^6° ST^
\T ST
^2
£
5
%
ui
(D
Model Clearinghouse: Operational Plan
-------
-------
EPA-454/B-16-008
December 2016
Model Clearinghouse:
Operational Plan
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Air Quality Assessment Division
Research Triangle Park, North Carolina
-------
-------
TABLE OF CONTENTS
Acknowledgements iii
1. Introduction 1
1.1 Need for a Model Clearinghouse 2
1.2 Purposes of the Model Clearinghouse 4
1.3 Scope of this Operational Plan 5
2. Functions of the Model Clearinghouse 7
2.1 Review of Proposed Regulatory Actions 7
2.2 Maintaining Awareness of Current Modeling Guidance and Historical Precedents 8
2.3 Communication of Decisions 10
2.4 Identifying Needs for Additional Modeling Guidance 11
3. Structure of the Model Clearinghouse 13
4. Model Clearinghouse Procedures 17
4.1 Clarification on Model Clearinghouse Formal Actions 17
4.2 Review of Alternative Models and Analytical Techniques 18
4.3 Review of Regulatory Packages and Policy Issues 21
4.4 Documentation and Communication of Case-Specific Reviews 22
5. References 25
Appendix A: The Guideline, Section 3.3 A-l
Appendix B: Model Clearinghouse Points of Contact B-l
Appendix C: Conceptual Flow Diagram for Alternative Model Approvals C-l
Appendix D: Example Model Clearinghouse Request/Response Memoranda D-l
Appendix E: Model Clearinghouse Formal Actions from 1981 to 2016 E-l
i
-------
ii
-------
Acknowledgements
We acknowledge the contributions since 1981 from the previous and current Directors of
the EPA's Model Clearinghouse (Dean Wilson, Dennis Doll, Warren Peters, Dennis Atkinson,
and George Bridgers) and the continual support of their management (Joseph Tikvart and Tyler
Fox) in providing the autonomy to develop the Model Clearinghouse into a respected and
integral part of the regulatory air quality modeling community. We also acknowledge the
instrumental assistance in this success from countless modeling scientists and other
representatives in all ten EPA Regional Offices, the Office of Air Quality Planning and
Standards, the Office of Research and Development, and the Office of Transportation and Air
Quality over the past three-and-a-half decades.
iii
-------
iv
-------
1. Introduction
The U.S. Environmental Protection Agency (EPA's) Model Clearinghouse, also known
as the Model Clearinghouse or MCH, is the central point of consultation and coordination within
the EPA for reviewing the use of air quality models and analytical techniques for demonstrating
compliance or attainment with the National Ambient Air Quality Standards (NAAQS) in
regulatory applications or implementation plans. All case-specific approvals of alternative
models by an EPA Regional Office, hereupon referred to as Regional Office or RO, require
consultation and concurrence by the MCH, per Section 3.2.2 of the Guideline on Air Quality
Models (40 CFR Part 51 Appendix W, "Guideline")1. When appropriately engaged, the MCH
ensures fairness, consistency, and transparency in regulatory air quality modeling decisions
across all of the ROs.
The MCH is also a nexus of information for the regulatory air quality modeling
community to gain knowledge of case-specific decisions by the EPA on the use of alternative
models and analytical techniques and to understand clarifications to the requirements and
recommendations of the Guideline. To facilitate a broad sharing of this information, the MCH
conducts monthly coordination conference calls with the ROs, hosts annual Regional, State, and
Local Modelers' Workshops for the co-regulating agencies and the triennial Conference of Air
Quality Models for the entire regulatory air quality modeling community as required by Section
320 of the Clean Air Act (CAA)2, maintains the Model Clearinghouse Information Storage and
Retrieval System (MCHISRS) on the EPA's Support Center for Regulatory Atmospheric
Modeling (SCRAM) website, and periodically produces summary reports.
1
-------
1.1 Need for a Model Clearinghouse
Section 165 of the CAA states that with regard to Prevention of Significant Deterioration
(PSD) analyses, "The Administrator... shall specify with reasonable particularity each air quality
model or models to be used under specified sets of conditions for the purposes of this part. Any
models or models designated under such regulations may be adjusted upon a determination, after
notice and opportunity for public hearing, by the Administrator that such adjustment is necessary
to take into account unique terrain or meteorological characteristics of an area potentially
affected by emissions from a source applying for a permit required under this part." In response
to this requirement and other regulatory needs, the Office of Air Quality Planning and Standards
(OAQPS) issued the Guideline of Air Quality Models. The Guideline established preferred air
quality models and recommends analytical techniques that may be applied to air pollution
control strategy evaluations and new source reviews, including PSD. The Guideline is intended
for use by the ROs in judging the adequacy of modeling analyses performed by the EPA, by
state, local, and tribal permitting authorities, and by industry. It is appropriate for use by other
federal government agencies and by state, local, and tribal agencies with air quality and land
management responsibilities (co-regulating agencies). The Guideline serves to identify, for all
interested parties, those modeling techniques and databases that the EPA considers acceptable.
The Guideline provides requirements and makes specific recommendations concerning air
quality models, databases, and general requirements for concentration estimates.
However, the Guideline also recognizes that: (1) there are situations where the
adjustment of the "preferred" air quality models is necessary to take into account unique terrain
or meteorological characteristics of an area; (2) the developing state of modeling science may
provide the opportunity for application of a new or revised model which is more appropriate than
2
-------
the recommended model; (3) for some classes of modeling problems the state of the modeling
science does not provide a basis for identifying appropriate refined models; and (4) database
availability in particular situations may warrant deviations from the Guideline. To allow for these
situations, the Guideline states in Section 3.2.2(a) that when a preferred model or database is not
used, the Regional Administrator may approve the use of other techniques that are demonstrated
to be more appropriate. The Guideline then provides general criteria for determining the
technical acceptability of alternative techniques. To assist the RO personnel in making such
technical judgments, a mechanism is needed by which in-depth review of alternate models can
be performed. Access to Agency personnel who have specialized knowledge about specific types
of modeling techniques is highly desirable.
Section 301(a) of the CAA authorizes the Administrator to delegate authority for carrying
out regulations and policies to the ROs. However, this Section also requires the Administrator to
"... promulgate regulations establishing general applicable procedures and policies for regional
officers and employees (including the Regional Administrator) to follow in carrying out a
delegation... Such regulations shall be designed-
(A) to assure fairness and uniformity in the criteria, procedures, and policies applied
by the various regions in implementing and enforcing the Act;
(B) to assure at least an adequate quality audit of each State's performance and
adherence to the requirements of this Act in implementing and enforcing the Act,
particularly in the review of new sources and in enforcement of the act; and
(C) to provide a mechanism for identifying and standardizing inconsistent or varying
criteria, procedures, and policies being employed by such officers and employees in
implementing and enforcing the Act."
3
-------
Thus, although the Regional Administrator has the authority to specify models that are
appropriate for use in a given situation, there is a need to provide for a mechanism that promotes
fairness and consistency in modeling decisions among the various ROs and the co-regulating
agencies.
1.2 Purposes of the Model Clearinghouse
To fulfill the needs described above in Section 1.1, the MCH has been established in
OAQPS with the primary purposes to provide a mechanism whereby the proposed case-specific
acceptance by a RO of a non-guideline or alternative model or analytical technique can be
reviewed for national consistency before final approval by the Regional Administrator.
Interrelated to ensuring this national consistency, the MCH provides a mechanism whereby the
in-depth technical evaluation and/or performance evaluation of a proposed alternative model or
analytical technique can be reviewed by those EPA personnel who are most familiar with the
types of models or analytical techniques to be employed. Finally, the MCH provides a
communication outlet for EPA's experience with the use of alternative models and analytical
techniques, databases, or other deviations from the Guideline and current guidance.
The establishment and purpose of the MCH are formally declared in Section 3.3 of the
Guideline. For convenient reference in the context of this Operational Plan, the text of the
Guideline, Section 3.3 is provided in Appendix A.
4
-------
1.3 Scope of this Operational Plan
The remainder of this Operational Plan describes the MCH in greater detail and how it
most efficiently and effectively operates.
• Section 2 presents the functions of the MCH with expanded explanation.
• Section 3 presents the principal structure of the MCH.
• Section 4 provides the procedures to be followed in engaging with and submitting
material to the MCH and how the review of material, development of comments, and the
communication of relevant important information to all parties are accomplished.
• Section 5 is a list of references cited throughout this document.
• Appendix A provides the relative text from Section 3.3 of the Guideline concerning the
MCH.
• Appendix B lists the MCH points of contact, including respective OAQPS and RO
personnel.
