                                                                        
                                                        Technical
Support Document for the Modeling Portions of the State of Maryland’s
Regional Haze State Implementation Plan (SIP) Entitled “State of
Maryland Regional Haze 

State Implementation Plan                                               
                          SIP Number: 12-01                             
                                   February 9, 2012”

TSD Prepared February, 2012

Todd A. Ellsworth

Office of Air Monitoring and Analysis, 3AP40

U.S. Environmental Protection Agency, Region 3

1650 Arch Street

Philadelphia, Pennsylvania 19103

_____________________/s/_______________________

Reviewed by Andrew Hass, Acting Associate Director,

Office of Air Monitoring and Analysis (3AP40)

_______2/15/12__________

Date Signed

Technical Support Document for the Modeling Portions of the State of
Maryland’s Regional Haze State Implementation Plan (SIP) Entitled
“State of Maryland Regional Haze 

State Implementation Plan SIP Number: 12-01                             
                                   February 9, 2012”

Purpose of the Technical Support Document

This Technical Support Document (TSD) describes the Environmental
Protection Agency’s (EPA’s) evaluation of the modeling portions of
Maryland’s State Implementation Plan (SIP) revision entitled “State
of Maryland Regional Haze State Implementation Plan SIP Number: 12-01
February 9, 2012”.  This SIP revision will hereafter be referred to as
the Maryland (MD) Haze SIP.

The purpose of this TSD is to provide more detailed information than can
be contained in the official notice published in the Federal Register. 
Readers who need more information than we provide in this TSD or want to
review the modeling in more detail should read the above referenced MD
Haze SIP.  

The Regulatory Framework 

In section 169A(a)(1) of the 1977 Amendments to the CAA, Congress
created a program for protecting visibility in the nation’s national
parks and wilderness areas.  This section of the CAA establishes as a
national goal the “prevention of any future, and the remedying of any
existing, impairment of visibility in mandatory Class I Federal areas
which impairment results from manmade air pollution.”  On December 2,
1980, EPA promulgated regulations to address visibility impairment in
Class I areas that is “reasonably attributable” to a single source
or small group of sources, i.e., “reasonably attributable visibility
impairment” (RAVI).  See 45 FR 80084.  These regulations represented
the first phase in addressing visibility impairment.  EPA deferred
action on regional haze that emanates from a variety of sources until
monitoring, modeling and scientific knowledge about the relationships
between pollutants and visibility impairment were improved. 

Congress added section 169B to the CAA in 1990 to address regional haze
issues.  EPA promulgated a rule to address regional haze on July 1, 1999
(see 64 FR 35713), the Regional Haze Rule (RHR).  The RHR revised the
existing visibility regulations to integrate into the regulation
provisions addressing regional haze impairment and established a
comprehensive visibility protection program for Class I areas.  The
requirements for regional haze, found at 40 CFR 51.308 and 51.309, are
included in EPA’s visibility protection regulations at 40 CFR
51.300-309.  

The RHR addressed the combined visibility effects of various pollution
sources over a wide geographic region.  40 CFR 51.308(b) requires states
to submit the first implementation plan addressing regional haze
visibility impairment no later than December 17, 2007.  Consequently,
all 50 states, including those without Class I areas, Washington, D.C.,
and the Virgin Islands, are required to submit Regional Haze SIPs. The
USEPA designated five Regional Planning Organizations (RPOs) to assist
with the coordination and cooperation needed to address the visibility
issue. The Mid-Atlantic Region Air Management Association (MARAMA), the
Northeast States for Coordinated Air Use Management (NESCAUM) and the
Ozone Transport Commission (OTC) established the Mid-Atlantic/Northeast
Visibility Union (MANE-VU)8 regional planning organization to coordinate
efforts to address visibility impairment at seven Class I areas located
in the Mid-Atlantic and Northeast corridor: Acadia National Park, ME;
Brigantine Wilderness, NJ; Great Gulf Wilderness, NH; Lye Brook
Wilderness, VT; Moosehorn Wilderness, ME; Presidential Range – Dry
River Wilderness, NH; and Roosevelt Campobello International Park, New
Brunswick. 

Figure 1. Geographical Areas of Regional Planning Organizations

 

                                                                        
                                                                        
                

             

                 

The State of Maryland is a member of the Mid-Atlantic / Northeast
Visibility Union (MANE-VU) RPO. Members of MANE –VU are listed in
Table 1.

Table 1 MANE-VU Members

Connecticut 

	Pennsylvania 



Delaware	Penobscot Nation 



District of Columbia 

	Rhode Island 



Maine	St. Regis Mohawk Tribe 



Maryland 

	Vermont 



Massachusetts 

	U.S. Environmental Protection Agency*

New Hampshire 

	U.S. National Park Service*

New Jersey 

	U.S. Fish and Wildlife Service* 



New York	U.S. Forest Service*



*Non-voting members

MANE-VU’s work is managed by the Ozone Transport Commission (OTC) and
carried out by OTC, the Mid-Atlantic Regional Air Management Association
(MARAMA), and the Northeast States for Coordinated Air Quality
Management (NESCAUM). The states along with federal agencies and
professional staff from OTC, MARAMA and NESCAUM are members of the
various committees and workgroups. 

