AQ_GEN_PLNG_20220404_SIP_RH-SIP_AppF4

Appendix F-4

North Carolina’s Consultation with MANE-VU

This page intentionally left blank.

October 16, 2017
Secretary Michael Regan
Mid-Atlantic/Northeast Visibility Union
MANE-VU
Reducing Regional Haze for
Improved Visibility and Health
North Carolina Department of Environment and Natural Resources
1601 Mail Service Center
Raleigh , NC 27699-1601
Dear Secretary Regan,
The mid-Atlantic and northeast region i s home to seven national parks or wilderness areas
designated as Class I areas under Section 169A of the federal Clean Air Act. These areas are
among 156 Class I areas located throughout the United States . States with Class I areas are
required to maintain and improve visibility in these areas to achieve natural background
conditions by the year 2064. Existing visibility impairment in these Class I areas, also called
regional haze, is caused by many sources located over a wide region .
All States, regardless of whether they are home to a Class I area, must prepare a State
Implementation Plan (SIP) for Regional Haze by July 31, 2021 to meet the United State s
Environmental Protection Agency (EPA) rules implementing Section 169A of the Clean Air
Act (40 CFR 51.300). This Regional Haze SIP must demonstrate that reasonable progress
towards improved visibility at the nearby Class I areas will be made by certain milestone
years. The milestone year for the second planning period is 2028 . The regulations at
40 CFR 51.308 require States with Class I areas to develop reasonable progress goals in
consultation with any State that may reasonably cause or contribute to visibility
impairment in the Class I area . The Mid-Atlantic/Northeast Visibility Union (MANE-VU)
coordinate s regional haze planning activities for the mid-Atlantic and northeast states and
tribes. This letter is part of MANE-VU's consultation process for improving visibility.
For the purpose of establishing reasonable progress goals for the second Regional Haze SIP,
MANE-VU has identified several States, including all MANE-VU member states, that may
reasonably contribute to visibility impairment at MANE -VU Class I areas . The list of States
identified for inclusion in the consultation process is in the attached Table 1.
MANE-VU is seeking your consultation to exchange information and analyses with regard
to visibility issues at MANE-VU Class I areas in Maine, New Jersey, New Hampshire and
444 North Capitol Street , NW -Suite 322 -Washington, DC 20001
202.508.3840 p -202 .508 .3841 f
Vermont, and to review the implementation process and development of a coordinated emissions
management strategy.
We invite you to join our upcoming State-to-State consultation meetings. These meetings are intended
to comply with the consultation requirements specified in the Clean Air Act (40 CFR Part 51), and are in
accordance with the MANE-VU Inter-Regional Planning Organization (RPO) Consultation Framework.
Our goal for these consultation meetings is to exchange information and analyses with regard to visibility
issues in MANE-VU Class I areas, and to review the control strategies MANE-VU is pursuing (i.e ., the MANE-
VU Inter-RPO "Ask"). We invite your state to share your analyses and to provide feedback on comments
regarding the MANE-VU Inter-RPO Ask.
MANE-VU states intend to submit Regional Haze SIPs by July 31, 2018. We anticipate that this date may
be ahead of your goals and planning schedule but would appreciate your early consultation at this time
and we believe it will be advantageous to you by expanding time for your state's planning and
implementation.
MANE-VU proposes to host a series of three webinar meetings to allow for information exchange and
policy discussion. Additional consultation sessions can be scheduled as needed prior to the policy
di scussion session .
1. Inter-RPO Consultation #1 -Introduction, overview and initial discussion (Air Director and
technical staff)
2. Inter-RPO Consultation #2 -State presentation s and discussion of issues (Air Director and
technical staff)
3. Inter-RPO Consultation #3 -Review and policy discussion (Commissioner/Secretary and Air
Director)
These consultation webinars are intended for the participation of all states and tribal nations located
within the state borders of the states included in the table below. We also welcome the participation of
the multi-jurisdictional agencies that represent the invited states and tribal nations, as well as
representatives from the U.S. Environmental Protection Agency and the Federal Land Manager agencies.
MANE-VU Class 1 Area States (Maine, New Hampshire, New Jersey and Vermont) request consultation
with:
Alabama Illinois Missouri Tennessee
Connecticut* Kentucky New York* Texas
Delaware * Louisiana North Carolina Virginia
District of Columbia* Maryland * Ohio West Virginia
Florida Massachusetts* Pennsylvania*
Indiana Michigan Rhode Island*
* MANE-VU State
Con sultation meeting dates, agendas, and discussion materials will be coordinated through the RPO
leads and will be circulated prior to each consultation.
On behalf of MANE-VU, I look forward to working with you in an informative and productive
consultation process of that yields mutually beneficial results .
Sincerely,
~~~F
Dave Foerter, Executive Director
MANE-VU/OTC
Cc: Michael Abraczinskas

(This page intentionally left blank)

.

