input on appropriate definitions of both the ``air pollution''
and the ``air pollutants.'' With regard to section 108, it is important
to note that EPA has clear precedents for listing related compounds as
groups rather than as individual pollutants. For example, photochemical
oxidants, oxides of nitrogen, and particulate matter all comprise
multiple compounds, but the listing under section 108 is for the group
of compounds, not the individual elements of the group. The Agency is
soliciting comment on the relevance of these precedents for GHGs. In
addition, as discussed later, there would be increased complexity in
setting NAAQS for individual GHGs than for GHGs as a group. We are
particularly interested in comments on how to apply the terms ``air
pollution'' and/or ``air pollutants'' under sections 108 and 109 in the
context of GHGs, and the implications of taking consistent or different
approaches under other Titles or sections of the Act.
    A positive endangerment finding for GHGs under section 202(a) or
other sections of the CAA could have significant and direct impacts on
EPA's consideration of the first two criteria for listing the
pollutant(s) under section 108, as explained in section IV.B.2 of this
notice. The third criterion for listing under section 108, however, may
be unrelated to the issues involved in any motor vehicle or other
endangerment finding. Moreover, this third criterion could provide EPA
discretion to decide whether to list those pollutants under section 108
for purposes of regulating them via the NAAQS.\229\ EPA requests
comment on the effect of a positive finding of endangerment for GHGs
under section 202(a) of the Act on potential listing of the
pollutant(s) under section 108.
---------------------------------------------------------------------------

    \229\ With respect to the third criterion, while there is a
decision of U.S. Court of Appeals for the Second Circuit to the
contrary, NRDC v. Train, 545 F.2d 320 (2nd Cir. 1978), EPA notes
that that decision was rendered prior to the Supreme Court's
decision in Chevron v. Natural Resources Defense Council, 467 U.S.
837 (1984). Thus, a proper and reasonable question to ask is whether
this criterion affords EPA discretion to decide whether it is appropriate 
to apply the NAAQS structure to a global air pollution problem like GHGs.
---------------------------------------------------------------------------

    Section 108 also requires that once a pollutant is listed, EPA
issue ``air quality criteria'' encompassing ``all identifiable effects
on public health or welfare,'' including interactions between the
pollutant and other types of pollutants in the atmosphere. We are
interested in commenters' views on whether and how developing air
quality criteria for GHGs would differ from developing such criteria
for other pollutants such as ozone and particular matter, given the
long-lived nature of GHGs and the breadth of impacts and other special
issues involved with global climate change. EPA also invites comment on
the extent to which it would be appropriate to use the most recent IPCC
reports, including the chapters focusing on North America, and the U.S.
government Climate Change Science Program synthesis reports as
scientific assessments that could serve as an important source or as
the primary basis for the Agency's issuance of ``air quality criteria.''
    Finally, section 108 requires EPA to issue information on air
pollution control techniques at the same time it issues air quality
criteria. This would include information on the cost of installation
and operation, energy requirements, emission reduction benefits, and
environmental impacts of these techniques. Generally, the Agency defers
this obligation until the time a standard is actually issued. As
required under Executive Order 12866, EPA must issue a Regulatory
Impact Analysis (RIA) for major rulemaking actions, and it is in this
context that EPA has previously described the scope and effectiveness
of available pollution control techniques. EPA requests comment on
whether this approach is appropriate in the case of GHGs.
b. Section 109: Standard-Setting
    Section 109 requires that the Administrator establish NAAQS for any
air pollutant for which air quality criteria are issued under section
108. Both the air quality criteria and the standards are to be reviewed
and, as appropriate, revised by the Administrator, every five years.
These decisions are to be informed by an

[[Page 44478]]

independent scientific review committee, a role which has been
fulfilled by the Clean Air Scientific Advisory Committee (CASAC) of
EPA's Science Advisory Board. The committee is charged with reviewing
both the air quality criteria for the pollutant(s) and the standards,
and recommending any revisions deemed appropriate.
    The statute specifically provides that primary NAAQS ``shall be
ambient air quality standards the attainment and maintenance of which
in the judgment of the Administrator, based on such criteria and
allowing an adequate margin of safety, are requisite to protect the
public health,'' including the health of sensitive groups. The
requirement that primary standards provide an adequate margin of safety
was intended to address uncertainties associated with inconclusive
scientific and technical information available at the time of standard
setting. It was also intended to provide a reasonable degree of
protection against hazards that research has not yet identified. Lead
Industries Association v. EPA, 647 F.2d 1130, 1154 (DC Cir 1980), cert.
denied, 449 U.S. 1042 (1980); American Petroleum Institute v. Costle,
665 F.2d 1176, 1186 (DC Cir 1981), cert. denied, 455 U.S. 1034 (1982).
The selection of any particular approach to providing an adequate
margin of safety is a policy choice left specifically to the
Administrator's judgment. Lead Industries Association v. EPA, 647 F.2d
at 1161-62.
    With regard to secondary NAAQS, the statute provides that these
standards ``specify a level of air quality the attainment and
maintenance of which in the judgment of the Administrator * * * is
requisite to protect the public welfare from any known or anticipated
adverse effects associated with the presence of such air pollutant in
the ambient air.'' Welfare effects as defined in CAA section 302(h)
include, but are not limited to, ``effects on soils, water, crops,
vegetation, manmade materials, animals, wildlife, weather, visibility
and climate, damage to and deterioration of property, and hazards to
transportation, as well as effects on economic values and on personal
comfort and well-being.''
    One of the central issues posed by potential regulation of GHGs
through the NAAQS is the nature of the health and environmental effects
to be addressed by the standards and, thus, what effects should be
addressed when considering a primary (public health) standard and what
effects should be addressed when considering a secondary (public
welfare) standard. This issue has implications for whether it would be
appropriate to establish a primary standard as well as a secondary
standard for these pollutants. As discussed above in section V, the
direct effects of GHG emissions appear to be principally or exclusively
welfare-related. GHGs are unlike other current NAAQS pollutants in that
direct exposure to GHGs at current or projected ambient levels appears
to have no known adverse effects on human health. Rather, the health
impacts associated with ambient GHG concentrations are a result of the
changes in climate at the global, regional, and local levels, which
trigger myriad ecological and meteorological changes that can adversely
affect public health (e.g., increased viability or altered geographical
range of pests or diseases; increased frequency or severity of severe
weather events including heat waves) (see section V above). The effects
on human health are thus indirect impacts resulting from these
ecological and meteorological changes, which are effects on welfare.
This raises the question of whether it is more appropriate to address
these health effects as part of our consideration of the welfare
effects of GHGs when setting a secondary NAAQS rather than a primary
NAAQS. Control of GHGs would then occur through implementation of the
secondary NAAQS rather than the primary NAAQS. EPA invites comment on
whether and how these indirect human health impacts should be addressed
in the context of setting a primary or a secondary NAAQS.
    Past experience suggests EPA may have discretion to decline to set
either a primary or a secondary standard for a pollutant if the
evidence shows that there are no relevant adverse effects at or near
current ambient concentrations, and therefore that no standard would be
requisite to protect public health or welfare. In 1985, for example,
EPA determined that it was appropriate to revoke the secondary standard
for carbon monoxide (CO) after a review of the scientific evidence
indicated that there was no evidence of known or anticipated adverse
welfare effects associated with CO at or near ambient levels. 50 FR
37484, 37494 (September 13, 1985). This decision was reaffirmed by the
Agency in the 1994 CO NAAQS review, and there remains only a primary
standard for this pollutant. EPA requests comment on whether it would
be necessary and/or appropriate for the Agency to establish both
primary and secondary NAAQS for GHGs if those pollutants were listed
under section 108.
    It is also important to consider how a NAAQS for GHGs would
interface with existing NAAQS for other pollutants, particularly oxides
of nitrogen (NO_X) and ozone (O3), as well as particulate
matter. EPA's approach in other NAAQS reviews has been to consider
climate impacts associated with any pollutant as part of the welfare
impacts evaluated for that pollutant in setting secondary standards for
the pollutant. If separate NAAQS were established for GHGs, EPA would
likely address the climate impacts of each specific GHG in the NAAQS
for GHGs, and would not need to address the climate impacts of that GHG
when addressing other NAAQS, thus avoiding duplication of effort.
    In considering the application of section 109 to GHGs and whether
it would be appropriate to regulate GHGs through the NAAQS, EPA must
evaluate a number of other standard-setting issues, as discussed below.
i. Level
    For potential GHG standards, EPA would face special challenges in
determining the level of the NAAQS. As noted above, the primary
standard must be ``requisite to protect public health with an adequate
margin of safety'' and the secondary standard ``requisite to protect
public welfare against any known or anticipated adverse effects.''
EPA's task is to establish standards that are neither more nor less
stringent than necessary for the purposes of protecting public health
or welfare. Whitman v. American Trucking Associations, 531 U.S. 457,
473. Under established legal interpretation, the costs of
implementation associated with various potential levels cannot be
factored into setting a primary or secondary standard.\230\ Any
determinations by the EPA Administrator regarding the appropriate level
(and other elements of) of a NAAQS for GHGs must based on the available
scientific evidence of adverse public health and/or public welfare
impacts, without consideration of the costs of implementation.
---------------------------------------------------------------------------

    \230\ The Supreme Court has confirmed EPA's long-standing
interpretation and ruled that ``[t]he text of Sec.  109(b),
interpreted in its statutory and historical context and with
appreciation for its importance to the CAA as a whole, unambiguously
bars cost considerations from the NAAQS-setting process.'' The court
also noted that consideration of costs occurs in the state's
formulation of the implementation plan with the aid of EPA cost
data. Whitman v. American Trucking Associations, 531 U.S. at 472.
---------------------------------------------------------------------------

    EPA expects it would be difficult to determine what levels and
other elements of NAAQS would meet these criteria for GHGs, given that
the full effects associated with elevated atmospheric concentrations of
these

[[Page 44479]]

pollutants occur over a long period of time and there are significant
uncertainties associated with the health or welfare impacts at any
given concentration. The delayed nature of effects and the complex
feedback loops associated with global climate change would require EPA
to consider both the current effects and the future effects associated
with current ambient concentrations. In making a determination of what
standard is sufficient but not more stringent than necessary, EPA would
also have to grapple with significant scientific uncertainty. As with
other NAAQS, however, the iterative nature of the 5-year review cycle
means the standards could be revised as appropriate in light of new
scientific information as it becomes available. EPA requests comment on
the scientific, technical, and policy challenges of determining
appropriate levels for NAAQS for GHG pollutants, for both primary and
secondary standards.
    As with all pollutants for which EPA establishes NAAQS, EPA would
need to evaluate what constitutes an ``adverse'' impact in the climate
context. EPA notes that the 1992 UNFCCC calls for the avoidance of
``dangerous anthropogenic interference with the climate system.''
However, it is possible that the criteria for setting a NAAQS may call
for protection against risks and effects that are less egregious than
``dangerous interference.'' Furthermore, international agreement has
not been reached on either the metric (e.g., atmospheric concentrations
of the six major directly emitted anthropogenic GHGs, radiative
forcing, global average temperature increase) or the level at which
dangerous interference would occur. EPA requests comment on whether it
would be appropriate, given the unique attributes of GHGs and the
significant contribution to total atmospheric GHG contributions from
emissions emanating outside the United States, to establish a level for
a GHG NAAQS based on an internationally agreed-upon target GHG level,
considering legal and policy factors.
    Another key question is the geographical extent of the human health
and welfare effects that should be taken into consideration in
determining what level and other elements of a standard would provide
the appropriate protection. The pollutants already subject to NAAQS are
typically local and/or regional in nature, so the standards are
designed to limit ambient concentrations of pollutants associated with
emissions typically originating in and affecting various parts of the
United States. In assessing what standard is requisite to protect
either public health or welfare, EPA has focused in the past on
analyzing and addressing the impacts in the United States. It may be
appropriate to interpret the Act as requiring standards that are
requisite for the protection of U.S. public health and welfare.
However, atmospheric concentrations of GHGs are relatively uniform
around the globe, the impacts of climate change are global in nature,
and these effects, as described in section V, may be unequally
distributed around the world. The severity of impacts in the U.S. might
differ from the severity of impacts in the rest of the world. In light
of these factors, EPA invites comment on whether it would be
appropriate to consider adverse effects on human health and welfare
occurring outside the U.S. Specifically, we invite comment on whether,
and if so, on what legal basis, it would be appropriate for EPA to
consider impacts occurring outside the U.S. when those impacts, either
in the short or long term, may reasonably be anticipated to have an
adverse effect on health or welfare in the U.S.
    As noted briefly above, if each GHG is listed as a separate
pollutant under section 108, rather than as a group or category of
pollutants, then EPA arguably would have to establish separate NAAQS
for each listed GHG. This scenario raises significant challenges for
determining which level of any particular standard is appropriate,
especially as the science of global climate change is generally focused
on the total radiative impact of the combined concentration of GHGs in
the atmosphere. Since for any one pollutant, the standard that is
requisite to protect public health with an adequate margin of safety or
public welfare from known or anticipated adverse effects is highly
dependent upon the concentration of other GHGs in the atmosphere, it
would be difficult to establish independent standards for any of the
six principal GHGs. EPA requests comments on possible approaches for
determining appropriate levels for GHG NAAQS if these pollutants are
listed individually under section 108.
ii. Indicator
    If each GHG is listed as an individual pollutant under section 108,
the atmospheric concentration of each pollutant could be measured
separately, and establishing an indicator for each pollutant would be
straightforward. However, if GHGs are listed as a group, it would be
more challenging to determine the appropriate indicator for use in
measuring ambient air quality in comparison to a GHG NAAQS. One
approach could be to measure the total atmospheric concentration of a
group of GHGs on a CO₂ equivalent basis, by assessing their
total radiative forcing (measured in W/m2).\231\ Radiative
forcing is a measure of the heating effect caused by the buildup of the
GHGs in the atmosphere. Estimating CO₂-equivalent
atmospheric concentrations, however, would not be a simple matter of
multiplying emissions times their respective GWP values. Rather, the
heating effect (radiative forcing) due to concentrations of each
individual GHG would have to be estimated to define CO₂-
equivalent concentrations. EPA invites comment on the extent to which
radiative forcing could be an effective metric for capturing the
heating effect of all GHGs in a group (or for each GHG individually).
For example, in the year 2005 global atmospheric CO₂
concentrations were 379 parts per million (ppm), but the
CO₂-equivalent concentration of all long-lived GHGs was 455
ppm. This approach would not require EPA to specify the allowable level
of any particular GHG, alone or in relation to the concentration of
other GHGs present in the atmosphere.
---------------------------------------------------------------------------

    \231\ See footnote 13 for an explanation of CO₂ equivalency.
---------------------------------------------------------------------------

    A second option would be to select one GHG as the indicator for the
larger group of pollutants intended to be controlled under the
standard. This kind of indicator approach is currently used in
regulating photochemical oxidants, for which ozone is the indicator,
and oxides of nitrogen, for which NO₂ has been used as an
indicator. There are several reasons, however, that this approach may
not be appropriate for GHGs. For example, in the instances noted above,
the indicator species is directly related to the other pollutants in
the group, either through common precursors or similar chemical
composition, and there is a basis for expecting that control of the
indicator compound will lead to the appropriate degree of control for
the other compounds in the listed pollutant. In the case of GHGs, it
would be more difficult to select one species as the indicator for the
larger group, given that the GHGs are distinct in origin, chemical
composition, and radiative forcing, and will require different control
strategies. Furthermore, this approach raises an issue regarding
whether states would have the appropriate incentive to address all
pollutants within the group. For example, there could be a focus on
controlling the single indicator species at the expense of other
species also associated with the adverse effects from

[[Page 44480]]

which the standard(s) are designed to offer protection.
    EPA seeks comment on the merits and drawbacks of these various
approaches, as well as suggestions for other possible approaches, to
defining an indicator for measuring allowable concentrations of GHGs in
the atmosphere.
c. Section 107: Area Designations
    After EPA establishes or revises a NAAQS, the CAA requires EPA and
the states to begin taking steps to ensure that the new or revised
NAAQS are met. The first step is to identify areas of the country that
do not meet the new or revised NAAQS. This applies to both the primary
and secondary NAAQS. EPA is required to identify each area of the
country as ``attainment,'' ``nonattainment,'' or ``unclassifiable.'' \232\
---------------------------------------------------------------------------

    \232\ CAA Section 107(d)(1) requires EPA to establish a deadline
for states to submit recommendations for area designations that is
no later than one year after promulgation of the new or revised
NAAQS. Section 107(d)(1) also directs states to recommend
appropriate area boundaries. A nonattainment area must consist of
that area that does not meet the new or revised NAAQS, and the area
that contributes to ambient air quality in a nearby area that does
not meet the new or revised NAAQS. Thus, a key factor in setting
boundaries for nonattainment areas is determining the geographic
extent of nearby source areas contributing to the nonattainment
problem. EPA then reviews the states' recommendations, collects and
assesses additional information as appropriate, and issues final
designations no later than 2 years following the date EPA
promulgated the new or revised NAAQS. EPA may take one additional
year (meaning final designations can be up to 3 years after
promulgation of new or revised NAAQS) if the Administrator has
insufficient information to promulgate the designations. Whether or
not a state or a Tribe provides a recommendation, EPA must
promulgate the designation that it deems appropriate.
---------------------------------------------------------------------------

    For a GHG NAAQS, the designations given to areas would depend on
the level of the NAAQS and the availability of ambient data to make
informed decisions for each area. For GHGs, in contrast to current
NAAQS pollutants, it would likely make sense to conduct the air quality
assessment at the national scale rather than at a more localized scale.
All of the potential indicators discussed above for measuring ambient
concentrations of GHGs for purposes of a NAAQS involve globally
averaged metrics. Therefore, the ambient concentrations measured across
all locations within the U.S. for purposes of comparison to the level
of the standard would not vary, and all areas of the country would have
the same designation--that is, the entire U.S. would be designated
either attainment or non-attainment, depending on the level of the
NAAQS compared to observed GHG ambient concentrations.
    If, in making decisions about the appropriate level of the GHG
NAAQS, EPA were to determine that current ambient concentrations are
not sufficient to cause known or anticipated adverse impacts on human
health or welfare now or in the future, then it is possible that the
NAAQS would be set at some level higher than current ambient
concentrations. In that case, the entire country would likely be
designated nonattainment. If, on the other hand, EPA were to set the
NAAQS at a level above current ambient concentrations, the entire
country would likely be designated attainment.
d. Section 110: State and Federal Implementation Plans
i. State Implementation Plans
    The CAA assigns important roles to EPA, states, and tribal
governments in implementing NAAQS and in ensuring visibility protection
in Class I areas. States have the primary responsibility for developing
and implementing state implementation plans (SIPs). A SIP is the
compilation of authorities, regulations, control programs, and other
measures that a state uses to carry out its responsibilities under the
CAA to attain, maintain, and enforce the NAAQS and visibility
protection goals, and to prevent significant deterioration of air
quality in areas meeting the standard. Additional specifics on SIP
requirements are contained in other parts of the CAA.
    EPA assists states and tribes in their efforts to clean the air by
promulgating national emissions standards for mobile sources and
selected categories of stationary sources. Also, EPA assists the states
in developing their plans by providing technical tools, assistance, and
guidance, including information on potentially applicable emissions
control measures.
    Historically, the pollutants addressed by the SIP program have been
local and regional pollutants rather than globally mixed pollutants
like GHGs. The SIP development process, because it relies in large part
on individual states, is not designed to result in a uniform national
program of emissions controls.
(1) Generic Requirements for All SIPs
    This section discusses the specific CAA requirements states must
address when implementing any new or revised NAAQS.\233\
---------------------------------------------------------------------------