Since its inception on July 24, 2001, MANE-VU established an active
committee structure to address both technical and non-technical issues
related to regional haze. The primary committees are the Technical
Support Committee (TSC) charged with assessing the nature and magnitude
of the regional haze problem within MANE-VU, interpreting the results of
technical work, and report on such work to the MANE-VU Board. In
addition to the formal working committees, there are also three standing
working groups of the TSC. They are broken down by topic area: Emissions
Inventory, Modeling, and Monitoring/Data Analysis Workgroups. 

Introduction to Maryland’s Regional Haze State Implementation Plan    
                                                                        
                                                 EPA promulgated a rule
to address regional haze on July 1, 1999 (see 64 FR 35713), the RHR. 
The RHR addressed the combined visibility effects of various pollution
sources over a wide geographic region.  40 CFR 51.308(b) requires states
to submit the first implementation plan addressing regional haze
visibility impairment no later than December 17, 2007.

MANE-VU states agreed upon a ≥ 2 percent sulfate attribution to a
Class I area in order for an upwind state to meet the definition of
“significantly contributing” to visibility impairment for that Class
I area. A contribution assessment analysis was completed by the
Northeast States for Coordinated Air Use (NESCAUM), entitled
Contributions to Regional Haze in the Northeast and Mid-Atlantic States
and can be found in Appendix A of the MD Haze SIP.    The MANE-VU
Contribution Assessment indicates that emission sources located in
Maryland may impact visibility at the following Class I areas in
MANE-VU:  Acadia National Park, Brigantine Wildlife Refuge, Great Gulf
Wilderness Area, Lye Brook Wilderness Area, Moosehorn Wildlife Refuge,
Presidential Range-Dry River Wilderness Area, and Roosevelt-Campobello
International Park as well as Dolly Sods Wilderness, Otter Creek
Wilderness and Shenandoah National Park which lie outside of the MANE-VU
region.

The VISTAS Contribution Assessment states that the MANE-VU states
contribute to SO2 emissions in both the Dolly Sods Wilderness and
Shenandoah National Park Class I areas.  The Contribution Assessment
states that reductions in SO2 emissions from electric generating units
would produce the greatest improvements in visibility.  Maryland will
fulfill its commitment to reduce SO2 emissions through reductions from
the Healthy Air Act as described in section 11 of the MD Haze SIP.  

                                                                        
                                                                 

Therefore, this SIP focuses on how Maryland control measures will
improve visibility at   

Acadia National Park, Brigantine Wildlife Refuge, Great Gulf Wilderness
Area, Lye Brook Wilderness Area, Moosehorn Wildlife Refuge, Presidential
Range-Dry River Wilderness Area, and Roosevelt-Campobello International
Park as well as Dolly Sods Wilderness, Otter Creek Wilderness and
Shenandoah National Park.   

Maryland believes their Regional Haze SIP demonstrates that it has met
its BART, RPG and LTS obligations for the first visibility impairment
planning period through existing Maryland/Federal regulations and
on-the-books/on-the-way federal emission controls. In addition to
extensive consultation with the MANE-VU states and the EPA, Maryland,
through MANE-VU, has consulted with Federal Land Managers (FLMs)
responsible for the Class I areas, Midwest RPO(MPRO) and VISTAS in the
development of this SIP.

What Are The Components Of A Modeled Regional Haze Demonstration?

Modeling Process Overview                                               
                                                                        
                  

The goal of the regional haze program is to return to natural conditions
by 2064, and

States are required to demonstrate, by the end of the first planning
period (by 2018), reasonable progress toward meeting that goal. 

Maryland is a member of the Mid-Atlantic/Northeast Visibility Union
(MANE-VU) Regional Planning Organization (RPO).  The MANE-VU RPO was
tasked with the assignment of preparing a PM2.5 modeling platform that
all member states could use to model their LTSs to demonstrate
reasonable progress by 2018 in meeting the ultimate goal of natural
visibility conditions by 2064.  

The New York State Department of Environmental Conservation (NYSDEC) was
the lead agency for coordinating and running the modeling platform used
by the MAVE-VU RPO.  Modeling centers responsible for running the
platform included the NYSDEC, the University of Maryland at College Park
(UMD), the Northeast States for Coordinated Air Use Management
(NESCAUM), the New Jersey Department of Environmental Protection (NJ
DEP), and the Virginia Department of Environmental Quality (VA DEQ). 
Each modeling center was responsible for installing the modeling
platform, conducting diagnostic tests and completing a benchmark run to
ensure accurate results.   

                                                                        
                                                              

The MANE-VU RPO used the Community Multi-scale Air Quality model (CMAQ)
version 4.5.1 as its photochemical grid model.  The model uses
simulations of chemical reactions, emission of PM2.5   and PM2.5
precursors and a sophisticated meteorological model (The Pennsylvania
State University/National Center for Atmospheric Research Mesoscale
Meteorological Model (MM5)) to produce speciated PM2.5 concentrations
over the eastern United States.  The meteorological data used in the
meteorological model was for the 2002 base year.  The photochemical grid
model was run with the base year meteorology and base year emissions to
determine if the model performance was satisfactory.  Once the model
performance was determined to be adequate, PM 2.5 concentrations were
modeled by running the model with projected emissions for 2018 and the
original 2002 meteorology. The meteorology was held constant so that the
results of changing the emissions would not be influenced by changing
meteorology.