• Appendix C presents a conceptual flow diagram of the pre-EPA submittal development
steps for an alternative model or analytical technique justification by an applicant and
reviewing authority and the post-EPA submittal approval steps by the RO and MCH.
• Appendix D gives an example of a RO alternative model concurrence request and
subsequent MCH concurrence response memorandum.
• Appendix E provides a graphical review of the frequency of MCH formal actions from
1981 to 2016.
5
-------
6
-------
2. Functions of the Model Clearinghouse
The three main functions of the MCH that address the purposes discussed above in
Section 1.2 are to: (1) review proposed regulatory actions that contain modeling issues and
alternative models or analytical techniques; (2) develop and maintain a historical record of
alternative model and analytical technique decisions; and (3) communicate decisions on
regulatory modeling issues to all users in the regulatory air quality modeling community. These
functions are more fully described below in Sections 2.1 through 2.3. As a byproduct of its
operation the MCH is also in a position to identify needs for clarification of and potentially
changes to modeling guidance. This aspect of the MCH operation is described below in Section
2.4.
2.1 Review of Proposed Regulatory Actions
The major function of the MCH is to review case-specific proposed actions which
involve interpretation of modeling guidance, deviations from strict interpretation of such
guidance, and the use of options in the guidance, e.g., RO acceptance of alternative models or
analytical techniques and databases. This is handled in two ways:
1. The MCH, on request from the RO, reviews the Region's position on proposed
(specific case) use of an alternative model or analytical technique or other deviation
from the modeling guidance for technical soundness and national consistency.
2. The Clearinghouse screens regulatory actions, either in advance (upon request) or
when formally submitted, for adherence to modeling policy and makes
recommendations for resolution for any issues identified.
7
-------
In the review of alternative models and analytical techniques, the MCH first attempts to
conduct the review of the RO's request within its own resources. The basis for the review is the
requirements and recommendations set forth in the Guideline, available relevant guidance,
historical records of previous analogous cases and MCH reviews, and the technical expertise of
MCH personnel. As the need arises, the MCH may call upon other EPA personnel, e.g., other
ROs, the Office of Research and Development (ORD), or the Office of Transportation and Air
Quality (OTAQ), with specific expertise to assist in resolving complex issues or in the review of
all or parts of the proposed alternative model or analytical technique. In all cases the final
outcome of the review takes the form of a single concurrence or recommendation from the MCH,
with supporting rationale, to the RO. This "formal action" of review and subsequent concurrence
or recommendation by the MCH promotes the use of equivalent acceptance criteria by all of the
ROs.
2.2 Maintaining Awareness of Current Modeling Guidance and Historical Precedents
In order for the MCH to properly judge consistency in the interpretation of modeling
guidance, it is necessary to maintain awareness of current modeling guidance and to be cognizant
of past decisions involving the interpretation of this guidance in specific cases. The location of
the MCH in OAQPS allows for easy access and awareness of current modeling guidance. The
primary basis for modeling policy considered by the MCH is the Guideline. Other written
material also constitutes part of the modeling guidance, including workshop and conference
reports, guidelines, Federal Register rules and regulations, and records of previous MCH
concurrences and recommendations.
8
-------
One very important aspect of this function is the development and maintenance of a
historical record of regulatory decisions that involved interpretation of modeling guidance.
Although most regulations, e.g., State Implementation Plans (SIPs), are relatable to a strict
interpretation of the Guideline and related guidance documents,3 there are still many situations
that involve a deviation from a strict reading for either technical or broad policy reasons. For
example, it might be expeditious to just concur with an alternative analytical technique in a
compliance demonstration for an isolated new or modifying source even though the
recommended databases or perhaps the recommended model were not used in the analysis.
However, it is essential to document the circumstances involved with a case-specific approval so
as not to set a precedent for all sources because the environmental setting or other criteria may be
different in other permitting situations.
In order for the MCH to maintain technical and policy consistency in its
recommendations, it is necessary that a current database of decisions involving interpretation of
or deviation from modeling guidance be maintained, easily referenceable, transparent, and
openly accessible to the public. To accomplish this database of information, the MCH has
established the Model Clearinghouse Information Storage and Retrieval System (MCHISRS) for
the archival of clarifications to modeling guidance and all formal actions by the MCH, including
all situations involving the case-specific approval by a RO with MCH concurrence of alternative
models and analytical techniques. The MCHISRS forms the primary basis for the
communication of decisions described in the following section.
9
-------
2.3 Communication of Decisions
It is important that the regulatory air quality modeling community be made aware of
significant decisions involving the interpretations of modeling guidance. To fulfill this function,
the MCH utilizes numerous communication pathways, including the MCHISRS database which
is publically available on the EPA's SCRAM website at https://www.epa.gov/scram/air-qualitv-
model-clearinghouse.
Whenever there is a formal action involving the MCH, a new record in the MCHISRS
database is created and includes: (1) a brief description of the situation and of the resolution; (2)
the RO concurrence request memorandum along with the technical basis or justification material
from the applicant and/or co-regulating agency seeking the case-specific approval of an
alternative model or analytical technique; and (3) the MCH concurrence or response
memorandum along with any additional technical basis documentation deemed necessary to
support the EPA's decision. An electronic carbon copy of the MCH concurrence or response
memorandum, including all of the information contained in the new MCHISRS record, is shared
via email to all of the Regional Office Air Program Managers and appropriate Regional Office
Modeling Contacts51. This email is subsequently shared by the ROs with their respective co-
regulating agenciesb, as appropriate. Additionally, a notice of the MCH formal action is placed
prominently on the EPA's SCRAM website.
The MCH conducts monthly coordination conference calls with the ROs to assist in a
consistent exchange of pertinent information concerning the Guideline, modeling guidance and
clarifications, and MCH formal actions. The RO monthly coordination calls also provide a
a Current list of Regional Office Modeling Contacts: https://www.epa.gov/scram/air-modeling-regional-contacts.
b Current list of co-regulating agency contacts: https://www.epa.gov/scram/air-modeling-state-modeling-contacts.
10
-------
mechanism through which the Regional Office Modeling Contacts can share their experiences
with particular permit compliance demonstration and SIP modeling issues in their respective
regions to collectively broaden the knowledgebase and further promote consistency in modeling
related decisions throughout all of the ROs.
Annually, the MCH hosts a Regional, State, and Local Modelers' Workshops for the co-
regulating agencies. On a triennial basis, the MCH conducts a Conference of Air Quality Models
for the entire regulatory air quality modeling community as required by Section 320 of the CAA.
Both the workshops and conferences offer an opportunity for the MCH to summarize and present
all of the business of the MCH over the previous one or three years and to gain even further
feedback from the respective parts of the regulatory air quality modeling community.
Finally, the MCH is reinstituting the practice of producing a Model Clearinghouse
Annual Report to summarize significant decisions that have been made and the circumstances
involved over the previous year. The basis for this report is primarily details from the records
maintained in the MCHISRS database and also includes relevant details from any new rules and
regulations and information gleaned from the annual Regional, State, and Local Modelers'
Workshop, the Conference on Air Quality Models (if conducted in that year), and other
interactions with the co-regulating agencies or industrial stakeholders. The report can be used as
a reference to improve consistency in future decisions and as a source of technical information.
2.4 Identifying Needs for Additional Modeling Guidance
By the very nature of its business, the MCH is in a unique position to identify areas
where gaps exist and clarification is needed in EPA's modeling guidance and also in regulatory
policy related to modeling. The MCH is also a valuable resource for making recommendations
11
-------
and developing guidance to cover such situations because it is familiar with the circumstances
involved.
As necessary, the ROs may seek clarification from the OAQPS on technical issues and
areas of concern in a modeling protocol or PSD compliance demonstration. Through these
interactions and subsequent resolutions of the specific issues, clarifications of preferred modeling
procedures can ultimately become official EPA modeling guidance. This can happen in several
ways: 1) the preferred procedures are published as regulations or guidelines; 2) the preferred
procedures are formally transmitted as guidance to the Air Division Directors in the ROs; 3) the
preferred procedures are formally transmitted as guidance to the Regional Office Modeling
Contacts as a result of a regional consensus on technical issues; or 4) the preferred procedures
are relied upon in decisions and concurrences by the MCH that effectively establish national
precedent that the approach is technically sound.
12
-------
3. Structure of the Model Clearinghouse
This section describes the location of the MCH within the EPA and the key personnel
involved, including support staff.