January 27, 2018

David Foerter
Ozone Transport Commission
444 N Capitol St NW Ste 322
Washington DC 20001-1529

RE: MANE-VU Regional Haze Consultation

Dear Mr. Foerter:

This correspondence is being sent to you on behalf of the state air pollution control agencies in
Alabama, Florida, Kentucky, North Carolina, Tennessee, Virginia, and West Virginia (the seven
VISTAS states). Comments are offered herein in response to the following documents:

 Selection of States for MANE-VU Regional Haze Consultation (2018) – 9/5/2017
 Statement of the Mid-Atlantic/Northeast Visibility Union (MANE-VU) States Concerning a
Course of Action in Contributing States Located Upwind of MANE-VU Toward Assuring
Reasonable Progress for the Second Regional Haze Implementation Period (2018-2028)

As you know, the MANE-VU states have made available the documents above and have held
four consultation calls with the seven VISTAS states and other states. Thank you for sharing
your thoughts during these calls and especially for taking time to explain the technical
assessment in detail.

At this time, it is not possible for the seven VISTAS states to provide a detailed technical
response to the MANE-VU requests. However, this letter provides some initial thoughts and
concerns for your consideration.

Timing.

The MANE-VU states have indicated their intent to file their regional haze SIPs by the
original July 2018 deadline that EPA has more recently adjusted to July 31, 2021. The ten
VISTAS states are working toward completion of their regional haze technical analysis in
mid-2019 with the intention of submitting regional haze SIPs by July 2021. The differing
schedules have resulted in the seven VISTAS states being asked to assess the MANE-VU
analysis without the benefit of the forthcoming VISTAS technical work.

205 CORPORATE CENTER DR STE D ● STOCKBRIDGE GA 30281-7383
Phone: [phone] ● Web Site: www.metro4-sesarm.org
VISTAS Southeastern Regional
Haze Project – Phase II

David Foerter
January 27, 2018
Page 2 of 4

On January 18, 2018, EPA announced its decision to revisit aspects of the 2017 Regional
Haze Rule Revisions.1 While the extent of the new review is uncertain, the potential exists
that EPA could modify certain existing regional haze provisions prior to the SIP submittal
deadline; hence possibly affecting state obligations under the rule.

The MANE-VU states should allow time for EPA to complete its revisit to the rule and for the
VISTAS analysis to be completed and shared before submitting SIPs incorporating any new
emission control presumptions directed at the VISTAS states.

Technical Analysis – Inventories, Modeling, and Evaluation.

The MANE-VU states’ analysis used emission inventories that are inconsistent with the
recent EPA regional haze modeling platform. These inventories do not fully reflect emission
reductions expected from southeastern EGUs by 2028 and perhaps from other sources as
well. Modeling results derived from use of the outdated emissions inventory may not allow
conclusive determinations of impacts, if any, from VISTAS states on Class I areas in the
MANE-VU region. Additionally, the analyses may not meet EPA’s SIP approval criteria.

In many cases, the sources of the alleged contributions to downwind receptors are located
thousands of miles away from the MANE-VU Class I areas. The MANE-VU states used the
CALPUFF model and the Q/d screening approach to identify contributions that they allege
are significant. CALPUFF should not be used for transport distances greater than 300 km
since there are serious conceptual concerns with the use of puff dispersion models for very
long-range transport which can result in overestimations of surface concentrations by a
factor of three to four.2

The preamble to the recent Revisions to the Guideline on Air Quality Models that modified
appendix W of 40 CFR part 51 states, in part, “the EPA has fully documented the past and
current concerns related to the regulatory use of the CALPUFF modeling system and
believes that these concerns, including the well documented scientific and technical issues
with the modeling system, support the EPA’s decision to remove it as a preferred model in
appendix A of the Guideline.”3

1 https://www.epa.gov/visibility/epas-decision-revisit-aspects-2017-regional-haze-rule-revisions
2 Interagency Workgroup on Air Quality Modeling (IWAQM) Phase 2 Summary Report and Recommendations for
Modeling Long Range Transport Impacts (December 1998)
3 Federal Register, Vol. 82, No. 10, Tuesday, January 17, 2017, Page 5195

David Foerter
January 27, 2018
Page 3 of 4

The reliability of the Q/d screening approach diminishes over distance and especially
beyond 300 km. If the MANE-VU states wish to evaluate emission impacts more than 300
km downwind from sources, a scientifically reliable approach is essential such as the CAMx
model with the PSAT source apportionment method.

In response to our stated concerns about inaccuracies in the MANE-VU analysis during the
December 18, 2018 technical call, the MANE-VU states suggested that the seven VISTAS
states could reassess contributions using their own information to correct the MANE-VU
analysis. The VISTAS states intend to conduct a thorough technical review of emission
impacts during their forthcoming analysis. However, it is incumbent on the MANE-VU
states to correct the errors inherent in their own analysis and reassess the states with which
consultation is necessary.