    \233\ The visibility protection program required by CAA sections
169A and 169B, and as implemented through state compliance with
EPA's 1999 Regional Haze Rule, will only be raised again here in
this section of the ANPR in the context of a framework for
implementing a secondary GHG NAAQS.
---------------------------------------------------------------------------

    Under section 110(a)(1) and (2) of the CAA, all states are required
to submit plans to provide for the implementation, maintenance, and
enforcement of any new or revised NAAQS. Section 110(a)(1) and (2)
require states to address basic program elements, including
requirements for emissions inventories, monitoring, and modeling, among
other things. These requirements apply to all areas of the state regardless 
of whether those areas are designated nonattainment for the NAAQS.
    In general, every state is required to submit to EPA within 3 years
of the promulgation of any new or revised NAAQS a SIP demonstrating
that these basic program elements are properly addressed. Subsections
(A) through (M) of section 110(a)(2) enumerate the elements that a
state's program must contain. See the Stationary Source TSD for this list.
    Other statutory requirements for state implementation plans vary
depending on whether an area is in nonattainment or attainment. There
are four specific scenarios that could hypothetically apply, depending
on whether a primary or a secondary standard, or both, are established,
and on the level(s) set for those standards. Because we are proposing
no scientific determinations in this notice, our discussion of NAAQS
implementation addresses all four of these scenarios.
(2) Scenario 1: Primary GHG Standard With Country in Nonattainment
    If the entire country were designated nonattainment for a primary
GHG NAAQS, each state would be required to develop and submit a SIP
that provided for attainment and met the other specific requirements of
Part D of Title I of the Act by the specified deadline.
    Requirements for the general contents of a nonattainment area plan
are set forth in section 172 of the CAA. Section 172(c) specifies that
SIPs must, among other things: \234\
---------------------------------------------------------------------------

    \234\ For additional information about nonattainment area
planning requirements, please see the Technical Support Document.
---------------------------------------------------------------------------

     Include all Reasonably Available Control Measures (RACM)
(including, at a minimum, emissions reductions obtained through
adoption of Reasonably Available Control Technology (RACT)) and provide
for attainment of the NAAQS;
     Provide for Reasonable Further Progress (RFP), which means
reasonable interim progress toward attainment;
     Include an emissions inventory;
     Require permits for the construction and operation of
major new or modified stationary sources, known as

[[Page 44481]]

``nonattainment new source review'' (see also section 173 of the Act
and section VII.E. of this notice);
     Contain contingency measures that are to be implemented in
the event the air quality standard is not met by the area's attainment
deadline; and
     Meet the applicable provisions of section 110(a)(2) of the
CAA related to the general implementation of a new or revised NAAQS.
    In addition, all nonattainment areas must meet requirements of
section 176(c) known as ``general conformity'' and ``transportation
conformity.'' \235\ In brief, general conformity requires the federal
government only to provide financial assistance, issue a permit or
approve an activity that conforms to an approved SIP for a NAAQS.
Transportation conformity requires metropolitan planning organizations
and the U.S. Department of Transportation only to approve or fund
transportation plans, programs and projects that conform to an approved
SIP for a NAAQS. For the scenario of the country in nonattainment with
a GHG NAAQS, these requirements would apply nationwide one year after
the effective date of EPA's nonattainment designations.
---------------------------------------------------------------------------

    \235\ These requirements also apply to ``maintenance areas''--
former nonattainment areas that have met the standard and been
redesignated according to a formal EPA determination.
---------------------------------------------------------------------------

    For nonattainment areas, SIPs must provide for attainment of the
primary NAAQS as expeditiously as practicable, but no later than 5
years from the effective date of the nonattainment designation for the
area--or no later than 10 years if EPA finds additional time is needed
considering the severity of nonattainment and the availability and
feasibility of pollution control measures.
    At the outset, it would appear to be an inescapable conclusion that
the maximum 10-year horizon for attaining the primary NAAQS would be
ill-suited to GHGs. The long atmospheric lifetime of the six major
emitted GHGs means that atmospheric concentrations will not quickly
respond to emissions reduction measures (with the possible exception of
methane, which has an atmospheric lifetime of approximately a decade).
In addition, in the absence of substantial cuts in worldwide emissions,
worldwide concentrations of GHGs would continue to increase despite any
U.S. emission control efforts. Thus, despite active control efforts to
meet a NAAQS, the entire U.S. would remain in nonattainment for an
unknown number of years. If States were unable to develop plans
demonstrating attainment by the required date, the result would be
long-term application of sanctions, nationwide (e.g., more stringent
offset requirements and restrictions on highway funding), as well as
restrictions on approvals of transportation projects and programs
related to transportation conformity. EPA is currently evaluating the
extent to which section 179B might provide relief to states in this
circumstance. As further explained below, section 179B is a waiver
provision providing for SIP approval under certain circumstances when
international emissions affect a U.S. nonattainment area.
    In addition to submitting plans providing for attainment within the
state, each state would be required to submit, within 3 years of NAAQS
promulgation, a plan under section 110(a)(2)(D) prohibiting emissions
that would significantly contribute to nonattainment in another state.
EPA requests comments on what approaches could be utilized for purposes
of addressing this requirement as well as the general matter of
controlling GHGs to meet a NAAQS.
    Impact of section 179B on nonattainment requirements: States may
use section 179B of the CAA to acknowledge the impact of emissions from
international sources that may contribute to violations of a NAAQS.
Section 179B provides that EPA shall approve a SIP for a nonattainment
area if: (1) The SIP meets all applicable requirements of the CAA; and
(2) the submitting state can satisfactorily demonstrate that ``but for
emissions emanating from outside of the United States,'' the area would
attain and maintain the applicable NAAQS. EPA has historically
evaluated these ``but for'' demonstrations on a case-by-case basis,
based on the individual circumstances and the data provided by the
submitting state. These data might include ambient air quality
monitoring data, modeling scenarios, emissions inventory data, and
meteorological or satellite data. In the case of GHGs, however, where
global emissions impact all areas within the United States, the federal
government may be best suited for establishing whether a ``but for''
demonstration can be made for the entire country.
    If a ``but for'' conclusion is affirmed, section 179B would allow
EPA to approve a SIP that did not demonstrate attainment or maintenance
of the relevant NAAQS. Section 179B does not provide authority to
exclude monitoring data influenced by international transport from
regulatory determinations related to an area's status as an attainment
or nonattainment area. Thus, even if EPA approves a section 179B ``but
for'' demonstration for an area, the area would continue to be
designated as nonattainment and subject to certain applicable
nonattainment area requirements, including nonattainment new source
review, conformity, and other measures prescribed for nonattainment
areas by the CAA. EPA requests comment on the practical effect of
application of section 179B on the global problem of GHG emissions and
on the potential for controls based on the attainment plan requirement
and other requirements directly related to the attainment requirement,
including the reasonable further progress requirement and the RACM
requirement.\236\
---------------------------------------------------------------------------

    \236\ EPA has interpreted RACM as emissions reducing measures
that are technically and economically feasible, and considered
collectively would advance the nonattainment area's attainment date
by at least one year. RACT has been interpreted in two different
ways, depending on the applicable statutory requirements. In the
case of ozone, RACT consists of measures that are technically and
economically feasible, without regard to whether the measures would
result in earlier attainment. In recent rules on PM2.5, EPA
interpreted RACT for PM2.5 as essentially the same as RACM, with
RACT referring to the stationary source component of RACM, which
applies to all types of sources.
---------------------------------------------------------------------------

(3) Scenario 2: Secondary Standard With Country in Nonattainment (No
Primary Standard)
    As noted above in the NAAQS standard-setting discussion, depending
on the nature and bases of any endangerment finding under section 108,
EPA may be able to consider setting only a secondary NAAQS for GHGs and
not also a primary NAAQS.
    In general, the same nonattainment requirements that apply to SIPs
for a primary standard apply for a secondary standard, including
nonattainment new source review and the other programs listed under the
Scenario 1 subsection above.
    A notable difference in nonattainment requirements for primary and
secondary standards is the time allowed for attainment. Under a
secondary standard, state plans must achieve attainment as
expeditiously as practicable, but there is no statutory maximum date
for attainment. The general requirement to attain as expeditiously as
practicable includes consideration of required controls, including
``reasonably available control measures.'' These requirements do allow
for consideration of cost. What would constitute ``as expeditiously as
practicable'' would be determined based on the entire set of facts and
circumstances at issue. EPA requests comment on how to interpret

[[Page 44482]]

the requirement that state plans demonstrate that attainment will be
achieved ``as expeditiously as practicable'' in the context of a
secondary NAAQS for GHGs.
    Potential implementation approach based on regional haze model: For
a secondary GHG NAAQS with no prescribed attainment date, EPA requests
comment on the concept of implementing a GHG secondary NAAQS standard
in a way roughly analogous to an approach used in the long-term
regional visibility program, known as the regional haze program. This
program is based on a goal of achieving natural visibility conditions
in our nation's parks and wilderness areas (Class I areas) by 2064. The
program requires states to develop reasonable progress goals every 10
years and implement emissions control programs to achieve those goals,
ultimately achieving the 2064 natural condition goal in each Class I
area. At the midpoint of every 10-year period, states must assess the
progress being made and take corrective action if necessary to maintain
reasonable progress toward the 10-year progress milestone.
    The regional haze program's model for goal planning, control
strategy development, and control strategy implementation could offer a
possible framework for achieving a GHG secondary NAAQS. This framework
potentially could be designed to address the RACM, RACT and Reasonable
Further Progress requirements, as well as the attainment planning
requirement. This framework may also provide a mechanism for
implementing a nationwide GHG emissions cap and trade program adopted
and implemented through state plans. However, EPA recognizes that the
global nature of GHGs and their persistence in the atmosphere make an
approach based on ``reasonable'' progress more difficult to implement
than in the case of regional haze. For example, despite domestic
emissions reductions, it might not be possible to discern improvement
in atmospheric concentrations of GHGs due to their relatively long
atmospheric lifetimes or to growth in emissions from other countries
which could eclipse reductions made in the U.S. We note that using this
framework would not provide relief from any of the applicable
nonattainment area requirements of the Act. EPA requests comment on
whether, and if so how, the regional haze approach could be adapted for
use in the GHG context.
(4) Scenarios 3 and 4: Primary and/or Secondary Standard With Country
in Attainment
    If a primary or secondary GHG NAAQS were set at a level higher than
ambient GHG levels at the time of designations, then the country would
be in attainment. (See preceding section on NAAQS standard-setting for
discussion of this issue.) In this case, a much shorter list of
requirements would apply than if the country were in nonattainment.
    SIPs would be required to include PSD programs for GHGs, which
would require preconstruction permitting of new major sources and
significant modifications to existing major sources. (See section VII.D
on PSD.)
    EPA has identified two other requirements that potentially could
apply, both of which could provide authority for a nationwide cap-and-
trade program implemented at the state level. First, section 110(a)(1)
requires states to submit a SIP providing for ``implementation,
maintenance, and enforcement'' of primary and secondary NAAQS. Under
the scenario of a GHG NAAQS with the country in attainment, where
states may need more than PSD/NSR to maintain attainment, EPA could
consider using this provision to require SIPs to provide for
maintenance of air quality consistent with the GHG standard. This
requirement could be implemented through a nationwide cap-and-trade
program designed at the federal level and adopted by individual states
in their SIPs, a program similar but broader in scope than existing
programs such as the more limited NO_X SIP Call regional cap-
and-trade system for EGUs and selected industrial source categories. If
a state failed to submit an adequate maintenance SIP, EPA would be
required to develop and implement a federal implementation plan for
that state. EPA could design the FIP to enable the state to participate
in a nationwide cap-and-trade system.
    Second, section 110(a)(2)(D) requires SIPs to prohibit emissions
that would interfere with maintenance of the standard by other states.
Because GHGs are globally well-mixed, it may be that GHGs emitted from
any state could be found to interfere with maintenance of a GHG NAAQS
in every other state. In the past, EPA has issued rules that have
resulted in states adopting interstate cap-and-trade programs (e.g.,
the Clean Air Interstate Rule) implemented through SIPs to address the
requirements of this provision. In the case of GHGs, this authority
could potentially support a nationwide cap-and-trade program for GHGs,
adopted through SIPs. If a state failed to submit its section
110(a)(2)(D) SIP, EPA would be required to develop and implement a FIP
for that state. EPA could design the FIP to enable the state to
participate voluntarily in a nationwide cap-and-trade system. We
request comment on the suitability of adopting either of these
approaches under section 110(a).
ii. Additional CAA Provisions Affecting SIP Obligations and FIPs
(1) Section 179(a)
    The CAA requires states to submit SIPs to EPA for review, and EPA
must approve or disapprove them based on whether the state plan or
component meets the Act's requirements. An EPA finding that a state has
failed to submit a nonattainment plan or plan component, or an EPA
disapproval of such a plan because it does not meet the requirements of
the Act, would start a ``sanctions clock'' under section 179(a). This
means that sanctions would apply in the state if the deficiencies are
not corrected within prescribed deadlines. These sanctions include
additional requirements for major new sources (18 months after the
finding of failure) and restrictions on federal highway funds (6 months
after the offset sanction).\237\ EPA must promulgate a FIP for the
deficient component of the SIP if the state's plan component is not
approved within 2 years of EPA's finding or disapproval action. In the
case of GHGs, it is possible that EPA could design the FIP to enable
the state to participate in a nationwide cap-and-trade system.
---------------------------------------------------------------------------

    \237\ 40 CFR 52.31.
---------------------------------------------------------------------------

(2) Section 115
    CAA section 115 creates a mechanism through which EPA can require
states to amend their SIPs to address international transport issues.
It is designed to protect public health and welfare in another country
from air pollution emitted in the U.S. provided the U.S. is given
essentially reciprocal rights with respect to prevention and control of
air pollution originating in the other country. The Administrator could
exercise his authority under this provision if EPA were to promulgate a
NAAQS for GHG.
    To act under section 115, the Administrator would need to make a
finding that, based on information from any duly constituted
international agency, he has reason to believe that air pollutants
(GHGs) emitted in the U.S. causes or contributes to air pollution which
may reasonably be anticipated to endanger public health or welfare in a
foreign country. Upon making such a finding, the Administrator would give

[[Page 44483]]

formal notification to the Governor of the state (or in this case
potentially all of the states) where GHGs originate. A finding under
this section has the same regulatory consequences as a finding that the
state's existing SIP is inadequate to attain the NAAQS or otherwise
meet the requirements of the Act. This notification would require the
notified states to modify their SIPs to prevent or eliminate the
endangerment.
    Addressing GHGs under this authority could allow some flexibility
in program design, subject to limitations of the SIP development
process. Section 115 could not be used to require states to incorporate
into their SIPs measures unrelated to attainment or maintenance of a
NAAQS. A factor to consider is that this section of the Act only
applies where countries that suffer possible endangerment give
reciprocal rights to the U.S. However, reciprocity with one or more
affected countries may be sufficient to trigger section 115. We request
comment on the efficacy of using section 115 as a mechanism to
facilitate more effective regulation of GHGs through a NAAQS.
2. What Sources Would Be Affected?
    Sections 108 and 109 impose no controls directly on sources, but
instead establish the air quality benchmarks that control requirements
would be designed to meet. The precise nature of these controls would
be determined through federal and state programs, as established via
SIPs and, for states failing to submit an approvable plan, FIPs.
Considering that GHGs are emitted by a wide array of sources, it is
likely that NAAQS implementation would result in controls on numerous
stationary and mobile sources through sections 110 and 172.
    The federal government could have less flexibility under the NAAQS
approach to target control efforts toward particular groups of existing
stationary sources. Under the traditional SIP approach, emissions
controls on specific source categories would flow from independent
state-level decisions, and could result in a patchwork of regulations
requiring different types and levels of controls in different states.
However, the SIP approach could also be adapted for use in a more
coordinated strategy. As mentioned above, EPA has in the past issued
rules that have resulted in states adopting limited interstate cap-and-
trade programs (e.g., NO_X SIP Call and the Clean Air
Interstate Rule) implemented through state SIPs. Furthermore, the
federal government would also have flexibility to design a national
control program in the event that states did not adopt the required
programs and EPA were required to promulgate a FIP.
    EPA requests comment on whether and how the different
implementation provisions within the NAAQS program could be adapted to
be most suitable for application to control GHGs.
3. What Would Be the Key Milestones and Implementation Timeline?
    The key milestones that would apply if EPA were to regulate GHGs as
a NAAQS pollutant include: listing the pollutant(s); issuing air
quality criteria; issuing information on air pollution control
techniques; proposing primary and secondary NAAQS for the pollutants;
issuing final standards; designating areas; development of SIPs/FIPs;
and application of control measures.
    EPA has discretion with regard to the date of listing of a
pollutant under section 108. The statute does not prescribe any
specific deadline for listing, instead stating that EPA ``shall from
time to time * * * list * * * each air pollutant'' that EPA judges
meets the three criteria discussed above. This could provide the Agency
some latitude in determining the precise timing of any listing.
    Once a pollutant is listed, the CAA specifies a very ambitious
timeline for issuing the initial NAAQS for the pollutant. Section 108
allows 12 months between date of listing and issuance of air quality
criteria for the pollutant(s). Since these criteria are intended to
encompass ``all identifiable effects on public health or welfare,'' it
would be difficult to meet this timeline in the case of GHGs. In 1970,
when the NAAQS program was first established under the CAA, air quality
criteria either were in development or had already been issued for a
variety of pollutants, and the process involved consideration of a much
smaller body of science than is now available. Therefore, the 12-month
period allotted for the initial issuance of air quality criteria
appeared reasonable.\238\ However, based on recent NAAQS reviews for
ozone, particulate matter, lead, and other pollutants, it now generally
takes several years for the Agency to complete the thorough scientific
assessment necessary to issue air quality criteria.
---------------------------------------------------------------------------