The MCH is formally located within the Air Quality Modeling Group (AQMG), Air
Quality Assessment Division (AQAD) of the OAQPS. As such, the MCH exists within the
normal chain of command of the AQMG whose primary functions are to conduct regulatory air
quality modeling for EPA actions, address regulatory and technical issues related to regulatory
air quality modeling and develop regulatory air quality modeling guidance. This provides ready
access to modeling policy and technical expertise on air quality modeling and its implementation
in demonstrating compliance with the NAAQS and PSD increment. It also allows for efficient
hierarchical clearance concerning MCH determinations on sensitive issues. The Air Quality
Policy Division (AQPD) of the OAQPS and, when appropriate, the Office of General Counsel
(OGC) also participates in matters involving SIP attainment strategies, NSR/PSD referrals, and
related policy issues and other regulatory functions that involve broader policy decisions that
need to be model by the EPA.
In order to ensure that modeling issues contained in SIP submittals and related documents
are reviewed for consistency in policy and for technical credibility, there are two main modes of
access to the MCH, as described above in Section 2. Regional Office requests for review of
alternative models and analytical techniques and other proposed deviations from modeling
guidance are handled directly by MCH personnel in the AQMG. SIP submittals and related
documents, as well as requests from the Region Offices for assistance on NSR/PSD issues, are
screened by MCH personnel and then coordinated with personnel in the AQPD and OGC for
adherence to implementation and Agency policy. All proposed deviations from Agency policy
13
-------
are flagged for examination by the AQPD and OGC and may ultimately be resolved through
direct responses or clarification memoranda from the AQPD separate from any necessary formal
concurrence or recommendation from the MCH on the use of an alternative model or analytical
technique.
The primary responsibility for managing the MCH and ensuring that all of the functions
described above in Section 2 are carried out is performed by the Model Clearinghouse Director.
This individual is a full-time employee located within the AQMG and appointed by the Direction
of the AQAD based on recommendation by the Group Leader of the AQMG. The Model
Clearinghouse Director is responsible for ensuring that proper communication and coordination
are maintained in a timely fashion on all business of the MCH within the EPA, with the co-
regulating agencies, and with users in the regulatory air quality modeling community, as
appropriate. Additionally, the Model Clearinghouse Director maintains the MCHISRS database,
coordinates the annual Regional, State, and Local Modelers' Workshop and the triennial
Conference on Air Quality Models, and produces the Model Clearinghouse Annual Report.
In addition to the Model Clearinghouse Director, personnel or contacts with assigned
areas of specialty or technical expertise in the AQMG are included in the organizational structure
of the MCH. Each of the MCH contacts has, as part of his/her normal assignment outside of the
MCH, responsibility for a specific modeling or regulatory program, e.g., SO2, NO2, secondary
formation of Ozone or PM2.5, mobile air quality modeling, etc. Thus it is appropriate for these
individuals to also include technical support to the MCH function as part of their routine
responsibilities.
Although these personnel comprise the formal operation of the MCH and most of the
work of the MCH can be done by them, it may be necessary on occasion to draw upon modeling
14
-------
and policy expertise throughout the Agency to assist in resolving complex issues or parts of a
proposed alternative model or analytical technique. Thus, the MCH may call upon the ROs,
ORD, OTAQ, etc. with specific expertise to assist in resolving complex issues or in the review of
all or parts of the proposed alternative model or analytical technique. However as identified
earlier, the MCH is responsible for resolving and condensing all comments received into a single
concurrence or recommendation memorandum on the issue.
Appendix B provides a list of the central personnel included in the organizational
structure of the MCH and their contact information. Appendix B also identifies the key RO,
ORD, and OTAQ personnel most often involved in the business of the MCH. Periodically as
needed, Appendix B will be revised to reflect changes in these personnel or their contact
information.
15
-------
16
-------
4. Model Clearinghouse Procedures
This section describes the procedures that the MCH follows to review alternative models
and analytical techniques proposed for specific applications, screen regulatory packages/issues
for adherence to modeling policy, and document decisions and communicate results. For
reference and easy of understanding, Appendix C presents a conceptual flow diagram of the pre-
EPA submittal development steps for an alternative model or analytical technique justification by
an applicant and reviewing authority and the post-EPA submittal approval steps by the RO and
MCH. For additional clarity, the location of the "Formal Model Clearinghouse Process" within
the flow diagram is highlight.
4.1 Clarification on Model Clearinghouse Formal Actions
Throughout the year of its existence, there has been continual confusion by the co-
regulating agencies and broader regulatory air quality modeling community as to when the MCH
is formally reviewing a case-specific alternative model or analytical technique, often referred to
as "in the Clearinghouse." There are numerous occasions during which AQMG technical experts
through the MCH may participate in discussions and other levels of coordination on a case-
specific alternative model or analytical technique before it has been submitted by the co-
regulating agency for consideration and case-specific approval by the RO. However, the MCH is
not formally engaged in the review of a case-specific situation until the RO has formally
transmitted a concurrence request memorandum to the MCH stating the Region's position on
proposed use of an alternative model or analytical technique or other deviation from the
modeling guidance for technical soundness and national consistency.
The best example of such an occasion that the MCH may be involved in discussions on a
17
-------
case-specific situation is a model protocol coordination conference call that the RO and co-
regulating agency may invite the MCH to participate. During such a call, the MCH may offer
advice and help clarify requirements of Section 3.2.2 of the Guideline on what would be
necessary to appropriately justify the use of an alternative model or analytical technique in a
regulatory application. However, the participation by and advice offered during this conversation
does not constitute the MCH formally being engaged in a review. Rather, these occasions should
be viewed as opportunities for the MCH to gain situational awareness of a particular unique
situation, interrelated to the previous discussion in Section 2.2 above, that may eventually result
in formal engagement of the MCH by the RO and to provide additional guidance to the RO and
co-regulating agency and indirectly to the representatives of the new or modifying source on
developing a technically feasible alternative model or analytical technique justification. Often,
these early opportunities for technical collaboration between the RO, co-regulating agency,
representatives of the new or modifying source, and MCH will significantly reduce the amount
of time that the RO needs to develop their position on a case-specific application and,
subsequently, the amount of time that the MCH requires to review the formal RO concurrence
request.
4.2 Review of Alternative Models and Analytical Techniques
The Guideline provides criteria in Section 3.2.2 that allows the Regional Administrator
with consultation of the MCH to approve the use of alternative models or analytical techniques
not specifically recommended in the Guideline when it is determined: (1) that a preferred air
quality model is not appropriate for the particular application; (2) that a more appropriate model
or technique is available and applicable; or (3) that the Guideline does not require a specific
18
-------
technique.
The RO should first develop a position on the proposed application of the alternative
model or analytical technique and substantiate that position with its own thorough appropriate
analysis before formally requesting review by the MCH. Consistent with the example shown in
Appendix D, it is vitally important that the RO provide the MCH with: (1) a project overview;
(2) background information on the particular alternative model or analytical technique being
proposed; (3) key aspects of how the justification provided by the co-regulating agency and
facility fulfill the requirements of Section 3.2.2 of the Guideline; (4) additional analysis
performed by the RO to support their proposed regulatory action; (5) a conclusion with the RO's
recommended action; and (6) all pertinent information relative to the alternative model or
analytical technique and its application provided by the co-regulating agency and/or facility.
This information sharing should be accomplished in the RO's formal concurrence request
memorandum or an attached technical report to this memorandum (reference Appendix D for a
comprehensive example of a RO concurrence request memorandum). However, the RO should
reach out to the MCH at the start of their evaluation of a proposed alternative model or analytical
technique. Similar to a pre-application or modeling protocol meeting that discussed in Section
9.2.1 of the Guideline, early engagement with the MCH promotes broad situational awareness
within the Agency, ensures that the RO does not expend unnecessary resources on their
evaluation, avoids the applicant and/or co-regulating agency developing potentially unacceptable
justification material, and promotes timely resolution of the alternative model or analytical
technique request. Through early engagement, milestones and suitable timelines for all parties
involved can be discussed and agreed upon. Additionally, the RO is strongly encouraged to share
a draft form of their concurrence request and supporting technical evaluation with the MCH in
19
-------
advance of formal submittal such that additional collaboration on specific issues can occur prior
to the formal review by the MCH.
Formal requests to the MCH for review of alternative models and analytical techniques
should be sent by the ROs directly to:
George Bridgers, Model Clearinghouse Director
Air Quality Modeling Group, Air Quality Assessment Division
Office of Air Quality Planning and Standards
109 T.W. Alexander Drive
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711.
