FHWA Comments on EPA’s PM2.5 Implementation Rule:

[Note:   FHWA may have additional comments on the proposed rule and will
forward them by Monday, Nov. 8, 2004]

Major Comments: 

In general, we have major concerns on 5 issues and we would like to
discuss them further with EPA and OMB.

PM2.5 nonattainment designations:  The proposed rule states that EPA’s
plan to finalize the PM2.5 nonattainment designations by Nov. 2004 with
a effective date in early 2005.  However, it did not mention that for
conformity purposes, both for transportation and general conformity,
areas are required to demonstrate conformity one year after the
effective designation date.  That means areas need to start working on
their PM2.5 conformity analyses soon, in order to meet the early 2006
deadline.  However, given that part of the conformity requirements for
PM2.5 are not finalized yet, designations of PM2.5 areas without a final
conformity rule in place will effectively take away the intended grace
period as areas have no idea what analysis is necessary.  If the
conformity rule is finalized during the grace period, areas will have a
much shorter grace period to meet the requirements.  Therefore, we
strongly urge that the designation process coordinated with the
different related rule making to minimize potential impacts on many
areas’ transportation programs and projects.  This is significant
issue and must be addressed.

Hot spot issue:  On page 33, the proposal states, “…we believe that
PM2.5 nonattainment problems generally are not attributable to
“hot-spot” situations caused by a single facility or high traffic
roadway.”  In addition, the footnote on this page states, “Results
of research studies are not conclusive on the existence of PM2.5
hot-spots due to transportation-related sources.”  However, EPA’s
position in the preamble to the supplemental transportation conformity
NPRM for PM-2.5 hot-spots drafted is not consistent with this proposal. 
If EPA believes that hot spots are generally not a nonattainment
problem, then the supplemental conformity NPRM should be revised to be
consistent.  We tend to agree with the assessment in the proposal
implementation rule.

Also on Page 135, the proposal states, “A “hot spot analysis” is
also described which may be applied at other, unmonitored locations
where there is concern that concentrations of PM2.5 may be high.” 
However, it is not clear how this “hot spot analysis” relates to
attainment demonstrations, or conformity requirements, especially in
light of the early discussions of PM-2.5 as a regional, not localized,
pollutant.  Recommend that this be clarified or the statement deleted.

PM2.5 Precursors:  On page 81, the proposal states, “An additional
consideration is that the majority of potential PM2.5 nonattainment
areas are areas already designated as nonattainment for the 8-hour ozone
standard. Because the ozone areas will be required to address NOx for
ozone control strategy purposes (as well as for the NSR and
transportation conformity programs), the additional burden associated
with including NOx as a precursor under the PM2.5 implementation program
may only apply to 5 areas that are nonattainment for PM2.5 but not for
the 8-hour ozone standard. For PM2.5 areas that are also violating the
8-hour ozone standard, strategies to reduce NOx emissions will help
address both air pollution problems.”      

We strongly disagree with the above statement.  Ozone is predominately
formed in the summer.  However, as stated in the proposal itself,
Nitrates are formed primarily in the winter months.  Therefore, there
will be additional burden in all areas.  Air quality modeling will have
to be done for several seasons.  Motor vehicle emissions tests will have
to be done for two seasons.  Emission factor models will have to be run
for two seasons.  These are just examples.  In addition, emissions
factors vary by temperature, and a control strategy that may achieve
certain emissions reductions during one season, may not achieve those
reductions in another season.  Also, the nonattainment boundary for
ozone and PM2.5 may be different and they may have to run different
analysis to be consistent with the nonattainment area boundary. 
Therefore, this logic of no “additional burden” is not correct and
should not be used to justify requiring NOx control prior to SIPs.  

This is especially bothersome considering that this will add additional
burden to most areas outside of California, since nitrates are, by
statement in the proposal itself, only 6-8% of PM-2.5.  This simply does
not justify requiring that NOx be addressed prior to SIPs anywhere,
except possibly in California.

Finally, an analysis should be done of when the subject areas are
experiencing violations.  Since the proposal states that NOx is
primarily a winter issue, if these areas are only experiencing PM-2.5
violations during the summer, there seems to be little value or
justification in requiring them to comply with the additional burden of
considering NOx prior to SIPs.  

This comment applies also to a similar discussion on P. 81.  On page 81,
the proposal states, “An additional consideration is that the majority
of potential PM2.5 nonattainment areas are areas already designated as
nonattainment for the 8-hour ozone standard. Because the ozone areas
will be required to address NOx for ozone control strategy purposes (as
well as for the NSR and transportation conformity programs), the
additional burden associated with including NOx as a precursor under the
PM2.5 implementation program may only apply to 5 areas that are
nonattainment for PM2.5 but not for the 8-hour ozone standard. For PM2.5
areas that are also violating the 8-hour ozone standard, strategies to
reduce NOx emissions will help address both air pollution problems.”  
Again, based on the comments presented above, we strongly disagree with
this statement.

 

Conformity issues – 

When both general and transportation conformity should be addressed in
the proposal.  In many parts of the proposal, only transportation
conformity was discussed.