The MANE-VU “ask” includes year-round use of effective control technologies on EGUs; a
four-factor analysis on sources with potential for visibility impacts of 3.0 Mm -1 or greater at
any MANE-VU Class I area; establishment of an ultra-low sulfur fuel oil standard; updated
permits, enforceable agreements, and/or rules to lock in lower emission rates for EGUs and
other large emission sources that have recently reduced emissions or are scheduled to do
so; and efforts to decrease energy demand through use of energy efficiency and increased
use of combined heat and power and other clean distributed generation technologies. This
“ask” fails to recognize fully the improved controls, fuel switches, retirements, and energy
demand reductions that have already been achieved in the Southeast. Further, the MANE-
VU states suggest that the Southeast adopt control measures that would produce little if
any visibility improvement at MANE-VU Class I areas. The MANE-VU states should refine
their analyses and establish a sound basis for any actions requested of the seven VISTAS
states and incorporated such expectations in MANE-VU SIPs.

Permanent and Enforceable.

Regional haze SIPs (including the reasonable progress goals that are set for each Class I
area) should only include emission reductions that are permanent, quantifiable, and
enforceable. Therefore, the MANE-VU states should only include in their regional haze SIPs
emission control presumptions for the seven VISTAS states that are clearly necessary and
effective and have been made permanent and enforceable via state rulemaking or permit
revisions. To include emission controls that are not permanent and enforceable in MANE-
VU states’ SIPs would be inconsistent with the Clean Air Act and the Regional Haze Rule and
could result in adverse comments from the seven VISTAS states during the MANE-VU
regional haze SIP public comment period.

David Foerter
January 27, 2018
Page 4 of 4

The VISTAS states will be initiating technical work in the spring of 2018. When that work is
completed, we will provide the MANE-VU states with a summary of our analysis. Early results
may be available as early as late 2018 and certainly by the spring of 2019.

Please note that this letter is not intended to cover every issue that may be of concern to the
seven VISTAS states. Any or all states represented by this letter may submit state-specific
comments to you.

Thank you for your consideration of these concerns. We welcome further conversations at
appropriate times as our collective work progresses.

Sincerely,

John E. Hornback
Executive Director
Metro 4/SESARM/VISTAS

Copies: VISTAS States Air Pollution Control
Agency Directors
From:Healy, David
To:Strait, Randy P; Joseph Jakuta
Subject:RE: [External] CALPUFF & KapStone Kraft Paper
Date:Wednesday, January 31, 2018 2:33:29 PM
Attachments:image002.png
CAUTION: External email. Do not click links or open attachments unless verified. Send all suspicious email as an
attachment to [e-mail].
Hi Randy,

Sorry to take so long in getting back to you, but we’ve been examining your question in detail. But
the short answer is that the 6 Mm-1 extinction shown in the Inter-RPO Ask for Kapstone Unit ST-1,2
is wrong. The 6 Mm-1 actually belongs to an industrial source in the Intra-RPO Ask. But somehow it
got pulled into the Inter-RPO Ask and was somehow assigned to the Kapstone source. So I apologize
for that and thanks for the sharp eye in finding it. The spreadsheet that you put together for this
source summarizing the impacts is accurate…the max impact in MANE-VU is 0.28 at Brigantine and
the max anywhere is 0.47 at James River Face.

I believe Jeff Underhill here at our agency has alerted, or will alert, the MANE-VU visibility group that
this error was discovered in the Inter-RPO Ask.

To answer your other question, both SO2 and NOx emissions were modeled.

Thanks again for your question and your sharp eye. Sorry again for the inconvenience. Please contact
me anytime if you have any further questions,

Dave

David Healy
Air Quality Analyst/Modeler
New Hampshire Dept. of Environmental Services
29 Hazen Drive
P.O. Box 95
Concord, NH 03302-0095
Phone: [phone]
Email: [e-mail]
Web: www.des.nh.gov

From: Strait, Randy P [mailto:[e-mail]]
Sent: Tuesday, January 30, 2018 10:52 AM
To: Joseph Jakuta; Healy, David
Subject: RE: [External] CALPUFF & KapStone Kraft Paper

Joseph, thank you for connecting us.

Dave, for the Kapstone power boiler, I prepared the attached spreadsheet that shows the max
extinction values presented in Appendix F of the document, “2016 MANE-VU Source Contribution
Modeling Report, CALPUFF Modeling of Large Electrical Generating Units and Industrial Sources, April
4, 2017”.

The table in the MANE-VU Ask for upwind states indicates that the max extinction for this unit is 6
Mm-1 but does not indicate where the impact occurs (i.e., is this for one Class I area or total for all 7
of MANE-VU’s Class I areas?). Either way, the values presented in Appendix F are less than 0.3 Mm-1
for each MANE-VU Class I area.

As you know, 2011 emissions were used in the CalPuff modeling for this emissions source. Were
both SO2 and NOx modeled or just SO2?

I will appreciate any help you can provide in understanding the basis for the 6 Mm-1 impact for the
Kapstone boiler. Please call me when you can discuss.

As Joseph indicated in his email, controls on the power boiler have been improved such that in 2016
SO2 emissions are 95% lower than in 2011, and NOx emissions are 18% lower than in 2011. We are
developing a letter to respond to the Ask so we will send the updated emissions data with the letter
in a couple of weeks.