    \238\ For each air pollutant for which air quality criteria had
already been issued prior to enactment of the Clean Air Act
Amendments of 1970, section 109(a)(1) actually required EPA to issue
proposed NAAQS within 30 days of enactment and to finalize those
standards within 90 days of publication of the proposal. This included 
carbon monoxide, ozone, particulate matter, hydrocarbons, and sulfur oxides.
---------------------------------------------------------------------------

    Given the complexity of global climate change science, and the vast
amount of research that would be relevant to the Agency's scientific
assessment, EPA anticipates this task would be particularly time
consuming in the case of GHGs, though relying on synthesis reports such
as the Intergovernmental Panel on Climate Change's Fourth Assessment
Report and various reports of the U.S. Climate Change Science Program
could help expedite the process. The challenge of completing a thorough
scientific assessment for GHGs could result in a significant delay in
listing the pollutant(s) under section 108, since EPA would likely
choose to list GHGs only when the scientific assessment had progressed
sufficiently to enable the Agency to meet the statutory requirement to
issue ``air quality criteria'' within one year of listing, and to meet
the tight rulemaking timeframe, discussed below. To the extent that EPA
addresses GHGs through this CAA mechanism, EPA requests comments on the
issuance of ``air quality criteria'' following listing, as well as the
adequacy of the available scientific literature.
    Under section 109, EPA must propose NAAQS for any newly listed
pollutant at the same time it issues air quality criteria under section
108, and must finalize those standards within 90 days after proposal.
Thus, from the date of listing a pollutant(s) under section 108, the
Agency has only 12 months to propose standards, and only 3 additional
months to issue final NAAQS for the pollutant(s). This tight timeframe
would be particularly challenging in the case of GHGs, for which review
and synthesis of an enormous body of literature would be required
before a proposal could be issued. Furthermore, it is important to note
that while subsequent NAAQS reviews of existing standards are required
on a revolving 5-year cycle, EPA has found it challenging to meet even
this extended schedule, which generally allows 9-12 months between
issuance of the air quality criteria and proposal and an additional 6
months or more for issuance of final standards.
    Once a new standard has been established, the CAA allows EPA to
establish a deadline for states to submit designation recommendations
that is no later than one year after promulgation of the new or revised
NAAQS. EPA then reviews the states' recommendations, collects and
assesses additional information as appropriate, and issues final
designations no later than 2 years following the date EPA promulgated
the new or revised NAAQS. EPA may take up to one additional year if the
Administrator has insufficient

[[Page 44484]]

information to promulgate the designations, which could push the date
of final designations out to three years after promulgation of a new
GHG NAAQS.
    The timeline for SIP submittal and implementation of control
requirements depends an area's designation status (attainment,
nonattainment, unclassifiable) and whether there is only a secondary
NAAQS, or both a primary and a secondary standard. These various
scenarios are described above. As a first step, regardless of
attainment status of level of the standard, states must submit
infrastructure SIPs to EPA within 3 years of the promulgation of any
new or revised NAAQS. These SIPs demonstrate that certain basic program
elements (including emissions inventories, monitoring, and modeling)
are properly addressed. Areas that are designated attainment would face
a much shorter list of requirements, which are discussed above in the
context of, Scenarios 3 and 4.
    For areas designated nonattainment with a primary standard, states
must submit nonattainment SIPs no more than 3 years after the effective
date of designations, and must reach attainment no later than 5 years
after the effective date designations. EPA can extend the attainment
deadline by up to an additional 5 years--i.e., to no later than 10
years after the effective date of designations, if EPA finds additional
time is needed considering the severity of nonattainment and the
availability and feasibility of pollution control measures.
    As noted above, the maximum 10-year horizon for attaining the
primary NAAQS is ill-suited to pollutants such as GHGs with long
atmospheric residence times. It is probable that, despite active
control efforts, the entire U.S. would remain in nonattainment for an
indefinite number of years if the level of a NAAQS were set at or below
current atmospheric concentrations; whether attainment would ever be
reached would depend on the timing and stringency of GHG control
measures implemented on a global basis.
    For areas designated nonattainment with a secondary standard only,
the attainment schedule could be significantly longer. The CAA requires
that state plans under a secondary standard must provide for reaching
attainment as expeditiously as practicable, but there is no statutory
maximum date for attainment (e.g., up to 10 years). EPA requests
comment on the suitability of adapting this approach for use in the GHG
context, and specifically, on the schedule that could reasonably be
considered as ``expeditious as practicable.'' We also request comment
on how global emissions should be taken into consideration in this context.
    EPA requests comment on whether the avenues discussed in this
notice, or alternative approaches, could facilitate schedule
adjustments that would better enable use of the NAAQS approach for
regulating GHGs.
4. What Are Key Considerations Regarding Use of This Authority for GHGs?
a. Possible Cost and Emissions Impacts
    Listing GHGs as pollutants under section 108 and setting NAAQS
under section 109 would have no direct cost or emissions impacts.
However, these actions would trigger further federal actions, including
designations under section 107, and state or federal actions through
SIPs or FIPs developed under section 110 and other provisions in title
I of the CAA. Thus, the listing of GHGs as NAAQS pollutants would
likely lead to the adoption of a substantial control program affecting
sources across the nation.
    Because establishing NAAQS for a pollutant sets in motion a broad
and prescriptive implementation process that could affect a wide array
of stationary and mobile sources, it is likely to entail substantial
costs. The magnitude of these costs would depend, in part, on the
relative reliance on technologies which are not yet suitable for
commercial application or which have not yet been developed. Though
this problem affects other pollutants, it is more acute in the case of
GHGs. The timing and nature of controls instituted, and thus the costs,
would depend to a significant extent on an area's designation status
and whether EPA set only a secondary NAAQS (with a longer
implementation time horizon), or a primary standard as well (with a
more rapid and rigid compliance schedule, allowing less time for
technological advances and efficiency improvements). The standard set
and the nature of GHGs could also determine whether it is feasible to
attain a NAAQS in the near-term, or how costly attainment could be over
a longer term.
    One important aspect of the NAAQS approach is that the standards
themselves (both primary and secondary) are established without
consideration of these costs. EPA requests comment on the suitability
of establishing regulations to limit atmospheric concentrations of GHGs
through a statutory mechanism that prohibits consideration of the costs
such regulations might entail. EPA also requests comment on the extent
to which various implementation mechanisms in Title I are available for
addressing such costs.
    As mentioned above, CAA section 108 requires EPA to issue
information on air pollution control techniques at the same time it
issues air quality criteria. This would include information on the cost
of installation and operation, energy requirements, emission reduction
benefits, and environmental impacts. Generally, the Agency fulfills
this obligation at the time a standard is issued; as required under
Executive Order 12866, EPA must issue an RIA for major rulemaking
actions. A NAAQS RIA provides an illustrative analysis of control
options available to reduce emissions and ambient concentrations of the
regulated pollutant(s); evaluates the costs of these controls; and
estimates the human health and environmental benefits likely to accrue
from the improved air quality resulting from the standards.
    As required by EO 12866 and guidance from OMB, the analysis
generally compares control options and estimated costs and benefits of
multiple, specific standard options under consideration. While EPA
recognizes the cost estimates for future GHG control technologies would
potentially place more reliance on yet-to-be-developed options, the
precedent exists for consideration of future, unknown controls. EPA
requests comment on whether there are important distinctions between
GHGs and previously regulated criteria pollutants that would make it
appropriate in the case of a new NAAQS for GHG(s) to issue a separate
air pollution control techniques document earlier in the process,
specifically in conjunction with the air quality criteria as required
by section 108, or whether such information is more useful if tailored
to specific standard options under consideration, as in the RIA.
b. Technology Development and Leakage
    Two of the policy design considerations noted in section III.F.1
include the potential to promote technology development and to address
potential concerns about shifting emissions to other countries. The
NAAQS establish standards based on ambient concentrations that must be
attained and maintained everywhere, and are implemented through SIPs
that establish emissions budgets consistent with meeting the standards.
The limited emissions budget encourages state and local areas and
affected sources to work together to identify least-cost emissions

[[Page 44485]]

controls to meet their SIP obligations and reduce ambient
concentrations of the regulated pollutant(s). The NAAQS requirements
help create market demand for technologies that can assist in meeting
air quality standards at the least cost. As discussed in Section III.C
of this notice, this process has encouraged significant technological
innovation. EPA requests comment on the extent to which the NAAQS can
be an effective mechanism for encouraging technological innovation and
development of least-cost controls for GHG emissions.
    The 10-year maximum timeline for attaining a primary NAAQS would
allow some time for development and deployment of emerging
technologies, but longer timelines available under other forms of the
NAAQS would provide greater flexibility to provide continuous
incentives over a longer time period for major technology advances, and
more time to deploy new technologies that are developed. EPA requests
comment on the extent to which a GHG NAAQS could reasonably be expected
to advance new control technologies, and on what timeframe.
    With respect to the leakage issue, establishing a primary NAAQS
could lead to high costs among affected industries unless a viable
approach is identified to limit the control burden on U.S. sources.
Because the standards themselves are set without consideration of cost
or availability of control technologies, and because states would be
required to adopt a plan to attain a primary standard within 10 years
of designation, the NAAQS approach might offer less flexibility to
delay emissions reductions in the absence of effective control
technologies or when costs are prohibitive. This consideration may be
particularly relevant in the case of GHGs, where highly efficient
control technologies or mitigation options are currently limited, and
where critical new control strategies, such as carbon capture and
storage, are still in the early stages of development. In these
instances, industries that are unable to locate cost-effective control
strategies may consider relocating to non-regulated locations,
resulting in significant emissions leakage.
    We request comment on the cost-effectiveness of utilizing a NAAQS
approach to regulating GHGs, and on the extent to which this approach
might be expected to result in emissions leakage, especially as compared 
to other potential regulatory approaches outlined in this notice.
c. Summary of Opportunities and Challenges Afforded by NAAQS Pathway
    Regulating GHGs through a NAAQS offers certain opportunities;
however, there are also significant technological, legal and program
design challenges that would tend to limit the appropriateness of the
NAAQS program.
    NAAQS are based purely on preventing adverse health and
environmental impacts, rather than on considerations of cost,
feasibility, or availability of technology. Our expectation is that the
NAAQS approach would establish a goal tied to actual ambient
concentrations of GHGs. A NAAQS would call for assessment of potential
control strategies for a broad array of sources, rather than focusing
only on emissions reductions from a specified (but potentially limited)
list of sources. The NAAQS approach would allow for some flexibility in
the design of control strategies and requirements, including the
possibility of a cap-and-trade approach, and might spur significant
technological innovation. It would provide a mechanism for reducing GHG
emissions from current sources and limiting the growth of emissions
from new sources. If the facts supported adopting only a secondary
standard, this would somewhat reduce the specific obligations on
states, and would allow a suitably extended timeline for achieving the
emissions reductions necessary to stabilize and then reduce ambient GHG
concentrations.
    Though such an approach has the potential to be effective in
reducing emissions, there would be a number of obstacles to overcome.
Chief among these is that if worldwide (non-U.S.) emissons were to
continue increasing, global concentrations of GHGs would continue to
increase despite U.S. emission control efforts, and the NAAQS would be
unachievable (depending on the level of the standards) even if U.S.
emissions were reduced to zero. Unless viable legal approaches could be
identified for limiting the control burden on U.S. sources, such as by
defining a U.S. share of the emissions reductions needed to attain a
NAAQS, the NAAQS approach would result in an expensive program. It
would not achieve the adopted GHG NAAQS due to foreign emissions
growth, although U.S. emissions reductions would be achieved. If the
result of a NAAQS were stringent unilateral controls for vulnerable
industries, this would encourage emissions leakage in the absence of
comparable control efforts abroad.
    Especially if the Agency were to set a primary as well as a
secondary standard, a NAAQS would trigger a relatively rigid
implementation apparatus, limiting the Agency's flexibility to target
cost-effective emissions reductions and to shift the burden of control
requirements among different industries based on the availability of
new technological approaches. The lack of flexibility allowed under the
CAA for many of the NAAQS implementation requirements--especially those
affecting areas designated nonattainment with a primary standard--makes
them difficult to adapt effectively for application in the GHG context.
For example, it would be challenging to apply requirements for
transportation conformity under a GHG NAAQS, or for states to develop
attainment demonstration SIPs. As discussed in section IV.E, a
nonattainment new source review program requiring for GHGs would
dramatically expand the scope of the preconstruction permitting program
to include smaller sources and new types of sources such as apartment
buildings with natural gas heat, unless EPA were successful in applying
legal theories that justify deviating from statutory language. This
would pose substantial administrative feasibility and cost issues.
While implementation of an attainment-level NAAQS would involve fewer
specific requirements, this avenue would only apply if the standard set
by EPA under section 109 resulted in attainment designations. Section
109 calls for standards to be set based on science-based criteria,
which exclude consideration of the cost or efficiency of the
implementation requirements in determining the level of the standard.
    We note that while the NAAQS implementation system is state-based,
legislative proposals have focused on establishing federally administered 
national cap-and-trade strategies to address the global climate problem.
    In closing, we request comment on our assessment of NAAQS
approaches, and on how the NAAQS approach compares to other potential
CAA approaches in light of the policy principles enunciated in section
III.F.1.
5. Possible Implications for Other CAA Provisions
    Listing a pollutant under section 108(a)(1) would preclude listing
under section 112 or regulation under section 111(d), but would not
preclude listing and regulation under section 111(a)-(c) New Source
Performance Standards (NSPS) provisions as described below. Similarly,
regulation of GHGs under section 111(a)-(c) NSPS provisions, as
discussed further in other sections of

[[Page 44486]]

today's notice, would not preclude regulation of those pollutants
through a NAAQS, although controls implemented through these provisions
might influence the Agency's perspective on the appropriateness of
establishing air quality criteria for GHGs. EPA requests comment on the
extent to which regulatory action under section 111 could be considered
in the context of exercising authority under section 108 relevant to GHGs.

B. Standards of Performance for New and Existing Sources

    CAA section 111 provides EPA with authority to set national
performance standards for stationary sources. There are two alternative
pathways for using section 111 to regulate GHGs--as part of an
implementation program for a GHG NAAQS or as a freestanding program.
     In the event of a GHG NAAQS, section 111 authorizes EPA to
set emissions performance standards for new and modified sources but
not for unmodified existing sources.
     In the absence of a GHG NAAQS, section 111 offers the
potential for an independent, comprehensive program for regulating most
stationary sources of GHGs, except to the extent GHG emissions are
regulated under section 112
    Section 111 provides for consideration of cost, and allows
substantial discretion regarding the types and size of sources
regulated. As with most other CAA authorities, however, establishment
of a section 111 standard for any source category of GHGs would trigger
preconstruction permitting requirements for all types of GHG major
sources under the PSD program.
    The Stationary Source TSD for this ANPR identifies some specific
industry sectors that EPA has evaluated for their emissions of multiple
pollutants, including GHGs. EPA requests comment on this analysis. In
addition, EPA requests comment on GHG emissions from these and all
other categories and subcategories that have been subject to section
111 standards and on the relative costs that could be associated with
employing certain identified control technology or practices affecting
GHG emissions, including any positive or negative impacts on the
emissions of traditional pollutants.
1. What Does Section 111 Require?
    Section 111 establishes two distinct mechanisms for controlling
emissions of air pollutants from stationary sources. Section 111(b)
provides authority for EPA to promulgate New Source Performance
Standards (NSPS) which may be issued regardless of whether there is a
NAAQS for the pollutant being regulated, but apply only to new and
modified sources. Once EPA has elected to set an NSPS for new and
modified sources in a given source category, section 111(d) calls for
regulation of existing sources with certain exceptions explained below.
Taken together, the section 111 provisions could allow significant
flexibility in regulation that may not be available under other CAA
Title I provisions.
a. Section 111(b) New Source Performance Standards
    Section 111(b) of the CAA requires EPA to establish emission
standards for any category of new and modified stationary sources that
the Administrator, in his judgment, finds ``causes, or contributes
significantly to, air pollution which may reasonably be anticipated to
endanger public health or welfare.'' EPA has previously made
endangerment findings under this section for more than 60 stationary
source categories and subcategories that are now subject to NSPS.\239\
An endangerment finding would be a prerequisite for listing additional
source categories under section 111(b), but is not required to regulate
GHGs from source categories that have already been listed.
---------------------------------------------------------------------------

    \239\ EPA has developed NSPS for more than 70 source categories
and subcategories. However, endangerment findings apply to the
categories as a whole, while subcategories within them have been
established for purposes of creating standards that distinguish
among sizes, types, and classes of sources.
---------------------------------------------------------------------------

    For listed source categories, EPA must establish ``standards of
performance'' that apply to sources that are constructed, modified or
reconstructed after EPA proposes the NSPS for the relevant source
category.\240\ However, EPA has significant discretion to define the
source categories, determine the pollutants for which standards should
be developed, identify the facilities within each source category to be
covered, and set the level of the standards. In addition, EPA believes
that the NSPS program is flexible enough to allow the use of certain
market-oriented mechanisms to regulate emissions, as discussed below.
---------------------------------------------------------------------------

    \240\ Specific statutory and regulatory provisions define what
constitutes a modification or reconstruction of a facility. 40 CFR
60.14 provides that an existing facility is modified, and therefore
subject to an NSPS, if it undergoes ``any physical change in the
method of operation . . . which increases the amount of any air
pollutant emitted by such source or which results in the emission of
any air pollutant not previously emitted.'' 40 CFR 60.15, in turn,
provides that a facility is reconstructed if components are replaced
at an existing facility to such an extent that the capital cost of
the new equipment/components exceed 50 percent of what is believed
to be the cost of a completely new facility.
---------------------------------------------------------------------------

    As implemented over many years by EPA, the NSPS program has
established standards that do not necessarily set emission limits for
all pollutants or even all regulated pollutants emitted by sources
within the relevant source category. Rather, the NSPS generally focus
on specific pollutants of concern for a particular source category. Air
pollutants currently regulated through section 111(b) include the
criteria pollutants listed under section 108 and certain additional
pollutants. These additional pollutants are acid mist, fluorides,
hydrogen sulfide in acid gas, total reduced sulfur, and landfill gas.
EPA has discretion to revise an existing NSPS to add standards for
pollutants not currently regulated for that source category, but has
interpreted the section to not require such a result when an NSPS is
reviewed pursuant to section 111(b)(1)(B). That section requires EPA to
review and, if appropriate, revise NSPS every eight years unless the
Agency determines that such review is not appropriate in light of
readily available information on the efficacy of the standard.
    Further, in contrast to other provisions in the CAA which require
regulation of all sources above specific size thresholds, section 111
gives EPA significant discretion to identify the facilities within a
source category that should be regulated. To define the affected
facilities, EPA can use size thresholds for regulation and create
subcategories based on source type, class or size. Emission limits also
may be established either for equipment within a facility or for an
entire facility.
    EPA also has significant discretion to determine the appropriate
level for the standards. Section 111(a)(1) provides that NSPS are to
``reflect the degree of emission limitation achievable through the
application of the best system of emission reduction which (taking into
account the cost of achieving such reduction and any nonair quality
health and environmental impact and energy requirements) the
Administrator determines has been adequately demonstrated.'' This level
of control is commonly referred to as best demonstrated technology
(BDT). In determining BDT, we typically conduct a technology review
that identifies what emission reduction systems exist and how much they
reduce air pollution in practice. This allows us to identify potential
emission limits. Next, we evaluate each limit in conjunction with
costs, secondary air benefits (or disbenefits) resulting from energy