Transportation conformity discussions needs to be updated to reflect the
July 1, 2004 rule and the upcoming rulemaking efforts to address PM2.5
precursors and hot spot analysis.  

Page 287.  Recommend that the second paragraph reference the July 1,
2004 transportation conformity rulemaking instead of the 1997
rulemaking, as larger portions of the 1997 rulemaking are no longer
applicable.  

Section 4, on page 289 needs to be updated to reflect the July 1, 2004
rulemaking.   PM-2.5 is now addressed in the rule, except for PM-2.5
precursors and hot-spots.  This entire section is out of date, and needs
to be completely revised.

Section 6, on page 290 needs to be updated to reflect the July 1, 2004
rulemaking; in particular the last sentence of this section.

Section 8, on page 292.  The parenthetical at the end of this section
should be updated to reflect the July 1, 2004 rulemaking.

Page 294.  The citation on the last line of the page needs to be updated
to reflect the July 1, 2004 rulemaking.

Page 391 should be updated and revised to reflect MOBILE6.2 as the most
current emissions model outside of California, and EMFAC2002 for
California.

Page 461 should mention EPA’s upcoming guidance on adjusting road dust
inventories to more accurately reflect observed conditions.

 RACM and TCMs:  We are concerned with the many discussions of RACM and
TCMs in the preamble.  

On page 174.  A key component of reasonableness is missing.  A measure
is not considered to be RACM if it does not advance the attainment date.
 (See page 217 on RACT; highlight here.) This point has been upheld in
court decisions and should not be lost in the discussion here.  It is
vital that the preamble not imply that emissions reductions that do not
advance the attainment date are RACM.  

On page 252, the proposal states, “The demonstration should show that
there are no additional reasonable measures available that would
meaningfully advance the attainment date or contribute to RFP for the
area.”  The term “meaningfully advance” is not all that clear. 
Instead, it is recommended that the proposal reiterate here the language
used on page 217 that measures are only RACM if the advance the
attainment date a year.

Starting on page 255, the proposal states, “In the past under other
SIP programs, States may have rejected a single measure under
consideration as RACM because the emission reduction benefits from that
measure alone would not advance the attainment date by one year. The EPA
does not believe this approach is appropriate under section 172. For the
PM2.5 program, our approach is to emphasize the collective evaluation of
RACM measures and whether they will advance the attainment date when
taken together. We believe that the State should not reject an
individual measure unless the State can show that it has evaluated the
collective effect of that measure plus all other available control
measures to determine whether implementing those measures together would
advance the attainment date.”

Why is EPA changing this interpretation?  Is this consistent with the
RACM court decision (Sierra Club v. EPA, 294 F.3d 155 (D.C. Cir. 2002)? 
If this is not current policy for ozone and other pollutants, then we
would strongly object to any changes to this definition of RACM.  

On page 262, revise to state, “However, under current EPA policy we do
not presume that any or all of these measures are reasonably available
in any areas.”  As written, the proposal still implies that some of
the measures are reasonably available for all areas, and this is not the
case.

On page 263, revise to state, “…we do not believe it is appropriate
to include a presumption that any of these measures should be considered
RACM for any PM2.5 nonattainment area.”  See preceding comment.

On page 263, the proposal states, “PM2.5 monitoring data show that
organic and elemental carbon typically make up a significant percentage
(e.g., 25 percent or more) of the PM2.5 mass in most urban areas, and
analyses have shown that transportation sources are significant emitters
of these PM2.5 components. For this reason, we believe that States
should give serious consideration to TCMs in their RACM analyses.”  
We strongly disagree with this conclusion.  This recommendation should
be deleted from the proposal.  It is not needed anyway, as the proposal
already states that all measures must be considered.  There is no reason
to specifically call out TCMs here.  Although transportation sources may
be significant sources of PM2.5 in some areas, we are not sure whether
TCMs are the most cost-effective way to achieve significant emission
reductions.

On page 263, revise to state, “However, experience has shown that
local circumstances vary to such a degree from city to city that it
would not be reasonable to assume that any section 108(f) measures are
reasonably available in any areas.

Page 285 of the proposal states, “Examples of voluntary mobile source
measures include transportation control measures, ozone action plans,
and trip reduction strategies.”  This implies that all TCMs are
voluntary measures, and this is not correct.  Recommend deleting
“transportation control measures from this sentence, and if necessary
replace with a specific example of a voluntary measure.

Page 485. Is this consistent with the RACM court decision (Sierra Club
v. EPA, 294 F.3d 155 (D.C. Cir. 2002)?  If this is not current policy
for ozone and other pollutants, then we would strongly object to any
changes to this definition of RACM that require consideration of
cumulative effects.

Other comments:  

 On pages 55-58, the proposal discusses the sources of primary PM bases
on the National Emissions Inventory (NEI).  This is somewhat misleading,
and should focus just on PM-2.5.  For example, the proposal states that
one of the main sources of these particles is re-entrained dust.  It
then states that much of these particles are course rather than fine. 
But since this proposal deals with fine particles, it seems out of place
to highlight the sources of PM-10.  This could be confusing to the
reviewer and potential commenter.  Recommend that this section be
revised to focus on PM-2.5 sources.