Upon receipt of a formal concurrence request from a RO, the MCH first attempts to
conduct the review of the RO's request within its own resources. The basis for the review is the
requirements and recommendations set forth in the Guideline, available relevant guidance,
historical records of previous analogous cases and MCH reviews, and the technical expertise of
AQMG personnel. If any regulatory policy issues or proposed deviations from Agency policy are
also interrelated to the RO request, then the MCH will engage with the appropriate personnel in
the AQPD and OGC. As the need arises, the MCH may call upon other EPA personnel, e.g.,
other ROs, the Office of Research and Development (ORD), or the Office of Transportation and
Air Quality (OTAQ), with specific expertise to assist in resolving complex issues or in the
review of all or parts of the proposed alternative model or analytical technique. If the issue(s)
involved are of broad national significance, it may be necessary to obtain a consensus of all the
ROs before the MCH finalizes a response.
The MCH makes every effort to provide a written response to the RO within four weeks
of receipt of the formal concurrence request memorandum. As discussed above, early
20
-------
engagement with the MCH at the start of the RO evaluation increases the situational awareness
of the issues and will further expedite the review by the MCH. The primary exceptions to the
four-week estimated timeframe are those cases where additional collaboration is required with
other EPA personnel or a consensus opinion of all the ROs is needed.
Again, it should be remembered that the MCH is an internal service within the EPA and
primarily provided to the ROs. The MCH does not interact directly with the co-regulating
agencies or with industrial stakeholders on case-specific situations, since this would compromise
the MCH's function as an independent, second-level reviewer in the process of approving
alternative models and analytical techniques as defined in Sections 3.2 and 3.3 of the Guideline.
However, there are circumstances where it is important for the MCH, Regional, Office, co-
regulating agency, and representatives of the industrial facility in question to collaborate on a
specific case. In such circumstances, the co-regulating agency and respective RO should closely
work together to arrange a mutually agreed-upon flow of information and/or conversations.
4.3 Review of Regulatory Packages and Policy Issues
All Federal Register action packages (advanced opinion, proposal or final) or NSR/PSD
questions on specific applications submitted to OAQPS are screened by MCH for consistency
with current modeling guidance. If significant deviations from guidance are identified, these
issues are then discussed within the MCH and with the appropriate personnel in the AQPD and
OGC. (Where there are significant deviations from guidance, the RO should have presented a
position, with appropriate justification, in the Federal Register package supporting either
approval or disapproval.) In all cases final resolution of the regulatory action rests with AQPD,
with input from the AQAD and OGC. The MCH, if at all possible, reaches a decision on the
21
-------
acceptability of the approach being utilized in the regulatory action within the normal review
period for SIP processing. If particularly complex issues are involved, which require additional
evaluation and more lengthy coordination with personnel outside the MCH, the MCH
coordinates with the AQPD on an extension to the regular review period in accordance with the
"SIP Processing Manual."4 All significant decisions involving modeling in regulatory packages
are included in the MCHISRS database, presentations at the Regional, State, and Local
Modelers' Workshops and/or Conferences on Air Quality Models, and the Model Clearinghouse
Annual Reports.
4.4 Documentation and Communication of Case-Specific Reviews
As discussed above in Section 2.3, one of the three primary functions of the MCH is the
communication of decisions. Whether the MCH is engaged in the review of alternative models or
analytical techniques for specific applications or screening regulatory packages/issues for
adherence to modeling policy, there is a formal documentation process that the MCH adheres.
First, the MCH develops a concurrence or response memorandum on all reviews of
proposed regulatory actions and provide it to the requesting RO along with any additional
technical basis documentation deemed necessary to support the EPA's decision. An electronic
carbon copy of the MCH concurrence or response memorandum and associated supporting
material is shared via email to all of the Regional Office Air Program Managers and appropriate
Regional Office Modeling Contacts. As mentioned previously, this email is subsequently shared
by the ROs with their respective co-regulating agencies, as appropriate. A notice of the MCH
formal action is also be placed prominently on the EPA's SCRAM website.
All documentation related to any MCH formal action are archived in the MCHISRS
22
-------
database. A case-specific record is created to which all material associated with the MCH's
review is entered. This includes the RO concurrence request memorandum along with the
technical basis or justification material from the applicant and/or co-regulating agency seeking
the approval of an alternative model or analytical technique and also includes the MCH
concurrence or response memorandum along with any additional technical basis documentation
deemed necessary to support the EPA's decision. The case-specific MCHISRS record also
includes key words and other tracking information such that the MCHISRS database is broadly
searchable for relevant issues by the MCH, ROs, co-regulating agencies, and regulatory air
quality modeling community.
Annually, the MCH is recommitting to producing a Model Clearinghouse Annual Report.
This report summarizes significant decisions that have been made and the circumstances
involved over the previous year and also includes relevant details from any new rules and
regulations. Additional information gleaned from the annual Regional, State, and Local
Modelers' Workshop, the Conference on Air Quality Models (if conducted in that year), and
other interactions with the co-regulating agencies or industrial stakeholders will be included in
the Model Clearinghouse Annual Report, as appropriate.
Finally, as needed, the MCH will continue to provide other communications and
assistance services that have been previously found useful. These include (but are not limited to):
periodic visits to ROs, often during Region specific modeling meetings and workshops, to
exchange information and maintain the rapport of the MCH; status reports and presentations on
the MCH and its activities for stakeholder meetings and conferences; and identification of areas
where modeling guidance is lacking or ambiguous and make recommendations for clarifying
these problems.
23
-------
24
-------
5. References
1 U.S. Environmental Protection Agency, 2016. Guideline on Air Quality Models. 40 CFRPart
5 1 Appendix W. https://www3.epa.gov/ttn/scram/guidance/guide/AppendixW 2016.pdf.
2 U.S. Congress, 1990. Clean Air Act Amendments of1990. 104 Stat. 2468, P.L. 101-549.
3 U.S. Environmental Protection Agency, 2014. Modeling Guidance for Demonstrating
Attainment of Air Quality Goals for Ozone, PM2.5, and Regional Haze. Office of Air
Quality Planning and Standards, Research Triangle Park, NC.
https://www3.epa.gov/ttn/scram/guidance/guide/Draft Q3-PM-RH Modeling Guidance-
2014.pdf.
4 U.S. Environmental Protection Agency, 2005. SIP Procession Manual. Office of Air Quality
Planning and Standards, Research Triangle Park, NC.
https ://cfpub. epa. gov/oarwebadmin/ sipman/sipman/.
25
-------
26
-------
Appendix A: The Guideline, Section 3.3
The establishment and purpose of the MCH is formally declared in Section 3.3 of the
Guideline. For convenient reference in the context of this Operational Plan, the text of the
Guideline, Section 3.3 is provided:
3.3 EPA 's Model Clearinghouse
a. The Regional Administrator has the authority to select models that are appropriate for
use in a given situation. However, there is a need for assistance and guidance in the selection
process so that fairness, consistency, and transparency in modeling decisions are fostered
among the EPA Regional Offices and the state, local, and tribal agencies. To satisfy that
need, the EPA established the Model Clearinghouse to serve a central role of coordination
and collaboration between EPA headquarters and the EPA Regional Offices. Additionally,
the EPA holds periodic workshops with EPA Headquarters, EPA Regional Offices, and state,
local, and tribal agency modeling representatives.
b. The appropriate EPA Regional Office should always be consulted for information and
guidance concerning modeling methods and interpretations of modeling guidance, and to
ensure that the air quality model user has available the latest most up-to-date policy and
procedures. As appropriate, the EPA Regional Office may also request assistance from the
EPA's Model Clearinghouse on other applications of models, analytical techniques, or
databases or to clarify interpretation of the Guideline or related modeling guidance.
c. The EPA Regional Office will coordinate with the EPA's Model Clearinghouse after
an initial evaluation and decision has been developed concerning the application of an
alternative model. The acceptability and formal approval process for an alternative model is
described in section 3.2.
