United States
Environmental Protection
Agency
Office of Air Quality
Planning And Standards
Research Triangle Park, NC 27711
DRAFT
October 1990
AIR
&EPA
New Source Review
Workshop Manual
Prevention of Significant Deterioration
and
Nonattainment Area
Permitting
Additional
Impacts
EPA
RTF W
EPA LIBRARY SERVICES RTF NC
TECHNICAL DOCUMENT COLLECTION
-------
This manual is the cumulative result of hundreds of hours of preparation
and review by numerous people within the U.S. Environmental Protection Agency.
Although it was a group effort by the entire New Source Review Section (NSRS),
the contributions by the following authors of the principal chapters is
grateful 1y acknowledged:
Applicability - Dennis Grumpier and David Solomon
Best Available Control Technology (BACT) - David Solomon
Air Quality Analysis - Dan deRoeck
Additional Impacts Analysis and Class I Area
Impact Analysis - Eric Noble
In addition, Sam Duletsky, before transferring from the NSRS to another
agency, both authored portions and coordinated the development of the manual.
The administrative support for the manual was handled by JoAnn Allman,
secretary for the NSRS.
Finally the unsung, tedious task of reviewing drafts of and suggesting
improvements to the manual was conducted with particular concern and
dedication by the new source review staff in EPA's regional and headquarters
offices. A special note of thanks is extended to those individuals.
Gary McCutchen
Chief, NSRS
-------
D R A f I
OCTOBER 1990
TABLE OF CONTENTS
Page
PART I - PREVENTION OF SIGNIFICANT DETERIORATION (PSD) REVIEW
PREFACE 1
MANUAL ORGANIZATION 3
INTRODUCTION AND OVERVIEW 4
Chapter A - Applicability
I. Introduction A.I
II. New Source PSD Applicability Determination A.3
A. Definition of Source A.3
B. Potential to Emit A.5
1. Basic Requirements A.5
2. Enforceabil ity of Limits A.5
3. Fugitive Emissions. -. A.9
4. Secondary Emissions A. 16
5. Regulated Pollutants A. 18
6. Methods for Determining Potential to Emit ... A.19
C. Emissions Thresholds for PSD Applicability A.22
1. Major Sources A.22
2. Significant Emissions A.24
D. Local Air Quality Considerations A.25
E. Summary of Major New Source Applicability A.26
F. New Source Applicability Example A.28
III. Major Modification Applicability A.33
A. Activities That Are Not Modifications A.34
-------
DRAFT
OCTOBER 1990
TABLE OF CONTENTS - Continued
Page
B. Emissions Netting A.35
1. Accumulation of Emissions A-^
2. Contemporaneous Emissions Changes A.37
3. Creditable Contemporaneous Emissions Changes. . A.38
4. Creditable Amount A-J°
5. Suggested Emissions Netting Procedure A.44
6. Netting Example A-51
IV. General Exemptions A.56
A. Sources and Modifications After August 7, 1980. . . . A.56
B. Sources Constructed Prior to August 7, 1980 A.56
Chapter B - Best Available Control Technology
I. Introduction B.I
II. BACT Applicability B.4
III. A Step by Step Summary of the Top-Down Process B.5
A. STEP 1--Identify All Control Technologies B.5
B. STEP 2--Eliminate Technically Infeasible Options. . . B.7
C. STEP 3--Rank Remaining Control Technologies by Control
Effectiveness B.7
D. STEP 4--Evaluate Most Effective Controls and Document
Results B.8
E. STEP 5--Select BACT B.9
IV. Top-Down Analysis: Detailed Procedures B.10
A. Identify Alternatives Emission Control Techniques . . B.10
1. Demonstrated and Transferable Technologies. . . B.ll
2. Innovated Technologies B.12
3. Consideration of Inherently Lower Polluting
Processes B.13
4. Example B.14
-------
DRAFT
OCTOBER 1990
TABLE OF CONTENTS - Continued
Page
B. Technical Feasibility Analysis B.17
C. Ranking the Technically Feasible Alternatives to
Establish a Control Hierarchy B.22
1. Choice of Units of Emissions Performance to
Compare Levels Amongst Control Options B.22
2. Control Techniques With a Wide Range of
Emissions Performance Levels B.23
3. Establishment of the Control Options Hierarchy. B.25
D. The BACT Selection Process B.26
1. Energy Impacts Analysis B.29
2. Cost/Economic Impacts Analysis B.31
a. Estimating Control Costs B.32
b. Cost Effectiveness • B.36
c. Determining an Adverse Economic Impact. . B.44
3. Environmental Impacts Analysis B.46
a. Examples (Environmental Impacts) B.48
b. Consideration of Emissions of Toxic
and Hazardous Pollutants B.50
E. Selecting BACT B.53
F. Other considerations B.54
V. Enforceability of BACT B.56
VI. Example BACT Analyses for Gas Turbines B.57
A. Example I—Simple Cycle Gas Turbines Firing Natural
Gas B.58
1. Project Summary B.58
2. BACT Analysis Summary B.58
a. Control Technology Options B.58
b. Technical Feasibility Considerations. . . B.61
c. Control Technology Hierarchy B.62
d. Impacts Analysis Summary B.65
e. Toxics Assessment B.65
f. Rationale for Proposed BACT B.68
B. Example 2--Combined Cycle Gas Turbines Firing
Natural Gas B.69
-------
DRAFT
OCTOBER 1990
TABLE OF CONTENTS - Continued
Page
C. Example 3--Combined Cycle Gas Turbine Firing Distillate
Oil B-73
D. Other Considerations.
B.74
Chapter C - The Air Quality Analysis
I. Introduction " c-l
II. National Ambient Air Quality Standards and PSD Increments . C.3
A. Class I, II and III Areas and Increments C.3
B. Establishing the Baseline Date C.6
C. Establishing the Baseline Area C.9
D. Redefining Baseline Areas (Area Redesignation). ... C.9
E. Increment Consumption and Expansion C.10
F. Baseline Date and Baseline Area Concepts -- Examples. C.12
III. Ambient Data Requirements C.16
A. Pre-Appl ication Air Quality Monitoring C.16
B. Post-Construction Air Quality Monitoring C.21
C. Meteorological Monitoring C.22
IV. Dispersion Modeling Analysis C.24
A. Overview of the Dispersion Modeling Analysis C.24
B. Determining the Impact Area C.26
IV
-------
DRAFT
OCTOBER 1990
TABLE OF CONTENTS - Continued
C. Developing the Emissions Inventories C.31
1. The NAAQS Inventory , £•;?£
2. The Increment Inventory ^-^
3. Noncriteria Pollutants Inventory. C.37
D. Model Selection £.37
1. Meteorological Data [;•«
2. Receptor Network ]-'**
3. Good Engineering Practice (GEP) Stack Height. . C.42
4. Source Data c-44
E. The Compliance Demonstration c-51
V. Air Quality Analysis—Example c-54
A. Determining the Impact Air c-54
B. Developing the Emissions Inventories C.58
1. The NAAQS Inventory C.59
2. The Increment Inventory c-62
C. The Full Impact Analysis C.66
1. NAAQS Analysis L.b/
2. PSD Increment Analysis c-59
VI. Bibliography c-71
Chapter D - Additional Impacts Analysis
I. Introduction D>1
II. Elements of the Additional Impacts Analysis D.3
A. Growth Analysis D-3
B. Ambient Air Quality Analysis D.3
C. Soils and Vegetation Analysis D.4
-------
DRAFT
OCTOBER 199°
TABLE OF CONTENTS - Continued
Page
D. Visibility Impairment Analysis D.5
1. Screening Procedures: Level 1 D.6
2. Screening Procedures: Level 2 D.6
3. Screening Procedures: Level 3 D.7
E. Conclusions D.7
III. Additional Impacts Analysis Example D.8
A. Example Background Information D.8
B. Growth Analysis D.9
1. Work Force D.9
2. Housing D.9
3. Industry D.10
C. Ambient Air Quality Analysis D.ll
D. Soils and Vegetation D.ll
E. Visibility Analysis D.13
F. Example Conclusions D.13
IV. Bibliography D.15
Chapter E - Class I Area Impact Analysis
I. Introduction E.I
II. Class I Areas and Their Protection E.2
A. Class I Increments E.8
B. Air Quality Related Values (AQRV's) E.10
C. Federal Land Manager E.12
VI
-------
DRAFT
OCTOBER 1990
TABLE OF CONTENTS - Continued
Page
III. Mandatory Federal Class I Area Impact Analysis and Review. . E.16
A. Source Applicability E.16
B. Pre-Appl ication Stage E.17
C. Preparation of Permit Application E.18
D. Permit Application Review E.19
IV. Visibility Impact Analysis and Review E.22
A. Visibility Analysis E.22
B. Procedural Requirements E.23
V. Bibliography E.24
PART II - NONATTAINMENT AREAS
Chapter F - Nonattainment Area Applicability
I. Introduction F.I
II. Definition of Source F.2
A. "Plantwide" Stationary Source Definition F.2
B. "Dual Source" Definition of Stationary Source .... F.3
III. Pollutants Eligible for Review and Applicability
Thresholds F.7
A. Pollutants Eligible for Review (Geographic
Considerations) F.7
B. Major Source Threshold F.7
C. Major Modification Thresholds F.8
-------
DRAFT
OCTOBER 1990
TABLE OF CONTENTS - Continued
Page
IV. Nonattainment Applicability Example F.9
Chapter G - Nonattainment Area Requirement
I. Introduction G.I
II. Lowest Achievable EMission Rate (LAER) G.2
III. Emissions Reductions "Offsets" G.5
A. Criteria for Evaluating Emissions Offsets G.6
B. Available Sources of Offsets G.7
C. Calculation of Offset Baseline G.7
D. Enforceabil ity of Proposed Offsets G.8
IV. Other Requirements G.9
PART III - EFFECTIVE PERMIT WRITING
Chapter H - Elements of an Effective Permit
I. Introduction H.I
II. Typical Permit Elements H.3
A. Legal Authority H.3
B. Technical Specifications H.5
C. Emissions Compliance Demonstrations H.6
D. Definition of Excess Emissions H.7
E. Administrative Procedures H.8
F. Other Conditions H.9
vn i
-------
DRAFT
OCTOBER 1990
TABLE OF CONTENTS - Continued
Page
III. Summary H.9
Chapter I - Permit Drafting
I. Recommended Permit Drafting Steps I.I
II. Permit Worksheets and File Documentation 1.5
III. Summary 1.5
-------
DRAFT
OCTOBER 1990
TABLE OF CONTENTS - Continued
Page
TABLES
A.I. PSD Source Categories With 100 tpy Major Source
Thresholds A.11
A-2. NSPS and National Emissions Standards for Hazardous Air
Pollutants Proposed Prior to August 7, 1980 A. 12
A-3. Suggested References for Estimating Fugitive Emissions. . . A. 17
A-4. Significant Emission Rates of Pollutants Regulated Under
the Clean Air Act A.20
A-5. Procedures for Determining the Net Emissions Change at a
Source A. 45
B-l Key Steps in the "Top-Down" BACT Process B.6
B-2 Sample BACT Control Hierarchy B.27
B-3 Sample Summary of Top-Down BACT Impact Analysis Results . . B.28
B-4 Example Control System Design Parameters B.34
B-5 Example 1 -- Combustion Turbine Design Parameters B.59
B-6 Example 1 -- Summary of Potential NOX Control Technology
Options B.60
B-7 Example 1 -- Control Technology Hierarchy B.63
B-8 Example 1 -- Summary of Top-Down BACT Impact Analysis
Results for NOX B.66
B-9 Example 2 -- Combustion Turbine Design Parameters B.70
B-10 Example 2 -- Summary of Top-Down BACT Impact Analysis
Results B.71
-------
DRAFT
OCTOBFR 1990
TABLE OF CONTENTS - Continued
Page
TABLES - Continued
B-ll Example of a Capital Cost Estimate for an Electrostatic
Precipitator b.5
B-12 Example of a Annual Cost Estimate for an Electrostatic
Precipitator Applied to a Coal-Fired Boiler b-9
C-l. National Ambient Air Quality Standards C.4
C-2. PSD Increments C.7
C-3. Significant Monitoring Concentrations C.17
C-4. Significance Levels for Air Quality Impacts in Class II
Areas C.28
C-5. Point Source Model Input Data (Emissions) for NAAQS
Compliance Demonstrations C.46
C-6. Existing Baseline Dates for S02, TSP, and N02 for Example
PSD Increment Analysis C.64
E.I. Mandatory Class I Areas E.3
E.2. Class I Increments E.9
E-3. Examples of Air Quality-Related Values and Potential Air
Pollution Caused Changes E.ll
E-4. Federal Land Manager E.14
E-5. USDA Forest Service Regional Offices and States They Serve. E.15
H-l. Suggested Minimum Contents of Air Emission Permits H.4
H-2. Guidelines for Writing Effective Specific Conditions in
NSR Permits H.10
1-1. Five Steps to Permit Drafting 1.2
XI
-------
DRAFT
OCTOBER 1990
TABLE OF CONTENTS - Continued
FIGURES
A-l. Creditable Reductions in Actual Emissions A.43
A-2. Establishing "Old" and "New" Representative Actual S02
Emissions : A.50
B-l Least-Cost Envelope B.42
B-2 Least-Cost Envelope for Example 1 B.67
B-3 Least-Cost Envelope for Example 2 B.72
B-4 Elements of Total Capital Cost b.2
B-5 Elements of Total Annual Cost b.6
C-l. Establishing the Baseline Area C.13
C-2. Redefining the Baseline Area C.15
C-3. Basic Steps in the Air Quality Analysis (NAAQS and PSD
Increments) C.27
C-4. Determining the Impact Area C.29
C-5. Defining the Emissions Inventory Screening Area C.33
C-6. Examples of Polar and Cartesian Grid Networks C.41
C-7. Counties Within 100 Kilometers of Proposed Source C.57
C-8. Point Sources Within 100 Kilometers of Proposed Source. . . C.60
Xll
-------
DRAFT
OCTOBER 1990
TABLE OF CONTENTS - Continued
Page
APPENDICES
A. Definition of Selected Terms a.l
B. Estimating Control Costs b.l
I. Capital Costs b.l
II. Total Annual Cost b.4
III. Other Cost Items b.ll
C. Potential to Emit c.l
xm
-------
DRAFT
OCTOBER 1990
PREFACE
This document was developed for use in conjunction with new source
review workshops and training, and to guide permitting officials in the
implementation of the new source review (NSR) program. It is not intended to
be an official statement of policy and standards and does not establish
binding regulatory requirements; such requirements are contained in the
statute, regulations and approved state implementation plans. Rather, the
manual is designed to (1) describe in general terms, and illustrate by
examples the requirements of the new source review regulations and existing
policies interpreting those regulations; and (2) provide suggested methods of
meeting the regulatory requirements as they have been interpreted by EPA.
Should there be any inconsistency between this manual and the regulations
(including any interpretational policy statements made pursuant to those
regulations), the regulations, interpretations, and policies shall govern.
This document also may be used to assist those who are unfamiliar with the NSR
program and its implementation to gain a working understanding of the program.
The principal focus of this manual is the prevention of significant
deterioration (PSD) portion of the NSR program found in the Code of Federal
Regulations at 40 CFR 52.21. Although state PSD programs are largely
identifial or very similar to the Federal PSD program, the specific
requirements applicable in those areas where the PSD program is conducted
under a State implementation plan (SIP) which has been developed and approved
in accordance with 40 CFR 51.166 may differ in some respects from the
requirements of 40 CFR 52.21. Accordingly, this manual may not describe the
specific State requirements in those respects. The reader is cautioned to
keep this in mind when using this manual for general program guidance. In
most cases where portions of an approved SIP are different from the Federal
PSD program described in this manual, the State program is more restrictive.
Consequently, it is suggested that the reader also obtain program information
from a State or local agency to determine all requirements that may apply in a
given area.
-------
DRAFT
OCTOBER 1990
The examples provided in this manual are for illustration purposes only.
They are designed to impart a basic understanding of the NSR regulations and
requirements.
A number of terms and acronyms used in this manual have specific
meanings within the context of the NSR program. Since this manual is intended
for use by those persons generally familiar with NSR these terms are used
throughout this document, often without definition. To aid users of the
document who are unfamiliar with these terms, general definitions can be found
in Appendix A. The specific regulatory definitions for most of the terms can
be found in 40 CFR 52.21. Should there be any inconsistency between the
definitions contained in Appendix A and the regulatory definitions or
requirements found in Part 40 of the Code of Federal Regulations (including
any interpretations and policy statements made pursuant to those regulations),
the regulations, interpretations, and policies shall govern.
-------
DRAFT
OCTOBER 1990
MANUAL ORGANIZATION
The manual is organized into three parts. Part I contains five chapters
(Chapters A - E) covering the PSD program requirements. Chapter A describes
the PSD applicability criteria and process used to determine if a proposed new
or modified stationary source is required to obtain a PSD permit. Chapter B
discusses the process by which best available control technology (BACT) is
determined for new or modified emissions units. Chapter C discusses the PSD
air quality analysis used to demonstrate that the proposed construction will
not cause or contribute to a violation of any applicable National Ambient Air
Quality Standard or PSD increment. Chapter D discusses the PSD additional
impacts analyses which assess the impact of air, ground, and water pollution
on soils, vegetation, and visibility caused by an increase in emissions at the
subject source. Chapter E identifies class I areas and describes the
procedures involved in preparing and reviewing a permit application for a
proposed source with potential class I area air quality impacts.
Part II of the manual (Chapters F and G) covers the nonattainment area
(NAA) permit program requirements for new major sources and major
modifications. Chapter F describes the NAA applicability criteria for major
new sources or modifications locating in a nonattainment area. Chapter G
provides a basic overview of the NAA preconstruction review requirements.
Part III (Chapters H and I) covers the major source permit itself.
Chapter H discusses the elements of an effective and enforceable permit.
Chapter I discusses permit drafting.
-------
DRAFT
OCTOBER 1990
INTRODUCTION AND OVERVIEW
New major stationary sources of air pollution and major modifications to
major stationary sources are required by the Clean Air Act to a obtain an air
pollution permit before commencing construction. The process is called new
source review (NSR) and is required whether the major source or modification
is planned for an area where the national ambient air quality standards
(NAAQS) are exceeded (nonattainment areas) or an area where air quality is
acceptable (attainment and unclassifiable areas). Permits for sources in
attainment areas are referred to as prevention of significant air quality
deterioration (PSD) permits; while permits for sources located in
nonattainment areas are referred to as nonattainment area (NAA) permits. The
entire program, including both PSD and NAA permit reviews, is referred to as
the NSR program.
The PSD and NAA requirements are pollutant-specific. For example,
although a facility may emit many air pollutants, only one or a few may be
subject to the PSD or NAA permit requirements, depending on the magnitude of
the emissions of each pollutant. Also, a source may have to obtain both PSD
and NAA permits if the source is in an area which is designated nonattainment
for one or more of the pollutants.
On August 7, 1977, Congress substantially amended the Clean Air Act.
These amendments added detailed PSD and NAA permitting programs. On June 19,
1978, EPA revised the PSD regulations to comply with the 1977 Amendments. The
June 1978 regulations were challenged in court and, as a result of the
judicial review, on August 7, 1980, EPA extensively revised both the PSD and
NAA regulations. Five sets of regulations resulted from those revisions.
These regulations and subsequent modifications represent the current NSR
regulatory requirements.
The first set of regulations, 40 CFR 51.166, specifies the minimum
requirements that a PSD air quality permit program under Part C of the Act
must contain in order to obtain approval by EPA as a revision to a State
implementation plan (SIP). The second set, 40 CFR 52,21, delineates the
-------
DRAFT
OCTOBER 1990
federal PSD permit program, which currently applies as part of the SIP for
States that have not submitted a PSD program meeting the requirements of
40 CFR 51.166. Roughly two thirds of the States are implementing their own
PSD program which has been approved by EPA under 40 CFR 51.166. The 40 CFR
52.21 applies in the remaining States, most of which have been delegated the
authority to implement the federal PSD program.
The basic goals of the PSD regulations are: (1) to ensure that economic
growth will occur in harmony with the preservation of existing clean air
resources; (2) to protect the public health and welfare from any adverse
effect which might occur even at air pollution levels better than the national
ambient air quality standards (NAAQS); and (3) to preserve, protect, and
enhance the air quality in areas of special natural recreational, scenic, or
historic value, such as national parks and wilderness areas. The primary
provisions of the PSD regulations require that major new stationary sources
and major modifications be carefully reviewed prior to construction to ensure
compliance with the NAAQS, the applicable PSD air quality increments, and the
requirement to apply BACT to minimize the project's emissions of air
pollutants.
The remaining regulations apply to the NAA program. The third set of
regulations, 40 CFR 51.165(a) and (b), specifies the elements of an approvable
State permit program for preconstruction review for nonattainment purposes
under Part D of the Act. A major new source or major modification that would
be located in an area designated as nonattainment and subject to a NAA permit
must meet stringent conditions designed to ensure that the new source's
emissions will be controlled to the greatest degree possible; that more than
equivalent offsetting emissions reductions ("emission offsets") will be
obtained from existing sources; and that there will be progress toward
achievement of the NAAQS.
The fourth and fifth sets, 40 CFR Part 51, Appendix S (Offset Ruling)
and 40 CFR 52.24 (construction moratorium) respectively, apply in certain
circumstances where a nonattainment area SIP has not been fully approved by
EPA as meeting the requirements of Part D of the Act.
-------
DRAFT
OCTOBER 1990
Briefly, a "major stationary source" is any source type belonging to a
list of 28 source categories which emits or has the potential to emit 100 tons
per year or more of any pollutant subject to regulation under the Act, or any
other source type which emits or has the potential to emit such pollutants in
amounts equal to or greater than 250 tons per year, A stationary source
generally includes all pollutant-emitting activities which belong to the same
industrial grouping, are located on contiguous or adjacent properties, and are
under common control.
A "major modification" is generally a physical change or a change in the
method of operation of a major stationary source which would result in a
contemporaneous significant net emissions increase in the emissions of any
regulated pollutant. In determining if a proposed increase would cause a
significant net increase to occur, several detailed calculations must be
performed.
If a proposed source or modification qualifies as major, it must be
located in a PSD area in order for PSD review to apply. A PSD area is one
formally designated by the state as "attainment" or "unclassifiable" for any
pollutant for which a national ambient air quality standard exists.
No source or modification subject to PSD review may be constructed
without a permit. To obtain a PSD permit an applicant must:
1. apply the best available control technology (BACJ);
A BACT analysis is done on a case-by-case basis, and
considers energy, environmental, and economic impacts in
determining the maximum degree of reduction achievable for the
proposed source or modification. In no event can the
determination of BACT result in an emission limitation which would
not meet any applicable standard of performance under 40 CFR Parts
60 and 61.
-------
DRAFT
OCTOBER 1990
2. conduct an ambient air quality analysis;
Each PSD source or modification must perform an air quality
analysis to demonstrate that its new pollutant emissions would not
violate either the applicable NAAQS or the applicable PSD
increment.
3. analyze impacts to soils, vegetation, and visibility;
An applicant is required to analyze whether its proposed
emissions increases would impair visibility, or adversely affect
soils or vegetation. Not only must the applicant look at the
direct effect of source emissions on these resources, but it also
must consider the indirect impacts from general commercial,
residential, industrial, and other growth associated with the
proposed source or modification.
4. not adversely impact a Class I area; and
If the reviewing authority receives a PSD permit application
for a source that could have an impact on a Class I area, it must
notify the Federal Land Manager and the federal official charged
with direct responsibility for managing these lands. These
officials have an affirmative responsibility to protect the air
quality-related values (including visibility) in Class I areas and
for consulting with the reviewing authority to determine whether
any proposed construction will adversely affect such values. If
the Federal Land Manager determines that emissions from a proposed
source or modification would impair air quality-related values,
even though the emissions levels would not cause a violation of
the allowable air quality increment, the Federal Land Manager may
recommend that the reviewing authority deny the permit.
5. undergo adequate public participation by applicant.
Specific public notice requirements and a public comment
period are required before the PSD review agency takes final
action on a PSD application.
-------
DRAFT
OCTOBER 1990
' The preconstruction review requirements for major new sources or major
modifications locating in areas designated nonattainment pursuant to section
107 of the Act differ from prevention of significant deterioration (PSD)
requirements. First, the emissions control requirement for nonattainment
areas, lowest achievable emission rate (LAER), is defined differently than the
best available control technology (BACT) emissions control requirement.
Second, the source must obtain any required emissions reductions (offsets) of
the nonattainment pollutant from other sources which impact the same area as
the proposed source. Third, the applicant must certify that all other sources
owned by the applicant in the State are complying with all applicable
requirements of the CAA, including all applicable requirements in the State
implementation plan (SIP). Fourth, such sources impacting visibility in
mandatory class I Federal areas must be reviewed by the appropriate Federal
land manager (FLM).
-------
PART I
PREVENTION OF SIGNIFICANT DETERIORATION (PSD) REVIEH
Chapter A - Applicability
Chapter B - Best Available Control Technology
Chapter C - The Air Quality Analysis
Chapter D - Additional Impact Analysis
Chapter E - Class I Area Impact Analysis
-------
DRAFT
OCTOBER 1990
CHAPTER A
PSD APPLICABILITY
I. INTRODUCTION
An applicability determination, as discussed in this section, is the
process of determining whether a preconstruction review should be conducted
by, and a permit issued to, a proposed new source or a modification of an
existing source by the reviewing authority, pursuant to prevention of
significant deterioration (PSD) requirements.
There are three basic criteria in determining PSD applicability. The
first and primary criterion is whether the proposed project is sufficiently
large (in terms of its emissions) to be a "major" stationary source or "major"
modification. Source size is defined in terms of "potential to emit," which
is its capability at maximum design capacity to emit a pollutant, except as
constrained by federally-enforceable conditions (which include the effect of
installed air pollution control equipment and restrictions on the hours of
operation, or the type or amount of material combusted, stored or processed).
A new source is major if it has the potential to emit any pollutant
regulated under the Act in amounts equal to or exceeding specified major
source thresholds [100 or 250 tons per year (tpy)] which are predicated on the
source's industrial category. A major modification is a physical change or
change in the method of operation at an existing major source that causes a
significant "net emissions increase" at that source of any pollutant regulated
under the Act.
The second criterion for PSD applicability is that a new major source
would locate, or the modified source is located, in a PSD area. A PSD area is
one formally designated, pursuant to section 107 of the ACT and 40 CFR 81, by
a State as "attainment" or "unclassifiable" for any criteria pollutant, i.e.,
an air pollutant for which a national ambient air quality standard exists.
A.I
-------
DRAFT
OCTOBER 1990
The third criterion is that the pollutants emitted in, or increased by,
"significant" amounts by the project are subject to PSD. A source's location
can be attainment or unclassified for some pollutants and simultaneously
nonattainment for others. If the project would emit only pollutants for which
the area has been designated nonattainment, PSD would not apply.
The purposes of a PSD applicability determination are therefore:
(1) to determine whether a proposed new source is a "major stationary
source," or if a proposed modification to an existing source is a
"major modification;"
(2) to determine if proposed conditions and restrictions, which will
limit emissions from a new source or an existing source that is
proposing modification to a level that avoids preconstruction review
requirements, are legitimate and federally-enforceable; and
(3) to determine for a major new source or a major modification to an
existing source which pollutants are subject to preconstruction
review.