Thank you,
Randy

Randy Strait
Supervisor, Attainment Planning Branch
Division of Air Quality, Planning Section
North Carolina Department of Environmental Quality

919 707 8721 office
919 724 8080 mobile
[e-mail]

1641 Mail Service Center
217 West Jones Street
Raleigh, NC 27699-1641

Email correspondence to and from this address is subject to the
North Carolina Public Records Law and may be disclosed to third parties.

From: Joseph Jakuta [mailto:[e-mail]]
Sent: Tuesday, January 30, 2018 10:23 AM
To: [e-mail]
Cc: Strait, Randy P <[e-mail]>
KapStone Kraft Paper Corporation, No. 1 Power Boiler (Facility ID 8048011; Unit ID ST-1,2; ES-11-CU-001)
Reference: 2016 MANE-VU Source Contribution Modeling Report, CALPUFF Modeling of Large Electrical Generating Units
and Industrial Sources, April 4, 2017
The following data are taken from Appendix F of the reference.
Est. Extinction (Mm-1)
Region Class I Area
Source
Contribution
Ranking Max Met Yr 2002 Met Yr 2011 Met Yr 2015
IMPROVE
Monitor
MANE-VU Acadia National Park 99 0.08 0.076 0.07 0.07 Yes
Brigantine 47 0.28 0.22 0.24 0.28 Yes
Great Gulf 97 0.07 0.05 0.04 0.07 Yes
Lye Brook 79 0.12 0.05 0.08 0.12 Yes
Moosehorn 102 0.07 0.07 0.06 0.06 Yes
Presidential/Dry River 93 0.08 0.058 0.05 0.08 No
Campobello 100 0.06 0.06 0.06 0.05 No
Totals 0.76 0.584 0.6 0.73
Near MANE-VU Dolly Sods 92 0.15 0.1 0.15 0.11 Yes
James River Face 21 0.47 0.47 0.26 0.3 Yes
Otter Creek 104 0.12 0.12 0.11 0.1 No
Shenandoah 56 0.32 0.25 0.32 0.23 Yes
Totals 1.06 0.94 0.84 0.74
Subject: [External] CALPUFF & KapStone Kraft Paper

CAUTION: External email. Do not click links or open attachments unless verified. Send all suspicious email as an
attachment to [e-mail].

Dave,

Randy Strait called me yesterday to ask about the inclusion of KapStone Kraft Paper on our ask list.
What he told me was that he wasn’t able to recreate the results that we calculated for the impact
that unit had on our Class I areas. I only understand at the high level what we did so couldn’t provide
any insights and told him I would get you two in touch to figure this out.

It sounds like part of the issue might be that they had high emissions in 2011 and became controlled
since then since they installed some venturi scrubbers I think Randy said in 2013, though that could
also just be something we need to consider in the RPG modeling too.

Hopefully we can figure this all out.

Joseph Jakuta
Program Manager
OTC / MANE-VU
202.508.3839
444 North Capitol Street NW #322
Washington, DC 20001

Website:www.otcair.org
Twitter: @OzoneOTC

ROY COOPER
Go i•en,o,
M I CHAE L S. R EGAN
Secretm}'
Air Quality M I C H AEL A . A BR ACZ JNSKAS
ENVIRONMENTAL QUALI TY
David Foerter
Ozone Transport Commission
444 N Capitol St NW Ste 322
Washington DC 20001-1529
February 16, 2018
Re: MANE-VU Regional Haze Consultation
Dear Mr. Foerter:
Directo,
As you know, the Mid-Atlantic/Northeast Visibility Union (MANE-VU) identified North Carolina as one
of 14 upwind states that may reasonably contribute to visibility impairment at MANE-VU Federal Class I
areas located in Maine, New Hampshire, New Jersey, and Vermont (hereafter referred to as the "Inter-
Regional Planning Organization (Inter-RPO) Ask or Ask).1 At your invitation, the North Carolina
Division of Air Quality (DAQ) has participated in each of the consultation calls MANE-VU held with the
states identified in the Inter-RPO Ask.2 These consultation calls have been helpful for understanding the
technical analyses MANE-VU completed to identify states that may reasonably contribute to visibility
impairment at MANE-VU Class I areas. The DAQ has also reviewed the technical documentation
supporting the Ask. In the spirit of the consultation process, the DAQ is submitting this letter to share
information, and express N011h Carolina's concerns with MANE-VU 's analytical approach and
conclusions as well as the timing for regional haze state implementation plan (SIP) submittals.
I. KapStone Kraft Paper Corporation (Facility ID 8048011 (3708300007), Unit ID ST-1,2 (ES-ll-
CU-001) -No. 1 Power Boiler)
The power boiler at Kapstone was identified in the MANE-VU Ask as having the potential for a 6.0
inverse megameter (Mm-1) light extinction impact on MANE-VU Class I areas based on CALPUFF
modeling of the facility's 2011 sulfur dioxide (S02) and nitrogen oxide (NOx) emissions. The DAQ
reviewed the modeling documentation and found that the maximum potential light extinction impact
modeled for the power boiler was 0 .28 Mm-1 for MANE-VU Class I areas and 0.47 Mm-1 for Class I areas
near the MANE-VU region (see Table 1). On January 31, 2018, the DAQ confirmed with Mr. David
Healy, New Hampshire Department of Environmental Services , that the 6 .0 Mm-1 extinction value shown
in the Inter-RPO Ask for Kapstone Unit ST-1,2 is wrong . Mr. Healy confirmed that the extinction values
shown in Table l below are correct for the power boiler and that the unit should not be included in the
Ask. Therefore, we request that MANE-VU remove Kapstone from the Inter-RPO Ask.