A-l
-------
A-2
-------
Appendix B: Model Clearinghouse Points of Contact
Model Clearinghouse Points of Contact
Revised: December 15, 2016
Model Clearinghouse Primary Contacts (AQMG. AQAD)
Name
ResDonsibilitv
TeleDhone
Email
George Bridgers
Model Clearinghouse Director
919-541-5563
bridgers.george@epa.gov
Tyler Fox
AQMG Leader
919-541-5562
fox.tyler@epa.gov
George Bridge rs
03 and PM2 5 Secondary Formation Modeling
919-541-5563
bridgers.george@epa.gov
Roger Brode
AERMOD
919-541-3518
brode.roger@epa.gov
Chris Misenis
Prognostic Met (WRF/MMIF)
919-541-2046
misenis.chris@epa.gov
Chris Owen
CO, N02, Direct PM2 5, and Tranporation/Mobile Modeling
919-541-5312
owen.chris@epa.gov
James Thurman
AERSCREEN and AERMET, S02 and Pb Modeling
919-541-2703
thurman.james@epa.gov
Clint Tillerson
AERMOD
919-541-2051
tillerson.clint@epa.gov
Brian Timin
03/ PM25/ and Regional Haze SIP Modeling
919-541-1850
timin.brian@epa.gov
Additional EPA Contacts
Name
Resoonsibilitv
Teleohone
Email
Raj Rao
AQPD NSR/PSD
919-541-5344
rao.raj@epa.gov
Steven Perry
ORD - Atmospheric Model Application & Analysis Branch
919-541-1896
perry.steven@epa.gov
Meg Patulski
OTAQ-Transportation Conformity
734-214-4842
patulski.meg@epa.gov
Brian Doster
OGC - Air and Radiation Law Office
202-564-1932
doster.brian@epa.gov
Regional Office Modeling Contacts*
Reeional Office
Reeional Modeline Contact
TeleDhone
Email
1
Lerian Biton
617-918-1267
biton.lerian@epa.gov
II
Annamaria Colecchia
212-637-4016
colecchia.annamaria@epa.gov
III
Tim Leon-Guerrero
215-814-2192
leon-guerre ro.tim@epa.gov
IV
Rick Gillam
404-562-9049
gilliam.rick@epa.gov
V
Randy Robinson
312-353-6713
robinson.randall@epa.gov
VI
Erik Snyder
214-665-7305
snyder.erik@epa.gov
VII
Andy Hawkins
913-551-7179
hawkins.andy@epa.gov
VIII
Rebecca Matichuck
303-312-6867
matichuck.rebecca@epa.gov
IX
Carol Bohnenkamp
415-947-4130
bohnenkamp.carol@epa.gov
X
Jay McAlpine
206-553-0094
mcalpine.jay@epa.gov
* Most Regional Offices have several modeling contacts, but only the lead Regional Modeling Contacts are listed here.
B-l
-------
B-2
-------
Appendix C: Conceptual Flow Diagram for Alternative Model Approvals
Pre-application or Modeling
Protocol Meeting between
Applicant and Reviewing
Authority
(Reference thq Guideline, Section 9.2.1)
Modeling Protocol Established
and/or Agreed Upon by Applicant
and Reviewing Authority
N
>
f
Reviewing Authority Follows
Routine Pathway for Permit
Issuance, including Public Notice
and Comment
Note:This Page Includes the Pre-EPA
Submittal Development Steps for an
Alternative Model or Analytical
Technique Justification by an Applicant
and Reviewing Authority
Issues requiring EPA
Regional Office Consultation
Concurrence, or Approval?
Initial Interaction Between
Applicant, Reviewing Authority, and
EPA Regional Office to Discuss
Modeling Protocol and Identify
Additional Modeling Issues that
Need Res olution
N
f
Modeling Protocol Established
and/or Agreed Upon by Applicant
and Reviewing Authority with
Input and Considerations from
EPA Regional Office
>
f
Reviewing Authority Follows
Routine Pathway for Permit
Issuance,including Public Notice
and Comment, with Appropriate
Documentation of all EPA
Regional Office Required
Engagements and Decisions
Alternative Model or
Analytical Technique to be
Considered?
Initial Interaction Between
EPA Regional Office and MCH to
Discuss Modeling Protocol and
that Need Justification
Modeling Protocol Established
and/or Agreed Upon by Applicant
and Reviewing Authority with
Input and Considerations from
EPA Regional Office and MCH
Development of Alternative Model
Justification with Supporting
Technical Material Consistent with
the Guideline, Section 3.2.2 and
Submittal to EPA Regional Office
j
See Following Page for Next
Step in Process
c-i
-------
Process Continuation from
Previous Page
EPA Regional Office Review of
Alternative Model Justification and
Supporting Technical Material
EPA Regional Office Interactions with
MCH, Including the Sharing of
Alternative Model Justification and
Sup p orting T echnical M aterial from
Reviewing Authority and the Regional
Office's Draft Concurrence Request
Memorandum
Formal Model Clearinghouse Process
EPA Regional Office Submits Formal
Concurrence Request Memorandum to
MCH with Alternative Model
Justification and Supp ortingTechnical
Material from Reviewing Authority and
Any Additional Material Necessary to
Support Regional Office Decision
I
MCH Review of Alternative Model
Justification and All Supporting
Material. Additional Engagements
with EPA Personnel with Specific
Expertise, as Needed
EPA Regional Office Formal
Concurrence Request, MCH
Formal Response, and All
Supporting Technical Material
Entered into MCHISRS
EPA Regional Office Issues Formal
Alternative Model Approval to
Reviewing Authority with Any
Additional Information or Caveats
Resulting From MCH Formal
Review
' Reviewing Authority Follows '
Routine Pathway for Permit
Issuance, including Public Notice
and Comment, with Alternative
Model Justification and Approval
Documentation and Appropriate
Documentation of All Other EPA
Regional Office Required
Engagements and Decisions ,
Note: This Page Includes the Post-EPA
Submittal Approval Steps for an
Alternative Model or Analytical
Technique Justification by the EPA
Regional Office and MCH
C-2
-------
Appendix D: Example Model Clearinghouse Request/Response Memoranda
This appendix contains an example of a concurrence request on a proposed regulatory
action by a RO and the resulting MCH response. Both the request and the response are formal
memoranda signed by the appropriate personnel for this situation. This example is provided for
reference of a MCH formal action and the types of information that should be included in the
memoranda from the RO and MCH. The particular facility and alternative model justification in
this example is not significant and included only for illustrative purposes.
Page D-2 is the RO concurrence request memorandum. Pages D-3 through D-9 are a
technical report developed by the RO to provide the MCH with a project overview, background
information on the particular alternative model or analytical technique being proposed, key
aspects of how the justification provided by the co-regulating agency and facility fulfill the
requirements of Section 3.2.2 of the Guideline, additional analysis performed by the RO to
support their proposed regulatory action, and a conclusion with the RO's recommended action.
Pages D-10 through D12 are the MCH concurrence response memorandum.
It should be noted that additional information from the co-regulating agency and facility
were attached to the RO request, including the facility's full modeling report, but are not include
here for space/printing conservation purposes. Anyone wishing to review these additional
materials can search the MCHISRS database, record 16-1-01, on the EPA's SCRAM website at
https://www.epa.gov/scram/air-qualitv-model-clearinghouse.
D-l
-------
o* ^ 4% UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
% REGION 1
I S 5 POST OFFICE SQUARE. SUITE 100
\ -VKr- - BOSTON, MA 02109-3912
^ PROlfe&
APR 0 7 2016
MEMORANDUM
SUBJECT: Request to approve the use of the beta alternative formulation of surface friction
velocity (u*) non-regulatory default option in AERMET version 15181; alternative
refined model demonstration
FROM: Lei ran Biton, Physical Scientist
Air Permits. Toxics and Indoor Programs Unit, Air Programs Branch, Region 1
TO: George Bridgers, Director of Model Clearinghouse
Air Quality Modeling Group, Office of Air Quality Planning and Standards
THRU: David Conroy, Chief
Air Programs Branch. Region
EPA Region 1 seeks concurrence from the Model Clearinghouse on approval of the use of the
beta alternative formulation of surface friction velocity (u*) non-regulatory default option
(ADJ_U*) in AERMET version 15181. EPA Region 1 has concluded that the second condition
of Section 3.2.2(b) of Appendix W has been satisfied by the submittal from the New Hampshire
Department of Environmental Services (DES), and would like to approve the use of the beta
ADJ_U* option (either with or without the Bulk Richardson option) in AERMET version 15181
as the meteorological preprocessor for AERMOD version 15181 for this modeling. In support of
this requested approval, we have prepared a technical report that reviews the DES submittal and
documents our basis for decision-making. The technical report is attached for your reference.
Thank you for your careful attention to this matter and we look forward to your response. Please
feel free to contact me at 617-918-1267 with any questions about this request or the attachment.