The EPA modeling guidance recommends that modeling be used to develop
relative response factors for each of 6 components of particulate matter
between a base period (2000-2004) and a future 5-year period which will
be reviewed in 2018.  PM components used for regional haze-related
applications are:

- mass associated with sulfates;

- mass associated with nitrates;

- mass associated with organic carbon;

- mass associated with elemental carbon;

- mass associated with fine soil (i.e., crustal material);

- mass associated with coarse particulate matter (i.e., PM10 - PM2.5).

Current speciated measurements in a Class I area are used in an
empirically derived equation to estimate light extinction for each day
with measurements. Days are ranked according to their resulting light
extinction (measured in deciviews). This ranking is used to identify the
20% of days with worst and 20% of days with best visibility during each
year in the base period. The 20% worst and best days are examined to
estimate appropriate observed concentrations for the components of PM on
“best” and “worst” days. Observed component concentrations are
multiplied by the corresponding relative response factors to estimate
future concentrations for each component on “best” and “worst”
days. Future component concentrations are then inserted into the
equation relating light extinction to concentrations of particulate
matter. The resulting estimates for future light extinction on
“best” and “worst” days are compared with observations made
during the base period to assess the future year visibility improvement
and demonstrate reasonable progress by 2018 in meeting the ultimate goal
of natural visibility conditions by 2064.  

                                                                        
                                                        

Steps Required in Modeling Future Year Visibility Improvement

The modeling guidance lists nine steps for preparing modeling to
demonstrate reasonable progress toward visibility improvement goals.    

1. Develop a conceptual description of the problem to be addressed.

2. Select an appropriate model to support the demonstration.

3. Select appropriate meteorological time periods to model.

4. Choose an appropriate area to model with appropriate
horizontal/vertical resolution                                          
   and establish the initial and boundary conditions that are suitable
for the application.

5. Generate meteorological inputs to the air quality model.

6. Generate emissions inputs to the air quality model.

7. Run the air quality model with base case emissions and evaluate the
performance.

    Perform diagnostic tests to improve the model, as necessary.

8. Perform future year modeling (including additional control
strategies, if necessary) and 

    use the results to calculate future year visibility and visibility
improvement.

How Did Maryland Address All Of The Components a Modeled Demonstration
Of  Future Year Visibility Improvement?

The MD Haze SIP addresses each of the required elements of a modeling
analysis used to predict visibility improvement that is expected by
2018. 

Conceptual Description of the Problem

A conceptual model describes how weather patterns affect the formation
and transport of PM2.5, accounting for emissions and photochemistry. 

The conceptual model for the MD Haze SIP is described in the document
entitled: The Nature of the Fine Particle and Regional Haze Air Quality
Problems in the Mid-Atlantic Northeast Visibility Union(MANE-VU) Region:
A Conceptual Description (NESCAUM), November 2006 (see Appendix I of the
MD Haze SIP). This document was prepared by Northeast States for
Coordinated Air Use Management (NESCAUM) for use by the MANE-VU member
States, and it provides the conceptual description of the fine particle
issues in these states, consistent with the EPA’s guidance.

The NSCAUM document contains a conceptual description that explains how
elevated regional PM2.5 peak concentrations in the summer differ
significantly from the largely urban peak concentrations observed during
winter. On average, summertime concentrations of sulfate in the
northeastern United States are more than twice that of the next most
important fine particle constituent, organic carbon (OC), and more than
four times the combined concentration of nitrate and black carbon (BC)
constituents. Episodes of high summertime sulfate concentrations are
consistent with stagnant meteorological flow conditions upwind of the
Philadelphia-Wilmington, PA-NJ-DE PM2.5 NAA and the accumulation of
airborne sulfate (via atmospheric oxidation of SO2) followed by
long-range transport of sulfur emissions from industrialized areas
within and outside the area.  National assessments have indicated that
in the winter, sulfate levels in urban areas are higher than background
sulfate levels across the eastern U.S., indicating that the local urban
contribution to wintertime sulfate levels is significant relative to the
regional sulfate contribution from long-range transport. A network
analysis for the winter of 2002 suggests that the local enhancement of
sulfate in urban areas of the MANE-VU region ranges from 25 to 40% and
that the long-range transport component of PM2.5 sulfate is still the
dominant contributor in most eastern cities.

In the winter, urban organic carbon (OC) and sulfate each account for
about a third of the overall PM2.5 mass concentration observed in
Philadelphia and New York City areas. Nitrate also makes a significant
contribution to urban PM2.5 levels observed in the northeastern United
States during the winter months. Wintertime concentrations of OC and
nitrate in urban areas can be twice the average regional concentrations
of these pollutants, indicating the importance of local source
contributions. This is likely because winter conditions are more
conducive to the formation of local inversion layers which prevent
vertical mixing. Under these conditions, emissions from tailpipe,
industrial and other local sources become concentrated near the
Earth’s surface, adding to background pollution levels associated with
regionally transported emissions. 

In summary, the presence of fine particulate matter in ambient air
significantly obscures visibility during most parts of the year at sites
across the MANE-VU region. Particle pollution generally, and its sulfate
component specifically, constitute the principle driver for regional
visibility impacts. While the broad region experiences visibility
impairment, it is most severe in the southern and western portions of
MANE-VU that are closest to large power plant SO2 sources in the Ohio
River and Tennessee Valleys.

The presence or absence of regional sulfate almost exclusively drives
summer visibility impairment, whereas winter visibility depends on a
combination of regional and local influences coupled with local
meteorological conditions (inversions) that lead to the concentrated
build-up of pollution.