In order to perform a satisfactory applicability determination, numerous
pieces of information must be compiled and evaluated. Certain information and
analyses are common to applicability determinations for both new sources and
modified sources; however, there are several major differences. Consequently,
two detailed discussions follow in this section: PSD applicability
determinations for major new sources and PSD applicability determinations for
modifications of existing sources. The common elements will be covered in the
discussion of new source applicability. They are the following:
* defining the source;
* determining the source's potential to emit;
* determining which major source threshold the source is subject
to; and
* assessing the impact on applicability of the local air quality,
i.e., the attainment designation, in conjunction with the
pollutants emitted by the source.
A.2
-------
DRAFT
OCTOBER 1990
II. NEW SOURCE PSD APPLICABILITY DETERMINATIONS
II.A. DEFINITION OF SOURCE
For the purposes of PSD a stationary source is any building, structure,
facility, or installation which emits or may emit any air pollutant subject to
regulation under the Clean Air Act (the Act). . "Building, structure, facility,
or installation" means all the pollutant-emitting activities which belong to
the same industrial grouping, are located on one or more contiguous or
adjacent properties and are under common ownership or control. An emissions
unit is any part of a stationary source that emits or has the potential to
emit any pollutant subject to regulation under the Act.
The term "same industrial grouping" refers to the "major groups"
identified by two-digit codes in the Standard Industrial Classification (SIC)
Manual, which is published by the Office of Management and Budget. The 1972
edition of the SIC Manual, as amended in 1977, is cited in the current PSD
regulations as the basis for classifying sources. Sources not found in that
edition or the 1977 supplement may be classified according to the most current
edition.
For example a chemical complex under common ownership manufactures
polyethylene, ethylene dichloride, vinyl chloride, and numerous other
chlorinated organic compounds. Each product is made in separate
processing equipment with each piece of equipment containing several
emission units. All of the operations fall under SIC Major Group 28,
"Chemicals and Allied Products;" therefore, the complex and all its
associated emissions units constitute one source.
In most cases, the property boundary and ownership are easily determined.
A frequent question, however, particularly at large industrial complexes, is
how to deal with multiple emissions units at a single location that do not
fall under the same two-digit SIC code. In this situation the source is
classified according to the primary activity at the site, which is determined
by its principal product (or group of products) produced or distributed, or by
A.3
-------
DRAFT
OCTOBER 1990
the services it renders. Facilities that convey, store, or otherwise assist
in the production of the principal product are called support facilities.
For example, a coal mining operation may include a coal cleaning
plant, which is located at the mine. If the sole purpose of the
cleaning plant is to process the coal produced by the mine, then it
is considered to be a support facility for the mining operation. If,
however, the cleaning plant is collocated with a mine, but accepts
more than half of its feedstock from other mines (indicating that the
activities of the collocated mine are incidental) then coal cleaning
would be the primary activity and the basis for the classification.
Another common situation is the collocation of power plants with
manufacturing operations. An example would be a silicon wafer and
semiconductor manufacturing plant that generates its own steam and
electricity with fossil fuel-fired boilers. The boilers would be
considered part of the source because the power plant supports the
primary activity of the facility.
An emissions unit serving as a support facility for two or more primary
activities (sources) is to be considered part of the primary activity that
relies most heavily on its support.
For example, a steam boiler jointly owned and operated by two sources
would be included with the source that consumes the most steam.
As a corollary to the examples immediately above, suppose a power
plant, is co-owned by the semiconductor plant and a chemical
manufacturing plant. The power plant provides 70 percent of its total
output (in Btu's per hour) as steam and electricity to the
semiconductor plant. It sells only steam to the chemical plant. In
the case of co-generation, the support facility should be assigned
to a primary activity based on pro rata fuel consumption that is
required to produce the energy bought by each of the support
facility's customers, since the emission rates in pounds per Btu are
different for steam and electricity. In this example then, the power
plant would be considered part of the semiconductor plant.
It is important to note that if a new support facility would by itself be
a major source based on its source category classification and potential to
emit, it would be subject to PSD review even though the primary source, of
which it is a part, is not major and therefore exempt from review. The
A.4
-------
DRAFT
OCTOBER 1990
conditions surrounding such a determination is discussed further in the
section on major source thresholds (see Section II.C.).
II.B. POTENTIAL TO EMIT
II.B.I. BASIC REQUIREMENTS
The potential to emit of a stationary source is of primary importance in
establishing whether a new or modified source is major. Potential to emit is
the maximum capacity of a stationary source to emit a pollutant under its
physical and operational design. Any physical or operational limitation on
the capacity of the source to emit a pollutant, provided the limitation or its
effect on emissions is federally-enforceable, shall be treated as part of its
design. Example limitations include:
(1) Requirements to install and operate air pollution control
equipment at prescribed efficiencies;
(2) Restrictions on design capacity utilization [note that these
types of limitations are not explicitly mentioned in the
regulations, but in certain instances do meet the criteria for
limiting potential to emit];
(3) Restrictions on hours of operation; and
(4) Restrictions on the types or amount of material processed,
combusted or stored.
II.B.2. ENFORCEABILITY OF LIMITS
For any limit or condition to be a legitimate restriction on potential to
emit, that limit or condition must be federally-enforceable, which in turn
requires practical enforceability (see Appendix A) [see U.S. v. Louisiana-
Pacific Corporation. 682 F. Supp. 1122, Civil Action No. 86-A-1880
(D. Colorado, March 22, 1988). Practical enforceability means the source
and/or enforcement authority must be able to show continual compliance (or
A.5
-------
DRAFT
OCTOBER 1990
noncompliance) with each limitation or requirement. In other words, adequate
testing, monitoring, and record-keeping procedures must be included either in
an applicable federally issued permit, or in the applicable federally approved
SIP or the permit issued under same.
For example, a permit that limits actual source emissions on an
annual basis only (e.g., the facility is limited solely to 249 tpy)
cannot be considered in determining potential to emit. It contains
none of the basic requirements and is therefore not capable of
ensuring continual compliance, i.e., it is not enforceable as a
practical matter.
The term "federally-enforceable" refers to all limitations and conditions
which are enforceable by the Administrator, including:
requirements developed pursuant to any new source performance
standards (NSPS) or national emission standards for hazardous
air pollutants (NESHAP),
requirements within any applicable federally-approved State
implementation plan, and
any requirements contained in a permit issued pursuant to
federal PSD regulations (40 CFR 52.21), or pursuant to PSD or
operating permit provisions in a SIP which has been federally
approved in accordance with 40 CFR 51 Subpart I.
Federally-enforceable permit conditions that may be used to limit
potential to emit can be expressed in a variety of terms and usually include a
combination of two or more of the following four requirements in conjunction
with appropriate record-keeping requirements for verification of compliance:
(1) Installation and continuous operation and maintenance of air
pollution controls, usually expressed as both a required
abatement efficiency of the maximum uncontrolled emission rate
and a maximum outlet concentration or hourly emission rate
(flow rate x concentration);
A typical example might be a 255 tpy limit on a stone crushing
operation. The enforceable permit conditions could be a maximum
emission rate of 58 Ibs/hr, a maximum concentration of 0.1 grains
per dry standard cubic foot (gr/dSCF) and a maximum flow rate of
A.6
-------
DRAFT
OCTOBER 1990
67,000 dSCFM based on nameplate capacity and 8760 hours per year.
In addition, the permit should also stipulate a minimum 90
percent overall reduction of particulate matter (PM) emissions
on an hourly basis via capture hoods and a baghouse.
(2) Capacity limitations;
The stone crusher decides to limit its potential to emit to 180
tpy by limiting the feed rate to 70 percent of the nameplate
capacity. One of the enforceable limits becomes a stone feed
rate (tons/hr.) based on 70 percent of nameplate capacity with
a federally-enforceable requirement for a method or device for
measuring the feed rate on an hourly basis. Another approach is
to limit the PM emissions rate to 41 Ibs/hr. A third
alternative is to retain a maximum concentration of 0.1 gr./dSCF,
but limit the maximum exhaust rate to 47,000 dSCFM due to the
decrease in feed rate. In all these cases, the 90 percent
overall reduction of particulate matter (PM) emissions on an
hourly basis via capture hoods and baghouse would also be
maintained.
In another example, the potential to emit of a boiler with a
design input capacity of 200 million Btu/hour is limited to a
100-million-Btu/hr fuel input rate by the permit, which requires
that the boiler's heat input not exceed 50 percent of its rated
capacity. The permit would further require that compliance be
demonstrated with a continuously recording fuel meter and
concurrent monitoring and recording of fuel heating value to show
that the fuel input does not exceed 100-million-Btu/hr.
(3) Restrictions on hours of operation, including seasonal operation;
and
In the stone crusher example, the operator may choose to limit
the hours of operation per year to keep the potential to emit
below the major source threshold of 250 tpy. For example, using
the same maximum concentration and flow rate and minimum overall
control efficiency limitations as in (1) above, a restriction on
the number of 8-hour shifts to two, i.e., 16 hours per day would
reduce the potential uncontrolled emissions by 33 percent to
170 tpy.
In another example, a citrus dryer that only operates during the
growing season could have its potential to emit limited by a
permit restriction on the hours of operation, and further, by
prohibiting the dryer from operating between March and November.
A.7
-------
DRAFT
OCTOBER 1990
(4) Limitations on raw materials used (including fuel combusted) and
stored.
An example of this type of limit would be a maximum 1 percent
sulfur content in the coal feed for a power plant. Another would
be a condition that a surface coater only use water-based or
higher solids coatings with a maximum VOC content of 2.0 pounds
VOC per gallon solids deposited on the substrate with requisite
limits on coating usage (gallons/hr or gallons/yr on a 12-month
rolling time period).
In addition to limits in major source construction permits or federally
approved SIP limits for major sources, terms and conditions contained in State
operating permits will be considered federally-enforceable under the following
conditions:
(1) the State's operating permit program is approved by EPA and
incorporated into the applicable SIP under section 110 of the
Act;
(2) the operating permits are legally binding on the source under
the SIP and the SIP specifically provides that permits that are
not legally binding may be deemed not "federally-enforceable;"
(3) all emissions limitations, controls, and other requirements
imposed by such permits are no less stringent than any
counterpart limitations and requirements in the SIP, or in
standards established under sections 111 and 112 of the ACT;
(4) the limitations, controls and requirements in the operating
permits are permanent, quantifiable, and otherwise enforceable
as a practical matter; and
(5) the permits are issued subject to public participation, i.e.,
timely notice, opportunity for public comment, etc.
(See also, 54 FR 27281, June 28, 1989.)
A minor (i.e., a non-major) source construction permit issued to a source
by a State may be used to determine the potential to emit if:
the State program under which the permit was issued has been
approved by EPA as meeting the requirements of
40 C.F.R. Parts 51.160 through 51.164, and
A.8
-------
DRAFT
OCTOBER 1990
the provisions of the permit are federally-enforceable and
enforceable as a practical matter.
Note, however, that a permit condition that temporarily restricts
production to a level at which the source does not intend to operate for any
extensive time is not valid if it appears to be intended to circumvent the
preconstruction review requirements for major source by making the source
temporarily minor. Such permit limits cannot be used in the determination of
potential to emit. Another situation that should receive careful scrutiny is
the construction of a manufacturing facility with a physical capacity far
greater than the limits specified in a permit condition. See also 54 FR 27280,
which specifically discusses "sham" minor source permits.
An example is construction of an electric power generating unit, which
is proposed to be operated as a peaking unit but which by its nature
can only be economical if it is used as a base-load facility.
Remember, if the permit or SIP requirements, conditions or limits on a
source are not federally-enforceable (which includes enforceable as a
practical matter), potential to emit is based on full capacity and year-round
operation. For additional informaiton on federally enforceability and
limiting potential to emit see Appendix A.
II.B.3. FUGITIVE EMISSIONS
As defined in the federal PSD regulations, fugitive emissions are those
"...which could not reasonably pass through a stack, chimney, vent, or other
functionally equivalent opening." To the extent they are quantifiable,
fugitive emissions are included in the potential to emit (and increases in
same due to modification), if they occur at one of the following stationary
sources:
A.9
-------
DRAFT
OCTOBER 1990
Any belonging to one of the 28 named PSD source categories listed in
Table A-l, which were explicitly identified in Section 169 of the
Act as being subject to a 100-tpy emissions threshold for
classification of major sources;
Any belonging to a stationary source category that as of August 7,
1980, is regulated (effective date of proposal) by New Source
Performance Standards (NSPS) pursuant to Section 111 of the Act
(listed in Table A-2); and
Any belonging to a stationary source category that as of August 7,
1980, is regulated (effective date of promulgation) by National
Emissions Standards for Hazardous Air Pollutants (NESHAP) pursuant
to Section 112 of the Act (listed in Table A-2).
Note also that, if a source has been determined to be major, fugitive
emissions, to the extent they are quantifiable, are considered in any
subsequent analyses (e.g., air quality impact).
Fugitive emissions may vary widely from source to source. Examples of
common sources of fugitive emission include:
coal piles - particulate matter (PM);
road dust - PM;
quarries - PM; and
leaking valves and flanges at refineries and organic chemical
processing equipment - volatile organic compounds (VOC).
A.10
-------
DRAFT
OCTOBER 1990
TABLE A-l. PSD SOURCE CATEGORIES WITH
100 tpy MAJOR SOURCE THRESHOLDS
1. Fossil fuel-fired steam electric plants of more than 250 million Btu/hr
heat input
2. Coal cleaning plants (with thermal dryers)
3. Kraft pulp mills
4. Portland cement plants
5. Primary zinc smelters
6. Iron and steel mill plants
7. Primary aluminum ore reduction plants
8. Primary copper smelters
9. Municipal incinerators capable of charging more than 250 tons of refuse
per day
10. Hydrofluoric acid plants
11. Sulfuric acid plants
12. Nitric acid plants
13. Petroleum refineries
14. Lime plants
15. Phosphate rock processing plants
16. Coke oven batteries
17. Sulfur recovery plants
18. Carbon black plants (furnace plants)
19. Primary lead smelters
20. Fuel conversion plants
21. Sintering plants
22. Secondary metal production plants
23. Chemical process plants
24. Fossil fuel boilers (or combinations thereof) totaling more than 250
million Btu/hr heat input
25. Petroleum storage and transfer units with a total storage capacity
exceeding 300,000 barrels
26. Taconite ore processing plants
27. Glass fiber processing plants
28. Charcoal production plants
A.11
-------
DRAFT
OCTOBER 1990
TABLE A-2. NEW SOURCE PERFORMANCE STANDARDS PROPOSED AND
NATIONAL EMISSION STANDARDS FOR HAZARDOUS AIR
POLLUTANTS PROMULGATED PRIOR TO August 7, 1980
Hew Source Performance Standards 40 CFR 60
Source
Subpart
Affected Facility
Proposed
Date
Phosphate rock
plants
NN
Grinding, drying and
calcining facilities
09/21/79
Ammonium sulfate
manufacture
pp
Ammonium sulfate dryer
02/04/80
National Emission Standards for Hazardous Air Pollutants 40 CFR 61
Pollutant
Subpart
Affected Facility
Promulgated
Date
Beryllium
Extraction plants,
ceramic plants,
foundries, incinerators,
propellant plants,
machining operations
04/06/73
Beryllium, rocket
motor firing
D
Rocket motor firing
04/06/73
Mercury
Ore processing,
chloralkali manufacturing,
sludge incinerators
04/06/73
Vinyl chloride
Ethylene dichloride
manufacture via Q£ HC1,
vinyl chloride manufacture,
polyvinyl chloride manufacture
10/21/76
Asbestos
Asbestos mills; roadway04/06/73
surfacing (asbestos tailings);
demolition; spraying, fabri-
cation, waste disposal and
insulting
Manufacture of shotgun
shells, renovation,
fabrication, asphalt concrete,
products containing asbestos
06/19/78
A.12
-------
DRAFT
OCTOBER 1990
TABLE A-2. NEW SOURCE PERFORMANCE STANDARDS PROPOSED AND
NATIONAL EMISSION STANDARDS FOR HAZARDOUS AIR
POLLUTANTS PROMULGATED PRIOR TO August 7, 1980
New Source Performance Standards 40 CFR 60
Source Subpart
Fossil -fuel fired D
steam generators for
which construction
is commenced after
08/17/71 and before
09/19/78
Elect, utility steam Da
generating units for
which construction
is commenced after
09/18/78
Municipal incinerators E
(>50 tons/day)
Portland cement plants F
Nitric acid plants G
Sulfuric acid plants H
Asphalt concrete I
plants
Petroleum refineries J
Storage vessels for K
Affected Facility
Utility and industrial
(coal, oil, gas, wood,
1 ignite)
Utility boilers (solid,
liquid, and gaseous fuels)
Incinerators
Kiln, clinker cooler
Process equipment
Process equipment
Process equipment
Fuel gas combustion devices
Claus sulfur recovery
Gasoline, crude oil, and
Proposed
Date
08/17/71
09/19/78
08/17/71
08/17/71
08/17/71
08/17/71
06/11/73
06/11/73
06/11/73
petroleum liquids
construction after
06/11/73 and prior
to 05/19/78
distillate storage tanks
>40,000 gallons capacity
Storage vessels for
petroleum liquids
construction after
05/18/78
Ka
Gasoline, crude oil, and
distillate storage tanks
>40,000 gallons capacity,
vapor pressure ^1.5
05/18/78
Secondary lead
smelters and
refineries
Blast and reverberatory
furnaces, pot furnaces
06/11/73
A.13
-------
DRAFT
OCTOBER 1990
TABLE A-2. NEW SOURCE PERFORMANCE STANDARDS PROPOSED AND
NATIONAL EMISSION STANDARDS FOR HAZARDOUS AIR
POLLUTANTS PROMULGATED PRIOR TO August 7, 1980
New Source Performance Standards 40 CFR 60
Source
Subpart
Affected Facility
Proposed
Date
Secondary brass
and bronze ingot
production plants
M
Reverberatory and electric 06/11/73
furnaces and blast furnaces
Iron and steel mills N
Basic oxygen process furnaces 06/11/73
(BOPF)
Primary emission sources
Sewage treatment
plants
0
Sludge incinerators
06/11/73
Primary copper
smelters
Roaster, smelting furnace, 10/16/74
converter dryers
Primary zinc
smelters
Roaster sintering machine
10/16/74
Primary lead
smelters
Sintering machine, electric 10/16/74
smelting furnace, converter
Blast or reverberatory furnace,
sintering machine discharge end
Primary aluminum
reduction plants
Primary aluminum
reduction plants
lll(d)
Pot 1ines and anode bake
plants
Pot 1ines and anode bake
plants
10/23/74
04/11/79
Phosphate fertilizer
industry
T
U
V
W
X
Wet process phosphoric
Superphosphoric acid
Diammonium phosphate
Triple superphosphate products
Granular triple superphosphate
products
10/22/74
Coal preparation
plants
Ferroalloy Z
production facilities
Air tables and thermal dryers 10/24/74
Specific furnaces
10/21/74
A.14
-------
DRAFT
OCTOBER 1990
TABLE A-2. NEW SOURCE PERFORMANCE STANDARDS PROPOSED AND
NATIONAL EMISSION STANDARDS FOR HAZARDOUS AIR
POLLUTANTS PROMULGATED PRIOR TO August 7, 1980
New Source Performance Standards 40 CFR 60
Source
Subpart
Affected Facil ity
Proposed
Date
Steel plants: AA
electric arc furnaces
Electric arc furnaces
10/21/74
Kraft pulp mills
BB
Lead acid battery
manufacturing plants
KK
Digesters, lime kiln
recovery furnace, washer,
evaporator, strippers,
smelt and BLO tanks
Recovery furnace, lime,
kiln, smelt tank
09/24/76
Glass manufacturing CC
plants
Grain elevators DO
Stationary gas GG
turbines
Lime manufacturing HH
plants
Degreasers (organic JJ
solvent cleaners)
Glass melting furnace
Truck loading and unloading
stations, barge or ship
loading and unloading stations
railcar loading and unloading
stations, and grain handling
operations
Each gas turbine
Rotary kiln, hydrator
Cold cleaner, vapor
degreaser, conveyorized
degreaser
06/15/79
01/13/77
10/03/77
05/03/77
06/11/80
Lead oxide production grid
casting, paste mixing, three-
process operation and lead
reclamation
01/14/80
Automobile and
light-duty truck
surface coating
operations
MM
Prime, guide coat, and
top coat operations at
assembly plants
10/05/79
A.15
-------
DRAFT
OCTOBER 1990
Due to the variability even among similar sources, fugitive emissions
should be quantified through a source-specific engineering analysis.
Suggested (but by no means all of the useful) references for fugitive
emissions data and associated analytic techniques are listed in Table A-3.
Remember, if emissions can be "reasonably" captured and vented through a
stack they are not considered "fugitive" under EPA regulations. In such
cases, these emissions, to the extent they are quantifiable, would count
toward the potential to emit regardless of source or facility type.
For example, the emissions from a rock crushing operation that could
reasonably be equipped with a capture hood are not considered fugitive
and would be included in the source's potential to emit.
As another example, VOC emissions, even if in relatively small
quantities, coming from leaking valves inside a large furniture
finishing plant, are typically captured and exhausted through the
building ventilation system. They are, therefore, measurable and
should be included in the potential to emit.
As a counter example, however, it may be unreasonable to expect that
relatively small quantities of VOC emissions, caused by leaking valves
at outside storage tanks of the large furniture finishing operation,
could be captured and vented to a stack.
II.BA. SECONDARY EMISSIONS
Secondary emissions are not considered in the potential emissions
accounting procedure. Secondary emissions are those emissions which, although
associated with a source, are not emitted from the source itself. Secondary
emissions occur from any facility that is not a part of the source being
reviewed, but which would not be constructed or increase its emissions except
as a result of the construction or operation of the major stationary source or
major modification. Secondary emissions do not include any emissions from any
off-site facility which would be constructed or increase its emissions for
some reason other than the construction or operation of the major stationary
source or major modification.
A.16
-------
DRAFT
OCTOBER 1990
TABLE A-3. SUGGESTED REFERENCES FOR ESTIMATING FUGITIVE EMISSIONS
1. Emission Factors and Frequency of Leak Occurrence for Fittings in
Refinery Process Units. Radian Corporation. EPA-600/2-79-044. February
1979.
2. Protocols for Generating Unit - Specific Emission Estimates for Equipment
Leaks of VOC and VHAP. U.S. Environmental Protection Agency.
EPA-450/3-88-0100.
3. Improving Air Quality: Guidance for Estimating Fugitive Emissions From
Equipment. Chemical Manufacturers Association. January 1989.
4. Compilation of Air Pollutant Emission Factors, 3rd ed. U.S.
Environmental Protection Agency. AP-42 (including Supplements 1-8).
May 1978.
5. Technical Guidance for Control of Industrial Process Fugitive Particulate
Emissions. Pedco Environmental, Inc. EPA-450/3-77-010. March 1977.
6. Fugitive Emissions From Integrated Iron and Steel Plants. Midwest
Research Institute, Inc. EPA-600/2-78-050. March 1978.
7. Survey of Fugitive Dust from Coal Mines. Pedco Environmental, Inc.
EPA-908/1-78-003. February 1978.
8. Workbook on Estimation of Emissions and Dispersion Modeling for Fugitive
Particulate Sources. Utility Air Regulatory Group. September 1981.
9. Improved Emission factors for Fugitive Dust from Weston Surface Coal
Mining Sources, Volumes I and II. U.S. Environmental Protection Agency.
EPA-600/7-84-048.
10. Control of Open Fugitive Dust Sources. Midwest Research Institute.
EPA-450/3-88-008. September 1988.
A.17
-------
DRAFT
OCTOBER 1990
An example is the emissions from an existing quarry owned by one
company that doubles its production to supply aggregate to a cement
plant proposed for construction as a major source on adjacent
property by another company. The quarry's increase in emissions
would be secondary emissions which the cement plant's ambient
impacts analysis must consider.
Secondary emissions do not include any emissions which come directly from
a mobile source, such as emissions from the tailpipe of a motor vehicle or
from the propulsion unit of a train or a vessel. This exclusion is limited,
however, to only those mobile sources that are regulated under Title II of the
Act (see 43 FR 26403 - note #9). Most off-road vehicles are not regulated
under Title II and are usually treated as area sources. [As a result of a
court decision in NRDC v. EPA. 725 F.2d 761 (D.C. Circuit 1984), emissions
from vessels at berth ("dockside") not to be included in the determination of
secondary emissions but are considered primary emissions for applicability
purposes.]
Although secondary emissions are excluded from the potential emissions
estimates used for applicability determinations, they must be considered in
PSD analyses if PSD review is required. In order to be considered, however,
secondary emissions must be specific, well-defined, quantifiable, and impact
the same general area as the stationary source or modification undergoing
review.
II.B.5. REGULATED POLLUTANTS
The potential to emit must be determined separately for each pollutant
regulated by the Act and emitted by the new or modified source. Twenty-six
compounds, 6 criteria and 20 noncriteria, are regulated as air pollutants by
the Act as of December 31, 1989. They are listed in Table A-4. Note that EPA
has designated PM-10 (particulate matter with an aerodynamic diameter less
than 10 microns) as a criteria pollutant by promulgating NAAQS for this
A.18
-------
DRAFT
OCTOBER 1990
pollutant as a replacement for total PM. Thus, the determination of potential
to emit for PM-10 emissions as well as total PM emissions (which are still
regulated by many NSPS) is required in applicability determinations. Several
halons and chlorofluorocarbon (CFC) compounds have been added to the list of
regulated pollutants as a result of the ratification of the Montreal Protocol
by the United States in January 1989.
II.B.6. METHODS FOR DETERMINING POTENTIAL TO EMIT
In determining a source's potential to emit, two parameters must be
measured, calculated, or estimated in some way. They are:
the worst case uncontrolled emissions rate, which is based on
the dirtiest fuels, and/or the highest emitting materials and
operating conditions that the source is or will be permitted to
use under federally-enforceable requirements, and
the efficiency of the air pollution control system, if any, in
use or contemplated for the worst case conditions, where the
use of such equipment is federally-enforceable.
A.19
-------
DRAFT
OCTOBER 1990
TABLE A-4. SIGNIFICANT EMISSION RATES OF POLLUTANTS
REGULATED UNDER THE CLEAN AIR ACT
Pollutant Emissions rate (tons/year)
Pollutants listed at 40 CFR 52.21{b)(23)
* Carbon monoxide 100
* Nitrogen oxides" 40
* Sulfur dioxide" 40
* Particulate matter (PM/PM-10) 25/15
* Ozone (VOC) 40 (of VOC's)
* Lead 0.6
Asbestos 0.007
Beryllium 0.0004
Mercury 0.1
Vinyl chloride 1
Fluorides 3
Sulfuric acid mist 7
Hydrogen sulfide (H2S) 10
Total Reduced sulfur compounds
(including H2S) 10
Criteria Pollutants
Nitrogen dioxide is the compound regulated as a criteria pollutant;
however, significant emissions are based on the sum of all oxides of
nitrogen.
Sulfur dioxide is the measured surrogate for the- criteria pollutant
sulfur oxides. Sulfur oxides have been made subject to regulation
explicitly through the proposal of 40 CFR 60 Subpart J as of
August 17, 1989.
A.20
-------
DRAFT
OCTOBER 1990
TABLE A-4. (Concluded) SIGNIFICANT EMISSION RATES OF POLLUTANTS
REGULATED UNDER THE CLEAN AIR ACT
Pollutant Emissions rate (tons/year)
Other pollutants regulated by the Clean Air Act:cd
Benzene |
Arsenic |
Radionuclides | Any emission rate
Radon-222 j
Polonium-210 |
CFC's 11,12, 112, 114, 115 |
Halons 1211, 1301, 2402 |
c Significant emission rates have not been promulgated for these pollutants,
and until such time, any emissions by a new major sources or any increase
in emissions at an existing major source due to modification, are
"significant."
d Regulations covering several pollutants such as cadmium, coke oven
emissions, and municipal waste incinerator emissions have recently been
proposed. Applicants should, therefore, verify what pollutants have
been regulated under the Act at the time of application.