Attachment
Toll Free •1-088-372*7341
Internet Address (URL) • http./'/www epa.gov/region1
Recycle
-------
Technical Report
Evaluation of the request for use of the beta adjust u* option in AERMET for
modeling for the 2010 1-hour sulfur dioxide standard at Schiller Station in
Portsmouth, New Hampshire
Leiran Biton
EPA Region 1 Modeling Contact
April 7, 2016
The New Hampshire Department of Environmental Services (DES) submitted a request for the
use of an alternative model to the U.S. Environmental Protection Agency (EPA) Region 1 Office
in a letter dated March 18, 2016. DES proposed to use an alternative formulation for the surface
friction velocity (u*) in the AERMET meteorological preprocessor (version 15181) to the
AERMOD model (version 15181) in its modeling of Schiller Station, operated by Eversource
Energy, LLC, in Portsmouth, New Hampshire. Specifically, DES has submitted the request for
the adjusted surface friction velocity technique, known as the beta adjust u* option, for modeling
intended to demonstrate compliance with the 2010 1-hour sulfur dioxide (SO2) National Ambient
Air Quality Standards (NAAQS). The modeling demonstration would apply to the Response to
Order on Title V Petition VI-2014-04 regarding the issuance of a proposed Title V Operating
Permit TV-0053 for Schiller Station, and also to commitments made by DES in its January 5,
2016 submittal under the SO2 Data Requirements Rule.
This technical report provides an assessment of the submitted request for use of the beta adjust u*
option in AERMET and describes the rationale for a recommendation regarding the request. The
submittal by DES included an attachment prepared by Exponent Inc. on behalf of Eversource
Energy. DES has explicitly stated that it concurs with the request for use of the beta adjust u*
option in AERMET as an alternative model; therefore, this report treats the Exponent attachment
as DES's own justification. The DES submittal is attached to this report.
Project overview
Schiller is a 150-MW capacity wood and fossil fuel-fired electricity generating facility located in
Portsmouth, New Hampshire. Schiller Station consists of three 50-MW capacity electric utility
steam boilers, two of which (Units 4 and 6) burn coal or oil and one of which (Unit 5) burns
biomass. Schiller is owned and operated by Eversource Energy, LLC, previously known as
Public Service of New Hampshire. Emissions from Schiller are released through three tall
stacks—one per boiler—ranging approximately 68-70 m in height at elevations of 6.4-7.3 m. The
temperature of stack releases ranges from 431 K (316 °F) to 450 K (350 °F). Stack and emissions
specifications are described in Table 1 of the DES submittal.
Terrain in the immediate area (within around 10 km) around Schiller is simple, consisting of
water bodies and low-elevation, flat or gently rolling features. Beyond the immediate area,
terrain becomes increasingly complex, with complex terrain features (i.e., features with
elevations above the height of the stack) beginning at around 16 km from the source.
Specifically, Mount Agamenticus (elevation 211m; about 16 km from Schiller) in York, Maine
D-3
-------
is the nearest complex terrain feature, and some further features in the 50 km square domain
around the source are even higher in elevation.
In addition to Schiller, it is anticipated that Newington Station will be included in the modeling.
Newington, located within 1 km of Schiller, is a natural gas and oil-fired electricity generating
facility with rated capacity of approximately 400 MW and stack height of 125 m.
Based on these project details, the releases from Schiller and Newington are well characterized
as buoyant plumes emitted from tall stacks in a region with complex terrain.
Background on default surface friction velocity and the beta adjust u* option
Starting in version 12345, AERMOD has included non-regulatory default options (identified
with the "beta" keyword) to address concerns regarding model performance under low wind
speed conditions. Specifically, in the current formulation, the model routinely underpredicts u*
during stable boundary layer conditions under low wind speeds. The u* parameter is key in
determining the height of mechanical mixing. Therefore, underestimating u* results in
underestimates in mixing layer height, leading to overestimates in concentrations in the mixed
layer.
The beta adjust u* option, designated by the beta ADJU* keyword introduced in version 12345
of the AERMET meteorological processor and augmented in subsequent versions, is one of the
tools available to address these concerns. The beta adjust u* option has been developed based on
peer-reviewed work by Qian and Venkatram (2011) and Luhar and Rayner (2009), as described
in the AERMOD Model Formulation Document Addendum (EPA 2015a). Additional non-
regulatory default beta options—specifically LOWWIND1, LOWWIND2, and LOWWIND3—
are also available as keywords in the AERMOD model. However, DES has requested only the
use of the beta adjust u* option for this modeling analysis, so the beta low wind options are not
discussed further in this report.
EPA has conducted model performance evaluations of the beta adjust u* option and the current
regulatory default AERMOD system (EPA 2015b). The evaluations were performed against
results from monitoring field studies to investigate diffusion under low wind speed conditions,
and against results from a field study with a tall stack in complex terrain where stable and low
wind speed conditions can also be important. The results of these evaluations indicated
significant overprediction using the regulatory default AERMET/AERMOD, and better
performance—though still somewhat overpredicting—using the beta adjust u* option. Based in
part on the results of these evaluations, EPA has proposed to designate the beta adjust u* option
as the default regulatory formulation in AERMET for estimating u* under stable conditions with
low wind speeds in the Guideline on Air Quality Models (i.e., 40 CFR Part 51 Appendix W;
hereafter, Appendix W).1
1 Revision to the Guideline on Air Quality Models: Enhancements to the AERMOD Dispersion
Modeling System and Incorporation of Approaches To Address Ozone and Fine Particulate
Matter; Proposed Rule. Federal Register, Vol. 80, No. 145, 45340-45387, 2015 July 29.
Beta adjust u* option for Schiller Page 2 of 7 Biton - Technical Report
D-4
-------
Process for approving an alternative model
According to Section 3.2.2(a) of Appendix W, the EPA Regional Office is responsible for
determining the acceptability of a model. Specifically,
Where the Regional Administrator finds that an alternative model is more appropriate
than a preferred model, that model may be used subject to the recommendations of this
subsection. This finding will normally result from a determination that (1) a preferred air
quality model is not appropriate for the particular application; or (2) a more appropriate
model or analytical procedure is available and applicable.
Section 3.2.2(b) of Appendix W goes on to describe the approval process for an alternative
model:
There are three separate conditions under which such a model may normally be approved
for use: (1) If a demonstration can be made that the model produces concentration
estimates equivalent to the estimates obtained using the preferred model; (2) if a
statistical performance evaluation has been conducted using measured air quality data
and the results of that evaluation indicate the alternative model performs better for the
given application than [the preferred model]; or (3) if the preferred model is less
appropriate for the specific application, or there is no preferred model.
DES has indicated its intention to use the second condition as its justification, subject to the
procedures for determining the acceptability of the alternative model using "established
procedures and techniques" as described in Section 3.2.2(d) of Appendix W. This subsection also
states that preparation and implementation of the evaluation protocol should be acceptable to the
state regulatory agency and EPA, as well as the regulated entity. EPA Region 1 held a
conference call on March 2, 2016 with representatives from the EPA Model Clearinghouse,
DES, Eversource Energy, and Exponent Inc. to discuss the process for demonstrating
appropriateness of an alternative model. This discussion satisfied the requirements for state,
EPA, and industry participation in the development of an evaluation protocol described in
Section 3.2.2(d) of Appendix W.
In December 2015, EPA issued a memorandum that clarified the approval process for non-
regulatory beta options in AERMOD that have been proposed as regulatory options in the
proposed revision to Appendix W (EPA 2015c). This memorandum confirmed that the use of all
non-default beta options, including the beta adjust u* option, in regulatory modeling must receive
EPA Regional Office approval.
In response to a request for the use of the beta adjust u* option by the Alaska Department of
Conservation (ADEC) to characterize air quality resulting from the Donlin Mine, EPA Region 10
approved the request based on demonstration supplied by ADEC and the relevance of the model
evaluations described in the previous sections (EPA 2015d). In its approval, EPA Region 10
supplied additional analysis of the influence of adjust u* on other meteorological parameters.
This analysis showed relatively moderate impacts on parameters for power plant sources
compared to the effects on sources with lower release heights. Subsequently, the EPA Model
Clearinghouse concurred on that approval and indicated that the justification was well-reasoned,
Beta adjust u* option for Schiller Page 3 of 7 Biton - Technical Report
D-5
-------
thoroughly documented, and demonstrated that the beta adjust u* option performed better than
the regulatory default option for that application (EPA 2016).
Statistical performance evaluation
The DES submittal cites several published statistical analyses as the basis for justifying the use
of the beta adjust u* option in the Schiller modeling. Specifically, DES presents information from
Paine et al. (2015) and EPA model evaluation studies presented at the 11th Conference on Air
Quality Modeling,2 specifically the Cordero Rojo surface coal mine fugitive dust study, the 1974
NOAA Oak Ridge study for low-level release, and the 1974 NOAA Idaho Falls tracer study for
low-level release.