I Statement of the Mid-Atlantic/Northeast Visibility Union (MANE-VU) States Concerning a Course of Action in
Contributing States Located Upwind of MANE-VU Toward Assuring Reasonable Progres s for the Second Regional
Haze Implementation Period (2018-2028), August 25 , 2017 .
2 Letter from Foerter, Dave, Executive Director, MANE-VU/OTC to Regan , Michael , Secretary, NCDEQ , October
16 , 2017 .
State of North Carolina I Environmenta l Qua lity I A ir Qua li ty
164 1 Mail Service Ce nte r I 2 17 W . Jo nes Street, Su ite 4000 I Raleigh, NC 27609-1 641
9 19 707 8400 T
David Foerter
February 16, 2018
Page 2
Table 1. Summary of Potential Visibility Impacts on MANE-VU and Nearby Federal
Class I Areas Modeled for the KapStone Kraft Paper Corporation Power Boiler'
Estimated Extinction (Mm-1)
Maximum
Potential
Visibility Met Year Met Year Met Year
Region Class I Area Impact 2002 2011 2015
MANE-VU Acadia National Park , :ME 0.08 0.076 0.07 0.07
Brigantine Wilderness Area, NJ 0.28 0.22 0.24 0.28
Great Gulf Wilderness Area, NH 0.07 0 .05 0 .04 0.07
Lye Brook Wilderness Area, VT 0.12 0 .05 0.08 0.12
Moosehorn Wilderness Area, :ME 0.07 0.07 0 .06 0.06
Presidential Range Dry River 0 .08 0.058 0 .05 0.08 Wilderness Area, NH
Roosevelt Campobello International 0.06 0 .06 0.06 0.05 Park, :ME/NB Canada
Near Dolly Sods Wilderness Area, WV 0.15 0.1 0.15 0.11 MANE-VU
James River Face Wilderness Area, 0.47 0.47 0.26 0.3 VA
Otter Creek Wilderness Area, WV 0.12 0.12 0.11 0 .1
Shenandoah National Park, VA 0.32 0.25 0.32 0 .23
I Reference : 2016 MANE-VU Source Contribution Modeling Report, CALPUFF Modeling of Large Electrical
Generating Units and Industrial Sources, Appendix F, April 4 , 2017 . CALPUFF modeling was performed using
meteorological data for 2002, 2011, and 2015 and the highest light extinction impact was used as the maximum
potential visibility impact.
In addition, the Kapstone facility has significantly reduced its S02 and NOx emissions since 2011 . This
would result in extinction values much lower than the modeling showed based on 2011 emissions.
• From 2011 to 2016 , total facility S02 emissions have decreased by 94% (from 881 tons in 2011 to 55
tons in 2016) primarily due to S02 reductions from the No. 1 power boiler. The No. 1 power boiler
accounted for 91 % (803 tons) of total facility S02 emissions in 2011, and 68% (37 tons) in 2016.
• From 2011 to 2016, total facility NOx emissions have decreased by 13% (from 1,413 tons in 2011 to
1,232 tons in 2016). The No. 1 power boiler accounted for 71 % (1,005 tons) of total facility NOx
emissions in 2011, and 67% (820 tons) in 2016.
The DAQ will submit the latest 2016 emissions data for this facility to MANE-VU to support future
modeling updates.
II. Statewide Contribution Assessment
The DAQ reviewed the following two documents in an effort to understand MANE-VU's statewide
contribution assessment:
1. Selection of States for MANE-VU Regional Haze Consultation (2018), MANE-VU Technical
Support Committee, Sept. 5, 2017.
2 . MANE-VU Updated Q /d*C Contribution Assessment, MANE-VU Technical Support Committee,
April 6 , 2016.
David Foerter
February 16 , 2018
Page 3
As noted in these documents , MANE-VU considered the results of a weight-of-evidence approach based
on emissions (tons per year) divided by distance (kilometers) (Q/d) calculations, CALPUFF modeling,
and HYSPLIT back trajectories to determine which upwind states may reasonably contribute to visibility
impairment at a MANE-VU Class I area. States that contributed 2 percent or more of the visibility
impairment to a Class I area, and had an average mass impact of over 1 percent (0.01 microgram per
cubic meter (µg/m 3)), were identified for consultation, and, therefore, included in the Inter-RPO Ask.
Sulfur dioxide and NOx emissions for 2015 for all anthropogenic sources were considered in the
assessment. The results for North Carolina are provided in Tables 2 and 3 .