A.21
-------
DRAFT
OCTOBER 1990
Sources of the worst-case uncontrolled emissions and applicable control
system efficiencies could be any of the following:
Emissions data from compliance tests or other source tests,
Equipment vendor emissions data and guarantees;
Emission limits and test data from EPA documents, including
background information documents for new source performance
standards, national emissions standards for hazardous air
pollutants, and Section llld standards for designated
pollutants;
AP-42 emission factors (see Table A-3, Reference 2);
Emission factors from technical literature; and
State emission inventory questionnaires for comparable sources.
The effect of other restrictions (federally-enforceable and practically-
enforceable) should also be factored into the results. The potential to emit
of each pollutant, including fugitive emissions if applicable, is estimated
for each individual emissions unit. The individual estimates are then summed
by pollutant over all the emissions units at the stationary source.
II.C. EMISSIONS THRESHOLDS FOR PSD APPLICABILITY
II.C.I. MAJOR SOURCES
A source is a "major stationary source" or "major emitting facility" if:
(1) It can be classified in one of the 28 named source categories
listed in Section 169 of the CM (see Table A-l) and it emits
or has the potential to emit 100 tpy or more of any pollutant
regulated by the Act, or
(2) it is any other stationary source that emits or has the
potential to emit 250 tons per year or more of any pollutant
regulated by the CM.
For example, one of the 28 PSD source categories subject to the 100-
tpy threshold is fossil fuel-fired steam generators with a heat input
A.22
-------
DRAFT
OCTOBER 1990
greater than 250 million Btu/hr. Consequently, a 300 million Btu/hr
boiler that is designed and permitted to burn any fossil fuel, i.e.,
coal, oil, natural gas or lignite, that emits 100 tpy or more of any
regulated pollutant, e.g., S02, is a major stationary source. If,
however, the boiler were designed and permitted to burn wood only, it
would not be classified as one of the 28 PSD sources and would instead
be subject to the 250 tpy threshold.
A single, fossil fuel-fired boiler with a maximum heat input capacity
of 300 million Btu/hr takes a federally-enforceable design limitation
that restricts heat input to 240 million Btu/hr. Consequently, this
source would not be classified within one of the 28 categories and
would therefore be subject to the 250-tpy, rather than the 100-tpy,
emissions threshold.
A situation frequently occurs in which an emissions unit that is included
in the 28 listed source categories (and so is subject to a 100 tpy threshold),
is located within a parent source whose primary activity is not on the list
(and is therefore subject to a 250 tpy threshold). A source which, when
considered alone, would be major (and hence subject to PSD) cannot "hide"
within a different and less restrictive source category in order to escape
applicability.
As an example, a proposed coal mining operation will use an on-site
coal cleaning plant with a thermal dryer. The source will be defined
as a coal mine because the cleaning plant will only treat coal from
the mine. The mine's potential to emit (including emissions from the
thermal dryer) is less than 250 tpy for every regulated pollutant;
therefore, it is a "minor" source. The estimated emissions from the
thermal dryer, however, will be 150 tpy particulate matter. Thermal
dryers are included in the list of 28 source categories that are
subject to the 100 tpy major source threshold. Consequently, the
thermal dryer would be considered an emissions unit that by itself is
a major source and therefore is subject to PSD review, even though the
primary activity is not.
Furthermore, when a "minor" source, i.e., one that does not meet the
definition of "major," makes a physical change or change in the method of
operation that is by itself a major source, that physical or operational
change constitutes a major stationary source that is subject to PSD review.
A.23
-------
DRAFT
OCTOBER 1990
To illustrate, consider the following scenarios at an existing glass
fiber processing plant, which proposes to add new equipment to
increase production. Glass fiber processing plants are included in
the list of 28 source categories that are subject to the 100-tpy major
source threshold. The existing plant emits 40 tpy particulate, which
is both its potential to emit and permitted allowable rate. It also
has a potential to emit all other pollutants in less than major
quantities; therefore it is a minor source.
Scenario 1 - The physical change will increase the source's potential
to emit particulate matter by 50 tpy. Since the plant is a minor
source and the increase is not major by itself, the change is not
subject to PSD review.
Scenario 2 - The physical change will increase the source's potential
to emit particulate matter by 65 tpy. Since the plant is a minor
source and the increase is not major by itself, neither is subject to
PSD review. However, the source's potential to emit after the change
will exceed the 100-tpy major source threshold, so future
modifications will be scrutinized under the netting provisions (see
section A.3.2).
Scenario 3 - The physical change will increase the source's potential
to emit particulate matter by 110 tpy. Since the existing plant is
a minor source and the change by itself results in an emissions
increase greater than the major source threshold, that change is
subject to PSD review. Furthermore, the physical change makes the
entire plant a major source, so future physical changes or changes in
the method of operation will be scrutinized against the criteria for
major modifications (see section II.A.3.2).
II.C.2. SIGNIFICANT EMISSIONS
A PSD review is triggered in certain instances when emissions associated
with a new major source or emissions increases resulting from a major
modification are "significant." "Significant" emissions thresholds are
defined two ways. The first is in terms of emission rates (tons/year).
Table A-4 listed the pollutants for which significant emissions rates have
been established.
Significant increases in emission rates are subject to PSD review in two
circumstances:
A.24
-------
DRAFT
OCTOBER 1990
(1) For a new source which is major for at least one regulated
attainment or noncriteria pollutant, i.e., is subject to PSD review,
all pollutants for which the area is not classified as nonattainment
and which are emitted in amounts equal to or greater than those
specified in Table A-4 are also subject to PSD review for its VOC
emissions.
For example, an automotive assembly plant is planned for an attainment area for
all criteria pollutants. The plant has a potential to emit 350 tpy VOC, 50 tpy
NOX, 60 tpy SO,,and 10 tpy PM including 5 tpy PM-10. The 350 tpy VOC exceeds
the major source threshold, and therefore subjects the plant to PSD review. The
"significant" emissions thresholds for NO, and SO, are 40 tpy; therefore, the NO,
and SO, emissions, also, will be subject to PSD review. The PM and PM-10
emissions will not exceed their significant emissions thresholds; therefore they
are not subject to review.
(2) For a modification to an existing major stationary source, if both the
potential increase in emissions due to the modification itself, and the
resulting net emissions increase of any regulated, attainment or
noncriteria pollutants are equal to or greater than the respective
pollutants' significant emissions rates listed in Table A-4, the
modification is "major," and subject to PSD review. Modifications are
discussed in detail in Section II.D.
The second type of "significant" emissions threshold is defined as any
emissions rate at a new major stationary source (or any net emissions increase
associated with a modification to an existing major stationary source) that is
constructed within 10 kilometers of a Class I area, and which would increase
the 24-hour average concentration of any regulated pollutant in that area by
1 fig/m3 or greater. Exceedence of this threshold triggers PSD review.
II.D. LOCAL AIR QUALITY CONSIDERATIONS FOR CRITERIA POLLUTANTS
The air quality, i.e., attainment status, of the area of a proposed new
source or modified existing source will impact the applicability determination
in regard to the pollutants that are subject to PSD review. As previously
stated, if a new source locates in an area designated attainment or
unclassifiable for any criteria pollutant, PSD review will apply to any
A.25
-------
DRAFT
OCTOBER 1990
pollutant for which the potential to emit is major (or significant, if the
source is major) so long as the area is not nonattainment for that pollutant.
For example, a kraft pulp mill is proposed for an attainment area
for SOo, and its potential to emit SO? equals 55 tpy. Its potential
to emit total reduced sulfur (TRS) a noncriteria pollutant, equals
295 tpy. Its potential to emit VOC will be 45 tpy and PM/PM-10,
30/5 tpy; however, the area is designated nonattainment for ozone
and PM. Applicability would be assessed as follows:
The source would be major and subject to PSD review due to the
noncriteria TRS emissions.
The 50^ emissions would therefore be subject to PSD because
they are significant and the area is attainment for 50^.
The VOC emission and PM emissions would not be subject to PSD,
even though their emissions are significant, because the area
is designated nonattainment for those pollutants.
The PM-10 emissions are neither major nor significant and would
therefore not be subject to review.
Similarly, if the modification of an existing major source, which is located
in an attainment area for any criteria pollutant, results in a significant
increase in potential to emit and a significant net emissions increase, the
modification is subject to PSD, unless the location is designated as
nonattainment for that pollutant.
Note that if the source is major for a pollutant for which an area is
designated nonattainment, all significant emissions or significant emissions
increases of pollutants for which the area is attainment or unclassifiable are
still subject to PSD review.
II.E. SUMMARY OF MAJOR NEW SOURCE APPLICABILITY
The elements and associated information necessary for determining PSD
applicability to new sources are outlined as follows:
A.26
-------
DRAFT
OCTOBER 1990
Element 1 - Define the source
includes all related activities classified under the same 2-digit
SIC Code number
must have the same owner or operator
must be located on contiguous or adjacent properties
includes all support facilities
Element 2 - Define applicability thresholds for major source as a whole
(primary activity)
100 tpy for individual emissions units or groups of units that
are included in the list of 28 source categories identified in
Section 169 of the CAA
250 tpy for all other sources
Element 3 - Define project emissions (potential to emit)
Reflects federally-enforceable air pollution control efficiency,
operating conditions, and permit limitations
Determined for each pollutant by each emissions unit
Summed by pollutant over all emissions units
«
Includes fugitive emissions for 28 listed source categories and
sources subject to NSPS or NESHAPS as of August 7, 1980
Element 4 - Assess local area attainment status
Area must be attainment or unclassifiable for at least one criteria
pollutant for PSD to apply
A.27
-------
DRAFT
OCTOBER 1990
Element 5 - Determine if source is major by comparing its potential emissions
to appropriate major source threshold
Major if any pollutant emitted by defined source exceeds
thresholds, regardless of area designation, i.e., attainment,
nonattainment, or noncriteria pollutants
Individual unit is major if classified as a source in one of
the 28 regulated source categories and emissions exceed an
applicable 100-tpy threshold
Element 6 - Determine pollutants subject to PSD review
Each attainment area and noncriteria pollutant emitted in
"significant" quantities
Any emissions or emissions increase from a major source that
results in an increase of 1 jug/m3 (24 hour average) or more in
a Class I area if the major source is located or constructed
within 10 kilometers of that Class I area.
II.F. NEW SOURCE APPLICABILITY EXAMPLE
The following example provided is for illustration only. The example
source is fictitious and has been created to highlight many of the aspects of
the PSD applicability process for a new source.
In this example the proposed project is a new coal-fired electric plant.
The plant will have two 600-MW lignite-fired boilers. The proposed location
is near a separately-owned surface lignite mine, which will supply the fuel
requirements of the power plant, and will therefore, have to increase its
mining capacity with new equipment. The lignite coal will be mined and then
transported to the power plant to be crushed, screened, stored, pulverized and
fed to the boilers. The power plant has informed the lignite coal mine that
the coal will not have to be cleaned, so the mine will not expand its coal
cleaning capacity. The power plant will have on-site coal and limestone
A. 28
-------
DRAFT
OCTOBER 1990
storage and handling facilities. In addition, a comparatively small auxiliary
boiler will be installed to provide steam for the facility when the main
boilers are inoperable. The area is designated attainment for all criteria
pollutants.
The applicant proposes pollution control devices for the two 600-MW
boilers which include:
- an electrostatic precipitator (ESP) for PM/PM-10 emissions control,
- a limestone scrubber flue gas desulfurization (F6D) system for
$62 emissions control;
- low-nitrogen oxide (NOX) burners and low-excess-air firing for
NOV emissions control; and
A
- controlled combustion for CO emissions control.
The first step is to determine what constitutes the source (or sources).
A source is defined as all pollutant-emitting activities associated with the
same industrial grouping, located on contiguous or adjacent sites, and under
common control or ownership. Industrial groupings are generally defined by
two-digit SIC codes. The power plant is classified as SIC major group 49; the
nearby mine is SIC major group 12. They are neither under the same SIC major
group number nor have the same owners, so they constitute separate sources.
The second step is to establish which major source thresholds are
applicable in this case. The proposed power plant is a fossil fuel-fired
steam electric plant with more than 250 million Btu/hr of heat input, making
it a source included in one of the 28 PSD-listed categories. It is therefore
subject to both the 100 ton per year criterion for any regulated pollutant
used to determine whether a source is major and to the requirement that
quantifiable fugitive emissions be included in determining potential to emit.
A.29
-------
DRAFT
OCTOBER 1990
The emissions units at the mine are neither classified within one of the
28 PSD source categories nor regulated under Sections 111 or 112 of the Act.
Therefore, the mine is compared against the 250 tpy major source threshold and
fugitive emissions from the mining operations are exempt from consideration in
determining whether the mine is a major stationary source.
The third step is to define the project emissions. To arrive at the
potential to emit of the proposed power plant, the applicant must consider all
quantifiable stack and fugitive emissions of each regulated pollutant (i.e.,
S02, NOX, PM, PM-10, CO, VOC, lead, and the noncriteria pollutants).
Therefore, fugitive PM/PM-10 emissions from haul roads, disturbed areas, coal
piles, and other sources must be included in calculating the power plant's
potential to emit.
All stack and fugitive emissions estimates have been obtained through
detailed engineering analysis of each emissions unit using the best available
data or estimating technique. Fugitive emissions are added to the emissions
from the two main boilers and the auxiliary boiler in order to arrive at the
total potential to emit of each regulated pollutant. The auxiliary boiler in
this case is restricted by enforceable limits on operating hours proposed to
be included in the source's PSD permit. If the auxiliary boiler were not
limited in hours of operation, its contribution would be based on full,
continuous operation, and the resulting potential emissions estimates would be
higher.
The potential to emit S02, NOX, PM, CO, and sulfuric acid mist each
exceeds 100 tons per year. From data collected at other lignite fired power
plants it is known that emissions of lead, beryllium, mercury, fluorides,
sulfuric acid mist and arsenic should also be quantified. It is known that
fluoride compounds are contained in the coal in significant quantities;
however, engineering analyses show fluoride removal in the proposed limestone
scrubber will result in insignificant stack emissions. Similarly, liquid
absorption, absorption of fly ash removed in the ESP, and removal of bottom
A.30
-------
DRAFT
OCTOBER 1990
ash have been shown to maintain emissions of lead and the other regulated
noncriteria pollutants below significance levels.
The only emissions at the existing mine, and consequently the only
emissions increase that will occur from the expansion to serve the power
plant, are fugitive PM/PM-10 emissions from mining operations. The mine's
potential to emit, for PSD applicability purposes, is zero and the mine is not
subject to a PSD review. The increase in fugitive emissions from the mine,
however, will be classified as secondary emissions with respect to the power
plant and, therefore, must be considered in the air quality analysis and
additional impacts analysis for the proposed power plant if the power plant is
subject to PSD review.
The next step is to compare the potential emissions of the power plant to
the 100 ton per year major" source threshold. If the potential to emit of any
regulated pollutant is 100 tons per year or more, the power plant is
classified as a major stationary source for PSD purposes. In this case, the
plant is classified as a major source because SC^, NOX, PM, CO, and sulfuric
acid mist emissions each exceed 100 tons per year. (Note that emissions of
any one of these pollutants classifies the source as major.)
Once it has been determined that the proposed source is major, any
regulated pollutant (for which the location of the source is not classified as
nonattainment) with significant emissions is subject to a PSD review. The
applicant quantified, through coal and captured fly ash analyses and through
performance test results from existing sources burning equivalent coals,
emissions of fluorides, beryllium, lead, mercury, and the other regulated
noncriteria pollutants to determine if their emissions exceed the significance
levels (see Table A-4.)- Pollutants with less than significant emissions are
not subject to PSD review requirements (assuming the proposed controls are
accepted as BACT for S02, or the application of BACT for S02 results in
equivalent or lower noncriteria pollutant emissions).
A.31
-------
DRAFT
OCTOBER 1990
Note that, because the proposed construction site is not within 10
kilometers of a Class I area, the source's emissions are not subject to the
Class I area significance criteria.
A.32
-------
DRAFT
OCTOBER 1990
III. MAJOR MODIFICATION APPLICABILITY
A modification is subject to PSD review only if (1) the existing source
that is modified is "major," and (2) the net emissions increase of any
pollutant emitted by the source, as a result of the modification, is
"significant," i.e., equal to or greater than the emissions rates given on
Table A-4 (unless the source is located in a nonattainment area for that
pollutant). Note also that any net emissions increase in a regulated
pollutant at a major stationary source that is located within 10 kilometers of
a Class I area, and which will cause an increase of 1 /*g/m3 (24 hour average)
or more in the ambient concentration of that pollutant within that Class I
area, is "significant".
Typical examples of modifications include (but are not limited to)
replacing a boiler at a chemical plant, construction of a new surface
coating line at an assembly plant, and a switch from coal to gas
requiring a physical change to the plant, e.g., new piping, etc.
As discussed earlier, when a "minor" source, i.e:, one that does not meet
the definition of "major," makes a physical change or change in the method of
operation that is by itself a major source, that physical or operational change
constitutes a ma.lor stationary source that is subject to PSD review. Also, if
an existing minor source becomes a major source as a result of a SIP relaxation,
then it becomes subject to PSD requirements just as if construction had not yet
commenced on the source or the modification.
III.A. ACTIVITIES THAT ARE NOT MODIFICATIONS
The regulations do not define "physical change" or "change in the method
of operation" precisely; however, they exclude from those activities certain
specific types of events described below.
(1) Routine maintenance, repair and replacement.,;
[Sources should discuss any project thai will significantly
increase actual emissions to the atmosphere with their
A.33
-------
DRAFT
OCTOBER 1990
respective permitting authority, as to whether that project
is considered routine maintenance, repair or replacement.]
(2) A fuel switch due to an order under the Energy Supply and
Environmental Coordination Act of 1974 (or any superseding
legislation) or due to a natural gas curtailment plan under the
Federal Power Act.
(3) A fuel switch due to an order or rule under section 125 of the CAA.
(4) A switch at a steam generating unit to a fuel derived in whole or in
part from municipal solid waste.
(5) A switch to a fuel or raw material which (a) the source was
capable of accommodating before January 6, 1975, so long as the
switch would not be prohibited by any federally-enforceable
permit condition established after that date under a federally
approved SIP (including any PSD permit condition) or a federal
PSD permit, or (b) the source is approved to make under a PSD
permit.
(6) Any increase in the hours or rate of operation of a source, so
long as the increase would not be prohibited by any federally-
enforceable permit condition established after January 6, 1975
under a federally approved SIP (including any PSD permit
condition) or a federal PSD permit.
(7) A change in the ownership of a stationary source.
For more details see 40 CFR 52.21(b)(2)(iii).
Notwithstanding the above, if a significant increase in actual emissions
of a regulated pollutant occurs at an existing major source as a result of a
physical change or change in the method of operation of that source, the "net
emissions increase" of that pollutant must be determined.
III.B. EMISSIONS NETTING
Emissions netting is a term that refers to the process of considering
certain previous and prospective emissions changes at an existing major source
to determine if a "net emissions increase" of a pollutant will result from a
proposed physical change or change in method of operation. If a net emissions
A.34
-------
DRAFT
OCTOBER 1990
increase is shown to result, PSD applies to each pollutant's emissions for
which the net increase is "significant", as shown in Table A-4.
The process used to determine whether there will be a net emissions
increase will result uses the following equation:
Net Emissions Change
EQUALS
Emissions increases associated with the proposed modification
MINUS
Source-wide creditable contemporaneous emissions decreases
PLUS
Source-wide creditable contemporaneous emissions increases
Consideration of contemporaneous emissions changes is allowed only in cases
involving existing ma.ior sources. In other words, minor sources are not
eligible to net emissions changes. As discussed earlier, existing minor
sources are subject to PSD review only when proposing to increase emissions by
"major" (e.g., 100 or 250 tpy, as applicable) amounts, which, for PSD
purposes, are considered and reviewed as a major new source.
For example, an existing minor source (subject to the 100 tpy major
source cutoff) is proposing a modification which involves the shutdown
and removal -of an old emissions unit (providing an actual
contemporaneous reduction in NOx emissions of 75 tpy) and the
construction of two new units with total potential NOx emissions of
110 tpy. Since the existing source is minor, the 75 tpy reduction is
not considered for PSD applicability purposes. Consequently, PSD
applies to the new units because the emissions increase of 110 tpy is
itself "major". The new units are then subject to a PSD review for
NOx and for any other regulated pollutant with a "significant"
potential to emit.
The consideration of contemporaneous emissions changes is also source
specific. Netting must take place at the same stationary source; emissions
reductions cannot be traded between stationary sources.
A.35
-------
DRAFT
OCTOBER 1990
III.B.I. ACCUMULATION OF EMISSIONS
If the proposed emissions increase at a major source is by itself
(without considering any decreases) less than "significant", EPA policy does
not require consideration of previous contemporaneous small (i.e., less than
significant) emissions increases at the source. In other words, the netting
equation (the summation of contemporaneous emissions increases and decreases)
is not triggered unless there will be a significant emissions increase from
the proposed modification.
For example, a major source experienced less than significant
increases of NOX (30 tpy) and SO? (15 tpy) 2 years ago, and a decrease
of SO2 (50 tpyf 3 years ago. The source now proposes to add a new
process unit with an associated emissions increase of 35 tpy NOX and
80 tpy S02. For SOo, the proposed 80 tpy increase from the
modification by itselT (before netting) is significant. The
contemporaneous net emissions change is determined, by taking the
algebraic sum of (-50) and (+15) and (+80), which equals +45 tpy.
Therefore, the proposed modification is a major modification and a
PSD review for S02 is required. However, the NOX increase from the
proposed modification is by itself less than significant.
Consequently, netting for PSD applicability purposes is not performed
for NOX (even though the modification is major for S02) and a PSD
review is not needed for /I/O .
It is important to note that when any emissions decrease is claimed (including
those associated with the proposed modification), all source-wide creditable
and contemporaneous emissions increases and decreases of the pollutant subject
to netting must be included in the PSD applicability determination.
A deliberate decision to split an otherwise "significant" project into
two or more smaller projects to avoid PSD review would be viewed as
circumvention and would subject the entire project to enforcement action if
construction on any of the small projects commences without a valid PSD
permit.
For example, an automobile and truck tire manufacturing plant, an
existing major source, plans to increase its production of both types
A.36
-------
DRAFT
OCTOBER 1990
"debottlenecking" its production processes. For its passenger tire line,
the source applies for and is granted a "minor" modification permit for a
new extruder that will increase VOC emissions by 39 tons/yr. A few months
later, the source applies for a "minor" modification permit to construct a
new tread-end cementer on the same line which will increase VOC emissions
by 12 tons/yr. The EPA would likely consider these proposals as an attempt
to circumvent the regulations because the two proposals are related in
terms of an overall project to increase source-wide production capacity.
The important point in this example is that the two proposals are
sufficiently related that the PSD regulations would consider them a single
project.
Usually, at least two basic questions should be asked when evaluating the
construction of multiple minor projects to determine if they should have been
considered a single project. First, were the projects proposed over a
relatively short period of time? Second, could the changes be considered as
part of a single project?
III.B.2. CONTEMPORANEOUS EMISSIONS CHANGES
The PSD definition of a net emissions increase [40 CFR 52.21(b)(3)(i)3
consists of two additive components as follows:
(a) Any increases in actual emissions from a particular physical change
or change in method of operation at a stationary source; and
(b) Any other increase and decreases jn actual emissions at the source
that are contemporaneous with the particular change and are
otherwise creditable.
The first component narrowly includes only the emissions increases
associated with a particular change at the source. The second component more
broadly includes all contemporaneous, source-wide (occurring anywhere at the
entire source), creditable emission increases and decreases.
To be contemporaneous, changes in actual emissions must have occurred
after January 6, 1975. The changes must also occur within a period beginning
5 years before the date construction is expected to commence on the proposed
A.37
-------
DRAFT
OCTOBER 1990
modification (reviewing agencies may use the date construction is scheduled to
commence provided that it is reasonable considering the time needed to issue a
final permit) and ending when the emissions increase from the modification
occurs. An increase resulting from a physical change at a source occurs when
the new emissions unit becomes operational and begins to emit a pollutant. A
replacement that requires a shakedown period becomes operational only after a
reasonable shakedown period, not to exceed 180 days. Since the date
construction actually will commence is unknown at the time the applicability
determination takes place and is simply a scheduled date projected by the
source, the contemporaneous period may shift if construction does not commence
as scheduled. Many States have developed PSD regulations that allow different
time frames for definitions of contemporaneous. Where approved by EPA, the
time periods specified in these regulations govern the contemporaneous
timeframe.
III.B.3. CREDITABLE CONTEMPORANEOUS EMISSIONS CHANGES
There are further restrictions on the contemporaneous emissions changes
that can be credited in determining net increases or decreases. To be
creditable, a contemporaneous reduction must be federally-enforceable on and
after the date construction on the proposed modification begins. The actual
reduction must take place before the date that the emissions increase from any
of the new or modified emissions units occurs. In addition, the reviewing
agency must ensure that the source has maintained any contemporaneous decrease
which the source claims has occurred in the past. The source must either
demonstrate that the decrease was federally-enforceable at the time the source
claims it occurred, or it must otherwise demonstrate that the decrease was
maintained until the present time and will continue until it becomes
federally-enforceable. An emissions decrease cannot occur at, and therefore,
cannot be credited from an emissions unit which was never constructed or
operated, including units that received a PSD permit.
A.38
-------
DRAFT
OCTOBER 1990
Reductions must be of the same pollutant as the emissions increase from
the proposed modification and must be qualitatively equivalent in their
effects on public health and welfare to the effects attributable to the
proposed increase. Current EPA policy is to assume that an emissions decrease
will have approximately the same qualitative significance for public health
and welfare as that attributed to an increase, unless the reviewing agency has
reason to believe that the reduction in ambient concentrations from the
emissions decrease will not be sufficient to prevent the proposed emissions
increase from causing or contributing to a violation of any NAAQS or PSD
increment. In such cases, the applicant must demonstrate that the proposed
netting transaction will not cause or contribute to an air quality violation
before the emissions reduction may be credited. Also, in situations where a
State is implementing an air toxics program, proposed netting transactions may
be subject to additional tests regarding the health and welfare equivalency
demonstration. For example, a State may prohibit netting between certain
groups of toxic subspecies or apply netting ratios greater than the normally
required 1:1 between certain groups of toxic pollutants.
A contemporaneous emissions increase occurs as the result of a physical
change or change in the method of operation at the source and is creditable to
the extent that the new emissions level exceeds the old emissions level. The
"old" emissions level for an emissions unit equals the average rate (in tons
per year) at which the unit actually emitted the pollutant during the 2-year
period just prior to the physical or operational change which resulted in the
emissions increase. In certain limited situations where the applicant
adequately demonstrates that the prior 2 years is not representative of normal
source operation, a different (2 year) time period may be used upon a
determination by the reviewing agency that it is more representative of normal
source operation. Normal source operations may be affected by strikes,
retooling, major industrial accidents and other catastrophic occurrences. The
"new" emissions levels for a new or modified emissions unit which has not
begun normal operation is its potential to emit.