Sulfate is the key particle constituent from the standpoint of designing
control strategies to improve visibility conditions in the northeastern
United States. Significant further reductions in ambient sulfate levels
are achievable, though they will require more than proportional
reductions in SO2 emissions.

Long-range pollutant transport and local pollutant emissions are
important, especially along the eastern seaboard, so one must also look
beyond the achievement of further sulfate reductions. During the winter
months, in particular, consideration also needs to be given to reducing
urban sources of SO2, as well as NOX and OC (North American Research
Strategy for Tropospheric Ozone (NARSTO), 2003).

Modeling Platforms

Currently two regional-scale air quality models have been evaluated and
used by NESCAUM to perform air quality simulations. These are the
Community Multi-scale

Air Quality modeling system (CMAQ; Byun and Ching, 1999) and the
Regional Modeling System for Aerosols and Deposition (REMSAD; SAI,
2002). CMAQ was developed by USEPA, while REMSAD was developed by ICF
Consulting/Systems Applications International (ICF/SAI) with USEPA
support. CMAQ has undergone extensive community development and peer
review (Amar et al., 2005) and has been successfully used in a number of
regional air quality studies (Bell and Ellis, 2003;

Hogrefe et al., 2004; Jimenez and Baldasano, 2004; Mao and Talbot, 2003;
Mebust et al.,

2003). REMSAD has also has been peer reviewed (Seigneur et al., 1999)
and used by

USEPA for regulatory applications
(www.epa.gov/otaq/regs/hd2007/frm/r00028.pdf and 

  HYPERLINK "http://www.epa.gov/clearskies/air_quality_tech.html" 
www.epa.gov/clearskies/air_quality_tech.html ) to study ambient
concentrations and deposition of sulfate and other PM species.

Five modeling centers, worked collectively to maximize efficiency of
computing resources in MANE-VU for SIP modeling.  These centers include
NY DEC, NJ DEP/Rutgers, VA DEQ, UMD, and NESCAUM. 

Meteorological Time Periods Used in the Modeling

All of 2002, which represents the baseline period from 2000 to 2004, was
included in CMAQ modeling.  2002 was divided into five periods. UMD is
responsible for modeling the period from January 1 to February 28; NJ
DEP/Rutgers are responsible for the period from March 1 to May 14; NY
DEC is responsible for the period from May 15 to September 30; VA DEQ is
responsible for the period from October 1 to October 31; and NESCAUM is
responsible for the period from November 1 to December 31. Each period
uses a 15 day spin up run to minimize the impact of the default initial
concentration fields. Each group performs CMAQ simulations on its period
for a series of scenarios including 2002 Base Case, 2009 Base Case, 2018
Base Case, 2009 Control Case, and 2018 Control Case. All scenarios adopt
the same meteorological field (2002) and boundary conditions, varying
only emission inputs. To ensure consistency between modeling groups, a
benchmark test was conducted by each group.

Meteorological Data Used in the Air Quality Model

The MANE-VU states decided to use a prognostic meteorological model that
provides life-like meteorological inputs to the photochemical grid
model.  The Pennsylvania State University/National Center for
Atmospheric Research Mesoscale Meteorological Model (MM5) version 3.6
was chosen for the modeling analysis.  The MM5 model provides a
reasonable representation of weather conditions at the surface and
aloft.  

Domain of the Model, Horizontal/Vertical Resolution and the Initial and
Boundary Conditions

MANE-VU has adopted the Inter-RPO domain description for its modeling
runs.  

This 36-km domain covers the continental United States, southern Canada
and northern

Mexico. The dimensions of this domain are 145 and 102 cells in the
east-west and north-south directions, respectively. A 12-km inner domain
was selected to better characterize

air quality in MANE-VU and surrounding RPO regions. This domain covers
the eastern

region, which includes the northeastern, central, and southeastern U.S.,
as well as southeastern Canada. It extends from 66oW~94oW in longitude
and 29oN~50oN in

latitude with 172 × 172 grid cells (Figure 3).

Figure 3. Modeling domains used in MANE-VU air quality modeling studies
with

CMAQ. Outer (blue) domain grid is 36 km and inner (red) domain is 12 km
grid.

The gridlines are shown at 180 km intervals (5 × 5 36 km cells/15 × 15
12 km cells).

                                                                        
                                                           Vertical
resolution is the number of layers and the size of each layer in the
model.  The 

layers in the photochemical grid model were set up to be compatible with
the model that produced weather conditions for the photochemical grid
model.  The vertical resolution used in the modeling exercise followed
EPA’s modeling guidance and therefore adequately represents the
atmosphere where PM2.5 is emitted, forms and is transported.  

Baseline and Future Year Emission Inventories for Modeling 

Section 51.308(d)(3)(iii) of EPA’s Regional Haze Rule requires the
States to identify the baseline emission inventory on which strategies
are based. The baseline inventory is intended to be used to assess
progress in making emission reductions. Based on EPA guidance entitled,
2002 Base Year Emission Inventory SIP Planning: 8-hour Ozone, PM 2.5,
and Regional Haze Programs, which identifies 2002 as the anticipated
baseline emission inventory year for regional haze.  The MANE-VU states
are using 2002 as the baseline year. Future year inventories were
developed for the years 2009, 2012 and 2018 based on the 2002 base year.
These future year emission inventories include emissions growth due to
projected increases in economic activity as well as the emissions
reductions due to the implementation of control measures. 