[[Page 44487]]

requirements, and non-air quality impacts such as solid waste
generation. The resultant standard is commonly a numerical emissions
limit, expressed as a performance level (i.e., a rate-based standard).
While such standards are based on the effectiveness of one or more
specific technological systems of emissions control, unless certain
conditions are met, EPA may not prescribe a particular technological
system that must be used to comply with a NSPS. Rather, sources remain
free to elect whatever combination of measures will achieve equivalent
or greater control of emissions.
    It is important to note that under section 111, the systems on
which a standard is based need only be ``adequately demonstrated'' in
EPA's view such that it would be reasonable to apply them to the
regulated category. The systems, and corresponding emission rates, need
not be actually in use or achieved in practice at potentially regulated
sources or even at a commercial scale. Further, EPA believes that if a
technology is ``adequately demonstrated'' for use at a date in the
future, EPA could establish a future-year standard based on that
technology. This would allow EPA to develop two- or multi-phased
standards with more stringent limits in future years that take into
account and promote the development of technology.
    Costs are also considered in evaluating the appropriate standard of
performance for each category or subcategory. We generally compare
control options and estimated costs and emission impacts of multiple,
specific emission standard options under consideration. As part of this
analysis, we consider numerous factors relating to the potential cost
of the regulation, including industry organization and market
structure; control options available to reduce emissions of the
regulated pollutant(s); and costs of these controls. Frequently, much
of this information is presented in the Regulatory Impact Analysis
(RIA) that is required for all major rulemaking actions.
b. Section 111(d) Emissions Guidelines for Existing Sources
    Section 111(d) requires regulation of existing sources in specific
circumstances. Specifically, where EPA establishes a NSPS for a
pollutant, a section 111(d) standard is required for existing sources
in the regulated source category except in two circumstances. First,
section 111(d) prohibits regulation of a NAAQS pollutant under that
section. Second, ``where a source category is being regulated under
section 112, a section 111(d) standard of performance cannot be
established to address any HAP listed under 112(b) that may be emitted
from that particular source category.'' \241\
---------------------------------------------------------------------------

    \241\ See 70 FR 15994, 16029-32 (Mar. 29, 2005).
---------------------------------------------------------------------------

    Section 111(d) also uses a different regulatory mechanism to
regulate existing sources than section 111(b) uses for new and modified
sources in a source category. Instead of giving EPA direct authority to
set national standards applicable to existing sources in the source
category, section 111(d) provides that EPA shall establish a procedure
for states to issue performance standards for existing sources in that
source category. Under the 111(d) mechanism, EPA first develops
regulations known as ``emission guidelines.'' These may be issued at
the same time or after an NSPS for the source category is promulgated.
Although called ``guidelines,'' they establish binding requirements
that states are required to address when they develop plans to regulate
the existing sources in their jurisdictions. These state plans are
similar to state implementation plans and must be submitted to EPA for
approval. Historically, EPA has issued model standards for existing
sources that could then be adopted by states. Under this approach,
creating an interstate trading system would require adoption of
compatible state rules promoted by EPA rules and guidance. In the event
that a state does not adopt and submit a plan, EPA has authority to
then issue a federal plan covering affected sources.
    Section 111(d) guidelines, like NSPS standards, must reflect the
emission reduction achievable through the application of BDT. However,
both the statute and EPA's regulations implementing section 111(d)
recognize that existing sources may not always have the capability to
achieve the same levels of control at reasonable cost as new sources.
The statute and EPA's regulations in 40 CFR 60.24 permit states and EPA
to set less stringent standards or longer compliance schedules for
existing sources where warranted considering cost of control; useful
life of the facilities; location or process design at a particular
facility; physical impossibility of installing necessary control
equipment; or other factors making less stringent limits or longer
compliance schedules appropriate.
2. What Sources Could Be Affected?
    Section 111 has been used to regulate emissions of traditional and
nontraditional air pollutants from a broad spectrum of stationary
source categories. EPA has already promulgated NSPS for more than 70
source categories and subcategoriesand we could add GHG emission
standards, as appropriate, to the standards for existing source
categories.\242\ EPA has begun a review of the existing NSPS source
categories to determine whether it would be appropriate to regulate GHG
emissions from sources in each category. In addition, EPA is in the
process of responding to a remand from the D.C. Circuit requiring it to
consider whether to add standards for GHGs to the NSPS for utility
boilers, and EPA has received suggestions that it would be appropriate
to add such standards to the NSPS for Portland cement kilns.\243\
---------------------------------------------------------------------------

    \242\ Some of the existing source categories are very broad,
comprising an entire industrial process such as steel making, while
others are narrowly defined as a single piece of equipment within a
broader production process. Examples of source categories subject to
NSPS are fossil fuel-fired boilers, incinerators, sulfuric acid
plants, petroleum refineries, lead smelters, and equipment leaks of
VOCs in the synthetic organic chemicals manufacturing industry. A
complete list of the NSPS source categories is found at 40 CFR part 60.
    \243\ The NSPS for Petroleum Refineries were recently amended,
resulting in the promulgation of new Subpart Ja. These performance
standards include emission limitations and work practice standards
for fluid catalytic cracking units, fluid coking units, delayed
coking units, fuel gas combustion devices, and sulfur recovery
plants. As such, they regulate criteria pollutant emissions from the
processes that are also responsible for most of the refinery GHG
emissions. During the public comment period for Subpart Ja, we
received several comments in favor of developing new source
performance standards to address GHG emissions from refineries.
However, we declined to adopt standards for GHG emissions in that
rulemaking, in part because while doing so was within our
discretion, we believed that it was important to fully consider the
implications for programs under other parts of the CAA before
electing to regulate GHG under section 111. This is a fundamental
purpose for today's notice and request for comments.
---------------------------------------------------------------------------

    To determine whether regulation of GHGs is appropriate for existing
categories, we must evaluate whether it is reasonable to do so given
the magnitude of emissions and availability of controls, considering
the costs of control. Decisions in this regard could be influenced by
several factors, including the magnitude of the GHG emissions from a
source category; the potency of the particular GHG emitted; whether
emissions are continuous, seasonal or intermittent; the availability of
information regarding the category's GHG emissions; and whether
regulating GHG emissions from the source category would be beneficial.
EPA requests comment on the extent to which these factors should, if at
all, influence EPA's decisions whether to add standards to existing
NSPS and what additional factors should be taken into consideration.
EPA also requests

[[Page 44488]]

comment on which of the previously regulated categories might be
appropriate for GHG regulation and on the information on which such
judgments might be based.
    To inform the public of EPA's analytical work to date, we have
provided descriptions of key industrial sectors, their GHG emissions,
and information that we have collected to date on GHG control options
for those sectors in the Stationary Source TSD in the docket for
today's notice. It is important to note that, as described further in
the technical support materials, many near-term technologies or
techniques for reducing GHG, e.g., energy efficiency or process
efficiency improvements, are relatively cost effective and achieve
modest emission reductions when compared with the potential of some
add-on control techniques. Other controls may become available in the
future whose costs and emission reduction effectiveness may differ
substantially from what is discussed here today. The Stationary Source
TSD also discusses various mechanisms, such as cap-and-trade programs
or emissions averaging approaches across facilities or industries, that
can help reduce costs of reducing emissions. EPA requests comment on
the availability and extent of its legal authority for such mechanisms.
    In addition to regulating GHGs from previously listed source
categories, section 111 provides discretionary authority to list new
source categories, or reformulate listed source categories, for
purposes of regulating of GHG emissions. For example, such categories
could include sources of emissions covered by existing NSPS source
categories as well as sources not currently covered by any NSPS. One
option available to EPA is the reorganization of source categories for
purposes of GHG regulation. In creating new categories to be used for
regulation of GHGs, EPA could consider factors unique to GHG emissions.
For example, EPA could take into account concerns about emissions
leakage (discussed in section III.F.5 of this notice), and structure
categories to minimize opportunities for shifting emissions to other
source categories. EPA could also explore how the rearrangement of
source categories could facilitate netting arrangements through which a
more broadly defined ``source'' could avoid triggering an GHG NSPS by
off-setting its increased GHG emissions.\244\ In addition, EPA could
structure categories to take into account possible reductions from
improvements at non-emitting parts of the plants, for example, by
creating source categories that cover all equipment at particular
plants, instead of using categories that cover only specific types of
equipment at a plant. EPA invites comment on whether such rearrangement
would be appropriate and what type of rearrangement would be desirable.
We also solicit information on how rearrangement could facilitate
netting and how we might structure such netting.
---------------------------------------------------------------------------

    \244\ We recognize that the Court in Asarco Inc. v. EPA, 578
F.2d 326 (D.C. Cir. 1978) struck down an NSPS provision that allowed
netting. The provision at issue there, however, permitted netting
between sources, not within a source. See Alabama Power v. EPA, 636
F.2d 323, 401-02 (D.C. Cir. 1980).
---------------------------------------------------------------------------

    An alternative, or complementary, scenario would be to create
larger ``super-categories'' covering major groupings of stationary
sources of GHG emissions. For example, it might be possible to create
process-based categories (i.e., all sources emitting CO2 through a
stack as a result of combustion processes) or vertically integrated
categories which take more of a life-cycle approach to the control of
GHG emissions and reduce the possibility of leakage of GHG reductions
to other parts of the economy or other geographic regions.\245\ The
creation of such ``super-categories'' might provide additional
opportunities for the development of innovative control mechanisms such
as cap-and-trade programs covering multiple industry sectors. In light
of these considerations, EPA requests comment on whether the creation
of such ``super categories'' would be appropriate and what categories
would be most useful for regulating GHGs.
---------------------------------------------------------------------------

    \245\ For instance, a ``super-category'' could be created
encompassing all aspects of the production, processing, and
consumption of petroleum fuels, or to regulate the production and
consumption of fossil fuels for heat and power, addressing all
aspects of emissions-producing activity within a sector, including
fuel production, consumption, and energy conservation.
---------------------------------------------------------------------------

    Under either option, EPA possesses authority to distinguish among
classes, types and sizes of sources within existing categories for
purposes of regulating GHG emissions. For example, we have at times
distinguished between new and modified/reconstructed sources when
setting the standards. This may be appropriate, for instance, when a
particular new technology may readily be incorporated into a new
installation, but it may be technically infeasible or unreasonably
costly to retrofit this technology to an existing facility undergoing
modification or reconstruction. Alternatively, we have distinguished
among sources within a category, for instance fossil fuel-fired
boilers, for which we have subcategorized on the basis of fuel types
(e.g., coal, oil, natural gas). EPA requests comment on what
considerations are relevant to determining whether it is appropriate
and reasonable to establish subcategories for regulation under section 111.
3. What Are Possible Key Milestones and Implementation Timelines?
a. Priority Setting Among Source Categories
    If EPA were to pursue section 111 regulation of GHGs, timetables
for regulation would depend upon how EPA prioritized among source
categories to determine which categories should be regulated first. In
the near term, it may be possible to address GHGs under section 111 in
a limited fashion by establishing control requirements for new and
existing sources in some number of existing source categories, while
information is developed on other source categories. Actions under
other portions of the CAA may involve longer lead times to develop and
implement, so that standards under section 111 for certain source
categories could provide for emission reductions in the interim. We
have begun to examine source categories subject to existing NSPS and
other standards to consider how we might determine priorities among
them for review and revisions, and whether GHGs could be addressed for
specific sectors in a more coordinated, multi-pollutant fashion. EPA
requests comment on the availability of its legal authority, if any, to
prioritize among source categories in the event that regulation under
section 111 was pursued.
    Under a ``prioritization'' approach, EPA could seek to revise
standards earliest for those categories offering the greatest potential
for significant reductions in the emissions of covered pollutants, and
either deferring action or determining that no further action is
necessary or appropriate at this time for other categories. This
conclusion could be based, for example, on the lack of significant
improvements in technology since the last NSPS review or the fact that
no new sources are considered to be likely in the foreseeable future.
    Another possibility might be to schedule and structure the review
and revision of standards for source categories to account for the fact
that, in addition to the need to address GHG emissions, they may be
subject to multiple standards for different pollutants under several
sections of the CAA. Such standards may often be subject currently to
different review

[[Page 44489]]

timetables resulting from when these standards were last established or
revised. In addition, as discussed in section III.D of today's notice,
they may have the potential for positive or negative interactions with
one another and with opportunities for the control of GHG emissions.
    Still another approach might consider the impacts of future
reduction opportunities or enacted legislation so that standards under
section 111 might focus initially on source categories for which near-
term benefits might result largely from efficiency improvements which
do not result in ``stranded capital,'' or investment in systems that
will be superseded by more effective systems that we determine will be
available at some specific future date. Alternatively, standards could
focus on those sectors of the economy which will not likely be subject
to controls being addressed in enacted legislation.
    We request comment on EPA's available legal authority, if any, to
defer action with respect to any ``class'' of section 111 source
categories or subcategories as well as how and under what circumstances
EPA could also consider such approaches to the identification of source
categories for standards to address GHGs. Assuming the existence of
adequate authority, what, if any, additional criteria should be
considered in our priority-setting analysis efforts? In considering
such sector- or multi-pollutant-based approaches, we further request
comment on the extent to which we could establish new or revised source
categories which better accommodate these approaches, or whether we are
bound by existing source categories and their definitions.
b. Timetables for Promulgation and Implementation
    In our experience, collecting and analyzing information regarding
available control technologies, resulting emission reductions, and cost
effectiveness can take up to several years for a source category.
However, this time period can be shortened to 1\1/2\ to 2 years when
information is readily available or is presented to the Agency in a
form that facilitates efficient consideration. With respect to GHGs,
there has been significant effort devoted to identifying and evaluating
ways to reduce emissions within sectors such as the electricity
generating industry, and we are aware of the potential for GHG
reductions through energy efficiency and other means within other
industries. However, for many others, technologies for reducing GHG
emissions have not yet been identified or evaluated by EPA. EPA
requests comment on whether and how the availability of current
information should be considered when considering regulation under
section 111.
    As is the case with traditional pollutants, any new or revised NSPS
for new and modified sources of GHGs under section 111(b) would be
developed through a notice and comment rulemaking process and would be
effective upon promulgation. As noted previously, EPA is also required
to review, and if appropriate revise, existing NSPS every 8 years
unless the Administrator determines that ``such review is not
appropriate in light of readily available information on the efficacy
of such standard.'' Standards for pollutants not regulated by the
existing NSPS may be added concurrent with the 8-year review, but such
additions are not part of that review process.
    Any section 111(d) emission guidelines associated with the revised
NSPS standards would be promulgated either along with or after the
NSPS. States are generally required to submit the required state plans
containing the standards of performance applicable to existing sources
in their jurisdictions within 9 months of EPA's promulgation of the
guidelines.
    In the case of existing sources regulated under section 111(d),
affected sources are typically provided up to 3 years to comply with
any resulting requirements; however, states have flexibility to provide
longer or shorter compliance timeframes based on a number of source-
specific factors. In addition, where we determine that a technology has
been adequately demonstrated to be available for use by some particular
future date, we believe it is possible to establish timeframes for
compliance that reflect this finding.\246\
---------------------------------------------------------------------------

    \246\ See Portland Cement Association v. EPA, 486 F.2d 275 (D.C.
Cir. 1973).
---------------------------------------------------------------------------

    No explicit 8-year review requirement exists with regard to section
111(d) standards for existing sources. Nonetheless, it also may be
appropriate to require existing source plans to periodically revise
their control strategies to reflect changes in available technologies
and standards over time, particularly where the existing limitations
were based on more limited controls at the time they were established.
EPA requests comment on its authority and the advisability of such
periodic updating with respect to the possible control of GHG.
    The CAA and EPA's regulations implementing section 111(d) permit
states to consider a number of factors when determining the level of
stringency of controls, but do not establish a bright line test when
stricter requirements for existing sources are warranted. Many of these
sources may also be subject to requirements for the control of other
non-section 111(d) pollutants as part of implementation plans to attain
and maintain NAAQS for one or more pollutants, and in some cases, these
provisions may result in more stringent coincidental control of section
111(d) pollutants. We request comment on how and when we should
evaluate, review, and revise as appropriate any section 111(d)
standards that might be established in the future for GHGs.
4. What Are the Key Considerations Regarding Use of This Authority To
Regulate GHGs?
a. Key Attributes and Limitations of Section 111
    As noted above, section 111 possesses certain flexible attributes
that may be useful in tailoring emissions standards to address GHG
emissions. Yet, regulation under this section also has important
limitations. This section of today's notice briefly summarizes these
attributes and limitations. We request comment on how these attributes
and limitations relate to the policy design considerations set forth in
section III.F.1.
    Program scope: Section 111 provides EPA with authority to regulate
GHG emissions from stationary source categories, but does not require
EPA to regulate GHGs emitted by all source categories or even all
listed source categories. EPA has flexibility to identify the source
categories for which it is appropriate to establish GHG limits. For
example, EPA could decide to set GHG limits for those source categories
with the largest GHG emissions and reduction opportunities. EPA could
postpone or decline to set GHG limits for source categories for which
emissions contributions may be small or for which no effective means of
reducing emissions exist, currently or within the reasonably
foreseeable future. EPA also could consider traditional air pollutants
as well as GHGs in setting its overall priorities for the NSPS program.
    Source size: Section 111 does not require regulation of all sources
above a certain size. Instead, EPA has discretion to use rational
emission thresholds to identify which facilities within a source
category are covered by NSPS standards.
    Consideration of cost: Section 111 explicitly directs EPA to take
``into account the cost of achieving'' emission