A.39
-------
DRAFT
OCTOBER 1990
An emissions increase or decrease is creditable only if the relevant
reviewing authority has not relied on it in issuing a PSD permit for the
source, and the permit is still in effect when the increase in actual
emissions from the proposed modification occurs. A reviewing authority relies
on an increase or decrease when, after taking the increase or decrease into
account, it concludes that a proposed project would not cause or contribute to
a violation of an increment or ambient standard. In other words, an emissions
change at an emissions point which was considered in the issuance of a
previous PSD permit for the source is not included in the source's "net
emissions increase" calculation. This is done to avoid "double counting" of
emissions changes.
For example, an emissions increase or decrease already considered in
a source's PSD permit (state or federal) can not be considered a
contemporaneous increase or decrease since the increases or decrease
was obviously relied upon for the purpose of issuing the permit.
Otherwise the increase or decrease would not have been specified in
the permit. In another example, a decrease in emissions from having
previously switched to a less polluting fuel (e.g., oil to gas) at an
existing emissions unit would not be creditable if the source had, in
obtaining a PSD permit (which is still in effect) for a new emissions
unit, modeled the source's ambient impact using the less polluting
fuel.
Changes in PM (PM/PM-10), S02 and NOX emissions are a subset of
creditable contemporaneous changes that also affect the available increment.
For these pollutants, emissions changes which do not affect allowable PSD
increment consumption are not creditable.
III.B.4- CREDITABLE AMOUNT
As mentioned above, only contemporaneous and creditable emissions changes
are considered in determining the source-wide net emissions change. All
contemporaneous and creditable emissions increases and decreases at the source
must, however, be considered. The amount of each contemporaneous and
A.40
-------
DRAFT
OCTOBER 1990
creditable emissions increase or decrease involves determining old and new
actual annual emissions levels for each affected emission unit.
The following basic criteria should be used when quantifying the increase
or decrease:
*• For proposed new or modified units which have not begun normal
operations, the potential to emit must be used to determine the
increase from the units.
»• For an existing unit, actual emissions just prior to either a
physical or operational change are based on the lower of the actual
or allowable emissions levels. This "old" emissions level equals
the average rate (in tons per year) at which the unit actually
emitted the pollutant during the 2-year period just prior to the
change which resulted in the emissions increase. These emissions
are calculated using the actual hours of operation, capacity, fuel
combusted and other parameters which affected the unit's emissions
over the 2-year averaging period. In certain limited circumstances,
where sufficient representative operating data do not exist to
determine historic actual emissions and the reviewing agency has
reason to believe that the source is operating at or near its
allowable emissions level, the reviewing agency may presume that
source-specific allowable emissions [or a fraction thereof] are
equivalent to (and therefore are used in place of) actual emissions
at the unit. For determining the difference in emissions from the
change at the unit, emissions after the change are the potential to
emit from the units.
»> A source cannot receive emission reduction credit for reducing any
portion of actual emissions which resulted because the source was
operating out of compliance.
»• An emissions decrease cannot be credited from a unit that has not
been constructed or operated.
Examples of how to apply these creditability criteria for prospective
emissions reductions is shown in Figure A-l. As shown in Case I of
Figure A-l, the potential to emit for an existing emissions unit
(which is based on the existing allowable emission rate) is greater
than the actual emissions, which are based on actual operating data
(e.g., type and amount of fuel combusted at the unit) for the past 2
years. The source proposes to switch to a lower sulfur fuel. The
amount of the reduction in this case is the difference between the
actual emissions and the revised allowable emissions. (Recall that
A.41
-------
DRAFT
OCTOBER 1990
for reductions to be creditable, the revised allowable emission rate
must be ensured with federally-enforceable limits.)
Figure A-l also illustrates in Case II that the previous allowable
emissions were much higher than the potential to emit. Common
examples are PM sources permitted according to process weight tables
contained in most SIPs. Since process weight tables apply to a range
of source types, they often overpredict actual emission rates for
individual sources. In such cases, as in the previous case, the only
creditable contemporaneous reduction is the difference, between the
actual emissions and the revised allowable emission rate for the
existing emissions unit.
Case III in Figure A-l illustrates a potential violation situation
where the actual emissions level exceeds allowable limit. The
creditable reduction in this case is the difference between what the
emissions would have been from the unit had the source been in
compliance with its old allowable limits (considering its actual
operations) and its revised allowable emissions level.
Consider a more specific example, where a source has an emissions unit
with an annual allowable emissions rate of 200 tpy based on full
capacity year-round operation and an hourly unit-specific allowable
emission rate. The source is, however, out of compliance with the
allowable hourly emission rate by a factor of two. Consequently, if
the unit were to be operated year-round at full capacity it would emit
400 tpy. However, in this case, although the unit operated at full
capacity, it was operated on the average 75 percent of the time for
the past 2 years. Consequently, for the past 2 years average actual
emissions were 300 tpy. The unit is now to be shutdown. Assuming
the reduction is otherwise creditable, the reduction from the shutdown
is its allowable emissions prorated by its operating factor
(200 tpy x .75 = 150 tpy).
A.42
-------
Case I: Normal Existing Source
Potential to Emit
Equals Existing
Allowable Emissions
Actual
Emissions
DRAFT
OCTOBtR 1990
Creditable
Reduction
Revised Allowable
Emissions
Case II: Existing Source Where Allowable Exceeds Potential
Existing
Allowable
Emissions
•••I
•."•:V;.'S;::7V ";-.'""
•,;,-•;;.•:-;,,.
1
c
F
f
Creditable
Reduction
Potential to Emit Actual
at Maximum Capacity Emissions
Revised Allowable
Emissions
Case III: Existing Source in Violation of Permit
Existing
Allowable
Emissions
(at 70% Capacity)
Actual
Emissions
(at 70% Capacity)
Creditable
' Reduction
Revised Allowable
Emissions
Figure A-1. Creditable Reductions in Actual Emissions
A.43
-------
DRAFT
OCTOBER 1990
III.B.5. SUGGESTED EMISSIONS NETTING PROCEDURE
Through its review of many emissions netting transactions, EPA has found
that, either because of confusion or misunderstanding, sources have used
various netting procedures, some of which result in cases where projects
should have been subjected to PSD but were not. Some of the most common
errors include:
>• Not including contemporaneous emissions increases when considering
decreases;
»• Improperly using allowable emissions instead of actual emissions
level for the "old" emissions level for existing units;
»• Using prospective (proposed) unrelated emissions decreases to
counterbalance proposed emission increases without also examining
all previous contemporaneous emissions changes;
»• Not considering a contemporaneous increase creditable because the
increase previously netted out of review by relying on a past
decrease which was, but is no longer, contemporaneous. If
contemporaneous and otherwise creditable, the increase must be
considered in the netting calculus.
»• Not properly documenting all contemporaneous emissions changes; and
»• Not ensuring that emissions decreases are covered by federally-
enforceable restrictions, which is a requirement for creditability.
For the purpose of minimizing confusion and improper applicability
determinations, the six-step procedure shown in Table A-5 and described below
is recommended in applying the emissions netting equation. Already assumed in
this procedure is that the existing source has been defined, its major source
status has been confirmed and the air quality status in the area is attainment
for at least one criteria pollutant.
A.44
-------
DRAFT
OCTOBER 1990
TABLE A-5. Procedures for Determining
the Net Emissions Change at a Source
Determine the emissions increases (but not any
decreases) from the proposed project. If increases are
significant, proceed; if not, the sources is not subject
to review.
Determine the beginning and ending dates of the
contemporaneous period as it relates to the proposed
modification.
Determine which emissions units at the source •
experienced (or will experience, including any proposed
decreases resulting from the proposed project) a
creditable increase or decrease in emissions during the
contemporaneous period.
Determine which emissions changes are creditable.
Determine, on a pollutant-by-pollutant basis, the amount
of each contemporaneous and creditable emissions
increase and decrease.
Sum all contemporaneous and creditable increases and
decreases with the increase from the proposed
modification to determine if a significant net emissions
increase will occur.
A.45
-------
DRAFT
OCTOBER 1990
Step 1. Determine the emissions increases from the proposed project.
First, only the emissions increases expected to result from the
proposed project are examined. This includes emissions increases
from the new and modified emissions units and any other plant-wide
emissions increases (e.g., debottlenecking increases) that will
occur as a result of the proposed modification. [Proposed emissions
decreases occurring elsewhere at the source are not considered at
this point. Emission decreases associated with a proposed project
(such as a boiler replacement) are contemporaneous and may be
considered along with other contemporaneous emissions changes at the
source. However, they are not considered at this point in the
analysis.]
A PSD review applies only to those regulated pollutants with a
significant emissions increase from the proposed modification. If
the proposed project will not result in a significant emissions
increase of any regulated pollutant, the project is exempt from PSD
review and the PSD applicability process is completed. However, if
this is not the case, each regulated pollutant to be emitted in a
significant amount is subject to a PSD review unless the source can
demonstrate (using steps 2-6) that the sum of all other source-wide
contemporaneous and creditable emissions increases and decreases
would be less than significant.
Step 2 Determine the beginning and ending dates of the contemporaneous
period as it relates to the proposed modification.
The period begins on the date 5 years (some States may have a
different time period) before construction commences on the proposed
modification. It ends on the date the emissions increase from the
proposed modification occurs.
Step 3 Determine which emissions units at the source have experienced an
increase or decrease in emissions during the contemporaneous period.
Usually, creditable emissions increases are associated with a
physical change or change in the method of operation at a source
which did not require a PSD permit. For example, creditable
emissions increases may come from the construction of a new unit, a
fuel switch or an increase in operation that (a) would have
otherwise been subject to PSD but instead netted out of review (per
steps 1-6) or (b) resulted in a less than significant emissions
increase (per step 1).
Decreases are creditable reductions in actual emissions from an
emissions unit that are, or can be made, federally-enforceable. A
A.46
-------
DRAFT
OCTOBER 1990
physical change or change in the method of operation is also
associated with the types of decreases that are creditable.
Specifically, in the case of an emissions decrease, once the
decrease has been made federally-enforceable, any proposed increase
above the federally-enforceable level must constitute a physical
change or change in the method of operation at the source or the
reduction is not considered creditable. For example, a source could
only receive an emissions decrease for netting purposes from a unit
that has been taken out of operation if, due to the imposition of
federally-enforceable restrictions preventing the use of the unit, a
proposal to reactivate the unit would constitute a physical change
or change in the method of operation at the source. If operating
the unit was not considered a physical or operational change, the
unit could go back to its prior level of operation at any time,
thereby producing only a "paper" reduction, which is not creditable.
Step 4 Determine which emissions changes are creditable.
The following basic rules apply:
1) A increase or decrease is creditable only if the relevant
reviewing authority has not relied upon it in previously issuing a
PSD permit and the permit is in effect when the increase from the
proposed modification occurs. As stated earlier, a reviewing
authority "relies" on an increase or decrease when, after taking the
increase or decrease into account, it concludes in issuing a PSD
permit that a project would not cause or contribute to a violation
of a PSD increment or ambient standard.
2) For pollutants with PSD increments (i.e., S02, particulate matter
and NOx), an increase or decrease in actual emissions which occurs
before the baseline date in an area is creditable only if it would
be considered in calculating how much of an increment remains
available for the pollutant in question. An example of this
situation is a 39 tpy NOX emissions increase resulting from a new
heater at a major source in 1987, prior to the NOX increment
baseline date. Because these emissions do not affect the allowable
PSD increment, they need not be considered in 1990 when the source
proposes another unrelated project. The emissions increase for the
heater (up to 39 tpy) would be zero in the accounting exercise.
Likewise, decreases which occurred before the baseline date was
triggered cannot be credited after the baseline date. Such
reductions are included in the baseline concentration and are not
considered in calculating PSD increment consumption.
3) A decrease is creditable only to the extent that it is
"federally-enforceable" from the moment that the actual construction
begins on the proposed modification to the source. The decrease
A.47
-------
DRAFT
OCTOBER 1990
must occur before the proposed emissions increase occurs. An
increase occurs when the emissions unit on which construction
occurred becomes operational and begins to emit a particular
pollutant. Any replacement unit that requires shakedown becomes
operational only after a reasonable shakedown period not to exceed
180 days.
4) A decrease is creditable only to the extent that it has the same
health and welfare significance as the proposed increase from the
source.
5) A source cannot take credit for a decrease that it has had to
make, or will have to make, in order to bring an emissions unit into
compliance.
6) A source cannot take credit for an emissions reduction from
potential emissions from an emissions unit which was permitted but
never built or operated.
Step 5 Determine, on a pollutant-by-pollutant basis, the amount of each
contemporaneous and creditable emissions increase and decrease.
An emissions increase is the amount by which the new level of
"actual emissions" at the emissions unit exceeds the old level. The
old level of "actual emissions" is that which prevailed just prior
(i.e., prior 2 year average) to the physical or operational change
at that unit which caused the increase. The new level is that which
prevails just after the change. In most cases, the old level is
calculated from the unit's actual operating data from a 2 year
period which directly preceded the physical change. The new "actual
emissions" level us the lower of the unit's "potential" or
"allowable" emissions after the change. In other words, a
contemporaneous emission increase is calculated as the positive
difference between an emissions unit's potential to emit just after
a physical or operation change at that unit (not the unit's current
actual emissions) and the unit's actual emissions just prior to the
change.
An emissions decrease is the amount by which the old level of actual
emissions or the old level of allowable emissions, whichever is
lower, exceeds the new level of "actual" emissions. Like emissions
increases, the old level is calculated from the unit's actual
operating data from a 2 year period which preceded the decrease, and
the new emissions level will be the lower of the unit's "potential"
or "allowable" emissions after the change.
A.48
-------
DRAFT
OCTOBER 1990
Figure A-2 shows a example of how old and new actual 502 emissions
levels are established for an existing emissions unit at a source.
The applicant met with the reviewing agency in January 1988, proposing
to commence construction on a 'new emissions unit in mid-1988. The
contemporaneous time frame in this case is from mid-1983 (using EPA's
5-year definition) to the expected date of the new boiler start-up,
about January 1990.
In mid-1984 an existing boiler switched to a low sulfur fuel oil.
The applicant wishes to use the fuel switch as a netting credit. The
time period for establishing the old 502 emissions level for the fuel
switch is the 2 year period preceding the change [mid-1982 to mid-
1984, when emissions were 600 tpy (mid-1982 through mid-1983) and 500
tpy (mid-1982 through mid-1983)]. The new 502 emissions level, 300
tpy, is established by the new allowable emissions level (which will
be made federally-enforceable). The old level of emissions is 550 tpy
(the average of 600 tpy and 500 tpy). Thus, if this is the only
existing 502 emissions unit at the source, a decrease of 250 tpy 502
emissions (550 tpy minus 300 tpy) is creditable towards the emissions
proposed for the new boiler. This example assumes that the reduction
meets all other applicable criteria for a creditable emissions
decrease.
Step 6 Sum all contemporaneous and creditable increases and decreases with
the increase from the proposed modification to determine if a
significant net emissions increase will occur.
The proposed project is subject to PSD review for each regulated
pollutant for which the sum of all creditable emissions increases
and decreases results in a significant net emissions increase.
If available, the applicant may consider proposing additional
prospective and creditable emissions reductions sufficient to
provide for a less than significant net emissions increase at the
source and thus avoid PSD review. These reductions can be achieved
through either application of emissions controls or placing
restrictions on the operation of existing emissions units. These
additional reductions would be added to the sum of all other
creditable increases and decreases. As with all contemporaneous
emissions reductions, these additional decreases must be based on
actual emissions changes, federally-enforceable prior to the
commencement of construction and occur before the new unit begins
operation. They must also affect the allowable PSD increment, where
applicable.
A.49
-------
(5
03
03
0.
CO
CO
I
CO
o
CO
800 - r- "Old" allowable emissions: 700 tpy
600 -
400 -
200 -
0 -
rDate of fuel switch
Representative "old" actual emissions level: 550 tpy
(average actual emissions for mid-82 to mid-84)
Creditable
contemporaneous
emissions
decrease: 250 tpy
i i
1980 1981
I
T
"New" federally enforceable
allowable emissions: 300 tpy
Construction to
commence on
proposed change
— Emissions increase
from proposed change
1985 1986 1987
u Date of fuel switch
Contemporaneous time frame
1988 1989
I
Allowable emissions from the boiler
Date 5 years prior to the construction of the proposed change
Actual emissions from the boiler
Actual average emissions from the boiler for
the two years proir to the fuel switch in mid 1984 _
to -n
U9
0 —I
Figure A-2. Establishing "Old" and "New" Representative Actual SO? Emissions
-------
DRAFT
OCTOBER 1990
III.B.6. NETTING EXAMPLE
An existing source has informed the local air pollution control agency
that they are planning to construct a new emissions unit "G". The existing
source is a major source and the construction of unit G will constitute a
modification to the source. Unit G will be capable of emitting 80 tons per
year (tpy) of the pollutant after installation of controls. The PSD
significant emissions level for the pollutant in question is 40 tpy. Existing
emissions units "A" and "B" at the source are presently permitted at 150 tpy
each. The applicant has proposed to limit the operation of units A and B, in
order to net out of PSD review, to 7056 hours per year (42 weeks) by accepting
federally-enforceable conditions. The applicant has calculated that there
will be an emissions reduction of -29.2 tpy [150 - 150x(7056/8760)] per unit
for a total reduction of 58.4 tpy. Thus, the net emissions increase, as
calculated by the applicant, will be +21.6 tpy (80-58.36). The applicant
proposes to net out of PSD review citing the +21.6 tpy increase as less than
the applicable 40 tpy PSD significance level for the pollutant.
The reviewing agency informed the source that 1) the emissions reductions
being claimed from units A and B must be based on the prior actual emissions,
not their allowable emissions and (2) because the increase from the
modification will be greater than significant, all contemporaneous changes
must be accounted for (not just proposed decreases) in order to determine the
net emission change at the source.
To verify if, indeed, the source will be able to net out of PSD review,
the reviewing agency requested information on the other emissions points at
the source, including their actual monthly emissions. For illustrative
purposes, the actual annual emissions of the pollutant in question from the
existing emissions points (in this example all emissions points are associated
with an emissions unit) are given as follows:
A.51
-------
DRAFT
OCTOBER 1990
Actual Emissions ftov)
Unit A
70
75
80
110
115
105
90
Unit B
130
130
150
90
85
75
90
Unit C
60
75
65
0
0
0
0
Unit D
85
75
80
0
0
0
0
Unit E
50
60
65
70
75
65
60
Unit F
0
0
0
0
75
70
65
Year
1983
1984
1985
1986
1987
1988
1989
The applicant's response indicates that units A and B will not be
physically modified. However, the information does show that the modification
will result in the removal of a bottleneck at the plant and that the proposed
modification will result in an increase in the operation of these units.
The PSD baseline for the pollutant was triggered in 1978. The history of
the emissions units at the source is as follows:
Emissions
Unit(s) History
A and B Built in 1972 and still operational
C and D Built in 1972 and retired from operation 01/86
E Built in 1972 and still operational
F PSD permitted unit; construction commenced 01/86 and the unit
became operational on 01/87
G New modification; construction scheduled to commence 01/90
and the unit is expected to be operational on 01/92
The contemporaneous period extends from 01/85 (5 years prior to 01/90,
the projected construction date of the modification) until 01/92 (the date the
emissions increase from the modification). The net emissions change at the
source can be formulated in terms of the sum of the unit-by-unit emissions
changes which are creditable and contemporaneous with the planned
A.52
-------
DRAFT
OCTOBER 1990
modification. Emission changes that are not associated with physical/
operational chagnes are not considered.
In assessing the creditable contemporaneous changes the permit agency
considered the following (all numbers are in tpy):
+ Potential to emit is used for a new unit. The new unit will receive
a federally-enforceable permit restricting allowable emissions to 80
tpy, which then becomes its potential to emit. Therefore, the new
unit represents an increase of +80.
>• Even though units A and B will not be modified, their emissions are
expected to increase as a result of the modification and the
anticipated increase must be included as part of the increase from
the proposed modification. The emissions change for these units is
based on their allowable emissions after the change minus their
current actual emissions. Current actual emissions are based on the
average emissions over the last 2 years. [Note that only the
operations of exiting units A and B are expected to be affected by
the modification.] The emissions changes at A and B are calculated
as follows:
Unit A's change = +23.3
(new allowable [150x(7056/8760)1 - old actual [(105+90)/2]>
Unit B's change = +38.3
(new allowable [150x(7056/8760)1 - old actual [(75+90)/2]}
The federally-enforceable restriction on the hours of operation for units
A and B act to reduce the amount of the emissions increase at the units
due to the modification. However, contrary to the applicant's analysis,
the restrictions did not restrict the units' emissions sufficiently to
prevent an actual emissions increase.
»> The emissions increase from unit F was permitted under PSD.
Therefore, having been "relied upon" in the issuance of a PSD permit
which is still in effect, the permitted emissions increase is not
creditable and cannot be used in the netting equation.
* The operation of unit E is not projected to be affected by the
proposed modification. It has not undergone any physical or
operational change during the contemporaneous period which would
otherwise trigger a creditable emissions change at the unit.
A.53
-------
DRAFT
OCTOBER 1990
Consequently, unit E's emissions are not considered for netting
purposes by the reviewing agency.
The retirement (a physical/operational change) of units C and D
occurred within the contemporaneous period and may provide
creditable decreases for the applicant. However, if the retirement
of the units was relied upon in the issuance of the PSD permit for
unit f (e.g, if the emissions of units C or D were modeled at zero
in the PSD application) then the reductions would not be creditable.
If they were not modeled as retired (zero emissions), then the
reduction would be available as an emissions reduction. The
reduction credit would be based on the last 2 years of actual data
prior to retirement. As with all reductions, to be creditable the
retirement of the units must be made federally-enforceable prior to
construction of the modification to and start-up of the source.
Upon checking the PSD permit application for unit F, the reviewing
agency determined that units C and D were not considered retired
and their emissions were included in the ambient impact analysis for
unit F. Consequently, the emissions reduction from the retirement
of unit C and D (should the reductions be made federally-
enforceable) was determined as followed:
Unit C's change = -70
(its new allowable [0] - its old actual [(75+65)/2]}
Unit D's change = -77.5
(its new allowable [0] - its old actual [(75+80)/2]>
The netting transaction would not cause or contribute to a violation
of the applicable PSD increment or ambient standards.
The applicant, however, is only willing to accept federally-enforceable
conditions on the retirement of unit C. Unit D is to be kept as a standby
unit and the applicant is unwilling to have its potential operation limited.
Consequently, the reduction in emissions at unit D is not creditable.
The net contemporaneous emissions change at the source is calculated by
the reviewing agency as follows:
A.54
-------
DRAFT
OCTOBER 1990
Emissions Change ftpvl
+80.0 increase from unit G.
+23.3 increase at A from modification at source.
+38.8 increase at B from modification at source.
-70.0 creditable decrease from retirement of unit C
+72.1 total contemporaneous net emissions increase at the source.
The +72.1 tpy net increase is greater than the +40 tpy PSD significance level;
consequently the proposed modification is subject to PSD review for that
pollutant.
If the applicant is willing to agree to federally-enforceable conditions
limiting the allowable emissions from unit D (but not necessarily requiring
the unit's permanent retirement), a sufficient reduction may be available to
net unit G out of a PSD review. For example, the applicant could agree to
accept federally-enforceable conditions limiting the operation of unit D to
672 hours a year (4 weeks), which (for illustrative purposes) equates to an
allowable emissions of 15 tpy. The creditable reduction from the unit D would
then amount to -62.5 tpy (-77.5 +15). This brings the total contemporaneous
net emissions change for the proposed modification to +9.6 tpy (+72.1 - 62.5).
The construction of Unit G would then not be considered a major modification
subject to PSD review. It is important to note, however, that if unit D is
permanently taken out of service after January 1991 and had not operated in
the interim, the source would not be allowed an emissions reduction credit
because there would have been no actual emissions decrease during the
contemporaneous period. In addition, if the source later requests removal of
restrictions on units which allowed unit G to net out of review, unit G then
becomes subject to PSD review as though construction had not yet commenced.
A.55
-------
DRAFT
OCTOBER 1990
IV. GENERAL EXEMPTIONS
IV.A. SOURCES AND MODIFICATIONS AFTER AUGUST 7,1980
Certain sources may be exempted from PSD review or certain PSD
requirements. Nonprofit health or educational sources that would otherwise be
subject to PSD review can be exempted if requested by the Governor of the
State in which they are located. A portable, major stationary source that has
previously received a PSD permit and is to be relocated is exempt from a
second PSD review if (1) emissions at the new location will not exceed
previously allowed emission rates, (2) the emissions at the new location are
temporary, and (3) the source will not, because of its new location, adversely
affect a Class I area or contribute to any known increment or national ambient
air quality standard (NAAQS) violation. However, the source must provide
reasonable advance notice to the reviewing authority.
IV.B. SOURCES CONSTRUCTED PRIOR TO AUGUST 7,1980
The 1980 PSD regulations do not apply to certain sources affected by
previous PSD regulations. For example, sources for which construction began
before August 7, 1977 are exempt from the 1980 PSD regulations and are instead
reviewed for applicability under the PSD regulations as they existed before
August 7, 1977. Several exemptions also exist for sources for which
construction began after August 7, 1977, but before the August 7, 1980
promulgation of the PSD regulations (45 FR 52676). These exemptions and the
criteria associated nonapplicability are detailed in paragraph (i) of
40 CFR 52.21.
A.56
-------
DRAFT
OCTOBER 1990
CHAPTER B
BEST AVAILABLE CONTROL TECHNOLOGY
I. INTRODUCTION
Any major stationary source or major modification subject to PSD must
conduct an analysis to ensure the application of best available control
technology (BACT). The requirement to conduct a BACT analysis and
determination is set forth in section 165(a)(4) of the Clean Air Act (Act), in
federal regulations at 40 CFR 52.21(j), in regulations setting forth the
requirements for State implementation plan approval of a State PSD program at
40 CFR 51.166(j), and in the SIP's of the various States at 40 CFR Part 52,
Subpart A - Subpart FFF. The BACT requirement is defined as:
"an emissions limitation (including a visible emission standard)
based on the maximum degree of reduction for each pollutant
subject to regulation under the Clean Air Act which would be
emitted from any proposed major stationary source or major
modification which the Administrator, on a case-by-case basis,
taking into account energy, environmental, and economic impacts
and other costs, determines is achievable for such source or
modification through application of production processes or
available methods, systems, and techniques, including fuel
cleaning or treatment or innovative fuel combustion techniques for
control of such pollutant. In no event shall application of best
available control technology result in emissions of any pollutant
which would exceed the emissions allowed by any applicable
standard under 40 CFR Parts 60 and 61. If the Administrator
determines that technological or economic limitations on the
application of measurement methodology to a particular emissions
unit would make the imposition of an emissions standard
infeasible, a design, equipment, work practice, operational
standard, or combination thereof, may be prescribed instead to
satisfy the requirement for the application of best available
control technology. Such standard shall, to the degree possible,
set forth the emissions reduction achievable by implementation of
such design, equipment, work practice or operation, and shall
provide for compliance by means which achieve equivalent results."
During each BACT analysis, which is done on a case-by-case basis, the
reviewing authority evaluates the energy, environmental, economic and other
B.I
-------
DRAFT
OCTOBER 1990
costs associated with each alternative technology, and the benefit of reduced
emissions that the technology would bring. The reviewing authority then
specifies an emissions limitation for the source that reflects the maximum
degree of reduction achievable for each subject pollutant regulated under the
Act. In no event can a technology be recommended which would not meet any
applicable standard of performance under 40 CFR Parts 60 (New Source
Performance Standards) and 61 (National Emission Standards for Hazardous Air
Pollutants).