The 2002 emissions were first generated by the individual states in the
MANE-VU area. MARAMA then coordinated and quality assured the 2002
inventory data. The 2002 emissions from non-MANE-VU areas within the
modeling domain were obtained from other Regional Planning Organizations
for their corresponding areas. These Regional Planning Organizations
included the Visibility Improvement State and Tribal Association of the
Southeast (VISTAS), the Midwest Regional Planning Organization and the
Central Regional Air Planning Association. 

Version 3 of the 2002 base year emission inventory was used in the
regional modeling exercise. Technical support documentation for the
MANE-VU 2002 base inventory is presented in Appendix D of the MD Haze
SIP. This document explains the data sources, methods, and results for
preparing this version of the 2002 base year criteria air pollutant and
ammonia emissions inventory. Documentation for the future year
estimations of EGUs is presented in Appendix E of the MD Haze SIP.
Documentation for the future year estimations of the remaining source
sectors (non-EGU sectors) is also presented in Appendix E of the MD Haze
SIP.

Model Performance Evaluation

NESCAUM evaluated the 2002 annual 12 km resolution meteorological fields
generated by MM5 using ENVIRON's METSTAT program. Model results of
surface wind speed, wind direction, temperature, and humidity were
paired with measurements from EPA’s Clean Air  Status and Trends
Network (CASTNET) and National Center for Atmospheric Research’s  
Techniques Data Laboratory (TDL) network by hour and by location and
then statistically compared. Based on this statistical comparison
between model prediction and data from the two networks for wind speed,
wind direction, temperature, and humidity, MM5 performs well. An
acceptable small bias, high index of agreement and strong correlation
with CASTNET and TDL data are shown. Since MM5 uses TDL data for
nudging, the model predictions are in better agreement with TDL data
than with CASTNET data. MM5 performs better in Midwest and Northeast
than Southeastern US.

s are: Mean Fractional Error (MFE) ≤ +50%, and Mean Fraction Bias
(MFB) ≤ ±30%; while the criteria are proposed as: MFE ≤ +75%, and
MFB ≤ ±60%. CMAQ prediction of PM2.5 species from 40 STN sites and 17
IMPROVE sites within the MANE-VU Region were paired with measurements
and statistically analyzed to generate MFE and MFB values. Considering
CMAQ performance in terms of MFE and MFB goals, sulfate, nitrate, OC,
EC, and PM2.5 all had the majority of data points within the goal curve,
some were between the goal and acceptable criteria, and only a few were
outside the criteria curve. Only fine soil has the majority of points
outside the criteria curve, but there were some sites still within the
goal. For the MANE-VU region, CMAQ performs best for PM2.5 sulfate,
followed by PM2.5, EC, nitrate, OC, and then fine soil. Regional haze
modeling also requires CMAQ performance evaluation for aerosol
extinction coefficient (Bext) and the haze index.   Modeled daily
aerosol extinction at each improve site was calculated following the
IMPROVE formula with modeled daily PM2.5 species concentration and
relative humidity factors from IMPROVE. The approach used natural
background visibility estimates and the haze index following EPA
Guidance. The modeled Bext showed a near 1:1 linear relationship (slope
of 0.78 and r2 of 0.46) with IMPROVE observed Bext. The regression
excluded three points from July 7, 2002; the monitors were directly
impacted by Canadian fires whose emissions were not modeled. 

Uniform Rate of Progress Goals

The key difference between SIPs from States with Class I areas and those
States without Class I areas, but may have sources that impact
visibility on Class I areas, is the calculation of the baseline and
natural visibility for their Class I areas and the determination of
uniform rate of progress goals - expressed in deciviews - that provide
for reasonable progress towards achieving natural visibility by 2064. 
It is the Class I states responsibility assess these calculations. The
Class I States must also consult with those States, which may reasonably
be anticipated to cause or contribute to visibility impairment in their
Class I areas (40 CFR 51.308 (d)(1)(i-vi)). 

The baseline visibility conditions are calculated for the baseline
period between 2002 and 2004. The average impairment for the 20 most and
20 least impaired days are determined for each calendar year and
compiled into the average of three annual averages (40 CFR 51.308
(d)(2)(i)). The natural visibility conditions are determined for the
same baseline period with the most and least impaired days determined by
available monitoring data or an appropriate data analysis technique (40
CFR 51.308 (d)(iii-iv)). 

U.S. Environmental Protection Agency (EPA) released guidance on June 7,
2007 to use in setting reasonable progress goals. The goals must provide
improvement in visibility for the most impaired days, and ensure no
degradation in visibility for the least impaired days over the State
Implementation Plan (SIP) period. The State must also provide an
assessment of the number of years it would take to attain natural
visibility condition if improvement continues at the rate represented by
the reasonable progress goal. Figure 4 illustrates an example of how
Uniform Rate of Progress is calculated.  

 

                                                                        
                             

Figure 4.  Example calculation of Uniform Rate of Progress

Modeled Visibility Projections for 2018

The CMAQ air quality model was used to simulate base period emissions
and future emissions.  The modeling results for the base year period
(2002) and the year representing the end of the first planning period
(2018) are used to develop relative response factors (RRF) for each
component of particulate matter identified previously in this TSD.  The
relative response factors are multiplied by the measured species
concentration data during the base period (for the measured 20% best and
worst days). This results in daily future year species concentrations
data. The projected concentrations are then used to derive daily
visibility in deciviews and are averaged across all best and worst days
to create the projected future visibility. The results of this procedure
are plotted along with the uniform progress glide slope in Appendix M of
the MD Haze SIP.   