Table 2. Percent Mass-Weighted Sulfate and Nitrate Contributions from North Carolina to
MANE-VU Class I Areas in 2015
Maximum Acadia Bri2antine Great Gulf Lye Brook Moosehorn Mass Factor
2.7% 2.7% 2.7% 2.1% 2 .3% 2 .2% 0.34
Table 3. Percentage of Trajectories from North Carolina in 2015 on 20% Most Impaired Visibility
Days 1
Acadia Bri2antine Great Gulf Lye Brook Moosehorn
0.55% 2.00% 0 .00% 1.84% 1.22%
1 500 meter (m) trajectories were modeled using the HYSPLIT model, and 72-hour back
trajectories were created 4 times per day at 3AM & PM and 9AM & PM. 2015 trajectories
used the Eta Data Assimilation System (EDAS) 40-kilometer (km) meteorology. Trajectory
points were mapped and counted within 25 x 25 mile grid cells.
Based on these results , MANE-VU concluded that, "Modeling and trajectory analy s es appear to support
Alabama, North Carolina and Tenn essee as being 2% contribution states. Each ha s s ufficient emission s
to cause some d egree of visibility impact in the MANE-VU area and th e trajectories suggest a connection
on 20% mo st impaired visibility days, even if they are not as frequent as oth er states."
Although the DAQ was unable to fully understand the methodologies that MANE-VU applied due to a
lack of documentation in the two references reviewed, the following identifies serious technical
limitations with the information presented .
Ol d Screening Tool
The Q/d screening methodology yields conservatively high estimates of potential impacts for the
following reasons :
1. Q/d does not account for the formation of secondary particulate matter with an aerodynamic diameter
less than or equal to 2.5 micrometers (PM2.s) through chemical reactions as a function of distance .
Consequently, Q/d assumes I 00 percent conversion of S02 and NOx to ammonium sulfate
((NH4)2S04 and ammonium nitrate ((NH4)N03)), respectively, which is overly conservative and
yields unrealistic estimates. 3
2 . Q /d does not account for wind direction or residence time (i.e ., the amount of time a pollutant impacts
a given area). MANE-VU attempted to correct for this limitation, in part, by developing wind-
3 US EPA, lnteragency Work Group on Air Quality Modeling Phase 3 Summary Report: Near-Field Single Source
Secondary Impacts. U .S. Environmental Protection Agency, Office of Air Quality Planning and Standards, Air
Quality Analysis Division, Air Quality Modeling Group , Research Triangle Park, NC , EPA-454/P-15-002 , July
2015 , pages 23-24 .
David Foerter
February 16, 2018
Page 4
direction-specific constants for each IMPROVE monitor (based on prior CALPUFF modeling for
point sources) to "scale" Q/d results. However, the details of this methodology is not documented in
the references we reviewed; consequently, the DAQ cannot determine if this is a reasonable approach
for screening purposes. The Brigantine Wilderness Area in New Jersey, which is the closest MANE-
VU Class I area to North Carolina, is located about 507 km (315 miles) from the centroid of the
closest point to North Carolina, and 635 km (394 miles) from the centroid of North Carolina. The
DAQ does not believe that the MANE-VU screening methodology is robust enough to determine
visibility impairment attribution at these long distances .
3 . For the stationary non-point and mobile source sectors , MANE-VU did not provide documentation of
how it prepared 2015 year emissions. The DAQ requests that MANE-VU provide this documentation
for review and comment by the upwind states . Furthermore, MANE-VU cited several references
justifying the use of Q/d as a screening tool for assessing potential visibility impacts of these sources
on Class I areas. The DAQ reviewed these references and found that they all focus on using Q/d as a
screening tool for large point sources only; not surface emissions from stationary non-point and
mobile sources.3,4 ,s ,6 The DAQ believes that Q/d applied to the sum of total statewide annual
emissions for stationary non-point and mobile sources at the state centroid results in significantly high
impacts especially since Q/d does not account for atmospheric dispersion or residence time of
pollutants impacting a Class I area .
Back-Trajectory Analysis
MANE-VU modeled back trajectories for the 20 percent most impaired visibility days during 2002 , 2011
and 2015 at each of the MANE-VU Class I areas. MANE-VU used the back-trajectory results to
qualitatively cross-check with the screening results to justify including states in MANE-VU Inter-RPO
Ask . If an upwind state was determined to have a 2 percent or more impact on a MANE-VU Class I area
and it had at least one trajectory originating from the upwind state, MANE-VU included the state in the
Ask. The MANE-VU documentation does not identify the days during which or the number of
trajectories originating from North Carolina. Given the low percentage of trajectories originating from
North Carolina in 2015 (see Table 3), the DAQ believes that the back-trajectory analysis shows that North
Carolina should not be considered as reasonably attributing to visibility impairment in any of MANE-
VU ' s Class I areas, particularly when the screening analysis overestimates potential impacts. For distant-
source regions, the trajectory threshold should be much higher to definitively assign culpability.
The DAQ further questions why MANE-VU used the course Eta Data Assimilation System (EDAS) 40-
km meteorology for its 2011 and 2015 analysis , and EDAS 89-km meteorology for its 2002 analysis,
instead of using the North American Mesoscale Forecast System (NAM) model with a 12-km grid for
HYSPLIT trajectory modeling . The NAM model has become the model of choice not just for DAQ but
also for EPA and other air quality agencies and RPOs for HYSPLIT trajectory modeling. Furthermore ,
the DAQ questions MANE-VU 's selective use of meteorological years 2002 , 2011 and 2015 , instead of
across consecutive years (e .g ., 2011-2015). The DAQ believes that use of more current year emissions
and meteorology would significantly improve the contribution assessment for MANE-VU Class I areas .