In addition, if the reviewing authority determines that there is no
economically reasonable or technologically feasible way to accurately measure
the emissions, and hence to impose an enforceable emissions standard, it may
require the source to use design, alternative equipment, work practices or
operational standards to reduce emissions of the pollutant to the maximum
extent.
On December 1, 1987, the EPA Assistant Administrator for Air and
Radiation issued a memorandum that implemented certain program initiatives
designed to improve the effectiveness of the NSR programs within the confines
of existing regulations and state implementation plans. Among these was the
"top-down" method for determining best available control technology (BACT).
In brief, the top-down process provides that all available control
technologies be ranked in descending order of control effectiveness. The PSD
applicant first examines the most stringent--or "top"--alternative. That
alternative is established as BACT unless the applicant demonstrates, and the
permitting authority in its informed judgment agrees, that technical
considerations, or energy, environmental, or economic impacts justify a
conclusion that the most stringent technology is not "achievable" in that
case. If the most stringent technology is eliminated in this fashion, then
the next most stringent alternative is considered, and so on.
B.2
-------
DRAFT
OCTOBER 1990
The purpose of this chapter is to provide a detailed description of the
top-down method in order to assist permitting authorities and PSD applicants
in conducting BACT analyses.
B.3
-------
DRAFT
OCTOBER 1990
II. BACT APPLICABILITY
The BACT requirement applies to each individual new or modified affected
emissions unit and pollutant emitting activity at which a net emissions
increase would occur. Individual BACT determinations are performed for each
pollutant subject to a PSD review emitted from the same emission unit.
Consequently, the BACT determination must separately address, for each
regulated pollutant with a significant emissions increase at the source, air
pollution controls for each emissions unit or pollutant emitting activity
subject to review.
B.4
-------
DRAFT
OCTOBER 1990
III. A STEP BY STEP SUMMARY OF THE TOP-DOWN PROCESS
Table B-l shows the five basic steps of the top-down procedure, including
some of the key elements associated with each of the individual steps. A
brief description of each step follows.
III.A. STEP 1-IDENTIFY ALL CONTROL TECHNOLOGIES.
The first step in a "top-down" analysis is to identify, for the
emissions unit in question (the term "emissions unit" should be read to mean
emissions unit, process or activity), all "available" control options.
Available control options are those air pollution control technologies or
techniques with a practical potential for application to the emissions unit
and the regulated pollutant under evaluation. Air pollution control
technologies and techniques include the application of production process or
available methods, systems, and techniques, including fuel cleaning or
treatment or innovative fuel combustion techniques for control of the affected
pollutant. This includes technologies employed outside of the United States.
As discussed later, in some circumstances inherently lower-polluting processes
are appropriate for consideration as available control alternatives. The
control alternatives should include not only existing controls for the source
category in question, but also (through technology transfer) controls applied
to similar source categories and gas streams, and innovative control
technologies. Technologies-required under lowest achievable emission rate
(LAER) determinations are available for BACT purposes and must also be
included as control alternatives and usually represent the top alternative.
In the course of the BACT analysis, one or more of the options may be
eliminated from consideration because they are demonstrated to be technically
infeasible or have unacceptable energy, economic, or environmental impacts on
a case-by-case (or site-specific) basis. However, at the outset, applicants
B.5
-------
DRAFT
OCTOBER 1990
TABLE B-l. - KEY STEPS IN THE "TOP-DOWN" BACT PROCESS
STEP 1: IDENTIFY ALL CONTROL TECHNOLOGIES.
LIST is comprehensive (LAER included).
STEP 2: ELIMINATE TECHNICALLY INFEASIBLE OPTIONS.
A demonstration of technical infeasibility should be clearly
documented and should show, based on physical, chemical, and
engineering principles, that technical difficulties would preclude
the successful use of the control option on the emissions unit
under review.
STEP 3: RANK REMAINING CONTROL TECHNOLOGIES BY CONTROL EFFECTIVENESS.
Should include:
control effectiveness (percent pollutant removed);
expected emission rate (tons per year);
expected emission reduction (tons per year);
energy impacts (BTU, kWh);
environmental impacts (other media and the emissions of toxic and
hazardous air emissions); and
economic impacts (total cost effectiveness, incremental cost
effectiveness).
STEP 4: EVALUATE MOST EFFECTIVE CONTROLS AND DOCUMENT RESULTS.
Case-by-case consideration of energy, environmental, and economic
impacts.
If top option is not selected as BACT, evaluate next most
effective control option.
STEP 5: SELECT BACT
Most effective option not rejected is BACT.
B.6
-------
DRAFT
OCTOBER 1990
should initially identify all control options with potential application to
the emissions unit under review.
III.B. STEP 2--ELIMINATE TECHNICALLY INFEASIBLE OPTIONS.
In the second step, the technical feasibility of the control options
identified in step one is evaluated with respect to the source-specific (or
emissions unit-specific) factors. A demonstration of technical infeasibility
should be clearly documented and should show, based on physical, chemical, and
engineering principles, that technical difficulties would preclude the
successful use of the control option on the emissions unit under review.
Technically infeasible control options are then eliminated from further
consideration in the BACT analysis.
For example, in cases where the level of control in a permit is not
expected to be achieved in practice (e.g., a source has received a permit but
the project was canceled, or every operating source at that permitted level
has been physically unable to achieve compliance with the limit), and
supporting documentation showing why such limits are not technically feasible
is provided, the level of control (but not necessarily the technology) may be
eliminated from further consideration. However, a permit requiring the
application of a certain technology or emission limit to be achieved for such
technology usually is sufficient justification to assume the technical
feasibility of that technology or emission limit.
III.C. STEP 3--RANK REMAINING CONTROL TECHNOLOGIES BY CONTROL EFFECTIVENESS.
In step 3, all remaining control alternatives not eliminated in step 2
are ranked and then listed in order of over all control effectiveness for the
pollutant under review, with the most effective control alternative at the
top. A list should be prepared for each pollutant and for each emissions unit
(or grouping of similar units) subject to a BACT analysis. The list should
present the array of control technology alternatives and should include the
following types of information:
B.7
-------
DRAFT
OCTOBER 1990
• control efficiencies (percent pollutant removed);
• expected emission rate (tons per year, pounds per hour);
• expected emissions reduction (tons per year);
• economic impacts (cost effectiveness);
• environmental impacts [includes any significant or unusual
other media impacts (e.g., water or solid waste), and, at a
minimum, the impact of each control alternative on emissions of
toxic or hazardous air contaminants];
• energy impacts.
However, an applicant proposing the top control alternative need not
provide cost and other detailed information in regard to other control
options. In such cases the applicant should document, to the satisfaction of
the review agency and for the public record, that the control option chosen
is, indeed, the top, and review for collateral environmental impacts.
III.D. STEP 4--EVALUATE MOST EFFECTIVE CONTROLS AND DOCUMENT RESULTS.
After the identification of available and technically feasible control
technology options, the energy, environmental, and economic impacts are
considered to arrive at the final level of control. At this point the
analysis presents the associated impacts of the control option in the listing.
For each option the applicant is responsible for presenting an objective
evaluation of each impact. Both beneficial and adverse impacts should be
discussed and, where possible, quantified. In general, the BACT analysis
should focus on the direct impact of the control alternative.
If the applicant accepts the top alternative in the listing as BACT, the
applicant proceeds to consider whether impacts of unregulated air pollutants
or impacts in other media would justify selection of an alternative control
option. If there are no outstanding issues regarding collateral environmental
impacts, the analysis is ended and the results proposed as BACT. In the event
that the top candidate is shown to be inappropriate, due to energy,
B.8
-------
DRAFT
OCTOBER 1990
environmental, or economic impacts, the rationale for this finding should be
documented for the public record. Then the next most stringent alternative in
the listing becomes the new control candidate and is similarly evaluated.
This process continues until the technology under consideration cannot be
eliminated by any source-specific environmental, energy, or economic impacts
which demonstrate that alternative to be inappropriate as BACT.
III.E. STEP 5--SELECT BACT
The most effective control option not eliminated in step 4 is proposed as
BACT for the pollutant and emission unit under review.
B.9
-------
DRAFT
OCTOBER 1990
IV. TOP-DOWN ANALYSIS DETAILED PROCEDURE
IV.A. IDENTIFY ALTERNATIVE EMISSION CONTROL TECHNIQUES (STEP 1)
The objective in step 1 is to identify all control options with potential
application to the source and pollutant under evaluation. Later, one or more
of these options may be eliminated from consideration because they are
determined to be technically infeasible or to have unacceptable energy,
environmental or economic impacts.
Each new or modified emission unit (or logical grouping of new or
modified emission units) subject to PSD is required to undergo BACT review.
BACT decisions should be made on the information presented in the BACT
analysis, including the degree to which effective control alternatives were
identified and evaluated. Potentially applicable control alternatives can be
categorized in three ways.
• Inherently Lower-Emitting Processes/Practices, including
the use of materials and production processes and work
practices that prevent emissions and result in lower
"production-specific" emissions; and
• Add-on Controls, such as scrubbers, fabric filters, thermal
oxidizers and other devices that control and reduce emissions
after they are produced.
• Combinations of Inherently Lower Emitting Processes and Add-on
Controls. For example, the application of combustion and post-
combustion controls to reduce NOx emissions at a gas-fired
turbine.
The top-down BACT analysis should consider potentially applicable control
techniques from all three categories. Lower-polluting processes should be
considered based on demonstrations made on the basis of manufacturing
identical or similar products from identical or similar raw materials or
fuels. Add-on controls, on the other hand, should be considered based on the
physical and chemical characteristics of the pollutant-bearing emission
stream. Thus, candidate add-on controls may have been applied to a broad
B.10
-------
DRAFT
OCTOBER 1990
range of emission unit types that are similar, insofar as emissions
characteristics, to the emissions unit undergoing BACT review.
IV.A.I. DEMONSTRATED AND TRANSFERABLE TECHNOLOGIES
Applicants are expected to identify all demonstrated and potentially
applicable control technology alternatives. Information sources to consider
include:
EPA's BACT/LAER Clearinghouse and Control Technology Center;
• Best Available Control Technology Guideline - South Coast Air
Quality Management District;
• control technology vendors;
Federal/State/Local new source review permits and associated
inspection/performance test reports;
• environmental consultants;
• technical journals, reports and newsletters (e.g., Journal of
Air and Waste Management Association and the Mclvaine reports),
air pollution control seminars; and
• EPA's New Source Review (NSR) bulletin board.
The applicant is responsible to compile appropriate information from
available information sources, including any sources specified as necessary by
the permit agency. The permit agency should review the background search and
resulting list of control alternatives presented by the applicant to check
that it is complete and comprehensive.
In identifying control technologies, the applicant needs to survey the
range of potentially available control options. Opportunities for technology
transfer lie where a control technology has been applied at source categories
other than the source under consideration. Such opportunities should be
identified. Also, technologies in application outside the United States to
the extent that the technologies have been successfully demonstrated in
B.ll
-------
DRAFT
OCTOBER 1990
practice on full scale operations. Technologies which have not yet been
applied to (or permitted for) full scale operations need not be considered
available; an applicant should be able to purchase or construct a process or
control device that has already been demonstrated in practice.
To satisfy the legislative requirements of BACT, EPA believes that the
applicant must focus on technologies with a demonstrated potential to achieve
the highest levels of control. For example, control options incapable of
meeting an applicable New Source Performance Standard (NSPS) or State
Implementation Plan (SIP) limit would not meet the definition of BACT under
any circumstances. The applicant does not need to consider them in the BACT
analysis.
The fact that a NSPS for a source category does not require a certain
level of control or particular control technology does not preclude its
consideration for control in the top-down BACT analysis. For example, post
combustion NOx controls are not required under the Subpart GG of the NSPS for
Stationary Gas Turbines. However, such controls must still be considered
available technologies for the BACT selection process and be considered in the
BACT analysis. An NSPS simply defines the minimal level of control to be
considered in the BACT analysis. The fact that a more stringent technology
was not selected for a NSPS (or that a pollutant is not regulated by an NSPS)
does not exclude that control alternative or technology as a BACT candidate.
When developing a list of possible BACT alternatives, the only reason for
comparing control options to an NSPS is to determine whether the control
option would result in an emissions level less stringent than the NSPS. If
so, the option is unacceptable.
IV.A.2. INNOVATIVE TECHNOLOGIES
Although not required in step 1, the applicant may also evaluate and
propose innovative technologies as BACT. To be considered innovative, a
control technique must meet the provisions of 40 CFR 52.21(b)(19) or, where
appropriate, the applicable SIP definition. In essence, if a developing
B.12
-------
DRAFT
OCTOBER 1990
technology has the potential to achieve a more stringent emissions level than
otherwise would constitute BACT or the same level at a lower cost, it may be
proposed as an innovative control technology. Innovative technologies are
distinguished from technology transfer BACT candidates in that an innovative
technology is still under development and has not been demonstrated in a
commercial application on identical or similar emission units. In certain
instances, the distinction between innovative and transferable technology may
not be straightforward. In these cases, it is recommended that the permit
agency consult with EPA prior to proceeding with the issuance of an innovative
control technology waiver.
In the past, only a limited number of innovative control technology
waivers for a specific control technology have been approved. As a practical
matter, if a waiver has been granted to a similar source for the same
technology, granting of additional waivers to similar sources is highly
unlikely since the subsequent applicants are no longer "innovative."
IV.A.3. CONSIDERATION OF INHERENTLY LOWER POLLUTING PROCESSES/PRACTICES
Historically, EPA has not considered the BACT requirement as a means to
redefine the design of the source when considering available control
alternatives. For example, applicants proposing to construct a coal-fired
electric generator, have not been required by EPA as part of a BACT analysis
to consider building a natural gas-fired electric turbine although the turbine
may be inherently less polluting per unit product (in this case electricity).
However, this is an aspect of the PSD permitting process in which states have
the discretion to engage in a broader analysis if they so desire. Thus, a gas
turbine normally would not be included in the list of control alternatives for
a coal-fired boiler. However, there may be instances where, in the permit
authority's judgment, the consideration of alternative production processes is
warranted and appropriate for consideration in the BACT analysis. A
production process is defined in terms of its physical and chemical unit
operations used to produce the desired product from a specified set of raw
B.13
-------
DRAFT
OCTOBER 1990
materials. In such cases, the permit agency may require the applicant to
include the inherently lower-polluting process in the list of BACT candidates.
In some cases, a given production process or emissions unit can be made
to be inherently less polluting (e.g; the use of water-based versus solvent
based paints in a coating operation or a coal-fired boiler designed to have a
low emission factor for NOx). In such cases the ability of design
considerations to make the process inherently less polluting must be
considered as a control alternative for the source. Inherently lower-
polluting processes/practice are usually more environmentally effective
because lower amounts of solid wastes and waste water are generated when
compared with add-on controls. These factors are considered in the cost,
energy and environmental impacts analyses in step 4 to determine the
appropriateness of the additional add-on option.
Combinations of inherently lower-polluting processes/practices (or a
process made to be inherently less polluting) and add-on controls are likely
to yield more effective means of emissions control than either approach alone.
Therefore, the option to utilize an inherently lower-polluting process does
not, in and of itself, mean that no additional add-on controls need be
included in the BACT analysis. These combinations should be identified in
step 1 of the top down process for evaluation in subsequent steps.
IV.A.4- EXAMPLE
The process of identifying control technology alternatives (step 1 in the
top-down BACT process) is illustrated in the following hypothetical example.
B.14
-------
DRAFT
OCTOBER 1990
Description of Source
A PSD applicant proposes to install automated surface coating process
equipment consisting of a dip-tank priming stage followed by a two-step spray
application and bake-on enamel finish coat. The product is a specialized
electronics component (resistor) with strict resistance property
specifications that restrict the types of coatings that may be employed.
List of Control Options
The source is not covered by an applicable NSPS. A review of the
BACT/LAER Clearinghouse and other appropriate references indicates the
following control options may be applicable:
Option fl: water-based primer and finish coat;
[The water-based coatings have never been used in applications
similar to this.]
Option #2: low-VOC solvent/high solids coating for primer and
finish coat;
[The high solids/low VOC solvent coatings have recently been
applied with success with similar products (e.g., other types of
electrical components).]
Option #3: electrostatic spray application to enhance coating
transfer efficiency; and
[Electrostatically enhanced coating application has been applied
elsewhere on a clearly similar operation.]
Option 14: emissions capture with add-on control via incineration
or carbon adsorber equipment.
[The VOC capture and control option (incineration or carbon
adsorber) has been used in many cases involving the coating of
different products and the emission stream characteristics are
similar to the proposed resistor coating process and is identified
as an option available through technology transfer.]
B.15
-------
DRAFT
OCTOBER 1990
Since the low-solvent coating, electrostatically enhanced application,
and ventilation with add-on control options may be considered for use in
combination to achieve greater emissions reduction efficiency, a total of
eight control options are eligible for further consideration. The options
include each of the four options listed above and the following four
combinations of techniques:
Option 15: low-solvent coating with electrostatic applications
without ventilation and add-on controls;
Option #6: low-solvent coating without electrostatic applications
with ventilation and add-on controls;
Option #7: electrostatic application with add-on control; and
Option #8: a combination of all three technologies.
A "no control" option also was identified but eliminated because the
applicant's State regulations require at least a 75 percent reduction in VOC
emissions for a source of this size. Because "no control" would not meet the
State regulations it could not be BACT and, therefore, was not listed for
consideration in the BACT analysis.
Summary of Key Points
The example illustrates several key guidelines for identifying control
options. These include:
• All available control techniques must be considered in the BACT
analysis.
• Technology transfer must be considered in identifying control
options. The fact that a control option has never been applied
to process emission units similar or identical to that proposed
does not mean it can be ignored in the BACT analysis if the
potential for its application exists.
B.16
-------
DRAFT
OCTOBER 1990
Combinations of techniques should be considered to the extent
they result in more effective means of achieving stringent
emissions levels represented by the "top" alternative,
particularly if the "top" alternative is eliminated.
IV.B. TECHNICAL FEASIBILITY ANALYSIS (STEP 2)
In step 2, the technical feasibility of the control options identified
in step 1 is evaluated. This step should be straightforward for control
technologies that are demonstrated--if the control technology has been
installed and operated successfully on the type of source under review, it is
demonstrated and it is technically feasible. For control technologies that
are not demonstrated in the sense indicated above, the analysis is somewhat
more involved.
Two key concepts are important in determining whether an undemonstrated
technology is feasible: "availability" and "applicability." As explained in
more detail below, a technology is considered "available" if it can be
obtained by the applicant through commercial channels or is otherwise
available within the common sense meaning of the term. An available
technology is "applicable" if it can reasonably be installed and operated on
the source type under consideration. A technology that is available and
applicable is technically feasible.
Availability in this context is further explained using the following
process commonly used for bringing a control technology concept to reality as
a commercial product:
• concept stage;
• research and patenting;
• bench scale or laboratory testing;
• pilot scale testing;
B.17
-------
DRAFT
OCTOBER 1990
• licensing and commercial demonstration; and
• commercial sales.
A control technique is considered available, within the context presented
above, if it has reached the licensing and commercial sales stage of
development. A source would not be required to experience extended time
delays or resource penalties to allow research to be conducted on a new
technique. Neither is it expected that an applicant would be required to
experience extended trials to learn how to apply a technology on a totally
new and dissimilar source type. Consequently, technologies in the pilot scale
testing stages of development would not be considered available for BACT
review. An exception would be if the technology were proposed and permitted
under the qualifications of an innovative control device consistent with the
provisions of 40 CFR 52.21(v) or, where appropriate, the applicable SIP. In
general, if a control option is commercially available, it falls within the
options to be identified in step 1.
Commercial availability by itself, however, is not necessarily
sufficient basis for concluding a technology to be applicable and therefore
technically feasible. Technical feasibility, as determined in Step 2, also
means a control option may reasonably be deployed on or "applicable" to the
source type under consideration.
Technical judgment on the part of the applicant and the review authority
is to be exercised in determining whether a control alternative is applicable
to the source type under consideration. In general, a commercially available
control option will be presumed applicable if it has been or is soon to be
deployed (e.g., is specified in a permit) on the same or a similar source
type. Absent a showing of this type, technical feasibility would be based on
examination of the physical and chemical characteristics of the pollutant-
bearing gas stream and comparison to the gas stream characteristics of the
source types to which the technology had been applied previously. Deployment
of the control technology on an existing source with similar gas stream
B.18
-------
DRAFT
OC10BER 1990
characteristics is generally sufficient basis for concluding technical
feasibility barring a demonstration to the contrary.
For process-type control alternatives the decision of whether or not it
is applicable to the source in question would have to be based on an
assessment of the similarities and differences between the proposed source and
other sources to which the process technique had been applied previously.
Absent an explanation of unusual circumstances by the applicant showing why a
particular process cannot be used on the proposed source the review authority
may presume it is technically feasible.
In practice, decisions about technical feasibility are within the purview
of the review authority. Further, a presumption of technical feasibility may
be made by the review authority based solely on technology transfer. For
example, in the case of add-on controls, decisions of this type would be made
by comparing the physical and chemical characteristics of the exhaust gas
stream from the unit under review to those of the unit from which the
technology is to be transferred. Unless significant differences between
source types exist that are pertinent to the successful operation of the
control device, the control option is presumed to be technically feasible
unless the source can present information to the contrary.
Within the context of the top-down procedure, an applicant addresses the
issue of technical feasibility in asserting that a control option identified
in Step 1 is technically infeasible. In this instance, the applicant should
make a factual demonstration of infeasibility based on commercial
unavailability and/or unusual circumstances which exist with application of
the control to the applicant's emission units. Generally, such a
demonstration would involve an evaluation of the pollutant-bearing gas stream
characteristics and the capabilities of the technology. Also a showing of
unresolvable technical difficulty with applying the control would constitute a
showing of technical infeasibility (e.g., size of the unit, location of the
proposed site, and operating problems related to specific circumstances of the
source). Where the resolution of technical difficulties is a matter of cost,
B.19
-------
DRAFT
OCTOBER 1990
the applicant should consider the technology as technically feasible. The
economic feasibility of a control alternative is reviewed in the economic
impacts portion of the BACT selection process.
A demonstration of technical infeasibility is based on a technical
assessment considering physical, chemical and engineering principles, and/or
empirical data showing that the technology would not work on the emissions
unit under review, or that unresolvable technical difficulties would preclude
the successful deployment of the technique. Physical modifications needed to
resolve technical obstacles do not in and of themselves provide a
justification for eliminating the control technique on the basis of technical
infeasibility. However, the cost of such modifications can be considered in
estimating cost and economic impacts which, in turn, may form the basis for
eliminating a control technology (see later discussion at V.D.2).
Vendor guarantees may provide an indication of commercial availability
and the technical feasibility of a control technique and could contribute to a
determination of technical feasibility or technical infeasibility, depending
on circumstances. However, EPA does not consider a vendor guarantee alone to
be sufficient justification that a control option will work. Conversely, lack
of a vendor guarantee by itself does not present sufficient justification that
a control option or an emissions limit is technically infeasible. Generally,
decisions about technical feasibility will be based on chemical and
engineering analyses (as discussed above) in conjunction with information
about vendor guarantees.
A possible outcome of the top-down BACT procedures discussed in this
document is the evaluation of multiple control technology alternatives which
result in essentially equivalent emissions. It is not EPA's intent to
encourage evaluation of unnecessarily large numbers of control alternatives
for every emissions unit. Consequently, judgment should be used in deciding
what alternatives will be evaluated in detail in the impacts analysis (Step 4)
of the top-down procedure discussed in a later section. For example, if two
or more control techniques result in control levels that are essentially
B.20
-------
DRAFT
OCTOBER 1990
identical considering the uncertainties of emissions factors and other
parameters pertinent to estimating performance, the source may wish to point
this out and make a case for evaluation of only the less costly of these
options. The scope of the BACT analysis should be narrowed in this way only
if there is a negligible difference in emissions and collateral environmental
impacts between control alternatives. Such cases should be discussed with the
reviewing agency before a control alternative is dismissed at this point in
the BACT analysis due to such considerations.
It is encouraged that judgments of this type be discussed during a
preapplication meeting between the applicant and the review authority. In
this way, the applicant can be better assured that the analysis to be
conducted will meet BACT requirements. The appropriate time to hold such a
meeting during the analysis is following the completion of the control
hierarchy discussed in the next section.
Summary of Key Points
In summary, important points to remember in assessing technical
feasibility of control alternatives include:
A control technology that is "demonstrated" for a
given type or class of sources is assumed to be
technically feasible unless source-specific factors
exist and are documented to justify technical
infeasibility.
Technical feasibility of technology transfer control
candidates generally is assessed based on an
evaluation of pollutant-bearing gas stream
characteristics for the proposed source and other
source types to which the control had been applied
previously.
Innovative controls that have not been demonstrated on
any source type similar to the proposed source need
not be considered in the BACT analysis.
B.21
-------
DRAFT
OCTOBER 1990
The applicant is responsible for providing a basis for
assessing technical feasibility or infeasibility and
the review authority is responsible for the decision
on what is and is not technically feasible.
IV.C. RANKING THE TECHNICALLY FEASIBLE ALTERNATIVES TO ESTABLISH A CONTROL
HIERARCHY (STEP 3)
Step 3 involves ranking all the technically feasible control alternatives
which have been previously identified in Step 2. For the regulated pollutant
and emissions unit under review, the control alternatives are ranked-ordered
from the most to the least effective in terms of emission reduction potential.
Later, once the control technology is determined, the focus shifts to the
specific limits to be met by the source.
Two key issues that must be addressed in this process include:
• What common units should be used to compare emissions
performance levels among options?
• How should control techniques that can operate over a wide
range of emission performance levels (e.g., scrubbers, etc.)
be considered in the analysis?
IV.C.I. CHOICE OF UNITS OF EMISSIONS PERFORMANCE TO COMPARE LEVELS AMONGST
CONTROL OPTIONS
In general, this issue arises when comparing inherently lower-polluting
processes to one another or to add-on controls. For example, direct
comparison of powdered (and low-VOC) coatings and vapor recovery and control
systems at a metal furniture finishing operation is difficult because of the
different units of measure for their effectiveness. In such cases, it is
generally most effective to express emissions performance as an average steady
state emissions level per unit of product produced or processed. Examples
are:
B.22
-------
DRAFT
OCTOBER 1990
• pounds VOC emissions per gallons of solids applied,
• pounds PM emissions per ton of cement produced,
• pounds S02 emissions per million Btu heat input, and
• pounds S02 emissions per kilowatt of electric power produced,
Calculating annual emissions levels (tons/yr) using these units becomes
straightforward once the projected annual production or processing rates are
known. The result is an estimate of the annual pollutant emissions that the
source or emissions unit will emit. Annual "potential" emission projections
are calculated using the source's maximum design capacity and full year round
operation (8760 hours), unless the final permit is to include federally
enforceable conditions restricting the source's capacity or hours of
operation. However, emissions estimates used for the purpose of calculating
and comparing the cost effectiveness of a control option are based on a
different approach (see section V.D.2.b. COST EFFECTIVENESS).
IV.C.2. CONTROL TECHNIQUES WITH A WIDE RANGE OF EMISSIONS PERFORMANCE LEVELS
The objective of the top-down BACT analysis is to not only identify the
best control technology, but also a corresponding performance level (or in
some cases performance range) for that technology considering source-specific
factors. Many control techniques, including both add-on controls and
inherently lower polluting processes can perform at a wide range of levels.