[[Page 44490]]

reductions, as well as other nonair quality, health and environmental
impact and energy requirements.'' This gives EPA significant
flexibility to determine of appropriate levels of control, and can be
an important source of distinctions between requirements for new
sources and those for modified or reconstructed sources.
    Potential for emissions trading: As EPA has interpreted the NSPS
requirements in the past with respect to certain air pollutants, we
believe that the NSPS program could use emissions trading, including
cap-and-trade programs and rate-based regulations that allow emissions
trading, to achieve GHG emission reductions. EPA believes such programs
are consistent with the statutory requirements because they satisfy the
three substantive components of the section 111(a)(1) definition of
``standard of performance''--(1) a standard for emissions of air
pollutants; that (2) reflects that degree of emission limitation
available''; and (3) ``constitutes the best system of emission
reduction.'' A cap-and-trade program can constitute a ``standard for
emissions of air pollutants'' because it is a system created by EPA for
control of emissions. The use of emissions budgets does not make the
system less of a ``standard'' since the budgets must be met regardless
of the methodology used to allocate allowances to specific sources.
Further, any such system would be based on our assessment of the
overall degree of emission reduction available for the source category
and our analysis of the available systems of emission reductions. EPA
could select a market-oriented mechanism as the ``standard of
performance'' if these analyses (including cost analyses) indicate that
the system would ``reflect the degree of emission limitation
achievable'' and ``constitute the best system of emission reduction.''
EPA also believes that trading among new and existing sources could be
permitted, and could offer, at least in some cases, cost
efficiencies.\247\ EPA also believes that because of the potential cost
savings, it might be possible for the Agency to consider deeper
reductions through a cap-and-trade program that allowed trading among
sources in various source categories relative to other systems of
emission reduction. We request comment on the extent of EPA's available
legal authority in this area as well as the attributes such a program
must possess to qualify as a standard of performance under section 111.
---------------------------------------------------------------------------

    \247\ In the Clean Air Mercury Rule we concluded that new
sources needed to comply with a unit specific control requirement in
addition to participating in the trading program. We solicit comment
on whether section 111 requires such controls for new sources or if
it would be sufficient for them to participate in a trading program
or other market based mechanism without this restriction. While not
ensuring an equally stringent level of control at each new source,
the latter approach would be expected to achieve the same total
emissions reductions at a lower overall compliance cost.
---------------------------------------------------------------------------

    Potential for declining performance standards: EPA believes that
section 111 authority may be used to set both single-phase performance
standards based upon current technology and to set two-phased or multi-
phased standards with more stringent limits in future years. Future-
year limits may permissibly be based on technologies that, at the time
of the rulemaking, we find adequately demonstrated to be available for
use at some specified future date. Alternatively, it may be possible to
establish a goal based on future availability of a technology and to
revise the standard to reflect technological advancements at
appropriate intervals, such as the 8-year review cycles. We believe
these concepts could be applied to standards for new and modified
sources, as well as to standards for existing sources under section
111(d). In addition, this concept could be coupled with emissions trading.
    We recognize that various legal issues and questions concerning
legal authority may be involved in setting standards based on
technology only adequately demonstrated for use at a future date. For
example, there might be greater uncertainty regarding the cost of
technology for such standards than for standards based only on
technology that is already commercially demonstrated at the time of
promulgation. In the Clean Air Mercury Rule (CAMR), which was vacated
by the D.C. Circuit on other grounds, EPA interpreted section 111 to
allow a two-phased ``standard of performance'' to reduce mercury
emissions from existing sources. The compliance date for the more
stringent second phase was 2018. EPA believed that it had greater
flexibility to set such a standard for existing sources under section
111(d) because these standards, in contrast to section 111(b) standards
for new sources, are not subject to the requirements of section 111(e).
Section 111(e) makes unlawful to operate any new source in violation of
a standard of performance after its effective date. EPA requests
comment on this interpretation. We also request comment on the
circumstances under which the requirements of section 111(e) would be
satisfied by a standard requiring compliance with the initial
requirements of a multi-phase standard. More generally, EPA seeks
comment on its legal authority in this matter as well as the legal and
factual conditions that must be satisfied to support a multi-phase
standard with future-year standards based on technology adequately
demonstrated for use by that future date. EPA also seeks comment on how
far into the future multi-phase standards could extend and the degree
of certainty with which EPA must make its determinations of
availability for future use, considering the section 111 standard
setting language.
    Technology development: Section 111 also contains a waiver
provision that can be used to encourage the development of innovative
technologies, as described below.
    Standards tied to available technology: The fact that section 111
requirements are based upon a demonstration of the availability of
control technology could limit the amount of reductions achievable
through section 111 regulations to demonstrably feasible and cost-
effective levels. If a given level of overall emission reduction is
determined to be necessary and that level exceeds what is currently
demonstrated to be feasible now or by some future date, then section
111 may not provide adequate authority by itself to achieve needed
reductions. Although section 111 provides certain opportunities and
incentives for technology development, this feature may make it more
difficult to set ``stretch goals'' without other companion mechanisms.
    In light of these considerations, we request comment on whether and
to what extent section 111 provides an appropriate means for regulating
GHG emissions.
b. Additional Considerations
    We also request comment on the questions presented below which
relate to the manner in which EPA could or should exercise its
authority under this section to regulate GHGs.
i. What Regulatory Mechanisms Are Available?
    As noted above, NSPS standards and 111(d) emission guidelines most
commonly establish numerical emission standards expressed as a
performance level. Such rate-based limits, however, are not the only
mechanisms that could be used to regulate GHGs.
    Efficiency Standards: We believe that most reductions in stationary
GHG emissions may occur initially as the result of increased energy
efficiency, process efficiency improvements, recovery and beneficial
use of process gases, and certain raw material and product changes that
could reduce inputs of carbon or other GHG-

[[Page 44491]]

generating materials. Such emission reductions may range in the near
term (e.g., 5-10 years) from 1 to 10%. Thus, it could be possible to
utilize NSPS standards to ensure reductions from efficiency
improvements are obtained. For such standards to be effective, they
likely would generally need to apply to the entire facility, not just
specific equipment at the facility. EPA requests comment on the
availability of its legal authority in this area and whether and when
it might be appropriate to establish efficiency standards for source
categories as a way of reducing GHG emissions.
    Plant-wide standards: EPA also believes there may be benefits to
developing plant-wide or company-wide standards for GHG emissions.
Section 111, however, requires each affected facility to comply with
the standard. EPA believes that it could redefine the affected facility
for certain categories, for purposes of GHG regulation only, to include
an entire plant. EPA also requests comment on whether it would be
consistent with the statutory requirements to establish company-wide
limits.
    Work practice standards: In some circumstances, it may not be
possible to identify a specific performance level for sources in a
particular category; however, section 111(h) permits promulgation of
design, equipment, work practice, or operational standards but allows
such standards to be established only in specific circumstances.
Specifically, it provides that where we determine ``that (A) a
pollutant or pollutants cannot be emitted through a conveyance designed
and constructed to emit or capture such pollutant, or that any
requirement for, or use of, such a conveyance would be inconsistent
with any Federal, State, or local law, or (B) the application of
measurement methodology to a particular class of sources is not
practicable due to technological or economic limitations,'' we may
establish a ``design, equipment, work practice, or operational
standard, or combination thereof, which reflects the best technological
system of continuous mission reduction which . . . has been adequately
demonstrated.'' EPA requests comment on the circumstances under which
the section 111(h) criteria would be satisfied and when, and for which
source categories, work practice standards could be appropriate
standards to control GHGs.
    Market-oriented regulatory mechanisms: As mentioned above, EPA
believes that market-oriented regulatory approaches including emissions
trading are worthy of consideration for applying NSPS to GHG emissions.
Several market-oriented regulatory mechanisms are discussed in section
VII.G of today's notice. EPA requests comment on which of these
mechanisms are consistent with the section 111 definition of a
``standard of performance.''
ii. Request for Comment on Section 111 Regulatory Approaches
    This notice and the Stationary Source TSD describe possible
approaches for using section 111 to reduce GHG emissions, in general
and in regard to particular source categories. We request comment on
the following specific questions regarding potential regulatory
approaches under section 111:
     What are the overall advantages and disadvantages of the
regulatory approaches discussed above, in light of the policy design
considerations in section III.F.1? Please describe in detail any
approaches not discussed in today's notice that you think we should
consider.
     What are the industry-specific advantages and disadvantages 
of the regulatory approaches discussed above and in the TSD?
    In developing section 111 standards for a particular source
category (e.g., refineries, cement plants, industrial commercial
boilers, electric generating plants, etc.) we are requesting source
category-specific comments on the following additional issues:
     What data are available, or would need to be collected, to
support the development of performance standards, either by process,
subcategory, or for the facility?
     Should the standards be different for new and existing
sources, either in terms of the systems for emission reductions on
which they should be based and/or on the regulatory structure and
implementing mechanisms for such standards?
     To what extent, if any, should the standards be
technology-forcing for existing sources?
     Should the standards require additional reductions over
time? To what extent would such reductions be consistent with the
authority and purpose of section 111, and how should they be designed
and carried out to ensure consistency?
iii. What Reductions Could Be Achieved From Efficiency Improvements at
Existing Sources?
    Recognizing that existing sources do not have as much flexibility
in the levels of control that may realistically be achieved at a new
source, a section 111(d) standard regulating GHG from existing sources
would at this time most likely focus on currently available measures to
increase the energy efficiency at the facility, thereby reducing GHG
emissions. Examples of typical measures that promote energy efficiency
include the use of cleaner fuels and equipment replacement or process
improvements which reduce energy consumption. How well a measure, or
combination of measures, will reduce GHG emissions at an individual
facility will vary. A review of available literature suggests a range
of improvements for various industry sectors that may be achievable
through energy and process efficiency improvements, and some
representative examples are summarized below. This information is
illustrative, and does not represent any final technical determination
by the agency as to what emission reduction requirements might be
appropriate to require from the source categories discussed below.
    For example, reductions in emissions of GHG from cement plants
would most likely occur from fuel efficiency and electric energy
efficiency measures as well as raw material and product changes that
reduce the amount of CO₂ generated per ton of cement
produced. There are numerous efficiency measures generally accepted by
much of the U.S. industry, and many of these measures have been adopted
in recent cement plant improvements. Such measures may directly reduce
GHG emissions by cement plants, or they may indirectly reduce GHG
emissions at sources of power generation due to reduced electrical
energy requirements. The range of effectiveness of the individual
measures in reducing GHG is from less than 1% to 10%.\248\ Benchmarking
and other studies have demonstrated a technical potential for up to 40%
improvement in energy efficiency for a new cement plant using the most
efficient technologies compared to older plants using wet kilns.
---------------------------------------------------------------------------

    \248\ U.S. EPA (2008), Air Pollution Controls and Efficiency
Improvement Measures for Cement Kiln. Final Report.
---------------------------------------------------------------------------

    A number of opportunities may exist within refineries to increase
energy efficiency by optimizing utilities, fired heaters, heat
exchangers, motors, and process designs. Competitive benchmarking data
indicate that most petroleum refineries can economically improve energy
efficiency by 10 to 20%.\249\ Therefore, we would expect that a new
refinery could be designed to be at least 20% more efficient than an
existing one.
---------------------------------------------------------------------------

    \249\ Energy Efficiency Improvement and Cost Saving
Opportunities for Petroleum Refineries, LBNL, 2005.

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

[[Page 44492]]

    In the case of industrial boilers, measures applied to individual
facilities could result in energy savings and GHG reductions on the
order of 1% to 10%. Replacing an existing boiler with a combined heat
and power plant could improve the energy efficiently of an existing
plant by 10% to 33%.
    Existing coal-fired power plants can reduce their fuel consumption
(reduce heat rate) and reduce CO₂ emissions by performing
well known modifications and upgrades to plant systems. Heat rate
reductions of up to 10% may be feasible through various efficiency
improvements at individual coal units, depending on site specific
conditions. Because of plant age and other physical limitations, the
potential average heat rate reduction for the coal fleet would likely
not exceed about 5%. The existing fleet operates at an average net
efficiency of about 33%. If the corresponding coal fleet average net
heat rate were reduced by 5% via efficiency improvements, a potential
5% reduction in CO₂ emissions could be obtained as well.
    As older, less efficient coal power plants are retired, their
capacity may be replaced with new, more efficient coal-fired units. A
new, fully proven supercritical coal plant design can operate at a heat
rate 10-15% below the current coal fleet average, and therefore produce
10-15% less GHG than the average existing coal plant. Future more
advanced ultra-supercritical plant designs with efficiencies above 40%
would have heat rates that are 20-25% or more below the current coal
fleet average, and therefore produce that much less GHG than the
average existing coal plant.
    Technology to capture and geologically sequester CO₂ is
the subject of ongoing projects in the U.S. and other countries and is
a promising technology.\250\ The electric power sector will most likely
be the largest potential market for carbon capture and sequestration
(CCS) technologies, with the potential to reduce CO₂ by
approximately 80-90% at an individual plant.\251\ It may become
possible to apply CCS to some portion of the existing coal-fired fleet
by retrofit to achieve significant CO₂ reductions. Other
facilities that might be able to use CCS include refineries, chemical
manufacturing plants, ethanol production facilities, cement kilns and
steel mills. As advances in GHG reduction technologies continue,
section 111(d) standards would be expected to consider and reflect
those advances over time. We solicit comment on the criteria EPA should
use to evaluate whether CCS technology is adequately demonstrated to be
available for the electric power and other industrial sectors,
including the key milestones and timelines associated with the wide-
spread use of the technology.
---------------------------------------------------------------------------

    \250\ See http://www.netl.doe.gov/technologies/carbon_seq/
partnerships/partnerships.html for more information about the
Regional Carbon Sequestration Partnerships in the United States.
    \251\ IPCC Special Report on Carbon Dixoide Capture and Storage,
2005, pp.3, 22.
---------------------------------------------------------------------------

    iv. What Are the Possible Effects of Section 111 With Respect to
Innovation?
    As noted previously, whatever path may be pursued with respect to
the control of GHG through the CAA or other authority, we believe it is
likely that most early reductions in stationary GHG emissions may occur
as the result of increased energy efficiency, process efficiency
improvements, recovery and beneficial use of process gases, and certain
raw material and product changes that could reduce inputs of carbon or
other GHG-generating materials. Clearly, more fundamental technological
changes will be needed to achieve deeper reductions in stationary
source GHG emissions over time. We request general comments on how to
create an environment in which new, more innovative approaches may be
encouraged pursuant to section 111, or other CAA or non-CAA authority.
    Waiver authority under section 111(j) would be useful as one
element of broader policies to encourage development of innovative
technologies. Section 111(j) authorizes the Administrator to waive the
NSPS requirements applicable to a source if he determines that the
innovative technology the source proposes to use will operate
effectively and is likely to achieve greater emission reductions, or at
least equivalent reductions but at lower cost. Also, the Administrator
must determine that the proposed system has not yet been adequately
demonstrated (i.e. it is still an innovative technology), but that it
will not cause or contribute to an unreasonable risk to public health,
welfare, or safety in its operation, function, or malfunction. These
waivers can be given for up to 7 years, or 4 years from the date that a
source commences operation, whichever is earlier.
    We believe that effective GHG reduction techniques for many source
categories potentially subject to NSPS may at this time be limited and
that additional research and development will be necessary before these
controls are demonstrated to be effective. We ask for comment on how
the use of innovative technology waivers could conceivably be used to
foster the development of additional approaches for GHG reductions.
5. Possible Implications for Other CAA Provisions
    Regulation of GHGs under a section 111 standard for any industry
would trigger preconstruction permitting requirements for all types of
GHG sources under the PSD program. NSPS are also incorporated into
operating permits issued under Title V of the CAA. The consequences of
triggering and the options for addressing these permitting requirements
are addressed in detail in section VII.D of this notice.
    Whether GHGs were regulated individually or as a group in NSPS
standards would affect the definition of regulated pollutant for
stationary sources subject to preconstruction permitting under the PSD
program. Conversely, while the section 111 mechanisms are relatively
independent of other CAA programs, NSPS decision-making as a practical
matter would need to consider the pollutant definitions adopted under
other CAA authorities. It would be advantageous to maintain consistency
regarding the GHG pollutants subject to regulation elsewhere in the Act
to avoid the potential for PSD review requirements for individual GHGs
as well as for groups of the same GHGs.
    In considering the impact that decisions to list pollutants under
other authorities of the CAA might have on our use of section 111
authority, we note that some industries have processes that emit more
than one GHG and a potential may exist among some of these industries
to control emissions of one GHG in ways that may increase emissions of
others (e.g., collecting methane emissions and combusting them to
produce heat and/or energy, resulting in emissions of CO₂.)
While an overall reduction in GHGs may occur, as well as a reduction in
global warming potential, whether GHGs are regulated as a class of
compounds or as individual constituents could have implications for the
degree of flexibility and for the outcome of any regulatory decisions.
More specifically, if we were to regulate GHGs as a group, then
standards under section 111 might establish an overall level of
performance that could accommodate increases in emissions of some gases
together with reductions in others, so long as the overall performance
target was met. If we were to regulate individual GHGs, then we may be
less able to establish less stringent requirements for the control of
some gases, while setting more stringent requirements for others. The
extent to which we may be able to do so depends

[[Page 44493]]

on the significance of the emissions of each gas from the source
category in question as well as the feasibility and cost-effectiveness
of controlling each. One result of this lessened flexibility may be the
preclusion of certain approaches that could yield greater net reduction
in GHG emissions. For this reason, we request comments on (1) the
extent to which we are limited in our flexibility to regulate GHG as a
class if listed individually under other CAA authorities, and (2)
whether regulation under section 111 should treat GHG emissions as a
class for determining the appropriate systems for emissions reduction
and resulting standards.
    Finally, we note that our authority to promulgate 111(d) standards
for existing sources depends on the two restrictions noted above.
First, section 111(d) prohibits regulation of a NAAQS pollutant under
that section. Second, ``where a source category is being regulated
under section 112, a section 111(d) standard of performance cannot be
established to address any HAP listed under 112(b) that may be emitted
from that particular source category.'' If we were to promulgate a
section 111(d) emission standard and then subsequently take action
under sections 108 or 112 such that we could not promulgate a section
111(d) standard had we not already done so, the continued validity of
the section 111(d) regulations might become unclear. We request comment
on the extent, if any, to which the requirements of section 111(d)
plans would, or could, remain in force under such circumstances.

C. National Emission Standards for Hazardous Air Pollutants

    Along with the NAAQS system and section 111 standards, section 112
is one of the three main regulatory pathways under the CAA for
stationary sources. Section 112 is the portion of the Act that Congress
designed for controlling hazardous air pollutant emissions from these
sources, including toxic pollutants with localized or more
geographically widespread effects. This focus is reflected in the
statutory provisions, which, for example, require EPA to regulate
sources with relatively small amounts of emissions. In comparison to
section 111, section 112 provides substantially less discretion to EPA
concerning the size and types of sources to regulate, and is specific
about when EPA may and may not consider cost.
    This section explores the implications if EPA were to list GHGs as
hazardous air pollutants under section 112.