4 National Association of Clean Air Agencies , PM2 .5 Modeling Implem entation for Projects Subject to National
Ambient Air Quality Demonstration Requirements Pursuant to New Source R eview , Report from NACAA PM2 .5
Modeling Implementation Workgroup , January 7, 2011 , page 2-4 and Appendix E .
5 Baker, K. R . and Foley, K. M ., "A Nonlinear Regress ion Model Estimating Single Source Concentrations of
Primary and Secondarily Formed PM2 .5," July 2011.
6 Federal Land Managers' Air Quality Related Values Work Group (FLAG) Phase I Report-Re vised (2010)
Natural Re source Report NPSINRPC/NRR-2010/232 , US Forest Service, National Park Service, and U.S . Fish and
Wildlife Service , October 2010 .
David Foerter
February 16 , 2018
Page 5
Basis for Determining Reasonable Attribution
The documentation the DAQ reviewed did not explain the technical basis for the visibility impairing
thresholds that MANE-VU used to include states in the Inter-RPO Ask. This is important for states such
as North Carolina to understand and to have the opportunity to address since MANE-VU is claiming that
North Carolina is reasonably attributing to visibility impairment in one or more ofMANE-VU's Class I
areas . Given the significant uncertainty associated with the Q/d screening tool , the weakness of the back-
trajectory analysis, and lack of documentation explaining how MANE-VU arrived at the contribution
results shown in Table 2, the DAQ believes it is inappropriate for MANE-VU to use these results to draw
any conclusions regarding North Carolina's contribution to visibility impairment in any of the MANE-
VU Class I areas . The DAQ requests that MANE-VU provide additional documentation explaining the
basis for the thresholds.
IV. Timing of SIP Submittals
We request that MANE-VU states seriously consider delaying submittal of their regional haze state
implementation plans (SIPs) from July 2018 to July 2021 . As EPA noted in its final regional haze rule, 7
extension of the SIP submittal date to July 2021 " ... will allow states to coordinate regional haze planning
with other regulatory programs, including but not limited to the Mercury and Air Toxics Standards,8 the
2010 I-hour S02 NAAQS ,9 the 2012 annual PM2.s NAAQS 10 and the Clean Power Plan,11 with the further
expectation that this cross-program coordination would lead to better overall policies and enhanced
environmental protection." In addition, EPA has yet to release its final regional haze guidance document
which, when released, may contain significant revisions to the draft guidance document released on June
30 , 2016 that would affect the process for identifying state(s) as reasonably attributing to visibility
impairment in downwind state Class I areas. 12 It is for these reasons that North Carolina is working with
the nine other Visibility Improvement -State and Tribal Association of the Southeast (VISTAS) states to
complete our regional haze modeling analysis in mid-2019 and regional haze SIP by July 2021. The
differing schedules have resulted in seven VISTAS states being asked to assess the MANE-VU analysis
without the benefit of the forthcoming VISTAS technical work. Accounting for the emission reduction
benefits associated with the federal programs EPA cited in its rule and following the final regional haze
guidance issued by EPA will help to ensure that upwind states such as North Carolina are not falsely
implicated as contributing to visibility impairment at MANE-VU Class I areas .
In addition, on January 18 , 2018, EPA announced its decision to revisit aspects of the 2017 regional haze
rule. 13 While the extent of the review is uncertain, the potential exists that EPA could modify certain
regional haze provisions prior to the July 2021 SIP submittal deadline that may affect state obligations
under the rule . The MANE-VU states should allow time for EPA to complete its revisit to the rule and for
the VISTAS analysis to be completed and shared before submitting SIPs incorporating any new emission
control presumptions directed at the VISTAS states.
7 82 FR 3116-3118 , January 10 , 2017 .
8 77 FR 9304, February 16 , 2012 .
9 75 FR 35520 , June 22 , 2010 .
10 78 FR 3086 , January 15, 2013 .
11 80 FR 64662 , October 23 , 2015 . The Clean Power Plan was stayed by the Supreme Court for the duration o f
litigation. Order in Pending Case, West Virginia v . EPA, No. 15A773 (February 9 , 2016). As a result, states have
no compliance obligations with respect to the Clean Power Plan at this time .
12 Draft Guidance on Progress Tracking Metrics, Long-term Strategies, Reasonable Progress Goals and Other
Requirements for Regional Haze State Implementation Plans for the Second Implementation Period , June 30, 2016.
13 EPA's Decis ion to Re visit Aspects of the 2017 Regional Haze Rule Revisions, https ://www.epa.go v/v isibility/epas-
dec is ion-re v isit-aspe cts-2017-reg ional-haze-ru le-rev isions .