On page 62, the proposal states, “The main sources of NOx are
combustion of fossil fuel in boilers and onroad mobile sources.”  It
also states that these sources account for 80% of NOx emissions with
each contributing about half.  This does not seem to be correct.  The
2002 and 2003 numbers show that fuel combustion electrical utilities and
highway vehicles only account for 57% of emissions.  Also, as alluded to
elsewhere in the proposal, non-road mobile sources are also a major
source of NOx.  And, over the past decades, while on-road emissions have
decreased significantly, non-road emissions have continued to grow. 
Also, according to the NEI, industrial fuel combustion is also a fairly
large source of NOx.  Therefore, this section needs to be updated to
reflect the recently release 2002 or 2003 numbers, and should also list
non-road sources as a major source.

Page 75.  Section 182(f) of the Clean Air Act can also apply to other
sources besides “major stationary sources,” and waive NOx
requirements for ozone.

Page 78. Conformity, both transportation and general conformity take
effect one year after the effective date of designations.

Page 80.  In the first full sentence on the page, shouldn’t this be
“attainment demonstrations SIPs” instead of “nonattainment
SIPs?”

Page 88.  The proposal has left blank the proposed design value for
which areas would be classified as serious under this option.  This is a
vital piece of information that is needed to evaluate this proposal.  In
addition, the proposal should clearly give a basis for this value.

Page 94.  Rural transport nonattainment areas would also be subject to
general conformity.

On page 96, the proposal states that only 1 area may meet the criteria
for a rural transport area.  What is the area?  Since there is only 1
area, it is recommended that the proposal go ahead and identify it,
providing a clearer example of the type of area this section is
referring to.

On page 97, the proposal states, “Under the current schedule,
nonattainment area designations are to be finalized in December 2004.”
 This is inconsistent with earlier in the document.

Page 102-104.  The dates in the example that begins on the bottom of
page 102 should be revised to reflect the actual proposed attainment
dates.  Areas will start with a presumed 2010 attainment date, not 2012.
 The years examined would be 2007, 2008, and 2009, with a 1-year
extension possible to 2011.  

Pages 144-146 of the proposal discuss trying to align modeling years. 
However, this does not address the issue of seasonal variation.  Even if
you look at ozone, PM-2.5, and regional haze for the same year, each
pollutant or the haze problem, could actually be worse during different
seasons.  How do areas look at this seasonal variation?  Would they need
to look at all three, for two or three different times of the year?  The
proposal seems to imply that one attainment modeling could cover both
ozone and PM-2.5, but depending on the nature of the PM-2.5 problem,
this may not be the case.  This needs to be addressed and clarified.

Page 149.  Would these RFP plans be required to contain motor vehicle
emissions budgets?  It is assumed that they would.  In that case, this
should be highlighted in this section and cross-referenced with the
conformity section.

Page 157.  2010 is proposed as the first RFP milestone year.  Will this
be consistent with the ozone implementation plan attainment dates and
RFP?  It seems that it would be beneficial to have these years aligned. 
Also, how does this fit with inventory requirements?  It seems to be
off-cycle.  Periodic inventories will be due in 2005, 2008, 2011, 2014,
etc.  Wouldn’t it be better and conservative of resources, if the RFP
milestone years aligned with the inventory years?

Page 187.  This section discusses using a different geographic area than
the nonattainment area for RFP purposes.  However, if the RFP SIP
establishes a motor vehicle emissions budget, it should only be for the
nonattainment area, as conformity only applies in nonattainment or
maintenances areas.  This issue should be discussed and cross-referenced
with the conformity section.

Page 202 begins a section on what precursors apply to RACT.  However, it
only discusses NOx, VOCs, and ammonia.  What about SO2, which is
mentioned in the first sentence of the section, but not discussed as
whether it applies to RACT.

On page 266, the proposal states, “Programs to expand use of clean
burning fuels (such as biodiesel).”  Can’t biodiesel actually
increase NOx emissions?  If this is the case, it is recommended that the
phrase “such as biodiesel” be deleted.

On page 267, “Reduce dust from paved and unpaved roads” is listed as
an area source.  This should be moved under mobile sources.  Dust from
paved and unpaved roads, if significant, will need to become part of the
on-road motor vehicle emissions budget.  There has been a great deal of
confusion in regards to this issue in the past.  As listed under areas
sources, this may confuse readers.

Section 11, on page 295 needs to be updated to reflect the withdrawal of
the CADC program proposal.

Page 301.  Please see above comments on designations and effective
dates.

On page 318, the proposal states, “We do not believe that this creates
a major regulatory burden because NOx is already a regulated NSR
pollutant. This is because it is a precursor for the ozone NAAQS and an
indicator for the NO2 NAAQS.”  Does this conclusion account for the
fact that ozone is a summer problem, whereas PM-2.5 may be a year-round
pollutant that the PM-2.5 NOx  may be a winter problem?  

Page 378 states, “The EPA plans to address these concerns in a future
Tribal NSR rule, which is scheduled for proposal in early 2004.”  This
should be updated.