The Cordero Rojo, Oak Ridge, and Idaho Falls studies are less directly applicable to the Schiller
scenario because the release heights from those studies are low-level, whereas Schiller (and
Newington) release buoyant plumes from tall stacks. The evaluation for the Gibson Generating
Station presented in Paine et al. (2015) is similarly limited in relevance to Schiller because of the
flat terrain of the area around Gibson.
The Lovett evaluation database, which is not explicitly mentioned by DES, but which is
presented in the most recent AERMOD model evaluation document (EPA 2015b), provides a
more comparable scenario to that of Schiller. The Lovett database consists of 2,595 hours of
ambient SO2 monitoring data from 12 monitors near the Lovett Power Plant, located in a rural
area with mountainous terrain along the Hudson River in New York. Some of the monitors had
elevations above the release height of Lovett's 145 m stack, and at distances from the source of
2-3 km. For the Lovett evaluation database, correlation is better with the beta adjust u* option
than the regulatory default option at relevant concentrations.3 In fact, the relevant modeled
concentrations at Lovett are actually higher using the beta adjust u* option compared with those
using the regulatory default. This suggests greater modeled impacts using the beta adjust u*
option at near-source locations (i.e., within several kilometers) than at more remote locations,
where impacts have been shown in the DES submittal to be lower. Therefore, it is likely that
impacts at nearer source impacts would be higher using the beta adjust u* option.
The Mercer County, ND evaluation described by Paine et al. (2015) is also highly relevant to the
Schiller scenario. The Mercer County database consists of approximately four years of SO2
monitoring data focused on two facilities in a region with complex terrain, and includes three
monitors at elevations near or above some stack release heights at distances of nearly 10 km. For
one of these monitoring locations (DGC#17), modeled concentrations were significantly closer
to monitored values, though still somewhat overpredicting, with the use of the beta adjust u*
option as compared to the regulatory default options; predictions at other monitoring locations
did change with use of the beta adjust u* option for this study.
2 https://www3.epa.gOv/scram001/l lthmodconf.htm
3 Because the form of the NAAQS is based on the three-year average of 99th percentile of daily
maximum SO2 concentrations, the 5-year average 4th highest modeled SO2 concentration is the
relevant comparison against the NAAQS. This process is described in detail in an EPA
memorandum on the subject (EPA 2010).
Beta adjust u* option for Schiller Page 4 of 7 Biton - Technical Report
D-6
-------
At Schiller, the relevant distances for impacts in complex terrain are 16 km or greater away from
the source. Though there is no evaluation database analysis for impacts in complex terrain at this
distance that match the precise characteristics of the Schiller scenario, the analyses cited above
provide a sufficient basis for making an assessment regarding the adequacy of the statistical
performance evaluation. Better model performance in the near field may translate into better
model performance at longer distances. However, no conclusive model performance evaluation
was available at the time of this review to confirm this notion, and this represents a data gap in
evidence provided for this alternative model justification.
Additional site-specific evidence
The DES submittal indicates that the regulatory default options in AERMET version 15181 and
AERMOD version 15181 lead to controlling concentrations at receptors on Mount Agamenticus
at elevations from 129 m to 147 m. These concentrations occur during low-wind speed and stable
boundary conditions. Table 2 of the submittal indicates that u* values are very low (0.033-0.077
m/s) for hours during which concentrations at the top ten receptors in the default modeling are
highest. At those receptor locations, using the beta adjust u* option increases 5-year average u*
values 62-96% (to 0.104-0.114 m/s). As a result of the increase in u* from the use of the beta u*
option, 5-year average 4th highest concentrations at these receptors dropped by 57-64%, from
93.9-100.6 |ig/m3 to 35.4-41.0 |ig/m3.
Significantly, the use of the beta adjust u* option shifted the controlling concentration from the
more remote ten receptors at Mount Agamenticus to a cluster of ten receptors in Eliot, Maine,
directly across the Piscatequa River within 1 km from Schiller (see Table 4 and Figure 3 in the
submittal). At these receptors, there were insignificant changes in u* and relevant concentration
values between the regulatory default and alternative modeling configurations; this indicates that
stable low wind speed conditions are not controlling at these receptors. For these ten receptors,
the 5-year average u* values are 0.62-0.76 m/s for relevant concentrations, which range from
51.4 to 54.1 |ig/m3.
The analysis in the DES submittal indicates that the beta adjust u* option only has significant
effects in the modeling domain at receptors with elevations at or above the height of release.
Specifically, the analysis showed that stable conditions with low wind speeds are the controlling
meteorological conditions for receptors with elevations above 85 m, and that concentrations at
these receptors are often lower by more than 50% under the beta adjust u* formulation than under
the regulatory default formulation. For receptors below 85 m, in the analysis, there is little to no
change in concentration indicating that stable conditions with low wind speeds are not
controlling at elevations below the release height.
In addition to the analysis of the effects from terrain height on controlling meteorological
conditions as described above, the DES submittal included a comparison of the results from the
two modeling techniques at the locations of nearby monitoring stations. The two monitoring
locations are the Pierce Island monitor, about 4 km from Schiller, and a temporary monitor at
Sawgrass Lane in Eliot, Maine, about 2 km from Schiller. The submitted analysis compares the
results of the model with regulatory default options versus with the beta adjust u* option at these
monitoring sites using Q-Q plots. The comparison presented in the submittal indicate nearly
identical predictions at monitor locations at values above 10 |ig/m3 for both monitoring sites. A
Beta adjust u* option for Schiller Page 5 of 7 Biton - Technical Report
D-7
-------
direct model to monitor comparison would provide an opportunity for direct model performance
evaluation against observations; however, the submitted analysis suggests that such a monitor-to-
model comparison would show nearly identical performance for each model.
The submittal included references to additional documents (i.e., Connors and Paine 2014,
Warren 2016), but this technical report did not rely on the analyses discussed in those documents
because they are not peer reviewed. The analyses discussed in this report comprise a sufficient
basis for determining the appropriateness of the beta adjust u* option for this modeling scenario
without these additional citations.
Conclusions and recommendations
Based on the strength of the Lovett analysis available from EPA (2015b) and the Mercer County
analysis described in Paine et al. (2015), and in light of the performance of the beta adjust u*
option as documented in other studies described by EPA (2015b, 2015d), and the additional case-
specific evidence presented in the DES submittal, the statistical evaluation is sufficient to
demonstrate that AERMET version 15181 with the beta adjust u* option is superior to the
regulatory default AERMET version 15181 for application in the Schiller Station modeling
analysis.
The condition of Section 3.2.2(d) of Appendix W in 40 CFR 51 has been adequately addressed
for justifying the use of the beta adjust u* option (with or without the Bulk Richardson option) in
AERMET version 15181 for the Schiller modeling for 1-hour SO2 under the Data Requirements
Rule and for the modeling demonstration in Response to Order on Title V Petition VI-2014-04.
References
EPA. 2010. Memorandum: Applicability of Appendix W Modeling Guidance for the 1-hour SO2
National Ambient Air Quality Standard. From Tyler Fox, EPA Air Quality Modeling
Group to EPA Regional Air Division Directors. August 23, 2010.
EPA. 2015a. AERMOD Model Formulation Document Addendum. Published online June 30,
2015.
EPA. 2015b. Addendum: User's Guide for the AMS/EPA Regulatory Model - AERMOD.
September 2004, updated June 2015. EPA-454/B-03-001. Appendix F. Evaluation of
Low Wind Beta Options.
EPA. 2015c. Memorandum: Clarification on the Approval Process for Regulatory Application of
the AERMOD Modeling System Beta Options. From Richard A. Wayland, EPA Air
Quality Assessment Division Director. December 10, 2015.
EPA. 2015d. Memorandum: Surface Friction Velocity (u*) Non-Default/Beta Option in
AERMET Version 15181; Alternative Refined Model Demonstration. From Herman
Wong, EPA Region 10 Modeling Contact/Regional Atmospheric Scientist to Alan
Schuler, ADEC Engineer. October 20, 2015.
Beta adjust u* option for Schiller Page 6 of 7 Biton - Technical Report
D-8
-------
EPA. 2016. Memorandum: Model Clearinghouse Review of the Use of the ADJU* Beta Option
in the AERMET Meteorological Processor (version 15181) for the Donlin Mine
Compliance Demonstration. From George Bridgers, EPA Model Clearinghouse Director
to Janis Hastings, ADEC Office of Air, Waste, and Toxics Acting Director. February 10,
2016.