1. What Does Section 112 Require?

a. Overview
    Section 112 contains a list of hazardous air pollutants (HAPs) for
regulation. EPA can add or delete pollutants from the list consistent
with certain criteria described below.
    EPA must list for regulation all categories of major sources that
emit one or more of the HAPs listed in the statute or added to the list
by EPA. A major source is defined as a source that emits or has the
potential to emit 10 tons per year or more of any one HAP or 25 tons
per year of any combination of HAPs.
    For each major source category, EPA must develop national emission
standards for hazardous air pollutants (NESHAP). Standards are required
for existing and new major sources. The statute requires the standards
to reflect ``the maximum degree of reduction in HAP emissions that is
achievable, taking into consideration the cost of achieving the
emission reduction, any nonair quality health and environmental
impacts, and energy requirements.'' This level of control is commonly
referred to as maximum achievable control technology, or MACT.
    The statute also provides authority for EPA to list and regulate
smaller ``area'' sources of HAPs. For those sources EPA can establish
either MACT or less stringent ``generally available control
technologies or management practices''.
    Section 112(d)(6), requires a review of these technology-based
standards every 8 years and requires that they be revised ``as
necessary taking into account developments in practices, processes and
control technologies.'' Additionally, EPA under section 112(f)(2)(C)
must reevaluate MACT standards within 8 years of their issuance to
determine whether MACT is sufficient to protect public health with an
ample margin of safety and prevent adverse environmental effects. If
not, EPA must promulgate more stringent regulations to address any such
``residual risk''.
b. How Are Pollutants and Source Categories Listed for Regulation Under
Section 112?
    Section 112(b)(1) includes an initial list of more than 180 HAPs.
Section 112(b)(2) requires EPA to periodically review the initial HAP
list and outlines criteria to be applied in deciding whether to add or
delete particular pollutants.
    A pollutant may be added to the list because of either human health
effects or adverse environmental effects. With regard to adverse human
health effects, the provision allows listing of pollutants ``including,
but not limited to, substances which are known to be, or may reasonably
be anticipated to be, carcinogenic, mutagenic, teratogenic, neurotoxic,
which cause reproductive dysfunction, or which are acutely or
chronically toxic.'' An adverse environmental effect is defined as
``any significant and widespread adverse effect, which may reasonably
be anticipated, to wildlife, aquatic life, or other natural resources,
including adverse impacts on populations of endangered or threatened
species or significant degradation of environmental quality over broad
areas.'' Section 112(b)(2) provides that ``no substance, practice,
process or activity regulated under [the Clean Air Act's stratospheric
ozone protection program] shall be subject to regulation under this
section solely due to its adverse effects on the environment.'' Thus,
section 112 may not be used to regulate certain chlorofluorocarbons and
other ozone-depleting substances, their sources, or activities related
to their production and use to address climate change unless we
establish that such regulations are necessary to address human health
effects in addition to any adverse environmental impacts. See section
602 of the Clean Air Act for a partial list of these substances.
    Section 112(b)(3) of the Act establishes general requirements for
petitioning EPA to modify the HAP list by adding or deleting a
substance. Although the Administrator may add or delete a substance on
his own initiative, if a party petitions the Agency to add or delete a
substance, the burden historically has been on the petitioner to
include sufficient information to support the requested addition or
deletion under the substantive criteria set forth in CAA section
112(b)(3)(B) and (C). The Administrator must either grant or deny a
petition within 18 months of receipt of a complete petition.
    The effects and findings described in section 112 are different
from other sections of the CAA addressing endangerment of public health
discussed in previous sections of today's notice. Given the nature of
the effects identified in section 112(b)(2), we request comment on
whether the health and environmental effects attributable to GHG fall
within the scope of this section. We also request comment on direct and
indirect GHG emissions from existing source categories currently
subject to regulation under section 112, any assessment of the relative
costs of regulating GHG under the authority of section 112, and any co-
benefits or co-detriments with regard to controlling GHG and the
emissions of HAP.

[[Page 44494]]

    The source categories to be regulated under section 112 are
determined based on the list of HAP. Section 112(c) requires EPA to
publish a list of all categories and subcategories of major sources of
one or more of the listed pollutants, and to periodically review and
update that list. In doing this, EPA also is required to list each
category or subcategory of area sources which the Administrator finds
presents a threat of adverse effects to human health or the environment
(by such sources individually or in the aggregate) warranting
regulation under section 112.
c. How Is MACT Determined?
    In essence, MACT standards are intended to ensure that all major
sources of HAP emissions achieve the level of control already being
achieved by the better controlled and lower emitting sources in each
category. This approach provides assurance to citizens that each major
source of toxic air pollution will be required to effectively control
its emissions. At the same time, this approach provides assurances that
facilities that employ cleaner processes and good emissions controls
are not disadvantaged relative to competitors with poorer controls.
    MACT is determined separately for new and existing sources. For
existing sources, MACT standards must be at least as stringent as the
average emissions limitation achieved by the best performing 12 percent
of sources in the category or subcategory (or the best performing five
sources for source categories with less than 30 sources). This level is
called the ``MACT floor.'' For new or reconstructed sources, MACT
standards must be at least as stringent as the control level achieved
in practice by the best controlled similar source.\252\ EPA also must
consider more stringent ``beyond-the-floor'' control options for MACT.
When considering beyond-the-floor options, EPA must consider not only
the maximum degree of reduction in emissions of the HAP, but also
costs, energy requirements and non-air quality health environmental
impacts of imposing such requirements.
---------------------------------------------------------------------------

    \252\ See CAA section 112(d)(3).
---------------------------------------------------------------------------

    MACT standards may require the application of measures, processes,
methods, systems, or techniques including, but not limited to, (1)
reducing the volume of, or eliminating emissions of, such pollutants
through process changes, substitution of materials, or other
modifications; (2) enclosing systems or processes to eliminate
emissions; (3) collecting, capturing, or treating such pollutants when
released from a process, stack, storage or fugitive emissions point;
(4) design, equipment, work practice, or operational standards
(including requirements for operator training or certification) as
provided in subsection (h); or (5) a combination of the above. (See
section 112(d)(2) of the Act.)
    For area sources, CAA section 112(d)(5) provides that the standards
may reflect generally available control technology or management
practices (GACT) in lieu of MACT.
d. What Is Required To Address Any Residual Risk?
    Section 112(f)(2) of the CAA requires us to determine for each
section 112(d) source category whether the MACT standards protect
public health with an ample margin of safety. If the MACT standards for
a HAP ``classified as a known, probable, or possible human carcinogen
do not reduce lifetime excess cancer risks to the individual most
exposed to emissions from a source in the category or subcategory to
less than 1-in-1-million,'' EPA must promulgate residual risk standards
for the source category (or subcategory) as necessary to protect public
health with an ample margin of safety. EPA must also adopt more
stringent standards if needed to prevent an adverse environmental
effect, but must consider cost, energy, safety, and other relevant
factors in doing so. EPA solicits comments on the extent to which these
programs could apply with respect to the possible regulation of sources
of GHG under section 112, including the relevance of any carcinogenic
effects of individual GHG.
2. What Sources Would Be Affected if GHGs Were Regulated Under This
Authority?
    If GHGs were listed as HAP, EPA would be required to regulate a
very large number of new and existing stationary sources, including
smaller sources than if alternative CAA authorities were used to
regulate GHG. This is the result of three key requirements. First, the
section 112(a) major sources thresholds of 10 tons for a single HAP and
25 for any combination of HAPs would mean that very small GHG emitters
would be considered major sources. Second, section 112(c) requires EPA
to list all categories of major sources. Third, section 112(d) requires
EPA to issue MACT standards for all listed categories.
    We believe that most significant stationary source categories of
GHG emissions have already been listed under section 112 (although the
10-ton threshold in the case of GHGs would be expected to bring in
additional categories such as furnaces in buildings, as explained
below). To date we have adopted standards for over 170 categories and
subcategories of major and area sources. This is a significantly
greater number than the categories for which we have adopted NSPS
because under section 112 we must establish standards for all listed
categories, whereas section 111 requires that we identify and regulate
only those source categories that contribute ``significantly'' to air
pollution endangering public health and welfare.
3. What Are the Key Milestones and Expected Timeline if Section 112
Were Used for GHG Controls?
    One possible timetable for addressing GHG under this part of the
Act would be to incorporate GHG emission control requirements
concurrent with the mandatory 8-year technology reviews for each
category, collecting information on emissions and control technologies
at the time the existing MACT standards are reviewed to determine
whether revisions are needed. If we were to list new source categories
under section 112, EPA would be required to adopt MACT standards for
those categories within 2 years of the date of category listing.
    EPA must require existing sources to comply within 3 years of a
standard's promulgation, although states and EPA are authorized in
certain circumstances to extend the period of compliance by one
additional year. Most new sources must comply as soon as a section 112
standard is issued; however, there is an exception where the final rule
is more stringent than the proposal.
    Because of the more detailed requirements for identifying
appropriate levels of control to establish a level for MACT,
significantly more information on the best performing sources is needed
under section 112 than under section 111, making the development of
such standards within 2 years after listing a source category
difficult. We request comment on this and other approaches for
addressing GHG under section 112, both for categories already listed
for regulation and for any that might appropriately be added to the
section 112 source category list if we were to elect to regulate GHGs
under this section.
4. What Are the Key Considerations Regarding Use of This Authority for
GHGs (and How Could Potential Issues Be Addressed)?
    A key consideration in evaluating use of section 112 for GHG
regulation is that

[[Page 44495]]

the statutory provisions appear to allow EPA little flexibility
regarding either the source categories to be regulated or the size of
sources to regulate. As described above, EPA would be required to
regulate a very large number of new and existing stationary sources,
including smaller sources than if alternative CAA authorities were used
to regulate GHG. For example, in calculating CO2 emissions based on
fossil-fuel consumption, we believe that small commercial or
institutional establishments and facilities with natural gas-fired
furnaces would exceed this major source threshold; indeed, a large
single-family residence could exceed this threshold if all appliances
consumed natural gas. EPA requests comment on the requirement to
establish standards for all sources under section 112 relevant to GHG
emissions and whether any statutory flexibility is or is not available
with respect to this requirement and GHGs.
    A section 112 approach for GHGs would require EPA to issue a large
number of standards based on assessments for each source category.
Determining MACT based on the best-controlled 12 percent of similar
sources for each category would present a difficult challenge, owing to
our current lack of information about GHG control by such sources and
the effort required to obtain sufficient information to establish a
permissible level of performance.
    GHG regulation under section 112 would likely be less cost
effective than under some CAA authorities, in part because section 112
was designed to ensure a MACT level of control by each major source,
and thus provides little flexibility for market-oriented approaches.
Given the structure and past implementation of section 112, this
section may not provide EPA with authority to allow emissions trading
among facilities or averaging across emitting equipment in different
source categories. This is because the statutory terms of section 112
provide that emission standards must be established for sources within
``each category'' and those standards must be no less stringent than
the ``floor,'' or the level of performance achieved by the best-
performing sources within that category. Each source in the category
must then achieve control at least to this floor level. Trading would
allow sources to emit above the floor. In addition, it may not be
possible to assess individual source fence line risk for section 112(f)
residual risk purposes if the sources did not each have fixed limits.
Finally, the section 112 program is in part designed to protect the
population in the vicinity of each facility, which trading could
undermine (in contrast to an ambient standard). Given the global nature
of GHGs and the lack of direct health effects from such emissions at
ambient levels, EPA requests comments on the extent to which the CAA
could be interpreted to grant flexibility to consider such alternative
implementation mechanisms, and what, if any, limitations should be
considered appropriate in conjunction with them.
    Another reason that section 112 regulation of GHGs would be
expected to be less cost effective than other approaches is that the
statute limits consideration of cost in setting MACT standards. As
described above, the statute sets minimum stringency levels, or
``floors,'' for new and existing source standards. Cost can only be
considered in determining whether to require standards to be more
stringent than the floor level.
    A further consideration is that the short compliance timetables--
immediate for most new sources, and within 3-4 years for existing
sources--appear to preclude setting longer compliance timeframes to allow 
for emerging GHG technologies to be further developed or commercialized.
5. What Are the Possible Implications for Other Provisions of the Clean
Air Act?
    As provided under section 112(b)(6), pollutants regulated under
section 112 of the Act are exempt from regulation under the PSD
program. Also, a section 111(d) standard of performance for existing
sources cannot be established to address any HAP listed under section
112(b) that that is emitted from a source category regulated under
section 112.\253\
---------------------------------------------------------------------------

    \253\ It is important to note that many sources may be subject
to standards under both section 111 and 112; however these standards
establish requirements for the control of different pollutants.
---------------------------------------------------------------------------

    If EPA were to list GHGs under section 108 of the CAA for purposes
of establishing NAAQS, we would be prevented by section 112(b)(2) from
listing and regulating them as HAPs under this section of the Act.
However, it is less clear that the reverse is true; that is, if a
pollutant were first listed under section 112 and then EPA decided to
list and regulate it under section 108, the statute does not clearly
say whether that is permissible, or whether EPA would then have to
remove the pollutant from the section 112 pollutant list. We request
comment on the extent to which this apparent ambiguity in the Act poses
an issue regarding possible avenues for regulating GHG and if so, how
it should be addressed.
    In light of the foregoing, we request comment on the
appropriateness of section 112 as a mechanism for regulating stationary
source emissions of GHGs under the CAA. If commenters believe use of
section 112 would be appropriate, we further request comments on which
GHGs should be considered, what additional sources of emissions should
be listed and regulated, and how MACT should be determined for GHG
emission sources.

D. Solid Waste Combustion Standards

1. What Does Section 129 Require?
    Section 129 of the CAA requires EPA to set performance standards
under section 111 to control emissions from solid waste incineration
units of at least 9 specific air pollutants. It directs EPA to develop
standards which include emission limitations and other requirements for
new units and guidelines and other requirements applicable to existing
units.
    Section 129 directs EPA to set standards for ``each category'' of
such units, including those that combust municipal, hospital, medical,
infectious, commercial, or industrial waste, and ``other categories''
of solid waste incineration units, irrespective of size. The pollutants
to be addressed by these standards include the NAAQS pollutants
particulate matter (total and fine), sulfur dioxide, oxides of
nitrogen, carbon monoxide, and lead; and the hazardous air pollutants
hydrogen chloride, cadmium, mercury, and dioxins and dibenzofurans. EPA
is authorized to regulate additional pollutants under these provisions,
but section 129 includes no endangerment test or other criteria for
determining when it is appropriate to do so.
    Although the emission standards called for by section 129 are to be
established pursuant to section 111, the degree of control required
under those standards more closely resembles that of section 112(d).
For new sources the level of control is required to be no less
stringent than that of the best performing similar source, while for
existing sources the level of control is to be no less stringent than
the average of the top 12% of best-performing sources. For both new and
existing source standards, beyond these ``floor'' levels EPA must
consider the cost of achieving resulting emission reductions and any
non-air quality health and environmental impacts and energy
requirements in determining what is achievable for units within each
category. The performance standards must be reviewed every 5 years.
Additionally, for those pollutants that

[[Page 44496]]

are listed under section 112 as a HAP, EPA must reevaluate the
standards in accordance with section 112(f) to determine whether they
are sufficient to protect public health with an ample margin of safety
and prevent adverse environmental effects, and must promulgate more
stringent regulations if necessary to address any such ``residual
risk.'' Thus, for this particular class of source categories, section
129 merges important elements of both sections 111 and 112.
    EPA has established standards for a variety of solid waste
incinerator categories and is in the process of developing additional
standards and revising others.\254\ In the absence of statutory
criteria for determining whether and under what circumstances EPA
should regulate additional pollutants under this section of the CAA, we
request comment on whether emissions of GHG could fall within the scope
of this section. We also request comment on direct and indirect GHG
emissions from existing source categories currently subject to
regulation under section 129, any assessment of the relative costs of
regulating GHGs under the authority of section 129, and any co-benefits
or co-detriments with regard to controlling GHG and the emissions of
pollutants specifically listed for regulation under section 129.
---------------------------------------------------------------------------

    \254\ Rules have been promulgated for large and small municipal
waste combustors; medical waste incinerators; other solid waste
incinerators; and commercial, institutional, and industrial solid
waste incinerators. EPA is also currently reevaluating and revising
certain standards under section 129 in response to decisions by the
U.S. Court of Appeals for the D.C. Circuit.
---------------------------------------------------------------------------

2. What Sources Would Be Affected if GHGs Were Regulated Under This
Authority?
    Standards required by section 129 are applicable to ``any facility
which combusts any solid waste material from commercial or industrial
establishments or the general public (including single and multiple
residences, hotels, and motels).'' Thus the provisions of this section
are limited to a specific type of emission source, although there are
many such units in existence that are subject to regulation. To date we
have adopted standards for five categories of incinerators and are
currently in the process of developing revised standards on remand for
several of these categories, which may involve the inclusion of several
additional subcategories of incineration units. We anticipate that when
completed these rules will establish standards of performance for as
many as five hundred or more units.
    Because section 129 does not require, but authorizes EPA to
establish requirements for other air pollutants, we request comment on
whether and for what categories or subcategories of incinerators EPA
could address GHG emissions control requirements.
a. How Are Control Requirements Determined?
    As noted above, the control requirements for sources regulated
under section 129 are similar to the MACT standards mandated under
section 112(d). However, whereas section 112(d)(3) provides that
standards are to be based on the best performing sources ``for which
the Administrator has emissions information,'' section 129 contains no
such limitation. Consequently, it appears that EPA is obligated to
obtain information from all potentially affected sources in order to
determine the appropriate level of control.
    Section 129(a)(2) provides authority for EPA to distinguish among
classes, types, and sizes of units within a category in establishing
standards. This provision is similar to authorities provided in
sections 111( b)(2) and 112(b)(2). Because section 129 directs that EPA
establish standards for affected source categories under sections
111(b) and (d), we believe that the provisions governing the creation
of design, equipment, work practice, or operational standards are also
available for standards required by section 129. For existing sources,
we believe that provisions for consideration of remaining useful life
and other related factors are relevant to EPA and States when
determining the requirements and schedules for compliance for
individual affected sources.
b. What Is Required To Address Any Residual Risk?
    For each of the air pollutants named in section 129 that are listed
as HAP under section 112, section 129 requires EPA to evaluate and
address any residual risk remaining after controls established under
the initial emission standards.\255\ In so doing, it requires EPA to
determine for each affected source category whether the performance
standards protect public health with an ample margin of safety. EPA
must also adopt more stringent standards if needed to prevent an
adverse environmental effect, but must consider cost, energy, safety,
and other relevant factors in doing so.
---------------------------------------------------------------------------

    \255\ Section 129(h)(3) provides that for purposes of
considering residual risk the standards under section 129(a) and
section 111 applicable to categories of solid waste incineration
units are to be ``deemed standards under section 112(d)(2).''
---------------------------------------------------------------------------