The modeling results presented in Appendix M of the MD Haze SIP show all
MANE-VU sites are projected to meet or exceed the uniform rate of
progress goals for 2018 on the 20 percent worst days. In addition, no
site anticipates increases in visibility impairment relative to the
baseline on the 20 percent best days. 

Summary of Photochemical Grid Modeling Results

In summary, the photochemical grid modeling, presented in the MD Haze
SIP, follow EPA’s modeling guidance and is acceptable to EPA.  All
MANE-VU sites are projected to meet or exceed the uniform rate of
progress goal for 2018 on the 20 percent worst days. In addition, no
site anticipates increases visibility impairment relative to the
baseline on the 20 percent best days (see Appendix M of the MD Haze
SIP). 

Contribution Assessment

The 1999 Regional Haze Rule requires States and Tribes to submit State
Implementation Plans (SIPs) to the U.S. Environmental Protection Agency
(USEPA) for approval by January 2008 at the latest. The haze SIPs must
include a “contribution assessment” to identify those states or
regions that may be influencing specially protected federal lands known
as Federal Class I areas. These states or regions would then be subject
to the consultation provisions of the Haze Rule. The Haze Rule also
requires a “pollution apportionment” analysis as part of the
long-term emissions management strategy for each site.

As described in the Conceptual Description portion of this TSD, sulfate
alone accounts for anywhere from one-half to two-thirds of total fine
particle mass on the 20 percent haziest days at MANE-VU Class I sites.
As a result of the dominant role of sulfate in the formation of regional
haze in the Northeast and Mid-Atlantic region, MANE-VU concluded that an
effective emissions management approach would rely heavily on
broad-based regional SO2 control efforts in the eastern United States.

Area of Influence for MANE-VU Class I Areas 

The key difference between SIPs from States with Class I areas and
States without Class I areas is the calculation of the baseline and
natural visibility for their Class I areas, and the determination of
reasonable progress goals. Class I States calculate baseline visibility
conditions for the period between 2002 and 2004. The average impairment
for the most and least impaired days are determined for each calendar
year and compiled into the average of three annual averages (40 CFR
51.308 (d)(2)(i)). The natural visibility conditions are determined for
the same baseline period with the most and least impaired days
determined by available monitoring data or an appropriate data analysis
technique (40 CFR 51.308 (d)(iii-iv)). In contrast, States without Class
I areas are responsible for doing their fair share to help meet the
reasonable progress goals established by the impacted Class I States,
and for maintaining their emissions monitoring network.  

There are seven Class I areas located in the Mid-Atlantic and Northeast.
 As a result, the Regional Haze Rule requires Maryland, in consultation
with MANE-VU and others, to identify where its emissions are most likely
to influence visibility in Class I areas. In order to identify states
whose emissions are most likely to influence visibility in MANE-VU Class
I areas, MANE-VU prepared the Contributions to Regional Haze in the
Northeast and Mid-Atlantic United States (Contribution Assessment). The
full report can be found in Appendix A of the MD Haze SIP. 

Based on that work, MANE-VU concluded that it was appropriate to define
an “Area of Influence” (AOI) including all of the states
participating in MANE-VU plus other states outside MANE-VU for which
modeling indicated they contributed at least two percent (2%) of the
sulfate ion in MANE-VU Class I areas in 2002. The Visibility Improvement
State and Tribal Association of the Southeast (VISTAS) also conducted an
AOI analysis, which used a level of one percent (1%) to assess whether
an upwind state significantly contributed. 

The primary contribution assessment tool used in the MD Haze SIP was the
Regional Modeling System for Aerosols and Deposition (REMSAD) (SAI,
2002). A significant feature of the REMSAD work used to evaluate
regional contributions is that NESCAUM reprocessed the SO2 emission data
from each state to take advantage of REMSAD’s tagging capabilities.
Thus, all SO2 emissions included in the model for the eastern half of
the country were tagged according to state of origin, and emissions from
Canada and the boundary conditions were also tagged. This allowed for a
rough estimation of the total contribution from elevated point sources
in each state to simulated sulfate concentrations at eastern receptor
sites. Using identical emission and meteorological inputs to those
prepared for the CMAQ SIP modeling platform, described earlier in this
TSD, REMSAD was used to simulate the annual average impact of each
state’s SO2emission sources on the sulfate fraction of PM2.5 over the
northeastern United States.  A more in-depth description of the REMSAD
modeling used for contribution assessment can be found in Appendix L of
the MD Haze SIP.

The REMSAD contribution assessment modeling contained in the MD Haze SIP
conforms to EPA modeling requirements and is acceptable to EPA.  

REMSAD Contribution Assessment Results

MANE-VU States decided that any state or region that contributed at
least 2 percent of total sulfate observed on 20 percent worst visibility
days in 2002 is contributing significantly to the haze problem in that
particular Class I area.  Based on the MANE-VU Contribution Assessment
and the application of the “≥ 2% SO2 rule,” emissions from
Maryland were determined to contribute to visibility degradation in the
following Class I Areas in MANE-VU: Acadia National Park, Maine;
Brigantine Wilderness (within the Edwin B. Forsythe National Wildlife
Refuge), New Jersey; Great Gulf Wilderness, New Hampshire; Lye Brook
Wilderness, Vermont; Moosehorn Wilderness (within the Moosehorn National
Wildlife Refuge), Maine; Presidential Range – Dry River Wilderness,
New Hampshire; and Roosevelt Campobello International Park, New
Brunswick. 