Scrubbers, high and low efficiency electrostatic precipitators (ESPs), and
low-VOC coatings are examples of just a few. It is not the EPA's intention to
require analysis of each possible level of efficiency for a control technique,
as such an analysis would result in a large number of options. Rather, the
applicant should use the most recent regulatory decisions and performance data
for identifying the emissions performance level(s) to be evaluated in all
cases.
The EPA does not expect an applicant to necessarily accept an emission
limit as BACT solely because it was required previously of a,similar source
B.23
-------
DRAFT
OCTOBER 1990
type. While the most effective level of control must be considered in the
BACT analysis, different levels of control for a given control alternative can
be considered.1 For example, the consideration of a lower level of control
for a given technology may be warranted in cases where past decisions involved
different source types. The evaluation of an alternative control level can
also be considered where the applicant can demonstrate to the satisfaction of
the permit agency that other considerations show the need to evaluate the
control alternative at a lower level of effectiveness.
Manufacturer's data, engineering estimates and the experience of other
sources provide the basis for determining achievable limits. Consequently, in
assessing the capability of the control alternative, latitude exists to
consider any special circumstances pertinent to the specific source under
review, or regarding the prior application of the control alternative.
However, the basis for choosing the alternate level (or range) of control in
the BACT analysis must be documented in the application. In the absence of a
showing of differences between the proposed source and previously permitted
sources achieving lower emissions limits, the permit agency should conclude
that the lower emissions limit is representative for that control alternative.
In summary, when reviewing a control technology with a wide range of
emission performance levels, it is presumed that the source can achieve the
same emission reduction level as another source unless the applicant
demonstrates that there are source-specific factors or other relevant
information that provide a technical, economic, energy or environmental
justification to do otherwise. Also, a control technology that has been
1 In reviewing the BACT submittal by a source the permit agency may
determine that an applicant should consider a control technology alternative
otherwise eliminated by the applicant, if the operation of that control
technology at a lower level of control (but still higher than the next control
technology alternative) would no longer warrant the elimination of the
alternative. For example, while a scrubber operating at 98% efficiency may be
eliminated as BACT by the applicant due to source specific economic
considerations, the scrubber operating in the 90% to 95% efficiency range may
not have an adverse economic impact.
B.24
-------
DRAFT
OCTOBER 1990
eliminated as having an adverse economic impact at its highest level of
performance, may be acceptable at a lesser level of performance. For example,
this can occur when the cost effectiveness of a control technology at its
highest level of performance greatly exceeds the cost of that control
technology at a somewhat lower level (or range) of performance.
IV.C.3. ESTABLISHMENT OF THE CONTROL OPTIONS HIERARCHY
After determining the emissions performance levels (in common units) of
each control technology option identified in Step 2, a hierarchy is
established that places at the "top" the control technology option that
achieves the lowest emissions level. Each other control option is then placed
after the "top" in the hierarchy by its respective emissions performance
level, ranked from lowest emissions to highest emissions (most effective to
least effective emissions control alternative).
From the hierarchy of control alternatives the applicant should develop a
chart (or charts) displaying the control hierarchy and, where applicable,:
• expected emission rate (tons per year, pounds per hour);
• emissions performance level (e.g., percent pollutant removed,
emissions per unit product, Ib/MMbtu, ppm);
• expected emissions reduction (tons per year);
The charts should also contain columns for the following information
(Section IV.D discusses procedures for generating this information):
• economic impacts (total annualized costs, cost effectiveness,
incremental cost effectiveness);
• environmental impacts [includes any significant or unusual
other media impacts (e.g., water or solid waste), and the
relative ability of each control alternative to control
emissions of toxic or hazardous air contaminants];
• energy impacts (indicate any significant energy benefits or
disadvantages).
B.25
-------
DRAFT
OCTOBER 1990
This should be done for each pollutant and for each emissions unit (or
grouping of similar units) subject to a BACT analysis. The chart is used in
comparing the control alternatives during step 4 of the BACT selection
process. Some sample charts are displayed in Table B-2 and Table B-3.
Completed sample charts accompany the example BACT analyses provided in
section VI.
At this point, it is recommended that the applicant contact the reviewing
agency to determine whether the agency feels that any other applicable control
alternative should be evaluated or if any issues require special attention in
the BACT selection process.
IV.D. THE BACT SELECTION PROCESS (STEP 4)
After identifying and listing the available control options the next step
is the determination of the energy, environmental, and economic impacts of
each option and the selection of the final level of control. The applicant is
responsible for presenting an evaluation of each impact along with appropriate
supporting information. Consequently, both beneficial and adverse impacts
should be discussed and, where possible, quantified. In general, the BACT
analysis should focus on the direct impact of the control alternative.
Step 4 validates the suitability of the top control option in the
listing for selection as BACT, or provides clear justification why the top
candidate is inappropriate as BACT. If the applicant accepts the top
alternative in the listing as BACT from an economic and energy standpoint, the
applicant proceeds to consider whether collateral environmental impacts (e.g.,
emissions of unregulated air pollutants or impacts in other media) would
justify selection of an alternative control option. If there are no
outstanding issues regarding collateral environmental impacts, the analysis is
ended and the results proposed to the permit agency as BACT. In the event
that the top candidate is shown to be inappropriate, due to energy,
B.26
-------
DRAFT
OCTOBER 1990
TABLE B-2. SAMPLE BACT CONTROL HIERARCHY
Pollutant Technology
Range
of
control
(%)
Control
level
for BACT
analysis
(*)
Emissions
1 imit
so.
First Alternative 80-95 95
Second Alternative 80-95 90
Third Alternative 70-85 85
Fourth Alternative 40-80 75
Fifth Alternative 50-85 70
Baseline Alternative
15 ppm
30 ppm
45 ppm
75 ppm
90 ppm
B.27
-------
TABLE B-3. SAMPLE SUMMARY OF TOP-DOWN BACT IMPACT ANALYSIS RESULTS
Pollutant/
Emissions
Unit
Economic Impacts
Control alternative
Emissions
(Ib/hr.tpi)
Emissions
reduction(a)
(tpy)
Total Average Incremental
annualized Cost cost
cost(b) effectiveness(c) effectiveness(d)
($/yr) ($/ton) ($/ton)
Environmental Impacts
Toxics
impact(e)
(Yes/No)
Adverse
environmental
impacts(f)
(Yes/No)
Energy
Impacts
Incremental
increase
over
baseline(g)
(MMBtu/yr)
CO
•
ro
CO
NOx/Unit ft
HOx/Unit B
S02/Unit A
S02/Unit B
Top Alternative
Other Alternative(s)
Baseline
Top Alternative
Other Alternative(s)
Baseline
Top Alternative
Other Alternative(s)
Baseline
Top Alternative
Other Alternative(s)
Baseline
(a) Emissions reduction over baseline level.
(b) Total annualized cost (capital, direct, and indirect) of purchasing, installing, and operating the proposed control alternative. A capital recovery
factor approach using a real interest rate (i.e., absent inflation) is used to express capital costs in present-day annual costs.
(c) Average Cost Effectiveness is total annualized cost for the control option divided by the emissions reductions resulting from the option.
(d) The incremental cost effectiveness is the difference in annualized cost for the control option and the next most effective control option divided by the 3
difference in emissions reduction resulting from the respective alternatives.
(e) Toxics impact means there is a toxics impact consideration for the control alternative.
(f) Adverse environmental impact means there is an adverse environmental impact consideration with the control alternative. §
(g) Energy impacts are the difference in total project energy requirements with the control alternative and the baseline expressed in equivalent millions of
Btus per year.
a
30
>
-------
DRAFT
OCTOBER 1990
environmental, or economic impacts, the rationale for this finding needs to be
fully documented for the public record. Then, the next most effective
alternative in the listing becomes the new control candidate and is similarly
evaluated. This process continues until the control technology, under
consideration cannot be eliminated by any source-specific environmental,
energy, or economic impacts which demonstrate that the alternative is
inappropriate as BACT.
The determination that a control alternative is inappropriate involves a
demonstration that circumstances exist at the source which distinguish it from
other sources where the control alternative may have been required previously,
or that argue against the transfer of technology or application of new
technology. Alternately, where a control technique has been applied to only
one or a very limited number of sources, the applicant can identify those
characteristic(s) unique to those sources that may have made the application
of the control appropriate in those case(s) but not for the source under
consideration. In showing unusual circumstances, objective factors dealing
with the control technology and its application should be the focus of the
consideration. The specifics of the situation will determine to what extent
an appropriate demonstration has been made regarding the elimination of the
more effective alternative(s) as BACT. In the absence of unusual
circumstance, the presumption is that sources within the same category are
similar in nature, and that cost and other impacts that have been borne by one
source of a given source category may be borne by another source of the same
source category.
IV.D.I. ENERGY IMPACTS ANALYSIS
Applicants should examine the energy requirements of the control
technology and determine whether the use of that technology results in any
significant or unusual energy penalties or benefits. A source may, for
example, benefit from the combustion of a concentrated gas stream rich in
volatile organic compounds; on the other hand, more often extra fuel or
electricity is required to power a control device or incinerate a dilute gas
B.29
-------
DRAFT
OCTOBER 1990
stream. If such benefits or penalties exist, they should be quantified.
Because energy penalties or benefits can usually be quantified in terms of
additional cost or income to the source, the energy impacts analysis can, in
most cases, simply be factored into the economic impacts analysis. However,
certain types of control technologies have inherent energy penalties
associated with their use. While these penalties should be quantified, so
long as they are within the normal range for the technology in question, such
penalties should not, in general, be considered adequate justification for
nonuse of that technology.
Energy impacts should consider only direct energy consumption and not
indirect energy impacts. For example, the applicant could estimate the direct
energy impacts of the control alternative in units of energy consumption at
the source ( e.g., Btu, kWh, barrels of oil, tons of coal). The energy
requirements of the control options should be shown in terms of total (and in
certain cases also incremental) energy costs per ton of pollutant removed.
These units can then be converted into dollar costs and, where appropriate,
factored into the economic analysis.
As noted earlier, indirect energy impacts (such as energy to produce raw
materials for construction of control equipment) generally are not considered.
However, if the permit authority determines, either independently or based on
a showing by the applicant, that the indirect energy impact is unusual or
significant and that the impact can be well quantified, the indirect impact
may be considered. The energy impact should still focus on the application of
the control alternative and not a concern over general energy impacts
associated with the project under review as compared to alternative projects
for which a permit is not being sought, or as compared to a pollution source
which the project under review would replace (e.g., it would be inappropriate
to argue that a cogeneration project is more efficient in the production of
electricity than the powerplant production capacity it would displace and,
therefore, should not be required to spend equivalent costs for the control of
the same pollutant).
B.30
-------
DRAFT
OCTOBER 1990
The energy impact analysis may also address concerns over the use of
locally scarce fuels. The designation of a scarce fuel may vary from region
to region, but in general a scarce fuel is one which is in short supply
locally and can be better used for alternative purposes, or one which may not
be reasonably available to the source either at the present time or in the
near future.
IV.D.2. COST/ECONOMIC IMPACTS ANALYSIS
Average and incremental cost effectiveness are the two economic criteria
that are considered in the BACT analysis. Cost effectiveness, is the dollars
per ton of pollutant emissions reduced. Incremental cost is the cost per ton
reduced and should be considered in conjunction with total average
effectiveness.
In the economic impacts analysis, primary consideration should be given
to quantifying the cost of control and not the economic situation of the
individual source. Consequently, applicants generally should not propose
elimination of control alternatives on the basis of economic parameters that
provide an indication of the affordability of a control alternative relative
to the source. BACT is required by law. Its costs are integral to the
overall cost of doing business and are not to be considered an afterthought.
Consequently, for control alternatives that have been effectively employed in
the same source category, the economic impact of such alternatives on the
particular source under review should be not nearly as pertinent to the BACT
decision making process as the average and, where appropriate, incremental
cost effectiveness of the control alternative. Thus, where a control
technology has been successfully applied to similar sources in a source
category, an applicant should concentrate on documenting significant cost
differences, if any, between the application of the control technology on
those other sources and the particular source under review.
Cost effectiveness (dollars per ton of pollutant reduced) values above
the levels experienced by other sources of the same type and pollutant, are
B.31
-------
DRAFT
OCTOBER 1990
taken as an indication that unusual and persuasive differences exist with
respect to the source under review. In addition, where the cost of a control
alternative for the specific source reviewed is within the range of normal
costs for that control alternative, the alternative, in certain limited
circumstances, may still be eligible for elimination. To justify elimination
of an alternative on these grounds, the applicant should demonstrate to the
satisfaction of the permitting agency that costs of pollutant removal for the
control alternative are disproportionately high when compared to the cost of
control for that particular pollutant and source in recent BACT
determinations. If the circumstances of the differences are adequately
documented and explained in the application and are acceptable to the
reviewing agency they may provide a basis for eliminating the control
alternative.
In all cases, economic impacts need to be considered in conjunction with
energy and environmental impacts (e.g., toxics and hazardous pollutant
considerations) in selecting BACT. It is possible that the environmental
impacts analysis or other considerations (as described elsewhere) would
override the economic elimination criteria as described in this section.
However, absent a concern over an overriding environmental impact or other
considerations, an acceptable demonstration of an adverse economic impact can
be an adequate basis for eliminating the control alternative.
IV.D.Z.a. ESTIMATING THE COSTS OF CONTROL
Before costs can be estimated, the control system design parameters must
be specified. The most important item here is to ensure that the design
parameters used in costing are consistent with emissions estimates used in
other portions of the PSD application (e.g., dispersion modeling inputs and
permit emission limits). In general, the BACT analysis should present vendor-
supplied design parameters. Potential sources of other data on design
parameters are BID documents used to support NSPS development, control
technique guidelines documents, cost manuals developed by EPA, or control data
B.32
-------
DRAFT
OCTOBER 1990
in trade publications. Table B-4 presents some example design parameters
which are important in determining system costs.
To begin, the limits of the area or process segment to be costed
specified. This well defined area or process segment is referred to as the
control system battery limits. The second step is to list and cost each major
piece of equipment within the battery limits. The top-down BACT analysis
should provide this list of costed equipment. The basis for equipment cost
estimates also should be documented, either with data supplied by an equipment
vendor (i.e., budget estimates or bids) or by a referenced source [such as the
OAQPS Control Cost Manual (Fourth Edition), EPA 450/3-90-006, January 1990,
Table B-4]. Inadequate documentation of battery limits is one of the most
common reasons for confusion in comparison of costs of the same controls
applied to similar sources. For control options that are defined as
inherently lower-polluting processes (and not add-on controls), the battery
limits may be the entire process or project.
Design parameters should correspond to the specified emission level. The
equipment vendors will usually supply the design parameters to the applicant,
who in turn should provide them to the reviewing agency. In order to
determine if the design is reasonable, the design parameters can be compared
with those shown in documents such as the OAOPS Control Cost Manual. Control
Technology for Hazardous Air Pollutants (HAPS) Manual (EPA 625/6-86-014,
September 1986), and background information documents for NSPS and NESHAP
B.33
-------
DRAFT
OCTOBER 1990
TABLE B-4. EXAMPLE CONTROL SYSTEM DESIGN PARAMETERS
Control
Example Design parameters
Wet Scrubbers
Carbon Absorbers
Condensers
Incineration
Electrostatic Precipitator
Fabric Filter
Selective Catalytic Reduction
Scrubber liquor (water, chemicals, etc.)
Gas pressure drop
Liquid/gas ratio
Specific chemical species
Gas pressure drop
Ibs carbon/1bs pollutant
Condenser type
Outlet temperature
Residence time
Temperature
Specific collection area (ft2/acfm)
Voltage density
Air to cloth ratio
Pressure drop
Space velocity
Ammonia to NOx molar ratio
Pressure drop
Catalyst life
B.34
-------
DRAFT
OCTOBER 1990
regulations. If the design specified does not appear reasonable, then the
applicant should be requested to supply performance test data for the control
technology in question applied to the same source, or a similar source.
Once the control technology alternatives and achievable emissions
performance levels have been identified, capital and annual costs are
developed. These costs form the basis of the cost and economic impacts
(discussed later) used to determine and document if a control alternative
should be eliminated on grounds of its economic impacts.
Consistency in the approach to decision-making is a primary objective of
the top-down BACT approach. In order to maintain and improve the consistency
of BACT decisions made on the basis of cost and economic considerations,
procedures for estimating control equipment costs are based on EPA's OAQPS
Control Cost Manual and are set forth in Appendix B of this document.
Applicants should closely follow the procedures in the appendix and any
deviations should be clearly presented and justified in the documentation of
the BACT analysis.
Normally the submittal of very detailed and comprehensive project cost
data is not necessary. However, where initial control cost projections on the
part of the applicant appear excessive or unreasonable (in light of recent
cost data) more detailed and comprehensive cost data may be necessary to
document the applicant's projections. An applicant proposing the top
alternative usually does not need to provide cost data on the other possible
control alternatives.
Total cost estimates of options developed for BACT analyses should be on
the order of plus or minus 30 percent accuracy. If more accurate cost data
are available (such as specific bid estimates), these should be used.
However, these types of costs may not be available at the time permit
applications are being prepared. Costs should also be site specific. Some
site specific factors are costs of raw materials (fuel, water, chemicals) and
labor. For example, in some remote areas costs can be unusually high. For
B.35
-------
DRAFT
OCTOBER 1990
example, remote locations in Alaska may experience a 40-50 percent premium on ,
installation costs. The applicant should document any unusual costing
assumptions used in the analysis.
IV.D.Z.b. COST EFFECTIVENESS
Cost effectiveness is the economic criterion used to assess the
potential for achieving an objective at least cost. Effectiveness is measured
in terms of tons of pollutant emissions removed. Cost is measured in terms of
annualized control costs.
The cost-effectiveness calculations can be conducted on an average, or
incremental basis. The resultant dollar figures are sensitive to the number
of alternatives costed as well as the underlying engineering and cost
parameters. There are limits to the use of cost-effectiveness analysis. For
example, cost-effectiveness analysis should not be used to set the
environmental objective. Second, cost-effectiveness should, in and of itself,
not be construed as a measure of adverse economic impacts. There are two
measures of cost-effectiveness that will be discussed in this section: (1)
average cost-effectiveness, and (2) incremental cost-effectiveness.
Average Cost Effectiveness
Average cost effectiveness (total annualized costs of control divided by
annual emission reductions, or the difference between the baseline emission
rate and the controlled emission rate) is a way to present the costs of
control. Average cost effectiveness is calculated as shown by the following
formula:
B.36
-------
DRAFT
OCTOBER 1990
Average cost Effectiveness (dollars per ton removed) =
Control option annualized cost
Baseline emissions rate - Control option emissions rate
Costs are calculated in (annualized) dollars per year ($/yr) and
emissions rates are calculated in tons per year (tons/yr). The result is a
cost effectiveness number in (annualized) dollars per ton ($/ton) of pollutant
removed.
Calculating Baseline Emissions
The baseline emissions rate represents a realistic scenario of upper
bound uncontrolled emissions for the source. The NSPS/NESHAP requriements or
the application of controls, including other controls necessary to comply with
State or local air pollution regulations, are not considered in calculating
the baseline emissions. In other words, baseline emissions are essentially
uncontrolled emissions, calculated using realistic upper boundary operating
assumptions. When calculating the cost effectiveness of adding post process
emissions controls to certain inherently lower polluting processes, baseline
emissions may be assumed to be the emissions from the lower polluting process
itself. In other words, emission reduction credit can be taken for use of
inherently lower polluting processes.
Estimating realistic upper-bound emissions does not mean one should
assume the emissions represent the potential emissions. For example, in
developing a realistic upper bound case, baseline emissions calculations can
also consider inherent physical or operational constraints on the source.
Such constraints should reflect the upper boundary of the source's ability to
physically operate and the applicant should verify these constraints. If the
applicant does not adequately verify these constraints, then the reviewing
agency should not be compelled to consider these constraints in calculating
baseline emissions. In addition, the reviewing agency may require the
applicant to calculate cost effectiveness based on values exceeding the upper
B.37
-------
DRAFT
OCTOBER 1990
boundary assumptions to determine whether or not the assumptions have a
deciding role in the BACT determination. If the assumptions have a deciding
role in the BACT determination, the reviewing agency should include
enforceable conditions in the permit to assure that the upper bound
assumptions are not exceeded.
For example, VOC emissions from a storage tank might vary significantly
with temperature, volatility of liquid stored, and throughput. In this case,
potential emissions would be overestimated if annual VOC emissions were
estimated by extrapolating over the course of a year VOC emissions based
solely on the hottest summer day. Instead, the range of expected temperatures
should be considered in determining annual baseline emissions. Likewise,
potential emisisons would be overestimated if one assumed that gasoline would
be stored in a storage tank being built to feed an oil-fired power boiler or
that such a tank will be continually filled and emptied. On the other hand,
an upper bound case for a storage tank being constructed to store and transfer
liquid fuels at a marine terminal should consider emissions based on the most
volatile liquids at a high annual throughput level since it would not be
unrealistic for the tank to operate in such a manner.
In addition, historic upper bound operating data, typical for the source
or industry, may be used in defining baseline emissions in evaluating the cost
effectiveness of a control option for a specific source. For example, if for
a source or industry, historical upper bound operations call for two shifts a
day, it is not necessary to assume full time (8760 hours) operation on an
annual basis in calculating baseline emissions. For comparing cost
effectiveness, the same upper bound assumptions must, however, be used for
both the source in question and other sources (or source categories) that will
later be compared during the BACT analysis.
For example, suppose (based on verified historic data regarding the
industry in question) a given source can be expected to utilize numerous
colored inks over the course of a year. Each color ink has a different VOC
content ranging from a high VOC content to a relatively low VOC content. The
B.38
-------
DRAFT
OCTOBLR 1990
source verifies that its operation will indeed call for the application of
numerous color inks. In this case, it is more realistic for the baseline
emission calculation for the source (and other similar sources) to be based on
the expected mix of inks that would be expected to result in an upper bound
case annual VOC emissions rather than an assumption that only one color (i.e,
the ink with the highest VOC content) will be applied exclusively during the
whole year.
In another example, suppose sources in a particular industry
historically operate at most at 85 percent capacity. For BACT cost
effectiveness purposes (but not for applicability), an applicant may calculate
cost effectiveness using 85 percent capacity. However, in comparing costs
with similar sources, the applicant must consistently use an 85 percent
capacity factor for the cost effectiveness of controls on those other sources.
Although permit conditions are normally used to make operating
assumptions enforceable, the use of "standard industry practice" parameters
for cost effectiveness calculations (but not applicability determinations) can
be acceptable without permit conditions. However, when a source projects
operating parameters (e.g., limited hours of operation or capacity
utilization, type of fuel, raw materials or product mix or type) that are
lower than standard industry practice or which have a deciding role in the
BACT determination, then these parameters or assumptions must be made
enforceable with permit conditions. If the applicant will not accept
enforceable permit conditions, then the reviewing agency should use the worst
case uncontrolled emissions in calculating baseline emissions. This is
necessary to ensure that the permit reflects the conditions under which the
source intends to operate.
For example, the baseline emissions calculation for an emergency standby
generator may consider the fact that the source does not intend to operate
more than 2 weeks a year. On the other hand, baseline emissions associated
with a base-loaded turbine would not consider limited hours of operation.
This produces a significantly higher level of baseline emissions than in the
B.39
-------
DRAFT
OCTOBER 1990
case of the emergency/standby unit and results in more cost effective
controls. As a consequence of the dissimilar baseline emissions, BACT for the
two cases could be very different. Therefore, it is important that the
applicant confirm that the operational assumptions used to define the source's
baseline emissions (and BACT) are genuine. As previously mentioned, this is
usually done through enforceable permit conditions which reflect limits on the
source's operation which were used to calculate baseline emissions.
In certain cases, such explicit permit conditions may not be necessary.
For example, a source for which continuous operation would be a physical
impossibility (by virtue of its design) may consider this limitation in
estimating baseline emissions, without a direct permit limit on operations.
However, the permit agency has the responsibility to verify that the source is
constructed and operated consistent with the information and design
specifications contained in the permit application.
For some sources it may be more difficult to define what emissions level
actually represents uncontrolled emissions in calculating baseline emissions.
For example, uncontrolled emissions could theoretically be defined for a spray
coating operation as the maximum VOC content coating at the highest possible
rate of application that the spray equipment could physically process (even
though use of such a coating or application rate would be unrealistic for the
source). Assuming use of a coating with a VOC content and application rate
greater than expected is unrealistic and would result in an overestimate in
the amount of emissions reductions to be achieved by the installation of
various control options. Likewise, the cost effectiveness of the options
could consequently be greatly underestimated. To avoid these problems,
uncontrolled emission factors should be represented by the highest realistic
VOC content of the types of coatings and highest realistic application rates
that would be used by the source, rather than by highest theoretical VOC based
coating materials or rate of application in general.
Conversely, if uncontrolled emissions are underestimated, emissions
reductions to be achieved by the various control options would also be
B.40
-------
DRAFT
OCTOBER 1990
underestimated and their cost effectiveness overestimated. For example, this
type of situation occurs in the previous example if the baseline for the above
coating operation was based on a VOC content coating or application rate that
is too low [when the source had the ability and intent to utilize (even
infrequently) a higher VOC content coating or application rate],
Incremental Cost Effectiveness
In addition to the average cost effectiveness of a control option,
incremental cost effectiveness between dominant control options should also be
calculated. The incremental cost effectiveness should be examined in
combination with the average cost effectiveness in order to justify
elimination of a control option. The incremental cost effectiveness
calculation compares the costs and emissions performance level of a control
option to those of the next most stringent option, as shown in the following
formula:
Incremental Cost (dollars per incremental ton removed) =
Total costs (annualized) of control option - Total costs (annualized) of next control option
Next control option emission rate - Control option emissions rate
Care should be exercised in deriving incremental costs of .candidate
control options. Incremental cost-effectiveness comparisons should focus on
annualized cost and emisison reduction differences between dominant
alternatives. Dominant set of control alternatives are determined by
generating what is called the envelope of least-cost alternatives. This is a
graphical plot of total annualized costs for a total emissions reductions for
all control alternatives identified in the BACT analysis (see Figure B-l).
For example, assume that eight technically available control options for
analysis are listed in the BACT hierarchy. These are represented as A through
H in Figure B-l. In calculating incremental costs, the analysis should only
be conducted for control options that are dominant among all possible options
B.41
-------
t
0)
o
o
Q
ill
N
O
z
LJJ
DC
O
Dominant controls (B, D, F, G, H) lie on envelope
Inferior controls (A,C,E)
DRAFT
OCTOBER 1990
H
"delta" Total Costs Annualized
'delta" Emissions Reduction
INCREASING EMISSIONS REDUCTION (Tons/yr)
Figure B-1. LEAST-COST ENVELOPE
B.42
-------
DRAFT
OCTOBER 1990
In Figure B-l, the dominant set of control options, B, D, F, G, and H,
represent the least-cost envelope depicted by the curvilinear line connecting
them. Points A, C and E are inferior options and should not be considered in
the derivation of incremental cost effectiveness. Points A, C and E represent
inferior controls because B will buy more emissions reduction for less money
than A; and similarly, D and F will by more reductions for less money than C
and E, respectively.
Consequently, care should be taken in selecting the dominant set of
controls when calculating incremental costs. First, the control options need
to be rank ordered in ascending order of annualized total costs. Then, as
Figure B-l illustrates, the most reasonable smooth curve of the control
options is plotted . The incremental cost effectiveness is then determined by
the difference in total annual costs between two contiguous options divided by
the difference in emissions reduction. An example is illustrated in
Figure B-l for the incremental cost effectiveness for control option F. The
vertical distance, "delta" Total Costs Annualized, divided by the horizontal
distance, "delta" Emissions Reduced (tpy), would be the measure of the
incremental cost effectiveness for option F.