    Section 129(h)(3) limits residual risk assessments and any
subsequent resulting regulations to ``the pollutants listed under
subsection (a)(4) of this section and no others.'' Consequently, if EPA
were to regulated GHG emissions from incineration units under section
129, we would not be required to conduct additional residual risk
determinations.
3. What Are the Key Milestones and Expected Timeline if Section 129
Were Used for GHG Controls?
    As stated above, we have adopted rules governing emissions from
certain categories of solid waste incineration units and are in the
process of revising or establishing new standards for others. Thus if
we were to elect to regulate GHG emissions under section 129, a
question arises concerning how to incorporate new requirements for
those categories for which standards have already been established. One
possible timetable for addressing GHG under this part of the Act would
be to incorporate GHG emission control requirements concurrent with the
mandatory 5-year reviews for each previously-regulated category,
collecting information on emissions and control technologies at the
time the existing standards are reviewed to determine whether revisions
are needed. Because of the more detailed requirements for identifying
appropriate levels of control to establish a level for these categories
of sources, significantly more information on the best performing
sources is needed under section 129 than even under section 112
(because of the absence of limitations for this analysis to those
sources ``for which the Administrator has information''), making the
development of such standards a more time-consuming effort. In the
event that we were to elect to regulate GHGd under this section, we
request comment on this and other approaches for addressing GHGd under
section 129, both for categories already regulated and for any for
which standards are currently under development.
4. What Are the Key Considerations Regarding Use of This Authority for
GHGs (and How Could Potential Issues Be Addressed)?
    If we were to elect to regulate GHG emissions from solid waste
incinerators under section 129, then we would need to establish
standards for at least some number of categories of such sources. We
request comment on the availability of authority to establish requirements

[[Page 44497]]

for controlling GHG emissions from subcategories of incineration units
based on size, type or class, as provided under section 111, and to
exclude from regulation other categories or subcategories.
    Given the structure of section 129 and its hybrid approach to the
use of authorities under sections 111 and 112, we question whether this
section provides EPA with available authority to establish alternative
compliance approaches, such as emissions trading or averaging across
sources within a category. This is because the statutory terms of
section 129 provide that emission standards must be established for
sources within ``each category'' and those standards must be no less
stringent than the level of performance achieved by the best-performing
sources within that category. Each source in the category must then
achieve control at least to this level. Trading would allow sources to
emit above the floor. As a practical matter, given that requirements
for control of specifically-listed pollutants may preclude trading for
those pollutants, and given that many of the controls applicable to
those pollutants would be the same as or similar to those that would be
applicable to GHGs, we believe that trading options would likely be
infeasible with respect to GHG control requirements. However, EPA
requests comments on the extent to which the CAA could be interpreted
to grant flexibility to consider such alternative implementation
mechanisms, to what extent, and what, if any, limitations should be
considered appropriate in conjunction with them.
5. What Are the Possible Implications for Other Provisions of the Clean
Air Act?
    Section 129 recognizes that many incineration units may also be
subject to prevention of significant deterioration or nonattainment new
source review requirements. It addresses potentially conflicting
outcomes of control determinations under those programs by providing
that ``no requirement of an applicable implementation plan . . . may be
used to weaken the standards in effect under this section.''
    If EPA were to list GHGs under section 108 for purposes of
establishing NAAQS, we would not be prevented from regulating them
under this section of the Act as well. If EPA were to list GHG under
section 112, a potential conflict arises in that section 112
establishes major and area source emissions thresholds, providing for
standards of different stringency for each, and requires analysis of
residual risk for major sources regulated under that section of the
Act. We request comments on how such apparent conflicts could be
reconciled if we were to elect to regulate emissions of GHGs from solid
waste incineration units under section 129.
    In light of the foregoing, we request comment on the
appropriateness of section 129 as a mechanism for regulating
incineration unit emissions of GHGs under the CAA. If commenters
believe that use of section 129 would be appropriate, we further
request comments on which GHGs should be considered, what source
categories or subcategories should be regulated, and how appropriate
control requirements should be determined for new and existing GHG
emission sources.

E. Preconstruction Permits Under the New Source Review (NSR) Program

1. What Are the Clean Air Act Provisions Describing the NSR Program?
    Under what is known as the New Source Review (NSR) program, the CAA
requires the owners and operators of large stationary sources of air
pollution to obtain construction permits prior to building or modifying
such a facility. The program is subdivided into the Prevention of
Significant Deterioration (PSD) and nonattainment NSR (NNSR) programs,
either of which may be applicable depending on the air quality for a
particular pollutant in the location of the source subject to permitting.
    The PSD program, set forth in Part C of Title I of the CAA, applies
in areas that are in attainment with the NAAQS (or are unclassifiable)
and has the following five goals and purposes:
     To protect public health and welfare from air pollution
beyond that which is addressed by the attainment and maintenance of NAAQS;
     To protect specially designated areas such as national
parks and wilderness areas from the effects of air pollution;
     To assure that economic growth will occur in a manner
consistent with the preservation of existing clean air resources;
     To assure emissions in one state will not interfere with
another state's PSD plan; and
     To assure that any decision to permit increased air
pollution is made only after evaluating the consequences of the
decision and after opportunities for informed public participation.
    The main element of the PSD program is the requirement that a PSD
permit be obtained prior to construction of any new ``major emitting
facility'' or any new ``major modification.'' Before a source can
receive approval to construct under PSD, the source and its permitting
authority (usually a state or local air pollution control agency, but
sometimes EPA) must follow certain procedural steps, and the permit
must contain certain substantive requirements. The most important
procedural step is providing an opportunity for the public to comment
when a permitting authority proposes to issue a permit.
    The PSD program primarily applies to all pollutants for which a
NAAQS is promulgated, but some of the substantive requirements of the
PSD program also apply to regulated pollutants for which there is no
NAAQS (except that there is an explicit statutory exemption from PSD
for HAPs).\256\ Since there is currently no NAAQS for GHGs and GHGs are
not otherwise subject to regulation under the CAA, the PSD program is
not currently applicable to GHGs.\257\ However, as discussed in section
IV of this notice, it is possible that EPA actions under other parts of
the CAA could make GHGs pollutants subject to regulation under the Act
and thus subject to one or more parts of the PSD program.
---------------------------------------------------------------------------

    \256\ CAA section 112(b)(6).
    \257\ In the Energy Independence and Security Act of 2007
(EISA), Congress provided that regulation of GHGs under CAA section
211(o) would not automatically result in regulation of GHGs under
other CAA provisions. Because of this provision, EISA does not
impact the interrelationship of other provisions of the CAA, and we
only reference the HAP exception in the text.
---------------------------------------------------------------------------

    If EPA were to promulgate a rule establishing limitations on GHG
emissions from mobile sources or stationary sources without
promulgating a NAAQS for GHGs, the PSD requirement of greatest
relevance would be the requirement that a permit contain emissions
limits that reflect the Best Available Control Technology (BACT). BACT
is defined as the maximum achievable degree of emissions reduction for
a given pollutant (determined by the permitting authority on a case-by-
case basis), taking into account energy, environmental, and economic
impacts. BACT may include add-on controls, but also includes
application of inherently lower-polluting production processes and
other available methods and techniques for control. BACT cannot be less
stringent than any applicable NSPS.
    Since emission control requirements will likely have the most
direct impact on new or modified stationary sources subject to PSD, our
focus in this notice is on the BACT requirement. However, we are also
interested in stakeholder input on the extent to which we should

[[Page 44498]]

evaluate other substantive PSD program elements which would be affected
by any possible EPA action to regulate GHGs under other parts of the
Act. These include the requirements to evaluate, in consultation with
the appropriate Federal Land Manager (FLM), the potential impact of
proposed construction on the Air Quality Related Values of any affected
``Class I area'' (national parks, wilderness areas, etc.) and
additional impacts analysis.\258\
---------------------------------------------------------------------------

    \258\ As codified at 40 CFR 51.166(o), the owner or operator
shall provide an analysis of the impairment to visibility, soils,
and vegetation that would occur as a result of the source or
modification and general commercial, residential, industrial, and
other growth associated with the source or modification.
---------------------------------------------------------------------------

    If EPA were to promulgate a NAAQS for GHGs, because of the
relatively uniform concentration of GHGs, we expect that the entire
country would be in nonattainment or attainment of the NAAQS. The
preconstruction permitting requirements that apply would depend on
whether the country is designated as nonattainment or attainment for
the GHG emissions that would increase as a result of a project being
constructed.
    If the entire country is designated attainment, and PSD applies,
the adoption of a NAAQS would trigger air quality analysis requirements
that are in addition to all the requirements described above. For
example, under CAA section 165(a)(3), permit applicants have to conduct
modeling to determine whether they cause or contribute to a NAAQS
violation. Following promulgation of a NAAQS, EPA may also promulgate a
PSD increment for GHGs, which would require additional analysis for
each new and modified source subject to PSD.\259\ However, this notice
does not address in detail the PSD elements that relate to increments.
---------------------------------------------------------------------------

    \259\ PSD increments are air quality levels which represent an
allowable deterioration in air quality as compared to the existing
air quality level on a certain baseline date for a given area.
---------------------------------------------------------------------------

    Under a GHG NAAQS with the country in nonattainment, the
nonattainment NSR permitting program would be triggered nationally. The
nonattainment NSR program requirements are contained in section 173 of
the Act. Like PSD, they apply to new and modified major stationary
sources, but they contain significantly different requirements from the
PSD program. A key difference is the requirement that the emissions
increases from the new or modified source in a nonattainment area must
be offset by reductions in existing emissions from the same
nonattainment area or a contributing upwind nonattainment area of equal
or higher nonattainment classification. The offsetting emissions
reductions must be at least equal to the proposed increase and must be
consistent with a SIP that assures the nonattainment area is making
reasonable progress toward attainment.\260\ Another key difference is
that instead of BACT, sources subject to nonattainment NSR must comply
with the Lowest Achievable Emission Rate (LAER), which is the most
stringent emission limitation that is (1) contained in any SIP for that
type of source, or (2) achieved in practice for sources of the same
type as the proposed source.\261\ Notably, if the rate is achievable,
LAER does not allow for consideration of costs or of the other factors
that BACT does. While LAER and offsets are likely of greatest
significance for GHG regulation under nonattainment NSR, there are
additional requirements for nonattainment NSR that would also apply.
The additional requirements include the alternatives analysis
requirement; the requirement that source owners and operators
demonstrate statewide compliance with the Act; and the prohibition
against permit issuance if the SIP is not being adequately implemented.
---------------------------------------------------------------------------

    \260\ CAA section 173(a)(1); limitations on offsets are set
forth in section 173(c).
    \261\ CAA section 173(a); LAER is defined in section 171(3)(A).
---------------------------------------------------------------------------

    For simplicity, the remainder of this notice describing affected
sources, impacts, and possible tailoring generally focuses on PSD,
raising issues specific to nonattainment NSR where applicable.
2. What Sources Would Be Affected if GHGs Were Regulated Under NSR?
    A PSD permit is required for the construction or modification of
``major emitting facilities,'' which are commonly referred to as
``major sources.'' A ``major emitting facility'' is generally any
source that emits or has the potential to emit 250 tons per year (tpy)
of a regulated NSR pollutant.\262\ \263\ A source that belongs to one
of several specifically identified source categories is considered a
major source if it emits or has the potential to emit 100 tpy of a
regulated NSR pollutant.\264\ Also, for nonattainment NSR, the major
source threshold is at most 100 tpy, and is less in some nonattainment
areas, depending on the pollutant and the nonattainment classification.
---------------------------------------------------------------------------

    \262\ 42 U.S.C. 7569(1). The PSD regulations use the term
``major stationary source.'' 40 CFR 51.166(b)(1) The definition of
``regulated NSR pollutant'' is at 40 CFR 51.166(b)(49).
    \263\ ``Potential-to-emit'', or PTE, is defined as the maximum
capacity of a source to emit any air pollutant under its physical
and operational design.
    \264\ These specific sources include major industrial categories
such as petroleum refining, fossil-fuel fired steam electric plants,
chemical process plants, and 24 other categories. The full list of
100 tpy major sources is promulgated at 40 CFR 51.166(b)(1)(i)(a).
---------------------------------------------------------------------------

    A ``major modification'' is any physical change or change in the
method of operation of a major source which significantly increases the
amount of emissions of any regulated NSR pollutant. EPA defines what
emissions levels of a pollutant are ``significant'' through regulation,
and the defined significance levels range from 0.3 tpy for lead to 100
tpy for CO. Currently there is no defined significance level for GHGs
(either individually or as a group) because they are not regulated NSR
pollutants, and thus, were GHGs to become regulated, the significance
threshold would be zero. Note that, when determining whether a facility
is ``major,'' a source need not count fugitive emissions (i.e.,
emissions which may not reasonably be vented through stacks, vents,
etc.) unless it is in a listed category.
    As noted in section IV, GHGs are not currently subject to
regulation under the Act, and therefore are not regulated NSR
pollutants. However, if GHG emissions become subject to regulation
under any of the stationary or mobile source authorities discussed
above (except sections 112 and 211(o)), GHGs could become regulated NSR
pollutants. Many types of new GHG sources and GHG-increasing
modifications that have not heretofore been subject to PSD would become
subject to PSD permitting requirements. This is particularly true for
CO₂ because, as noted in section III, the mass
CO₂ emissions from many source types are orders of magnitude
greater than for currently regulated pollutants. Thus, many types of
new small fuel-combusting equipment could become newly subject to the
PSD program if CO₂ becomes a regulated NSR pollutant. As
discussed below in the section on potential to emit, the extent to
which such equipment would become subject to PSD would depend upon
whether, for each type of equipment, its maximum capacity considering
its physical and operational design would involve constant year-round
operation or some lesser amount of operation. For example, the
calculated size of a natural gas-fired furnace that has a potential to
emit 250 tpy of CO₂, if year-round operation (8760 hours per
year) were assumed--would be only 0.49 MMBTU/hr, which is comparable to
the size of a very small commercial furnace. In practice, a furnace
like this would likely operate far less than year round and its actual
emissions would be well below 250 tpy. For example, such a furnace, if
used for

[[Page 44499]]

space heating, might only be burning gas for about 1000 hours per year,
meaning that it would need to be sized at over 4 MMBTU/hr--a size more
comparable to a small industrial furnace--to actually emit 250 tons of
CO₂. For sources such as these, the interpretation of the
term ``potential to emit'' and the availability of streamlined
mechanisms for smaller sources to limit their potential to emit would
determine whether they would be considered ``major'' for GHG emissions
under PSD.
    For sources already major for other pollutants, it is likely that
many more changes made by the source would also qualify as major
modifications and become subject to PSD as well, unless potential
approaches (including those discussed below) for raising applicability
thresholds were implemented. Relatively small changes in energy use
that cause criteria pollutant emissions too small to trigger PSD would
newly trigger PSD at such facilities because such changes would likely
result in greater CO₂ increases. For example, consider a
hypothetical 500 MW electric utility boiler firing a bituminous coal
that is well-controlled for traditional pollutants. Such a boiler,
operating more than 7000 hours per year (out of a possible 8760), can
emit approximately 4 million tons of CO₂ per year, or more
than 580 tons per hour. Assuming a 100 tpy significance level (rather
than the current zero level for GHGs), any change resulting in just 10
additional minutes of utilization over the course of a year at such a
source would be enough to result in an increase of 100 tons and
potentially subject the change to PSD. By contrast, to be considered a
modification for NO_X, the same change would require
approximately 36 additional hours of operation assuming that the
hypothetical source had a low-NO_X burner, and 90 additional
hours of operation assuming that the source also employed a selective
catalytic reduction add-on control device.
    Once a source is major for any NSR regulated pollutant, PSD applies
to significant increases of any other regulated pollutant, so
significant increases of GHGs would become newly subject to PSD at
sources that are now major for other regulated pollutants. Similarly,
significant increases of other pollutants would become subject to PSD
if they occur at sources previously considered minor, but which become
classified as major sources for GHG emissions.
    Currently, EPA estimates that EPA, state, and local permitting
authorities issue approximately 200-300 PSD permits nationally each
year for construction of new major sources and major modifications at
existing major sources. Under existing major source thresholds, we
estimate that if CO₂ becomes a regulated NSR pollutant
(either as an individual GHG or as a group of GHGs), the number of PSD
permits required to be issued each year would increase by more than a
factor of 10 (i.e. more than 2000-3000 permits per year), unless action
were taken to limit the scope of the PSD program under one or more of
the legal theories described below. The additional permits would
generally be issued to smaller industrial sources, as well as large
office and residential buildings, hotels, large retail establishments,
and similar facilities. These facilities consist primarily of equipment
that combusts fuels of various kinds and release their exhaust gases
through a stack or vent. Few of these additional permits would be for
source categories (such as agriculture) where emissions are
``fugitive,'' because, as noted above, fugitive emissions do not count
toward determining if a source is a major source except in a limited
number of categories of large sources.
    Because EPA and states have generally not collected emissions
information on sources this small, our estimate of the number of
additional permits relies on limited available information and
engineering judgment, and is uncertain. Our estimate of the number of
additional permits is also not comprehensive. First, it does not
include permits that would be required for modifications to existing
major GHG sources because the number of these is more difficult to
estimate.\265\ Nonetheless, we anticipate that, for modifications,
coverage of GHGs would increase because the larger universe of major
sources will bring in additional sources at which modifications could
occur and because for ``traditional'' major sources, many more types of
small modifications that were minor for traditional pollutants could
become major due to increases in GHG emissions that exceed the
significance levels. Second, EPA's estimate is uncertain because it is
based on actual emissions, and thus excludes a potentially very large
number of sources that would be major if they operated at their full
potential-to-emit (PTE) (i.e. they emitted at a level that reflects the
maximum capacity to emit under their physical and operational design),
but which in practice do not. Such sources could be defined as major
sources without an enforceable limitation on their PTE, but for the
purposes of this estimate, we assume they have options for limiting
their PTE and avoiding classification as a major source. (Nonetheless,
there are important considerations in creating such PTE limits, as
discussed below). Third, this estimate does not specifically account
for CO₂ from sources other than combustion sources. While we
know there are sources with significant non-combustion emissions of
GHGs, there are relatively few of these compared to the sources with
major amounts of combustion CO₂. These non-combustion
sources would likely be major for combustion CO₂ in any
event, and many of these are likely already major for other pollutants,
though GHG regulation would likely mean increases in the number of
major modifications at such sources.
---------------------------------------------------------------------------

    \265\ Among other things, any estimate of modifications must
take into account the netting provisions of NSR, in which sources
can avoid NSR if the increase of pollutant emissions from a project
is below the significance level for that pollutant, after taking
into account other increases and decreases of emissions that are
contemporaneous with the project.
---------------------------------------------------------------------------

    We request any available information that would allow us to better
characterize the number and types of sources and modifications that
would become subject to the PSD program if CO₂ becomes a
regulated NSR pollutant. As discussed below, we are particularly
interested in information that would allow us to analyze the effects of
different major source thresholds and significance levels.
    Finally, we note that our estimates above are for CO₂.
As described above in section IV, there are implications to regulating
additional GHGs as pollutants, or GHGs in the aggregate. Our estimates
of PSD program impacts do not include consideration of GHGs other than
CO₂ because we expect that at the vast majority of these
sources CO₂ will be the dominant pollutant. We ask for
comment on whether there are large categories of potentially newly
regulated PSD sources for individual GHGs besides CO₂. We
also ask for comment on the effects of aggregating GHGs for PSD
applicability. Aggregating GHGs could bring additional sources into PSD
to the extent that other GHGs are present and would add enough to a
source's PTE to make it a major source. On the other hand, under the
netting provisions of the CAA, it may be easier to facilitate
interpollutant netting if GHGs are aggregated (e.g., a source using
netting to avoid PSD for a CO₂ increase based on methane
decreases at the same source).