The MANE-VU Contribution Assessment indicates that emission sources
located in Maryland may also impact visibility at the following Class I
areas outside MANE-VU:  Dolly Sods Wilderness/Otter Creek Wilderness
Area (the Dolly Sods IMPROVE monitor is also representative of Otter
Creek) in West Virginia; and Shenandoah National Park. The Maryland
Department of the Environment has conferred with regional haze planning
staff in both Virginia and West Virginia regarding Maryland’s impacts
as described in sections 4.2.2 and 4.3.3 and Appendix Q of the MD Haze
SIP.   

BART 

On July 6, 2005 (70 FR 39104) EPA finalized 40 CFR 51 – Regional Haze
Regulations and Guidelines for Best Available Retrofit Technology (BART)
Determinations addressing the issues from the Circuit Court decisions.
The BART requirements were most recently updated on October 5, 2006.
BART is defined as an emission limitation based on the degree of
reduction achievable through the application of the best system of
continuous emission reduction for each pollutant, which is emitted by a
BART-eligible source. The changes to the rule included how the States
would identify the best system of continuous emission control technology
and by which States can consider an individual facility’s contribution
to regional haze when determining to require controls, and what the
level of control should be met. The rule changes also clarified the
requirements associated with demonstrating how emissions trading or
alternative programs may be used as an alternative to applying Best
Available Retrofit Technology (BART) Requirements. 

 

Congress defined sources potentially subject to BART - as major
stationary sources, including reconstructed sources; from one of 26
identified source categories which included utility and industrial
boilers, and large industrial plants such as pulp mills, refineries and
smelters; which have the potential to emit 250 tons per year or more of
any air pollutant, and which were placed in operation between August
1962 and August 1977. [CAA 169A (b)(2)(A) & (g)(7)]. A list of
BART-eligible sources within the MANE-VU region and in Maryland is
contained in Appendix G of the MD Haze SIP.

Maryland Sources Subject to BART

Source	Unit/Point ID(s)	Pollutant	Location

(County)	Facility I.D	Facility MW

EGU





	Mirant - Chalk Point	1, 2 & 3	NOX , SOX	Prince George’s 	033-0014
2647

Mirant – Morgantown	1 & 2	NOX , SOX	Charles	017-0014	1548

CPSG – Crane	2	NOX , SOX 	Baltimore	005-0079	415.8

CPSG – Wagner	3	NOX , SOX	Anne Arundel	003-0014	1058.5







	Non – EGU





	New Page/Westvaco/Luke Paper	25	SOX	Allegany	001-0011

	Independent/St. Lawrence Cement	24	SOX	Washington	043-0008

	Mettiki Coal	1	SOX	Garrett	023-0042

	

 

Five Factor Analysis for Each BART Source 

States are required to determine BART for each BART-eligible source.
According to 40 CFR 51.308(e)(1)(ii)(A) the determination of BART must
be based on an analysis of the best system of continuous emission
control technology available and associated emission reductions
achievable. 40 CFR 51.308(e)(1)(ii)(A) requires the analysis to take
into consideration the following five factors for the technology
available : 

1) The costs of compliance, 

2) The energy and non-air quality environmental impacts of compliance,
any 

3) Pollution control equipment in use at the source, 

4) The remaining useful life of the source, and 

5) The degree of improvement in visibility which may reasonably be
anticipated to result from use of the technology.

The BART-eligible sources were identified using the methodology in the
Guidelines for Best Available Retrofit Technology (BART) Determinations
under the Regional Haze Rule (40 CFR Part 51, Appendix Y.)  A detailed
description of each BART-eligible source and the identification analysis
is included in Appendix G of the MD Haze SIP.  

The Maryland Department of the Environmental (MDE) conducted BART
determinations for each BART-eligible source using the 5-Factor
Analysis, for PM only. Consistent with the MANE-VU Board (June, 2004)
decision, this analysis would include consideration of potential
visibility impacts as a result of installing various controls for
primary particulate matter. 

MANE-VU conducted modeling analyses of individual BART-eligible sources
using CALPUFF – a model preferred by EPA for assessing long range
transport of pollutants and their impacts - in order to provide a
regionally consistent foundation for assessing the degree of visibility
improvement (Factor 5) that could result from installation of BART
controls.  Summary spreadsheets of the MANE-VU CALPUFF modeling results
for the Maryland BART sources are included in Appendix G of the MD Haze
SIP.  CALPUFF was promulgated by the USEPA on April 15, 2003 as a
preferred dispersion model to assess long-range transport applications
(i.e. transport distances exceeding 50 km to approximately 300 km). Up
to this distance, a non-steady-state modeling approach which considers
spatial and time variations in meteorological conditions, such as
CALPUFF, is appropriate. For this modeling demonstration, CALPUFF was
used, consistent with the approved BART version. 

CALPUFF is a multi-layer, multi-species, non-steady state puff
dispersion model which can simulate the time and space varying
meteorological conditions on pollutant transport, transformation and
removal. CALPUFF uses three dimensional meteorological fields developed
by the meteorological processing program CALMET. 