A comparison of incremental costs can also be useful in evaluating a
specific control option over a range of efficiencies. For example, depending
on the capital and operational cost of a control device, total and incremental
cost may vary significantly (either increasing or decreasing) over the
operation range of a control device.
As a precaution, differences in incremental costs among dominant
alternatives cannot be used by itself to argue one dominant alternative is
preferred to another. For example, suppose dominant alternatives B, D. and F
on the least-cost envelope (see Figure B-l) are identified as alternaitves for
a BACT analysis. We may observe the incremental cost effectivenss between
dominant alternative B and D is $500 per ton whereas between dominant
alternative D and F is is $1000 per ton. Alternative D does not dominate
alternative F. Both alternatives are dominant and hence on the least cost
B.43
-------
DRAFT
OCTOBER 1990
envelope. Alternative D cannot legitimately be preferred to F on grounds of
incremental cost effectiveness.
In addition, when evaluating the average or incremental cost
effectiveness of a control alternative, reasonable and supportable assumptions
regarding control efficiencies should be made. As mentioned above,
unrealistically low estimates of the emission reduction potential of a certain
technology could result in inflated cost effectiveness figures.
The final decision regarding the reasonableness of calculated cost
effectiveness values will be made by the review authority considering previous
regulatory decisions. Study cost estimates used in BACT are typically
accurate to + 20 to 30 percent. Therefore, control cost options which are
within ± 20 to 30 percent of each other should generally be considered to be
indistinguishable when comparing options.
IV.D.2.C. DETERMINING AN ADVERSE ECONOMIC IMPACT
It is important to keep in mind that BACT is primarily a technology-
based standard. In essence, if the cost of reducing emissions with the top
control alternative, expressed in dollars per ton, is on the same order as the
cost previously borne by other sources of the same type in applying that
control alternative, the alternative should initially be considered
economically achievable, and therefore acceptable as BACT. However, unusual
circumstances may greatly affect the cost of controls in a specific
application. If so they should be documented. An example of an unusual
circumstance might be the unavailability in an arid region of the large
amounts of water needed for a scrubbing system. Acquiring water from a
distant location might add unreasonable costs to the alternative, thereby
justifying its elimination on economic grounds. Consequently, where unusual
factors exist that result in cost/economic impacts beyond the range normally
incurred by other sources in that category, the technology can be eliminated
provided the applicant has adequately identified the circumstances, including
B.44
-------
DRAFT
OCTOBER 1990
the cost or other analyses, that show what is significantly different about
the proposed source.
Where the cost effectiveness of a control alternative for the specific
source being reviewed is within the range of normal costs for that control
alternative, the alternative may also be eligible for elimination in limited
circumstances. This may occur, for example, where a control alternative has
not been required as BACT (or its application as BACT has been extremely
limited) and there is a clear demarcation between recent BACT control costs in
that source category and the control costs for sources in that source category
which have been driven by other constraining factors (e.g., need to meet a PSD
increment or a NAAQS).
To justify elimination of an alternative on these grounds, the applicant
should demonstrate to the satisfaction of the permitting agency that costs of
pollutant removal (e.g., dollars per total ton removed) for the control
alternative are disproportionately high when compared to the cost of control
for the pollutant in recent BACT determinations. Specifically, the applicant
should document that the cost to the applicant of the control alternative is
significantly beyond the range of recent costs normally associated with BACT
for the type of facility (or BACT control costs in general) for the pollutant.
This type of analysis should demonstrate that a technically and economically
feasible control option is nevertheless, by virtue of the magnitude of its
associated costs and limited application, unreasonable or otherwise not
"achievable" as BACT in the particular case. Average and incremental cost
effectiveness numbers are factored into this type of analysis. However, such
economic information should be coupled with a comprehensive demonstration,
based on objective factors, that the technology is inappropriate in the
specific circumstance.
The economic impact portion of the BACT analysis should not focus on
inappropriate factors or exclude pertinent factors, as the results may be
misleading. For example, the capital cost of a control option may appear
excessive when presented by itself or as a percentage of the total project
B.45
-------
DRAFT
OCTOBER 1990
cost. However, this type of information can be misleading. If a large
emissions reduction is projected, low or reasonable cost effectiveness numbers
may validate the option as an appropriate BACT alternative irrespective of the
apparent high capital costs. In another example, undue focus on incremental
cost effectiveness can give an impression that the cost of a control
alternative is unreasonably high, when, in fact, the cost effectiveness, in
terms of dollars per total ton removed, is well within the normal range of
acceptable BACT costs.
IV.D.3. ENVIRONMENTAL IMPACTS ANALYSIS
The environmental impacts analysis is not to be confused with the air
quality impact analysis (i.e., ambient concentrations), which is an
independent statutory and regulatory requirement and is conducted separately
from the BACT analysis. The purpose of the air qual ity'analysis is to
demonstrate that the source (using the level of control ultimately determined
to be BACT) will not cause or contribute to a violation of any applicable
national ambient air quality standard or PSD increment. Thus, regardless of
"the level of control proposed as BACT, a permit cannot be issued to a source
that would cause or contribute to such a violation. In contrast, the
environmental impacts portion of the BACT analysis concentrates on impacts
other than impacts on air quality standards due to emissions of the regulated
pollutant in question, such as solid or hazardous waste generation, discharges
of polluted water from a control device, visibility impacts, or emissions of
unregulated pollutants.
Thus, the fact that a given control alternative would result in only a
slight decrease in ambient concentrations of the pollutant in question when
compared to a less stringent control alternative should not be viewed as an
adverse environmental impact justifying rejection of the more stringent
control alternative. However, if the cost effectiveness of .the more stringent
alternative is exceptionally high, it may (as provided in section V.D.2.) be
considered in determining the existence of an adverse economic impact that
would justify rejection of the more stringent alternative.
B.46
-------
DRAFT
OCTOBER 1990
The applicant should identify any significant or unusual environmental
impacts associated with a control alternative that have the potential to
affect the selection or elimination of a control alternative. Some control
technologies may have potentially significant secondary (i.e., collateral)
environmental impacts. Scrubber effluent, for example, may affect water
quality and land use. Similarly, emissions of water vapor from technologies
using cooling towers may affect local visibility. Other examples of secondary
environmental impacts could include hazardous waste discharges, such as spent
catalysts or contaminated carbon. Generally, these types of environmental
concerns become important when sensitive site-specific receptors exist or when
the incremental emissions reduction potential of the top control is only
marginally greater than the next most effective option. However, the fact
that a control device creates liquid and solid waste that must be disposed of
does not necessarily argue against selection of that technology as BACT,
particularly if the control device has been applied to similar facilities
elsewhere and the solid or liquid waste problem under review is similar to
those other applications. On the other hand, where the applicant can show
that unusual circumstances at the proposed facility create greater problems
than experienced elsewhere, this may provide a basis for the elimination of
that control alternative as BACT.
The procedure for conducting an analysis of environmental impacts should
be made based on a consideration of site-specific circumstances. In general,
however, the analysis of environmental impacts starts with the identification
and quantification of the solid, liquid, and gaseous discharges from the
control device or devices under review. This analysis of environmental
impacts should be performed for the entire hierarchy of technologies (even if
the applicant proposes to adopt the "top", or most stringent, alternative).
However, the analysis need only address those control alternatives with any
significant or unusual environmental impacts that have the potential to affect
the selection or elimination of a control alternative. Thus, the relative
environmental impacts (both positive and negative) of the various alternatives
can be compared with each other and the "top" alternative.
B.47
-------
DRAFT
OCTOBER 1990
Initially, a qualitative or semi-quantitative screening is performed to
narrow the analysis to discharges with potential for causing adverse
environmental effects. Next, the mass and composition of any such discharges
should be assessed and quantified to the extent possible, based on readily
available information. Pertinent information about the public or
environmental consequences of releasing these materials should also be
assembled.
IV.D.S.a. EXAMPLES (Environmental Impacts)
The following paragraphs discuss some possible factors for consideration
in evaluating the potential for an adverse other media impact.
• Water Impact
Relative quantities of water used and water pollutants produced and
discharged as a result of use of each alternative emission control system
relative to the "top" alternative would be identified. Where possible, the
analysis would assess the effect on ground water and such local surface water
quality parameters as ph, turbidity, dissolved oxygen, salinity, toxic
chemical levels, temperature, and any other important considerations. The
analysis should consider whether applicable water quality standards will be
met and the availability and effectiveness of various techniques to reduce
potential adverse effects.
• Solid Haste Disposal Impact
The quality and quantity of solid waste (e.g., sludges, solids) that must
be stored and disposed of or recycled as a result of the application of each
alternative emission control system would be compared with the quality and
quantity of wastes created with the "top" emission control system. The
composition and various other characteristics of the solid waste (such as
permeability, water retention, rewatering of dried material, compression
strength, Teachability of dissolved ions, bulk density, ability to support
B.48
-------
DRAFT
OCTOBER 1990
vegetation growth and hazardous characteristics) which are significant with
regard to potential surface water pollution or transport into and
contamination of subsurface waters or aquifers would be appropriate for
consideration.
• Irreversible or Irretrievable Commitment of Resources
The BACT decision may consider the extent to which the alternative
emission control systems may involve a trade-off between short-term
environmental gains at the expense of long-term environmental losses and the
extent to which the alternative systems may result in irreversible or
irretrievable commitment of resources (for example, use of scarce water
resources).
• Other Environmental Impacts
Significant differences in noise levels, radiant heat, or dissipated
static electrical energy, or greenhouse gas emissions may be considered.
One environmental impact that could be examined is the trade-off
between emissions of the various pollutants resulting from the application of
a specific control technology. The use of certain control technologies may
lead to increases in emissions of pollutants other than those the technology
was designed to control. For example, the use of certain volatile organic
compound (VOC) control technologies can increase nitrogen oxides (NOx)
emissions. In this instance, the reviewing authority may want to give
consideration to any relevant local air quality concern relative to the
secondary pollutant (in this case NOx) in the region of the proposed source.
For example, if the region in the example were nonattainment for NOx, a
premium could be placed on the potential NOx impact. This could lead to
elimination of the most stringent VOC technology (assuming it generated high
quantities of NOx) in favor of one having less of an impact on ambient NOx
concentrations. Another example is the potential for higher emissions of
toxic and hazardous pollutants from a municipal waste combustor operating at a
B.49
-------
DRAFT
OCTOBER 1990
low flame temperature to reduce the formation of NOx. In this case the real
concern to mitigate the emissions of toxic and hazardous emissions (via high
combustion temperatures) may well take precedent over mitigating NOx emissions
through the use of a low flame temperature. However, in most cases (unless an
overriding concern over the formation and impact of the secondary pollutant is
clearly present as in the examples given), it is not expected that this type
impact would affect the outcome of the decision.
Other examples of collateral environmental impacts would include
hazardous waste discharges such as spent catalysts or contaminated carbon.
Generally these types of environmental concerns become important when site-
specific sensitive receptors exist or when the incremental emissions reduction
potential of the top control option is only marginally greater than the next
most effective option.
IV.D.3.b. CONSIDERATION OF EMISSIONS OF TOXIC AND HAZARDOUS AIR POLLUTANTS
The generation or reduction of toxic and hazardous emissions, including
compounds not regulated under the Clean Air Act, are considered as part of the
environmental impacts analysis. Pursuant to the EPA Administrator's decision
in North County Resource Recovery Associates. PSD Appeal No. 85-2 (Remand
Order, June 3, 1986), a PSD permitting authority should consider the effects
of a given control alternative on emissions of toxics or hazardous pollutants
not regulated under the Clean Air Act. The ability of a given control
alternative to control releases of unregulated toxic or hazardous emissions
must be evaluated and may, as appropriate, affect the BACT decision.
Conversely, hazardous or toxic emissions resulting from a given control
technology should also be considered and may, as appropriate, also affect the
BACT decision.
Because of the variety of sources and pollutants that may be considered
in this assessment, it is not feasible for the EPA to provide highly detailed
national guidance on performing an evaluation of the toxic impacts as part of
the BACT determination. Also, detailed information with respect to the type
B.50
-------
DRAFT
OCTOBER 1990
and magnitude of emissions of unregulated pollutants for many source
categories is currently limited. For example, a combustion source emits
hundreds of substances, but knowledge of the magnitude of some of these
emissions or the hazards they produce is sparse. The EPA believes it is
appropriate for agencies to proceed on a case-by-case basis using the best
information available. Thus, the determination of whether the pollutants
would be emitted in amounts sufficient to be of concern is one that the
permitting authority has considerable discretion in making. However,
reasonable efforts should be made to address these issues. For example, such
efforts might include consultation with the:
• EPA Regional Office;
• Control Technology Center (CTC);
• National Air Toxics Information Clearinghouse;
• Air Risk Information Support Center in the Office of Air
Quality Planning and Standards (OAQPS); and
• Review of the current literature, such as EPA-prepared
compilations of emission factors.
Source-specific information supplied by the permit applicant is often the best
source of information, and it is important that the applicant be made aware of
its responsibility to provide for a reasonable accounting of air toxics
emissions.
Similarly, once the pollutants of concern are identified, the permitting
authority has flexibility in determining the methods by which it factors air
toxics considerations into the BACT determination, subject to the obligation
to make reasonable efforts to consider air toxics. Consultation by the review
authority with EPA's implementation centers, particularly the CTC, is again
advise'd.
It is important to note that several acceptable methods, including risk
assessment, exist to incorporate air toxics concerns into the BACT decision.
B.51
-------
DRAFT
OCTOBER 1990
The depth of the toxics assessment will vary with the circumstances of the
particular source under review, the nature and magnitude of the toxic
pollutants, and the locality. Emissions of toxic or hazardous pollutants of
concern to the permit agency should be identified and, to the extent possible,
quantified. In addition, the effectiveness of the various control
alternatives in the hierarchy at controlling the toxic pollutants should be
estimated and summarized to assist in making judgements about how potential
emissions of toxic or hazardous pollutants may be mitigated through the
selection of one control option over another. For example, the response to
the Administrator made by EPA Region IX in its analysis of the North County
permitting decision illustrates one of several approaches (for further
information see the September 22, 1987 EPA memorandum from Mr. Gerald Emison
titled "Implementation of North County Resource Recover PSD Remand" and
July 28, 1988 EPA memorandum from Mr. John Calcagni titled " Supplemental
guidance on Implementing the North County Prevention of Significant
Deterioration (PSD) Remand").
Under a top-down BACT analysis, the control alternative selected as BACT
will most likely reduce toxic emissions as well as the regulated pollutant.
An example is the emissions of heavy metals typically associated with coal
combustion. The metals generally are a portion of, or adsorbed on, the fine
particulate in the exhaust gas stream. Collection of the particulate in a
high efficiency fabric filter rather than a low efficiency electrostatic
precipitator reduces criteria pollutant particulate matter emissions and
toxic heavy metals emissions. Because in most instances the interests of
reducing toxics coincide with the interests of reducing the pollutants subject
to BACT, consideration of toxics in the BACT analysis generally amounts to
quantifying toxic emission levels for the various control options.
In limited other instances, though, control of regulated pollutant
emissions may compete with control of toxic compounds, as in the case of
certain selective catalytic reduction (SCR) NOx control technologies. The SCR
technology itself results in emissions of ammonia, which increase, generally
speaking, with increasing levels of NOx control. It is the intent of the
B.52
-------
DRAFT
OCTOBER 1990
toxics screening in the BACT procedure to identify and quantify this type of
toxic effect. Generally, toxic effects of this type will not necessarily be
overriding concerns and will not likely affect BACT decisions. Rather, the
intent is to require a screening of toxics emissions effects to ensure that a
possible overriding toxics issue does not escape notice.
On occasion, consideration of toxics emissions may support the selection
of a control technology that yields less than the maximum degree of reduction
in emissions of the regulated pollutant in question. An example is the
municipal solid waste combustor and resource recovery facility that was the
subject of the North County remand. Briefly, BACT for S02 and PM was selected
to be a lime slurry spray drier followed by a fabric filter. The combination
yields good S02 control (approximately 83 percent), good PM control
(approximately 99.5 percent) and also removes acid gases (approximately 95
percent), metals, dioxins, and other unregulated pollutants. In this
instance, the permitting authority determined that good balanced control of
regulated and unregulated pollutants took priority over achieving the maximum
degree of emissions reduction for one or more regulated pollutants.
Specifically, higher levels (up to 95 percent) of S02 control could have been
obtained by a wet scrubber.
IV.E. SELECTING BACT (STEP 5)
The most effective control alternative not eliminated in Step 4 is
selected as BACT.
It is important to note that, regardless of the control level proposed by
the applicant as BACT, the ultimate BACT decision is made by the permit
issuing agency after public review. The applicant's role is primarily to
provide information on the various control options and, when it proposes a
less stringent control option, provide a detailed rationale and supporting
documentation for eliminating the more stringent options. It is the
responsibility of the permit agency to review the documentation and rationale
presented and; (1) ensure that the applicant has addressed all of the most
B.53
-------
DRAFT
OCTOBER 1990
effective control options that could be applied and; (2) determine that the
applicant has adequately demonstrated that energy, environmental, or economic
impacts justify any proposal to eliminate the more effective control options.
Where the permit agency does not accept the basis for the proposed elimination
of a control option, the agency may inform the applicant of the need for more
information regarding the control option. However, the BACT selection
essentially should default to the highest level of control for which the
applicant could not adequately justify its elimination based on energy*
environmental, and economic impacts. The permit agency should proceed to
establish BACT and prepare a draft permit based on the most effective control
option for which an adequate justification for rejection was not provided.
IV.F. OTHER CONSIDERATIONS
Once energy, environmental, and economic impacts have been considered,
BACT can only be made more stringent by other considerations outside the
normal scope of the BACT analysis as discussed under the above steps.
Examples include cases where BACT does not produce a degree of control
stringent enough to prevent exceedences of a national ambient air quality
standard or PSD increment, or where the State or local agency will not accept
the level of control selected as BACT and requires more stringent controls to
preserve a greater amount of the available increment. A permit cannot be
issued to a source that would cause or contribute to such a violation,
regardless of the outcome of the BACT analysis. Also, States which have set
ambient air quality standards at levels tighter than the federal standards may
demand a more stringent level of control at a source to demonstrate compliance
with the State standards. Another consideration which could override the
selected BACT are legal constraints outside of the Clean Air Act requiring the
application of a more stringent technology (e.g., a consent decree requiring a
greater degree of control). In all cases, regardless of the rationale for the
permit requiring a more stringent emissions limit than would have otherwise
been chosen as a result of the BACT selection process, the emission limit in
the final permit (and corresponding control alternative) represents BACT for
the permitted source on a case-by-case basis.
B.54
-------
DRAFT
OCTOBER 1990
The BACT emission limit in a new source permit is not set until the final
permit is issued. The final permit is not issued until a draft permit has
gone through public comment and the permitting agency has had an opportunity
to consider any new information that may have come to light during the comment
period. Consequently, in setting a proposed or final BACT limit, the permit
agency can consider new information it learns, including recent permit
decisions, subsequent to the submittal of a complete application. This
emphasizes the importance of ensuring that prior to the selection of a
proposed BACT, all potential sources of information have been reviewed by the
source to ensure that the list of potentially applicable control alternatives
is complete (most importantly as it relates to any more effective control
options than the one chosen) and that all considerations relating to economic,
energy and environmental impacts have been addressed.
B.55
-------
DRAFT
OCTOBER 1990
V. ENFORCEABILITY OF BACT
To complete the BACT process, the reviewing agency must establish an
enforceable emission limit for each subject emission unit at the source and
for each pollutant subject to review that is emitted from the source. If
technological or economic limitations in the application of a measurement
methodology to a particular emission unit would make an emissions limit
infeasible, a design, equipment, work practice, operation standard, or
combination thereof, may be prescribed. Also, the technology upon which the
BACT emissions limit is based should be specified in the permit. These
requirements should be written in the permit so that they are specific to the
individual emission unit(s) subject to PSD review.
The emissions limits must be included in the proposed permit submitted
for public comment, as well as the final permit. BACT emission limits or
conditions must be met on a continual basis at all levels of operation (e.g.,
limits written in pounds/MMbtu or percent reduction achieved), demonstrate
protection of short term ambient standards (limits written in pounds/hour) and
be enforceable as a practical matter (contain appropriate averaging times,
compliance verification procedures and recordkeeping requirements).
Consequently, the permit must:
• be able to show compliance or noncompliance (i.e., through
monitoring times of operation, fuel input, or other indices
of operating conditions and practices); and
• specify a reasonable compliance averaging time consistent with
established reference methods, contain reference methods for
determining compliance, and provide for adequate reporting and
recordkeeping so that the permitting agency can determine
the compliance status of the source.
B.56
-------
DRAFT
OCTOBER 1990
VI. EXAMPLE BACT ANALYSES FOR GAS TURBINES
Note: The following example provided is for illustration only. The example
source is fictitious and has been created to highlight many of the aspects of the
top-down process. Finally, it must be noted that the cost data and other numbers
presented in the example are used only to demonstrate the BACT decision making
process. Cost data are used in a relative sense to compare control costs among
sources in a source category or for a pollutant. Determination of appropriate
costs is made on a case-by-case basis.
In this section a BACT analysis for a stationary gas turbine project is
presented and discussed under three alternative operating scenarios:
• Example I—Simple Cycle Gas Turbines Firing Natural Gas
• Example 2--Combined Cycle Gas Turbines Firing Natural Gas
• Example 3--Combined Cycle Gas Turbines Firing Distillate Oil
The purpose of the examples are to illustrate points to be considered in
developing BACT decision criteria for the source under review and selecting
BACT. They are intended to illustrate the process rather than provide
universal guidance on what constitutes BACT for any particular source
category. BACT must be determined on a case-by-case basis.
These examples are not based on any actual analyses performed for the
purposes of obtaining a PSD permit. Consequently, the actual emission rates,
costs, and design parameters used are neither representative of any actual
case nor do they apply to any particular facility.
B.57
-------
DRAFT
OCTOBER 1990
VI.A. EXAMPLE 1-SIMPLE CYCLE GAS TURBINES FIRING NATURAL GAS
VI.A.I PROJECT SUMMARY
Table B-5 presents project data, stationary gas design parameters, and
uncontrolled emission estimates for the new source in example 1. The gas
turbine is designed to provide peaking service to an electric utility. The
planned operating hours are less than 1000 hours per year. Natural gas fuel
will be fired. The source will be limited through enforceable conditions to
the specified hours of operation and fuel type. The area where the source is
to be located is in compliance for all criteria pollutants. No other changes
are proposed at this facility, and therefore the net emissions change will be
equal to the emissions shown on Table B-5. Only NOx emissions are significant
(i.e., greater than or equal to the 40 tpy significance level for NOx) and a
BACT analysis is required for NOx emissions only.
VI.A.2. BACT ANALYSIS SUMMARY
VII.A.Z.a. CONTROL TECHNOLOGY OPTIONS
The first step in evaluating BACT is identifying all candidate control
technology options for the emissions unit under review. Table B-6 presents
the list of control technologies selected as potential BACT candidates. The
first three control technologies, water or steam injection and selective
catalytic reduction, were identified by a review of existing gas turbine
facilities in operation. Selective noncatalytic reduction was identified as a
potential type of control technology because it is an add-on NOx control which
has been applied to other types of combustion sources.
B.58
-------
DRAFT
OCTOBER 1990
TABLE B-5. EXAMPLE 1--COMBUSTION TURBINE DESIGN PARAMETERS
Characteristics
Number of emissions units
Unit Type
Cycle Type
Output
Exhaust temperature,
Fuel(s)
Heat rate, Btu/kw hr
Fuel flow, Btu/hr
Fuel flow, Ib/hr
Service Type
Operating Hours (per year)
Uncontrolled Emissions, tpy(a)
NOY
so;
CO^
VOC
PM
1
Gas Turbine
Simple-cycle
75 MW
1,000 °F
Natural Gas
11,000
1,650 mill ion
83,300
Peaking
1,000
282 (169 ppm)
4.6 (6 ppm)
1
5 (0.0097 gr/dscf)
(a) Based on 1000 hours per year of operation at full load.
B.59
-------
DRAFT
OCTOBER 1990
TABLE B-6.
EXAMPLE 1—SUMMARY OF POTENTIAL NOx CONTROL
TECHNOLOGY OPTIONS
Control technology(a)
Selective Catalytic
Reductions
Water Injection
Steam Injection
Low NOx Burner
Selective Noncatalytic
Reduction
Typical
control
efficiency
range
(% reduction)
40-90
30-70
30-70
30-70
20-50
Simple
cycle
turbines
No
Yes
No
Yes
No
In Service On:
•Combined
cycle
gas
turbines
Yes
Yes
Yes
Yes
Yes
Other
combustion
sources(c)
Yes
Yes
Yes
Yes
Yes
Technically
feasible on
simple cycle
turbines
Yes(b)
Yes
No
Yes
No
(a) Ranked in order of highest to lowest stringency.
(b) Exhaust must be diluted with air to reduce its temperature to 600-750°F.
(c) Boiler incinerators, etc.
B.60
-------
DRAFT
OCTOBER 1990
In this example, the control technologies were identified by the
applicant based on a review of the BACT/LAER Clearinghouse, and discussions
with State agencies with experience permitting gas turbines in NOx
nonattainment areas. A preliminary meeting with the State permit issuing
agency was held to determine whether the permitting agency felt that any other
applicable control technologies should be evaluated and they agreed on the
proposed control hierarchy.
VI.A.Z.b. TECHNICAL FEASIBILITY CONSIDERATIONS
Once potential control technologies have been identified, each technology
is evaluated for its technical feasibility based on the characteristics of the
source. Because the gas turbines in this example are intended to be used for
peaking service, a heat recovery steam generator (HRSG) will not be included.
A HRSG recovers heat from the gas turbine exhaust to make steam and increase
overall energy efficiency. A portion of the steam produced can be used for
steam injection for NOx control, sometimes increasing the effectiveness of the
net injection control system. However, the electrical demands of the grid
dictate that the turbine will be brought on line only for short periods of
time to meet peak demands. Due to the lag time required to bring a heat
recovery steam generator on line, it is not technically feasible to use a HRSG
at the facility. Use of an HRSG in this instance was shown to interfere with
the performance of the unit for peaking service, which requires immediate
response times for the turbine. Although it was shown that a HRSG was not
feasible and therefore not available, water and steam are readily available
for NOx control since the turbine will be located near an existing steam
generating powerplant.
The turbine type and, therefore, the turbine model selection process,
affects the achievability of NOx emissions limits. Factors which the customer
considered in selecting the proposed turbine model were outlined in the
application as: the peak demand which must be met, efficiency of the gas
turbine, reliability requirements, and the experience of the utility with the
operation and maintenance service of the particular manufacturer and turbine
B.61
-------
DRAFT
OCTOBER 1990
design. In this example, the proposed turbine is equipped with a combustor
designed to achieve an emission level, at 15 percent 02, of 25 ppm NOx with
steam injection or 42 ppm with water injection.2
Selective noncatalytic reduction (SNCR) was eliminated as technically
infeasible, and therefore not available, because this technology requires a
flue gas temperature of 1300 to 2100oF. The exhaust from the gas turbines
will be approximately lOOOoF, which is below the required temperature range.