[[Page 44500]]

3. What Are the Key Milestones and Expected Timeline if the PSD Program
Were Used for GHG Controls?
    Because PSD applies to all regulated pollutants except HAP, EPA's
interpretation of the Act is that PSD program requirements would become
applicable immediately upon the effective date of the first regulation
requiring GHG control under the Act.\266\ While existing PSD permits
would remain unaffected, from that point forward, each new major source
of GHGs and each major modification at an existing major source that
significantly increases GHGs would need to get a PSD permit before
beginning construction. Control requirements could take effect as the
first new and modified sources obtain their permits and complete
construction of the permitted projects. Because of the case-by-case
nature of the PSD permitting decisions, the complexity of the PSD
permitting requirements, and the time needed to complete the PSD
permitting process, it can take several months to receive a simple PSD
permit, and more than a year to receive a permit for a complex
facility. We ask for comment on whether there are additional timeline
considerations not noted here.
---------------------------------------------------------------------------

    \266\ Because PSD is implemented in many areas by states under
EPA-approved state regulations, there may be a lag time in a small
number of states if their PSD regulations are written in such a way
that revision of the regulations (and EPA approval) would be
required to give the state authority to issue permits for GHGs.
However this would not be the case for EPA's own regulations or for
any state delegated to implement EPA regulations on our behalf.
---------------------------------------------------------------------------

4. What Are Key Considerations Regarding Application of the PSD Program
to GHGs (and How Could Potential Issues Be Addressed?)
a. Program Scope
    As noted above, regulating GHGs under the PSD program has the
potential to dramatically expand the number of sources required to
obtain PSD permits, unless action is taken to limit the scope of the
program, as described below. Since major source thresholds were enacted
before this assessment of the application of the PSD program to GHGs,
it is reasonable to expect that Congress could consider legislative
alterations to account for the different aspects of GHGs versus
traditional air pollutants noted above (e.g., the relatively uniform
atmospheric concentrations of GHGs versus more localized effects of
traditional pollutants.) Possible ways to limit the scope of the
program without legislation are described later in this section.
    In the absence of such action, we would expect (assuming a 250 tpy
major source threshold, or 100 tpy for statutorily specified source
categories) at least an order-of-magnitude increase in the number of
new sources required to obtain PSD permits, and an expansion of the
program to numerous smaller sources not previously subject to it. While
such sources may emit amounts of GHGs that exceed statutory thresholds,
they have relatively small emissions of non-GHG pollutants (such that
they have not been regulated under PSD, and many have not been
regulated under any CAA program).\267\ Regulating GHGs under the PSD
program would also cause a large increase in the number of
modifications at existing sources that would be required to obtain PSD
permits. Such modifications may occur at existing sources that have
been long regulated as major for other pollutants, or at existing
sources that become classified as major solely due to their GHG emissions.
---------------------------------------------------------------------------

    \267\ Some fraction of these small sources are regulated, at
least in some areas, by SIPs and state minor source permit programs
under section 110 of the CAA.
---------------------------------------------------------------------------

    Permitting smaller sources and modifications is generally less
effective due to the fact that, while there are still administrative
costs borne by the source and permitting authority, the environmental
benefit of each permit is generally less than what results from
permitting a larger source. Congress excluded smaller sources from PSD
by adopting 100 and 250 tpy major source cutoffs in 1977 when PSD was
enacted, and EPA rules have long excluded smaller sources and
modifications from the program. This cutoff would not exclude many
smaller sources of GHGs because the mass emissions (i.e., tons per
year) of the relevant GHG may be substantially higher than the mass
emissions of traditional pollutants for the same process or activity.
Thus, while existing cutoffs for traditional pollutants capture a
relatively modest number of new and modified sources per year, applying
those same major source levels to CO₂, and possibly for
other GHG, would capture a very large number of sources, many of which
are comparatively smaller in size when compared to ``traditional''
sources. Similarly, for modifications, the current absence of a
significance level, or the future adoption of a significance level that
is below the current major source thresholds, would subject numerous
small changes to PSD permitting requirements.
b. Potential Program Benefits
    In the past, EPA has recognized that the PSD program can achieve
significant emissions benefits over time as emissions increases from
new major sources and major modifications are minimized through
application of state-of-the-art technology.\268\ As a result, other
programs designed to reduce emissions are not compromised by growth in
new emissions from PSD sources. Further emissions benefits are achieved
when sources limit or reduce emissions to avoid PSD applicability.
---------------------------------------------------------------------------

    \268\ See, for example, Section II of ``NSR Improvements:
Supplemental Analysis of the Environmental Impact of the 2002 Final
NSR Improvement Rules,'' U.S. EPA, November 21, 2002.
---------------------------------------------------------------------------

    A rationale for new source review since its inception has been that
it is generally more effective and less expensive to engineer and
install controls at the time a source (or major modification) is being
designed and built, as BACT does, rather than retrofitting controls
absent other construction.\269\ In addition, the BACT determination
process requires consideration of new emissions reduction technologies,
which provides an ongoing incentive to developers of these
technologies. There is the potential for avoiding or reducing GHG
emissions if ``traditional'' sources begin to install abatement
technologies for GHGs as they do for traditional pollutants. On the
other hand, as discussed in section III,F, some suggest that
regulations that apply stringent requirements to new sources and
``grandfather'' existing sources may create incentives to keep older
and inefficient sources in use longer than otherwise would occur,
diminishing the incentive for technological innovation and diffusion
and reducing the environmental effectiveness and cost effectiveness of
the regulation. Others believe that economic factors other than these
regulatory differences tend to drive business decisions on when to
build new capacity. EPA examined the effect of new source review on
utilities and refineries in a 2002 report, as described in section
III.F.4 of this notice.\270\
---------------------------------------------------------------------------

    \269\ Critics of this rationale suggest that under a market-
oriented system covering both new and existing sources, source
owners would be best placed to decide whether it is economic to
place state-of-the-art controls on new sources.
    \270\ See U.S. EPA, ``New Source Review: Report to the
President, June 2002.'' As noted in section III.F of this notice,
the report concluded (pp. 30-31) that, for existing sources,
``[c]redible examples were presented of cases in which uncertainty
about the exemption for routine activities has resulted in delay or
resulted in the cancellation of projects which sources say are done
for purposes of maintaining and improving the reliability,
efficiency and safety of existing energy capacity. Such
discouragement results in lost capacity, as well as lost
opportunities to improve energy efficiency and reduce air
pollution.'' With respect to new facilities, the report said,
``there appears to be little incremental impact of the program on
the construction of new electricity generation and refinery facilities.''

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

[[Page 44501]]

    EPA has not performed an analysis of the GHG emissions that might
be avoided or reduced under PSD preconstruction permitting, nor of
possible increases through unintended incentives. Such an analysis
would necessarily involve new analysis of potential BACT technologies,
considering costs and other factors, for GHGs emitted by numerous
sectors. The PSD program, through the BACT requirement, might result in
installation of such technologies as CCS, or the incorporation of other
CO₂ reducing technologies, such as more efficient combustion
processes.\271\ However, it is not possible at this time to estimate
these effects in light of the uncertainty surrounding the future trends
in construction at new and modified sources, demonstration of
commercial availability of various GHG control technology options,
their control effectiveness, costs, and the aforementioned incentives
to keep existing sources in operation and avoid modifying them. We ask
for comment on the nature (and to the extent possible, the magnitude)
of the potential effects of PSD on GHG emissions, and whether these
effects vary between new and existing sources.
---------------------------------------------------------------------------

    \271\ However, EPA notes that the BACT requirement does not
require consideration of technologies that would fundamentally
redefine a proposed source into a different type of source (e.g.,
BACT for a proposed coal-fired power plant need not reflect emission
limitations based on building a gas-fired power plant instead). See,
for example, In re: Prairie State Generating Company, PSD Appeal No.
05-05, slip op. at 19-37 (EAB 2006).
---------------------------------------------------------------------------

    Regarding the potentially large universe of smaller sources and
modifications that could become newly subject to BACT, as described
above, there are large uncertainties about the potential benefits of
applying BACT requirements to GHG emissions from such sources.
Individual emission reduction benefits from such sources would be
smaller; however, the cumulative effect could theoretically be large
because the requirement would cover many more sources. However, unless
there are ways to effectively streamline BACT determinations and
permitting for smaller sources (as discussed below), BACT would not
appear to be an efficient regulatory approach for many other types of
sources. We request comment on the potential overall benefit of
applying the BACT requirement to GHG emissions, and how this potential
benefit is distributed among categories of potentially regulated
sources and modifications. Below, we discuss and ask for comment on
possible tailoring of BACT for GHGs.
    Finally, in considering the potential for emissions reductions from
the PSD program, it is important to note that, historically, sources
generally have taken action to avoid PSD rather than seeking a permit,
where possible. Companies can reduce their PTE, for example, by
artificially capping production or forgoing efficiency improvements.
While these PSD avoidance strategies can sometimes reduce emissions
(e.g., limiting operating hours or installing other controls to net
out), they can sometimes result in forgone environmental benefits
(e.g., postponing an efficiency project). These effects are very
difficult to quantify. For example, the developer of a large apartment
building that would be a major source for CO₂ might elect to
provide electric space heat if it were determined that the direct and
indirect costs of PSD made installation of gas heat uneconomical. From
a lifecycle analysis standpoint, PSD could--depending upon the source
of the electricity--lead to either a better or a worse outcome for
overall emissions of GHGs. Similarly, because PSD is triggered based on
increases over a past baseline, a source considering a potential
modification may have an incentive to increase emissions (to the extent
that can be done without a modification) for the 2-year period before
the modification to artificially inflate the baseline. Similarly, in
the electricity sector, a desire to avoid PSD review could be a
disincentive for some projects to improve efficiency, because a small
increase in utilization of the more-efficient EGU would raise
CO₂ emissions sufficiently to trigger review. We solicit
comments on the potential indirect effects, adverse or beneficial, that
may arise from the incentive to avoid triggering PSD.
c. Administrative Considerations and Implications of Regulating
Numerous Smaller Sources
    The PSD program is designed to provide a detailed case-by-case
review for the sources it covers, and that review is customized to
account for the individual characteristics of each source and the air
quality in the particular area where the source will be located.
Although this case-by-case approach has effectively protected the
environment from emissions increases of traditional criteria
pollutants, there have been significant and broad-based concerns about
PSD implementation over the years due to the program's complexity and
the costs, uncertainty, and construction delays that can sometimes
result from the PSD permitting process. Expanding the program by an
order of magnitude through application of the 100/250-ton thresholds to
GHGs, and requiring PSD permits for numerous smaller GHG sources and
modifications not previously included in the program, would magnify
these concerns. EPA is aware of serious concerns being expressed by
sources and permitting authorities concerning the possible impacts of a
PSD program for GHGs.
    While the program would provide a process for reviewing and
potentially reducing GHG emissions through the BACT requirement as it
has done for other pollutants, we are concerned that without
significant tailoring (and possibly even with significant tailoring),
application of the existing PSD permitting program to these new smaller
sources would be a very inefficient way to address the challenges of
climate change. We ask for comment on how we should approach a
determination of (1) whether PSD permit requirements could be
appropriate and effective for regulating GHGs from the sources that
would be covered under the statutory thresholds, (2) whether PSD
requirements could at least be effective for particular groups of
sources (and if so, which ones), and (3) what tailoring of program
requirements (options for which are described in more detail below) is
necessary to maximize the program's effectiveness while minimizing
administrative burden and permitting delays. We are particularly
interested in how we might make such judgments in light of the
limitations on our ability to quantify the costs and emissions
reduction benefits of the PSD program, and whether there are specific
examples or other data that would help us with such an analysis.
    For example, if 100- and 250-ton thresholds were applied to GHGs,
the BACT requirement would need to be newly implemented for numerous
small sources and modifications that permitting authorities have little
experience with permitting. It would also likely involve, for both
large and small sources, consideration of new pollutants for which
there are limited add-on control options available at this time. Thus,
as with setting NSPS, a BACT determination for GHGs would likely
involve decisions on how proposed installations of equipment and
processes for a specific source category can be redesigned to make
those sources more energy efficient while taking cost considerations
into account. However,

[[Page 44502]]

unlike NSPS, because BACT is typically determined on a case-by-case
basis for each facility and changes as technology improves, these
decisions would have to take into account case-specific factors and
constantly evolving technical information \272\. Due to the more-than-
tenfold increase in the number of PSD permits that would be required if
the 100- and 250-ton thresholds were applied to GHGs, and the potential
complexity of those permitting decisions, state, local, federal, and
tribal permitting authorities would likely face significant new costs
and other administrative burdens in implementing the BACT requirement
for GHGs. Large investments of resources would be required by
permitting authorities, sources, EPA, and members of the public
interested in commenting on these decisions. Also under this scenario,
sources would likely face new costs, uncertainty, and delay in
obtaining their permits to construct.
---------------------------------------------------------------------------

    \272\ The NSPS program does take into account improvements in
technology, but does so during the 8-year review of the NSPS under
111(b)(1)(B) rather than on a permit-by-permit basis.
---------------------------------------------------------------------------

d. Definition of Regulated Pollutant for GHGs
    We also note, as described above, that decisions on the definition
of regulated pollutant for GHGs--whether GHGs would be regulated as
individual gases or as a class--has implications for BACT
determinations under the PSD program. If GHGs are regulated separately,
it is possible that a control project for one GHG could trigger PSD for
another (e.g., controlling methane in a way that increases
CO₂). In addition, the economic and other impacts for BACT
would need to be evaluated on a pollutant-by-pollutant basis. While
regulating GHGs as a class would provide additional flexibility in this
area, each BACT analysis would be more extensive because it would have
to include combined consideration of all GHGs in the class. We ask for
comment on the relative strengths and weaknesses of the various ways to
define the regulated pollutant for GHGs as related to the BACT requirement.
e. Other PSD Program Requirements
    Other parts of the CAA PSD provisions and EPA regulations that
could be affected by bringing GHGs into the program include the
requirement to evaluate, in consultation with the Federal Land Manager
(FLM), impacts on Air Quality Related Values (AQRVs) in any affected
``Class I area'' (national parks, wilderness areas, etc.), and the need
to conduct additional analysis of the proposed source's impacts on
ambient air quality, climate and meteorology, terrain, soils and
vegetation, and visibility, as provided for in section 165(e) of the
Act. These requirements can result in adjustments to the permit (for
example, permit conditions may be added if a FLM demonstrates to a
permitting authority that additional mitigation is necessary to address
the impacts of GHG emissions on the AQRVs of a Class I area). Due to
the increase in number of permits, permitting authorities may have to
make significant programmatic changes to deal with the increased
workload to conduct these analytical requirements of the PSD program,
and many additional applicants will have to devote resources to
satisfying these requirements. In addition, given the uneven geographic
distribution of new source growth, some permitting authorities may be
required to conduct more permit analyses than others.
f. GHG NAAQS Nonattainment Scenario
    If nonattainment NSR were triggered under a GHG NAAQS, the most
significant requirement would be the LAER requirement. Because LAER
does not allow consideration of costs, energy, and environmental
impacts of the emissions reduction technology, the LAER requirement
would have the potential to act as a strong technology forcing
mechanism in GHG nonattainment areas. On the other hand, once a
technology is demonstrated, this mechanism does not allow consideration
of the costs, competitiveness effects, or other related factors
associated with the new technology. As with PSD requirements, the
application of LAER to numerous smaller sources nationwide would raise
new issues on which we request comment. For example, with LAER, any
demonstrated technology for reducing CO₂ emissions, such as
a new efficient furnace or boiler design, could become mandated as LAER
for all future construction or modification involving furnaces or
boilers. Manufacturers would have to supply technologies that could
meet LAER or face regulatory barriers to the market, and could face a
constantly changing regulatory level that may result in newly designed
products being noncompliant shortly after, or even before, they are
produced and sold. New and modified sources would be required to apply
the new technology even if it is a very expensive technology that may
not necessarily have been developed for widespread application at
numerous smaller sources, and even if a relatively small emissions
improvement came with significant additional cost. We request comment
on how EPA should evaluate the LAER requirement under a NAAQS approach
for GHGs. In particular, we ask for information about whether the
relatively inflexible nature of the LAER requirement would lead to
economic disruption for certain types of sources (and if so which
ones), and whether the benefits of a NAAQS approach including LAER
would warrant further evaluation and possible tailoring of LAER to
address GHGs.
    We also ask for comment on any other NSR program issues particular
to a NAAQS approach, should EPA decide to establish a NAAQS for GHGs.
Although we have not provided a comprehensive discussion of such
issues, a number of questions arise that are particular to the NSR
requirements that flow from a NAAQS approach. For example, if the
entire country were designated nonattainment for GHGs, would the offset
requirement function as a national cap-and-trade program for GHG
emissions for all major sources? If so, how would such a program be
administered, and would the numerous small sources described above be
covered? Would the offset requirement argue for regulating GHGs as a
group, rather than individually, to facilitate offset trading? What
would be an appropriate offset ratio to ensure progress toward
attainment? Similarly, for the air quality analysis requirements of
PSD, how would a single source determine whether its contribution to
nonattainment is significant? When must such a source mitigate its
emissions impact, and what options are available to do so? Should EPA
set a PSD increment for GHGs if a NAAQS is established? Are there
additional issues of interest that we have not raised in this notice?
5. What Are the Possible Implications on Other Provisions of the Clean
Air Act?
    If PSD for GHGs applied to the same sources as a new market-
oriented program to regulate GHGs under the Act, the interaction of the
two programs would be a key issue. PSD would ensure that new and
modified sources were built with the best available technology to
minimize GHG emissions. A traditional argument for NSR is that it
ensures that new sources are built with state-of-the-art technology
that will reduce emissions throughout the lifetime of that source,
which can be several decades. However if the market-oriented program is
a cap-and-trade system with sufficiently stringent caps, PSD would not
result in more stringent control of new GHG sources than the

[[Continued on page 44503]]
Notices For
2009
2008
2007
2006
2005
2004
2003
2002
2001
2000
1999
1998
1997
1996
1995
1994
Regulating Greenhouse Gas Emissions Under the Clean Air Act

Local Navigation