CALPUFF contain algorithms for near source effects such as building
downwash, traditional plume rise, partial plume penetration, sub-grid
scale terrain interactions, as well as long range effects such as
pollutant removal (dry and wet deposition), chemical transformation,
vertical wind shear, over-water transport, and coastal interaction
effects.

The CALPUFF modeling performed for all of Maryland’s BART sources
conforms to EPA modeling guidance.  A detailed description of the
CALPUFF modeling can be found in Appendix G of the MD Haze SIP. 

The BART determinations for the sources subject to BART in Maryland are
summarized in Table 8.6 of the MD Haze SIP.  BART is the emission limit
for each pollutant based on the degree of reduction achievable through
the application of the best system of continuous emission control
technology available, taking into consideration: the costs of
compliance, the energy and the non-air quality environmental impacts of
compliance, any pollution control equipment in use or in existence at
the source, the remaining useful life of the source, and the degree of
improvement in visibility which may reasonably be anticipated to result
from the use of such technology.  The Department’s BART analyses   for
each source subject to BART are included in Appendix G of the MD Haze
SIP.  The BART analyses identified the best system of continuous
emission control technology available and include the consideration of
these five factors: the costs of compliance, the energy and non-air
quality environmental impacts, any existing pollution controls at the
source, the remaining useful life of the source and the degree of
improvement in visibility, the latter of which is determined in the
Visibility Impacts portion of each review memo.  The consideration of
the degree of improvement in visibility is based on the maximum 24-hour
NESCAUM-modeled impact at the Class I area of maximum impact.  However,
if an affected BART source conducted a CALPUFF modeling analysis that
used three years of meteorological data input, a BART determination was
performed using the 98th percentile deciview modeled impact, as allowed
in the Regional Haze Regulations and Guidelines for Best Available
Retrofit Technology (BART) Determinations, Final Rule (40 CFR Part 51,
July 6, 2005).

The BART for each source subject to BART was determined based on the
methodology in the Guidelines for Best Available Retrofit Technology
(BART) Determinations Under the Regional Haze Rule (40 CFR Part 51,
Appendix Y).  Table 9.6, table 9.10 and Appendix G-2 of the MD Haze SIP
summarize the level of control determined to be BART for the non-EGU and
EGU BART-eligible sources in Maryland. 

 Summary of EPA’s Technical Findings

The technical analyses and modeling used to assess uniform rate of
progress and to support the LTS were successfully developed consistent
with EPA’s interim and final modeling guidance.  All MANE-VU sites are
projected to meet or exceed the uniform rate of progress goal for 2018
on the 20 percent worst days. In addition, no site anticipates increases
visibility impairment relative to the baseline on the 20 percent best
days. The technical analyses and modeling used in the MD Haze SIP’s
contribution assessment and prediction of visibility impacts from BART
controls comply with EPA modeling guidance and recommendations.  EPA
finds the MANE-VU technical modeling analyses presented in the MD Haze
SIP to be acceptable because the models that were used were applied
according to EPA modeling guidance.

Areas designated as mandatory Class I Federal areas consist of  national
parks exceeding 6000 acres, wilderness areas and national memorial parks
exceeding 5000 acres, and all international parks that were in existence
on August 7, 1977 (42 U.S.C. 7472(a)).  In accordance with section 169A
of the CAA, EPA, in consultation with the Department of Interior,
promulgated a list of 156 areas where visibility is identified as an
important value (see 44 FR 69122, November 30, 1979). The extent of a
mandatory Class I area includes subsequent changes in boundaries, such
as park expansions (42 U.S.C. 7472(a)).  Although states and tribes may
designate as Class I additional areas which they consider to have
visibility as an important value, the requirements of the visibility
program set forth in  section 169A of the CAA apply only to
‘‘mandatory Class I Federal areas.”  Each mandatory Class I
Federal area is the responsibility of a ‘‘Federal Land Manager’’
(FLM).  (42 U.S.C. 7602(i))  

Albuquerque/Bernalillo County in New Mexico must also submit a regional
haze SIP to completely satisfy the requirements of section 110(a)(2)(D)
of the CAA for the entire State of New Mexico under the New Mexico Air
Quality Control Act (section 74-2-4).

 h

 h

\

c

{

P

Q

U

]



%

S

T

U

n

z

{



ˆ

“

È

Í

Ù

Û

Ý

Þ

ß

é

í

ï

ð

ñ

ô

 hi

 hi

hi

hi

hi

hi

hi

h7

 h

 h

h7

愀Ĥ摧紫

kdÚ

 h

 hk

April 2007

 The Class I designation applies to national parks exceeding 6,000
acres, wilderness areas and national memorial parks exceeding 5,000
acres, and all international parks that were in existence prior to 1977.
In

the MANE-VU area, this includes: Acadia National Park, Maine; Brigantine
Wilderness (within the Edwin

B. Forsythe National Wildlife Refuge), New Jersey; Great Gulf
Wilderness, New Hampshire; Lye Brook

Wilderness, Vermont; Moosehorn Wilderness (within the Moosehorn National
Wildlife Refuge), Maine;

Presidential Range – Dry River Wilderness, New Hampshire; and
Roosevelt Campobello International

Park, New Brunswick.

 VISTAS is comprised of the following states: Alabama, Florida, Georgia,
Kentucky, Mississippi, North Carolina, South Carolina, Tennessee,
Virginia, West Virginia, the Eastern Band of Cherokee Indians, and Knox
County, TN 

 PAGE   

 PAGE   1 