Luhar AK and Rayner KN. 2009. "Methods to Estimate Surface Fluxes of Momentum and Heat
from Routine Weather Observations for Dispersion Applications under Stable
Stratification Boundary-Layer Meteorology. 132:437-454. DOI 10.1007/sl0546-009-
9409-z.
Qian W and Venkatram A. 2011. "Performance of Steady-State Dispersion Models Under Low
Wind-Speed Conditions " Boundary-Layer Meteorology. 138:475-491. DOI
10.1007/s 10546-010-9565-1.
Beta adjust u* option for Schiller Page 7 of 7 Biton - Technical Report
D-9
-------
i £* %
1222;
N PfiOi^
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK, NC 27711
MEMORANDUM
APR 2 9 2016
OFFICE OF
AIR QUALITY PLANNING
AND STANDARDS
SUBJECT: Model Clearinghouse Review of the Use of the ADJ U* Beta Option in the
AERMET Meteorological Processor (Version 15181) for the Schiller Station
Modeling Demonstration
INTRODUCTION
In response to your April 7, 2016 concurrence request memorandum, the Model Clearinghouse
has reviewed Region l's position on the proposed use of the ADJU* Beta option in the
AERMET meteorological processor (version 15181) for the Schiller Station energy generating
facility (Schiller Station) located in Portsmouth, New Hampshire. As noted in our February 10,
2016 response memorandum to Region 10', the ADJ U* Beta option was incorporated in
AERMET to address concerns regarding potential underprediction of the surface friction velocity
(u*) during low-wind/stable conditions that could contribute to overprediction of ambient air
impacts by the AERMOD dispersion model (version 15181) for some applications. In the case of
the Shiller Station energy generating facility, excessive 1-hour SO2 concentrations on distant
terrain, 16km from the source, were projected by the regulatory default version of the AERMOD
Modeling System specifically during low-wind/stable conditions when u* values were relatively
small. Given this model response, it was appropriate for the ADJ U* Beta option in AERMET to
be considered for this regulatory modeling application.
MODEL CLEARINGHOUSE RESPONSE
Application of ADJ U* Beta Option in AERMET
Appendix W, Section 3.2.2 provides three different conditions for which an alternative model is
approvable. These three conditions are briefly summarized as:
1) The alternative and preferred model provide equivalent estimates;
2) The alternative model outperforms the preferred model when comparing the results to
actual air quality data; or
1 http: cfpub.epa.tiov/oarweb MCHISRS/index.ct'm?tuseaction inain.resultdetails&recnum I6-X-01
FROM:
George Bridgers, Model Clearinghouse Director
Air Quality Modeling Group, C439-01
David Conroy, Chief
Air Programs Branch, Region 1
TO:
Internet Addresj^U^.) • http://www.epa.gov
Recycled/Recyclable .Printed with Vegetable Oil Based Inks on Recycled Paper (Minimum 25% Postconsumer)
-------
3) The preferred model is less appropriate or there is no preferred model for the given
scenario.
In reviewing the April 7. 2016 concurrence request memorandum from Region 1 and the
attached material from the New Hampshire Department of Environmental Services (NHDES), it
is noted that Region 1 and NHDES were following the second condition2 for the basis of this
alternative model approval. The Model Clearinghouse concurs that a well-reasoned justification
was thoroughly documented and demonstrates that the ADJU* Beta option in AERMET
selected for the Schiller Station modeling demonstration performs better than the default
regulatory version of AERMET for the given application, i.e., a tall stack located near complex
terrain, where high modeled concentrations are likely to occur under low wind, stable conditions
In this case, an isolated terrain feature, Mt. Agamenticas, is located about 15km north-northeast
from the Schiller Station with a peak elevation about 200m above the stack base, with relatively
flat terrain between the source and the mountain.
We appreciate the efforts of Region 1 in the Model Clearinghouse concurrence request
memorandum to highlight the additional evaluation databases, namely the Lovett3 and Mercer
County, ND4, that more directly represent the Schiller Station and surrounding terrain
circumstances. In both cases, the Lovett and Mercer County, ND evaluations demonstrate a
significant improvement of the modeled concentrations with the use of the ADJ U* Beta option
for a facility with tall stacks located near complex terrain, particularly during low wind, stable
conditions. Combined with the Qian and Venkatram5 and the Luhar and Rayner6 journal article
references in the NHDES alternative model submittal that provide a scientific basis for the
adjustment to u*, there is a reasonable justification for the application of the ADJ U* Beta
option in the Schiller Station modeling demonstration.
The NHDES alternative model submittal package included an additional source specific model
sensitivity and monitor evaluation that is worth noting in our concurrence memorandum. A
model sensitivity analysis was performed to further demonstrate the appropriateness and
applicability of the ADJ U* Beta option in the Schiller Station case. The sensitivity analysis
indicated that the most critical impacts at receptors on the distant terrain were only occurring at
hours with the u* values were substantially low, which is indicative of low wind, stable
conditions. These receptors were all at or above the emissions release height at the Shiller
Station. The application of the ADJ U* Beta option resulted in comparable increases in the u*
2 Appendix W to 40 CFR, Part 51, Section 3.2.2.b(2).
3 EPA's Addendum: User's Guide for the AMS/EPA Regulatory Model - AERMOD. September 2004, updated
June 2015. EPA-454/B-03-001. Appendix F. Evaluation of Low Wind Beta Options.
4 Paine. R., O. Samani. M. Kaplan, E. Knipping and N. Kumar. 2015. Evaluation of low wind modeling approaches
for two tall-stack databases", Journal of the Air & Waste Management Association, 65:11, 1341-1353, DOI:
10.1080/10962247.2015.1085924.
5 Qian, W. and A. Venkatram. 2010. "Performance of Steady-State Dispersion Models Under Low Wind-Speed
Conditions." Boundary-Layer Meteorology (2011) 138:475—491 DOI 10.1007/sl0546-010-9565-l. Published online
December 3, 2010. Accessed August 24, 2015.
6 Luhar AK and Rayner KN. 2009. "Methods to Estimate Surface Fluxes of Momentum and Heat from Routine
Weather Observations for Dispersion Applications under Stable Stratification." Boundary-Layer Meteorology.
132:437-454. DOI 10.1007/sl0546-009-9409-z.
2
D-ll
-------
values and reductions to the concentrations at these receptors as demonstrated in the
representative Lovett and Mercy County, ND evaluations. For the receptors below the emissions
release height, there was little to no change in concentrations with the application of the ADJU*
Beta option. Additionally for the nearby controlling receptors not associated with the distant
terrain feature, the critical impacts were occurring at times of much higher u* values, and these
u* values were relatively unchanged with the application of the ADJ U* Beta option. Therefore,
we support that the model sensitivity analysis is providing further evidence of the relevance and
appropriateness of the ADJ U* Beta option for the Schiller Station modeling demonstration.
Lastly, there was indication in our aforementioned February 10, 2016 response memorandum to
Region 10 that EPA has concerns that the use of the ADJ U* Beta option in combination with
site-specific meteorological data that includes the sigma-theta and/or sigma-w turbulence
parameters may introduce a bias toward concentration underprediction. We continue to evaluate
the potential for this concentration underprediction bias and caution anyone considering the use
of both the ADJ U* Beta option and meteorological data that includes the derived sigma-theta
and/or sigma-w turbulence parameters in regulatory applications without consultation and
approval from the appropriate permitting authority and the respective EPA Regional Office.
However, it is noted that the meteorological data used in the Schiller Station modeling
application were not site-specific and did not include any derived sigma-theta or sigma-w
turbulence information. So, the underprediction bias concern is not a factor in this case.
cc: Leiran Biton, EPA Region 1
Richard Wayland, C304-02
Anna Wood, C504-01
Tyler Fox, C439-01
Raj Rao, C504-01
EPA Air Program Managers
EPA Regional Modeling Contacts
3
D-12
-------
Appendix E: Model Clearinghouse Formal Actions from 1981 to 2016
EPA's Model Clearinghouse - Frequency of Formal Actions
30
.Establishment of Guideline
¦ Establishment of Model Clearinghouse
11st Revision to Guideline and ISC Promulgated
i , Supplement A
Supplement B, CTDMPLUS and EDMS Promulgated,
and Guideline Published in CFR
iSupplement C and ISC2 Promulgated
25
CFR Publication Fix and ISC3 Promulgated
20
CALPUFF Promulgated
AERMOD Promulgated
Current Proposed Revisions to
Guideline
10
E-l
-------
E-2
-------
-------
United States Office of Air Quality Planning and Standards Publication No. EPA-454/B-16-008
Environmental Protection Air Quality Assessment Division December 2016
Agency Research Triangle Park, NC
------- |