David Foerter
February 16 , 2018
Page 6
In our SIP, North Carolina will rely on VISTAS II regional-scale modeling for 2028 using the
Comprehensive Air Quality Model with Extensions (CAMx) model with the Particulate Matter Source
Apportionment Technology (PSAT) source apportionment method for assessing source contributions to
Class I areas . This work will also be used to determine if North Carolina has a significant anthropogenic
emissions source contribution to visibility impairment in each of MANE-VU 's Class I areas. By delaying
submittal of MANE-VU state regional haze SIPs until July 2021, North Carolina will be able to share
more current emissions and modeling data with the MANE-VU states to determine if North Carolina
emissions reasonably contribute to visibility impairment in any of the MANE-VU Class I areas.
IV. Summary and Conclusions
In closing, the DAQ welcomes the opportunity to consult with MANE-VU on the quality of data and
analytical techniques used to determine reasonable attribution in MANE-VU Class I areas. As previously
noted, the DAQ has serious concerns with the information included in the Inter-RPO Ask for North
Carolina . First, I request that MANE-VU revise the Inter-RPO Ask to exclude the power boiler at Kraft
Paper Corporation that was incorrectly included in the Ask.
Second, the statewide contribution assessment contains significant uncertainty associated with the Q/d
screening tool (especially applied to stationary non-point and mobile source emissions) and back-
trajectory analysis, and the technical documentation lacks clarity on how MANE-VU arrived at the
contribution results shown in Table 2. For these reasons, the DAQ believes it is inappropriate for MANE-
VU to use these results to draw any conclusions regarding North Carolina's contribution to visibility
impairment in any of the MANE-VU Class I areas . In addition, the DAQ believes that MANE-VU has
not demonstrated the need for North Carolina to pursue adoption and implementation of the emissions
management measures MANE-VU included in its Inter-RPO Ask.
Finally, North Carolina recommends that MANE-VU take the additional time allowed by EPA to conduct
CAMx and PSAT modeling such as VISTAS II is doing to determine if North Carolina reasonably
attributes to visibility impairment in MANE-VU 's Class I areas . Meanwhile, North Carolina is working
with the VISTAS states to complete its CAMx and PSAT modeling and will rely on this modeling to
assess its visibility impact on in-state and downwind state Class I areas . North Carolina will share this
information with MANE-VU when it becomes available in 2019.
Thank you for the opportunity to comment on the Inter-RPO Ask. I hope that these comments are helpful
and I look forward to continuing to work with you and the MANE-VU states to develop reasonable
regional haze SIPs.
MAA/rps
cc: Michael Pjetraj , DAQ
Sushma Masemore, DAQ
Randy Strait, DAQ
Sincerely,
Michael Abraczinskas, Director
Division of Air Quality, NCDEQ

Members
Connecticut
Delaware
District of Columbia
Maine Maryland
Massachusetts
New Hampshire
New Jersey
New York
Pennsylvania
Penobscot Indian Nation
Rhode Island
St. Regis Mohawk Tribe
Vermont

Nonvoting Members
U.S. Environmental
Protection Agency
National Park Service
U.S. Fish and Wildlife
Service
U.S. Forest Service

MANE-VU Class I Areas
ACADIA NATIONAL PARK ME
BRIGANTINE WILDERNESS
NJ
GREAT GULF WILDERNESS NH
LYE BROOK WILDERNESS
VT

MOOSEHORN WILDERNESS
ME

PRESIDENTIAL RANGE
DRY RIVER WILDERNESS
NH

ROOSEVELT CAMPOBELLO
INTERNATIONAL PARK
ME/NB, CANADA
Mid-Atlantic/Northeast Visibility Union
MANE-VU

Reducing Regional Haze for
Improved Visibility and Health
 May 8, 2018

Michael Regan
North Capitol Dept. of Environment and Natural Resources
1601 Mail Service Center
Raleigh, NC 27699

Dear Secretary Regan;
The MANE-VU members appreciate the feedback, time and effort the participants
dedicated for the consultation process. The collaborative effort is not only a Clean Air
Act (CAA) requirement but also essential to the success of the program.
Written comments were received from eight states and two regional planning
organizations. Each of the comments received were carefully considered. MANE-VU
has prepared a report documenting the consultation process, the feedback, and the
responses. This report will be publicly available on the MANE-VU website, with
finalization expected in early May.
The MANE-VU members were encouraged to find significant common ground and
look forward to continuing this collaboration in the next planning cycle. We are
appreciative of the general comments and technical updates that states and regional
planning organizations provided. Early communication is key in any collaborative
effort. This collaboration has yielded improved data, including emissions reductions, to
be accounted for in state implementation plans. While the success of this cycle of
regional haze planning and implementation is hopeful and rewarding, we all agree the
final goal set forth in the Section 169A of the CAA is far from complete. Therefore,
continued visibility improvement in this planning period is essential to the final goal.
MANE-VU is hopeful this consultation will result in real visibility improvements and
thereby continued success.
We look forward to consulting with you during your regional haze state
implementation planning process.
Sincerely,

Clark Freise, MANE-VU Chair (NH DES)

David Foerter, MANE-VU Executive Director

444 North
Capitol
Street, NW ~
Suite 322 ~
Washington,
DC 20001
2
0
2
.
5
0
8
.
3
8
4
0
p
~
2
0
2
.
5
0
8
.
3841 f