Selective catalytic reduction (SCR) was evaluated and no basis was found
to eliminate this technology as technically infeasible. However, there are no
known examples where SCR technology has been applied to a simple-cycle gas
turbine or to a gas turbine in peaking service. In all cases where SCR has
been applied, there was an HRSG which served to reduce the exhaust temperature
to the optimum range of 600-750oF and the gas turbine was operated
continuously. Consequently, application of SCR to a simple cycle turbine
involves special circumstances. For this example, it is assumed that dilution
air can be added to the gas turbine exhaust to reduce its temperature.
However, the dilution air will make the system more costly due to higher gas
flows, and may reduce the removal efficiency because the NOx concentration at
the inlet will be reduced. Cost considerations are considered later in the
analysis.
VI.A.2.C. CONTROL TECHNOLOGY HIERARCHY
After determining technical feasibility, the applicant selected the
control levels for evaluation shown in Table B-7. Although the applicant
2 For some gas turbine models, 25 ppm is not achievable with either water
or steam injection.
B.62
-------
DRAFT
OCTOBER 1990
TABLE B-7. EXAMPLE 1--CONTROL TECHNOLOGY HIERARCHY
Emissions Limits
Control Technology ppm(a) TRY
Steam Injection plus SCR 13 44
Steam Injection at maximum'*3) design rate 25 84
Water Injection at maximum' ' design rate 42 140
Steam Injection to meet NSPS ' 93 312
(a) Corrected to 15 percent oxygen.
(b) Water to fuel ratio.
B.63
-------
DRAFT
OCTOBER 1990
reported that some sites in California have achieved levels as low as 9 ppm,
at this facility a 13 ppm level was determined to be the feasible limit with
SCR. This decision is based on the lowest achievable level with steam
injection of 25 ppm and an SCR removal efficiency of 50 percent. Even though
the reported removal efficiencies for SCR are up to 90 percent at some
facilities, at this facility the actual NOx concentration at the inlet to the
SCR system will only be approximately 17 ppm (at actual conditions) due to the
dilution air required. Also the inlet concentrations, flowrates, and
temperatures will vary due to the high frequency of startups. These factors
make achieving the optimum 90 percent NOx removal efficiency unrealistic.
Based on discussions with SCR vendors, the applicant has established a
50 percent removal efficiency as the highest level achievable, thereby
resulting in a 13 ppm level (i.e., 50 percent of 25 ppm).
The next most stringent level achievable would be steam injection at the
maximum water-to-fuel ratio achievable by the unit within its design operating
range. For this particular gas turbine model, that level is 25 ppm as
supported by vendor NOx emissions guarantees and unit test data. The
applicant provided documentation obtained from the gas turbine manufacturers
verifying ability to achieve this range.
After steam injection the next most stringent level of control would be
water injection at the maximum water-to-fuel ratio achievable by the unit
within its design operating range. For this particular gas turbine model,
that level is 42 ppm as supported by vendor NOx emissions guarantees and
actual unit test data. The applicant provided documentation obtained from the
gas turbine manufacturer verifying ability to achieve this range.
The least stringent level evaluated by the applicant was the current
NSPS for utility gas turbines. For this model, that level is 93 ppm at
3 It should be noted that achievability of the NOx limits is dependent on
the turbine model, fuel, type of wet injection (water or steam), and system
design. Not all gas turbine models or fuels can necessarily achieve these
levels.
B.64
-------
DRAFT
OCTOBER 1990
15 percent 02. By definition, BACT can be no less stringent than NSPS.
Therefore, less stringent levels are not evaluated.
VI.A.Z.d. IMPACTS ANALYSIS SUMMARY
The next steps completed by the applicant were the development of the
cost, economic, environmental and energy impacts of the different control
alternatives. Although the top-down process would allow for the selection of
the top alternative without a cost analysis, the applicant felt cost/economic
impacts were excessive and that appropriate documentation may justify the
elimination of SCR as BACT and therefore chose to quantify cost and economic
impacts. Because the technologies in this case are applied in combination, it
was necessary to quantify impacts for each of the alternatives. The impact
estimates are shown in Table B-8. Adequate documentation of the basis for the
impacts was determined to be included in the PSD permit application.
The incremental cost impacts shown are the cost of the alternative
compared to the next most stringent control alternative. Figure B-2 is a plot
of the least-cost envelope defined by the list of control options.
VI.A.2.6. TOXICS ASSESSMENT
If SCR were applied, potential toxic emissions of ammonia could occur.
Ammonia emissions resulting from application of SCR could be as large as 20
tons per year. Application of SCR would reduce NOx by an additional 20 tpy
over steam injection alone (25 ppm)(not including ammonia emissions).
Another environmental impact considered was the spent catalyst which
would have to be disposed of at certain operating intervals. The catalyst
contains vanadium pentoxide, which is listed as a hazardous waste under RCRA
regulations (40 CFR 261.3). Disposal of this waste creates an additional
economic and environmental burden. This was considered in the applicant's
proposed BACT determination.
B.65
-------
TABLE B-8. EXAMPLE 1--SUMMARY OF TOP-DOWN BACT IMPACT ANALYSIS RESULTS FOR NOX
Emissions per Turbine Economic Impacts
Installed Total Average
Emissions capital annualized cost
Emissions reduction(a) cost(b) cost(c) efectiveness (d)
Control alternative (Ib/hr) (tpy) (tpy) ($) ($/yr) ($/ton)
13 ppm Alternative 44 22 260 11,470,000 1,717, OOO(g) 6,600
25 ppra Alternative 84 42 240 1,790,000 593,000 2,470
42 ppm Alternative 140 70 212 1,304,000 356,000 1,680
NSPS Alternative 312 156 126 927,000 288,000 2,285
Uncontrolled Baseline 564 282
Energy Impacts Environmental Impacts
Incremental Increase Adverse
cost over Toxics environmental
effectiveness(e) baseline(f) impact impact
($/ton) (MHBtu/yr) (Yes/No) (Yes/No)
56,200 464,000 Yes No
8,460 30,000 NO No
800 15,300 No No
8,000 No No
(a) Emissions reduction over baseline control level.
(b) Installed capital cost relative to baseline.
(c) Total annualized cost (capital, direct, and indirect) of purchasing, installing, and operating the proposed control alternative. A'capital
recovery factor approach using a real interest rate (i.e., absent inflation) is used to express capital costs in present-day annual costs.
(d) Average cost effectiveness over baseline is equal to total annualized cost for the control option divided by the emissions reductions resulting
from the uncontrolled baseline.
(e) The incremental cost effectiveness criteron is the same as the average cost effectiveness criteria except that the control alternative
is considered relative to the next most stringent alternative rather than the baseline control alternative.
(f) Energy impacts are the difference in total project energy requirements with the control alternative and the uncontrolled baseline expressed in ?
equivalent millions of Btus per year. j
(g) Assued 10 year catalyst life since this turbine operates only 1000 hours per year. Assumptions made on catalyst life lay have a profound affect ••
upon cost effectiveness.
-------
2,000,000
1,500,000
co
CD
-a
CD
_N
"co
3
I
1,000,000
500,000
0
DRAFT
OCTOBER 1990
13ppmi
NSPS
50 100 150 200 250 300
Emissions Reduction (tons per year)
Figure B-2. Least-Cost Envelope for Example 1
B.67
-------
DRAFT
OCTOBER 1990
VI.A.2.f. RATIONALE FOR PROPOSED BACT
Based on these impacts, the applicant proposed eliminating the 13 ppm
alternative as economically infeasible. The applicant documented that the
cost effectiveness is high at 6,600 $/ton, and well out of the range of recent
BACT NOx control costs for similar sources. The incremental cost
effectiveness of $56,200 also is high compared to the incremental cost
effectiveness of the next option.
The applicant documented that the other combustion turbine sources which
have applied SCR have much higher operating hours (i.e., all were permitted as
base-loaded units). Also, these sources had heat recovery steam generators so
that the cost effectiveness of the application of SCR was lower. For this
source, dilution air must be added to cool the flue gas to the proper
temperature. This increases the cost of the SCR system relative to the same
gas turbine with a HRSG. Therefore, the other sources had much lower cost
impacts for SCR relative to steam injection alone, and much lower cost
effectiveness numbers. Application of SCR would also result in emission of
ammonia, a toxic chemical, of possibly 20 tons per year while reducing NOx
emissions by 20 tons per year. The applicant asserted that, based on these
circumstances, to apply SCR in this case would be an unreasonable burden
compared to what has been done at other similar sources.
Consequently, the applicant proposed eliminating the SCR plus steam
injection alternative. The applicant then accepted the next control
alternative, steam injection to 25 ppmv. The use of steam injection was shown
by the applicant to be consistent with recent BACT determinations for similar
sources. The review authority concurred with the proposed elimination of SCR
and the selection of a 25 ppmv limit as BACT.
B.68
-------
DRAFT
OCTOBER 1990
VLB. EXAMPLE 2--COMBINED CYCLE GAS TURBINES FIRING NATURAL GAS
Table B-9 presents the design parameters for an alternative set of
circumstances. In this example, two gas turbines are being installed. Also,
the operating hours are 5000 per year and the new turbines are being added to
meet intermediate loads demands. The source will be limited through
enforceable conditions to the specified hours of operation and fuel type. In
this case, HRSG units are installed. The applicable control technologies and
control technology hierarchy are the same as the previous example except that
no dilution is required for the gas turbine exhaust because the HRSG serves to
reduce the exhaust temperature to the optimum level for SCR operation. Also,
since there is no dilution required and fewer startups, the most stringent
control option proposed is 9 ppm based on performance limits for several other
natural gas fired baseload combustion turbine facilities.
Table B-10 presents the results of the cost and economic impact analysis
for the example and Figure B-3 is a plot of the least-cost envelope defined by
the list of control options. The incremental cost impacts shown are the cost
of the alternative compared to the next most stringent control alternative.
Due to the increased operating hours and design changes, the economic impacts
of SCR are much lower for this case. There does not appear to be a persuasive
argument for stating that SCR is economically infeasible. Cost effectiveness
numbers are within the range typically required of this and other similar
source types.
In this case, there would also be emissions of ammonia. However, now the
magnitude of ammonia emissions, approximately 40 tons per year, is much lower
than the additional NOx reduction achieved, which is 270 tons per year.
Under these alternative circumstances, PM emissions are also now above
the significance level (i.e., greater than 25 tpy). The gas turbine
B.69
-------
DRAFT
OCTOBER 1990
TABLE B-9. EXAMPLE 2--COMBUSTION TURBINE DESIGN PARAMETERS
Characteristics
Number of emission units
Emission units
Cycle Type
Output
Gas Turbines (2 @ 75 MW each)
Steam Turbine (no emissions generated)
Fuel(s)
Gas Turbine Heat Rate, Btu/kw-hr
Fuel Flow per gas turbine, Btu/hr
Fuel Flow per gas turbine, Ib/hr
Service Type
Hours per year of operation
Uncontrolled Emissions per gas turbine, tpy (a)(b)
NOX
so2
CO
voc
PM
Gas Turbine
Combined-cycle
150 MW
70 MW
Natural Gas
11,000 Btu/kw-hr
1,650 mill ion
83,300
Intermediate
5000
1,410 (169 ppm)
<1
23 (6 ppm)
5
25 (0.0097 gr/dscf)
(a) Based on 5000 hours per year of operation.
(b) Total uncontrolled emissions for the proposed project is equal to the
pollutants uncontrolled emission rate multiplied by 2 turbines. For example
total NOX = (2 turbines) x 1410 tpy per turbine) = 2820 tpy.
B.70
-------
TABLE B-10. EXAMPLE 2--SUMMARY OF TOP-DOWN BACT IMPACT ANALYSIS RESULTS FOR NOX
Emissions per Turbine
Economic Impacts
Energy Impacts Environmental Impacts
Control alternative
Installed
Emissions capital
Emissions reduction(a,h) cost(b)
(Ib/hr) (tpy) (tpy) ($)
Total Average Incremental Increase Adverse
annualized cost cost over Toxics environmental
cost(c) effectiveness(d) effectiveness(e) baseline(f) impact impact
($/yr) ($/ton) ($/ton) (MHBtu/yr) (Yes/No) (Yes/No)
9 ppm Alternative
25 ppi Alternative
42 ppm Alternative
ro NSPS Alternative
-4
^ Uncontrolled Baseline
30
84
140
312
564
75
210
350
780
1,410
1,335
1,200
1,060
630
10,980,000
1,791,000
1,304,000
927,000
3,380, OOO(g)
1,730,000
883,000
805,000
2,531
1,440
833
1,280
12,200 160,000
6,050 105,000
181 57,200
27,000
Yes
No
No
NO
NO
No
No
NO
(a) Emissions reduction over baseline control level.
(b) Installed capital cost relative to baseline.
(c) Total annualized cost (capital, direct, and indirect) of purchasing, installing, and operating the proposed control alternative. A capital
recovery factor approach using a real interest rate (i.e., absent inflation) is used to express capital costs in present-day annual costs.
(d) Average cost Effectiveness over baseline is equal to total annualized cost for the control option divided by the emissions reductions resulting
from the uncontrolled baseline.
(e) The optional incremental cost effectiveness criteron is the same as the average cost effectiveness criteria except that the control alternative
is considered relative to the next most stringent alternative rather than the baseline control alternative.
(f) Energy impacts are the difference in total project energy requirements with the control alternative uncontrolled baseline expressed in
equivalent millions of Btus per year.
(g) Assumes a 2 year catalyst life. Assumptions made on catalyst life may have a profound affect upon cost effectiveness.
(h) Since the project calls for two turbines, actual project wide emissions reductions for an alternative will be equal to two times the reduction
listed.
-------
ca
CD
CD
Q.
O
o
T3
CD
.y
In
3
c
1
4,000,000
3,000,000
2,000,000
1,000,000
DRAFT
OCTOBER 1990
9ppml
NSPS
0 200 400 600 800 1,000 1,200 1,400 1,600
Emissions Reduction (tons per year)
Figure B-3. Least-Cost Envelope for Example 2
B.72
-------
DRAFT
OCTOBER 1990
combustors are designed to burn the fuel as completely as possible and
therefore reduce PM to the lowest possible level. Natural gas contains no
solids and solids are removed from the injected water. The PM emission rate
without add-on controls is on the same order (0.009 gr/dscf) as that for other
particulate matter sources controlled with stringent add-on controls (e.g.,
fabric filter). Since the applicant documented that precombustion or add-on
controls for PM have never been required for natural gas fired turbines, the
reviewing agency accepted the applicants analysis that natural gas firing was
BACT for PM emissions and that no additional analysis of PM controls was
required.
VI.C. EXAMPLE 3-COMBINED CYCLE GAS TURBINE FIRING DISTILLATE OIL
In this example, the same combined cycle gas turbines are proposed
except that distillate oil is fired rather than natural gas. The reason is
that natural gas is not available on site and there is no pipeline within a
reasonable distance. The fuel change raises two issues; the technical
feasibility of SCR in gas turbines firing sulfur bearing fuel, and NOx levels
achievable with water injection while firing fuel oil.
In this case the applicant proposed to eliminate SCR as technically
infeasible because sulfur present in the fuel, even at low levels, will poison
the catalyst and quickly render it ineffective. The applicant also noted that
there are no cases in the U.S. where SCR has been applied to a gas turbine
firing distillate oil as the primary fuel.4
A second issue would be the most stringent NOx control level achievable
with wet injection. For oil firing the applicant has proposed 42 ppm at
15 percent oxygen. Due to flame characteristics inherent with oil firing, and
limits on the amount of water or steam that can be injected, 42 ppm is the
lowest NOx emission level achievable with distillate oil firing. Since
4 Though this argument was considered persuasive in this case, advances
in catalyst technology have now made SCR with oil firing technically feasible.
B.73
-------
DRAFT
OCTOBER 1990
natural gas is not available and SCR is technically infeasible, 42 ppm is the
most stringent alternative considered. Based on the cost effectiveness of wet
injection, approximately 833 $/ton, there is no economic basis to eliminate
the 42 ppm option since this cost is well within the range of BACT costs for
NOx control. Therefore, this option is proposed as BACT.
The switch to oil from gas would also result in S02, CO, PM, and
beryllium emissions above significance levels. Therefore, BACT analyses would
also be required for these pollutants. These analyses are not shown in this
example, but would be performed in the same manner as the BACT analysis for
NOx.
VI.D. OTHER CONSIDERATIONS
The previous judgements concerning economic feasibility were in an area
meeting NAAQS for both NOx and ozone. If the natural gas fired simple cycle
gas turbine example previously presented were sited adjacent to a Class I
area, or where air quality improvement poses a major challenge, such as next
to a nonattainment area, the results may differ. In this case, even though
the region of the actual site location is achieving the NAAQS, adherence to a
local or regional NOx or ozone attainment strategy might result in the
determination that higher costs than usual are appropriate. In such
situations, higher costs (e.g., 6,600 $/ton) may not necessarily be persuasive
in eliminating SCR as BACT.
While it is not the intention of BACT to prevent construction, it is
possible that local or regional air quality management concerns regarding the
need to minimize the air quality impacts of new sources would lead the
permitting authority to require a source to either achieve stringent emission
control levels or, at a minimum, that control cost expenditures meet certain
cost levels without consideration of the resultant economic impact to the
source.
B.74
-------
DRAFT
OCTOBER 1990
Besides local or regional air quality concerns, other site constraints
may significantly impact costs of particular control technologies. For the
examples previously presented, two factors of concern are land and water
availability.
The cost of the raw water is usually a small part of the cost of wet
controls. However, gas turbines are sometimes located in remote locations.
Though water can obviously be trucked to any location, the costs may be very
high.
Land availability constraints may occur where a new source is being
located at an existing plant. In these cases, unusual design and additional
structural requirements could make the costs of control technologies which are
commonly affordable prohibitively expensive. Such considerations may be
pertinent to the calculations of impacts and ultimately the selection of BACT.
B.75
-------
DRAFT
OCTOBER 1990
CHAPTER C
THE AIR QUALITY ANALYSIS
I. INTRODUCTION
An applicant for a PSD permit is required to conduct an air quality
analysis of the ambient impacts associated with the construction and operation
of the proposed new source or modification. The main purpose of the air
quality analysis is to demonstrate that new emissions emitted from a proposed
major stationary source or major modification, in conjunction with other
applicable emissions from existing sources (including secondary emissions from
growth associated with the new project), will not cause or contribute to a
violation of any applicable NAAQS or PSD increment. Ambient impacts of
noncriteria pollutants must also be evaluated.
A separate air quality analysis must be submitted for each regulated
pollutant if the applicant proposes to emit the pollutant in a significant
amount from a new major stationary source, or proposes to cause a significant
net emissions increase from a major modification (see Table I-A-4, chapter A
of this part). [Note: The air quality analysis requirement also applies to
any pollutant whose rate of emissions from a proposed new or modified source
is considered to be "significant" because the proposed source would construct
within 10 kilometers of a Class I area and would have an ambient impact on
such area equal to or greater than 1 pg/m , 24-hour average.] Regulated
pollutants include (1) pollutants for which a NAAQS exists (criteria
pollutants) and (2) other pollutants, which are regulated by EPA, for which no
NAAQS exist (noncriteria pollutants).
Each air quality analysis will be unique, due to the variety of sources
and meteorological and topographical conditions that may be involved.
Nevertheless, the air quality analysis must be accomplished in a manner
consistent with the requirements set forth in either EPA's PSD regulations
under 40 CFR 52.21, or a State or local PSD program approved by EPA pursuant
to 40 CFR 51.166. Generally, the analysis will involve (1) an assessment of
existing air quality, which may include ambient monitoring data and air
quality dispersion modeling results, and (2) predictions, using dispersion
C.I
-------
DRAFT
OCTOBER 1990
modeling, of ambient concentrations that will result from the applicant's
proposed project and future growth associated with the project.
In describing the various concepts and procedures involved with the air
quality analysis in this section, it is assumed that the reader has a basic
understanding of the principles involved in collecting and analyzing ambient
monitoring data and in performing air dispersion modeling. Considerable
guidance is contained in EPA's Ambient Monitoring Guidelines for Prevention of
Significant Deterioration [Reference 1] and Guideline on Air Quality Models
(Revised) [Reference 2] . Numerous times throughout this chapter, the reader
will be referred to these guidance documents, hereafter referred to as the PSD
Monitoring Guideline and the Modeling Guideline, respectively.
In addition, because of the complex character of the air quality
analysis and the site-specific nature of the modeling techniques involved,
applicants are advised to review the details of their proposed modeling
analysis with the appropriate reviewing agency before a complete PSD
application is submitted. This is best done using a modeling protocol. The
modeling protocol should be submitted to the reviewing agency for review and
approval prior to commencing any extensive analysis. Further description of
the modeling protocol is contained in this chapter.
The PSD applicant should also be aware that, while this chapter focuses
primarily on compliance with the NAAQS and PSD increments, additional impact
analyses are required under separate provisions of the PSD regulations for
determining any impairment to visibility, soils and vegetation that might
result, as well as any adverse impacts to Class I areas. These provisions are
described in the following chapters D and E, respectively.
C.2
-------
DRAFT
OCTOBER 1990
II. NATIONAL AMBIENT AIR QUALITY STANDARDS AND PSD INCREMENTS
As described in the introduction to this chapter, the air quality
analysis is designed to protect the national ambient air quality standards
(NAAQS) and PSD increments. The NAAQS are maximum concentration "ceilings"
measured in terms of the total concentration of a pollutant in the atmosphere
(See Table C-l). For a new or modified source, compliance with any NAAQS is
based upon the total estimated air quality, which is the sum of the ambient
estimates resulting from existing sources of air pollution (modeled source
impacts plus measured background concentrations, as described in this section)
and the modeled ambient impact caused by the applicant's proposed emissions
increase (or net emissions increase for a modification) and associated growth.
A PSD increment, on the other hand, is the maximum allowable increase in
concentration that is allowed to occur above a baseline concentration for a
pollutant (see section II.E). The baseline concentration is defined for each
pollutant (and relevant averaging time) and, in general, is the ambient
concentration existing at the time that the first complete PSD permit
application affecting the area is submitted. Significant deterioration is
said to occur when the amount of new pollution would exceed the applicable PSD
increment. It is important to note, however, that the air quality cannot
deteriorate beyond the concentration allowed by the applicable NAAQS, even if
not all of the PSD increment is consumed.
II.A CLASS I, II, AND III AREAS AND INCREMENTS.
The PSD requirements provide for a system of area classifications which
affords States an opportunity to identify local land use goals. There are
three area classifications. Each classification differs in terms of the
amount of growth it will permit before significant air quality deterioration
would be deemed to occur. Class I areas have the smallest increments and thus
allow only a small degree of air quality deterioration. Class II areas can
C.3
-------
DRAFT
OCTOBER 1990
TABLE C-l. National Ambient Air Quality Standards
Primary Secondary
Pollutant/averaging time Standard Standard
Participate Matter
o PMin annual3. 50 ng/rn^ 50
o PMjjj; 24-hourD 150 fig/m6 150
Sulfur Dioxide
o S0?, annual0. 80 /zg/m3 (0.03 ppm)
o SO^, 24-hour.a 365 0g/nT (0.14 ppm) ,
o SO^, 3 -hour" 1,300 ng/m* (0.5 ppm)
Nitrogen Dioxide
o N02> annual0 0.053 ppm (100 #g/m3) 0.053 ppm (100 pg/m3
Ozone
o 03, l-hourb 0.12 ppm (235 #g/m3) 0.12 ppm (235 #g/m3)
Carbon Monoxide
o CO, 8-hourd 9 ppm (10 mg/m3)
o CO, 1-hour 35 ppm (40 mg/m3)
Lead
o Pb, calendar quarter0 1.5
a Standard is attained when the expected annual arithmetic mean is less than
or equal to 50 tig/m3.
b Standard is attained when the expected number of exceedances is less than or
equal to 1.
c Never to be exceeded.
d Not to be exceeded more than once per year.
C.4
-------
DRAFT
OCTOBER 1990
accommodate normal well-managed industrial growth. Class III areas have the
largest increments and thereby provide for a larger amount of development than
either Class I or Class II areas.
Congress established certain areas, e.g., .wilderness areas and national
parks, as mandatory Class I areas. These areas cannot be redesignated to any
other area classification. All other areas of the country were initially
designated as Class II. Procedures exist under the PSD regulations to
redesignate the Class II areas to either Class I or Class III, depending upon
a State's land management objectives.
PSD increments for S02 and particulate matter—measured as total
suspended particulate (TSP)--have existed in their present form since 1978.
On July 1, 1987, EPA revised the NAAQS for particulate matter and established
the new PM-10 indicator by which the NAAQS are to be measured. (Since each
State is required to adopt these revised NAAQS and related implementation
requirements as part of the approved implementation plan, PSD applicants
should check with the appropriate permitting agency to determine whether such
State action has already been taken. Where the PM-10 NAAQS are not yet being
implemented, compliance with the TSP-based ambient standards is still required
in accordance with the currently-approved State implementation plan.)
Simultaneously with the promulgation of the PM-10 NAAQS, EPA announced that it
would develop PM-10 increments to replace the TSP increments. Such new
increments have not yet been promulgated, however. Thus the national PSD
increment system for particulate matter is still based on the TSP indicator.
The EPA promulgated PSD increments for N02 on October 17, 1988. These
new increments become effective under EPA's PSD regulations (40 CFR 52.21) on
November 19, 1990, although States may have revised their own PSD programs to
incorporate the new increments for N02 on some earlier date. Until
November 19, 1990, PSD applicants should determine whether the N02 increments
are being implemented in the area of concern; if so, they must include the
necessary analysis, if applicable, as part of a complete permit application.
[NOTE: the "trigger date" (described below in section II.B) for the N02
increments has been established by regulation as of February 8, 1988. This
applies to all State PSD programs as well as EPA's Part 52 PSD program. Thus,
C.5
-------
DRAFT
OCTOBER 1990
consumption of the N02 increments may actually occur before the increments
become effective in any particular PSD program.] The PSD increments for S02,
TSP and N02 are summarized in Table C-2.
II.B ESTABLISHING THE BASELINE DATE
As already described, the baseline concentration is the reference point
for determining air quality deterioration in an area. The baseline
concentration is essentially the air quality existing at the time of the first
complete PSD permit application submittal affecting that area. In general,
then, the submittal date of the first complete PSD application in an area is
the "baseline date." On or before the date of the first PSD application, most
emissions are considered to be part of the baseline concentration, and
emissions changes which occur after that date affect the amount of available
PSD increment. However, to fully understand how and when increment is
consumed or expanded, three different dates related to baseline must be
explained. In chronological order, these dates are as follows:
the major source baseline date',
the trigger date; and
the minor source baseline date.
The major source baseline date is the date after which actual emissions
associated with construction (i.e., physical changes or changes in the method
of operation) at a major stationary source affect the available PSD increment.
Other changes in actual emissions occurring at any source after the major
source baseline date do not affect the increment, but instead (until after the
minor source baseline date is established) contribute to the baseline
concentration. The trigger date is the date after which the minor source
C.6
-------
TABLE C-2. PSD INCREMENTS
(//g/m3)
Class I Class II
Sulfur Dioxide
o S02, annual3 2 20
o S02, 24-hourb 5 91
o S02 3-hourb 25 512
Participate Matter
o TSP, annual3 5 19
o TSP, 24-hourb 10 37
Nitrogen Dioxide
o N02, annual3 2.5 25
DRAFT
OCTOBER 1990
Class III
40
182
700
37
75
50
a Never to be exceeded.
b Not to be exceeded more than once per year,
C.7
-------
DRAFT
OCTOBER 1990
baseline date (described below) may be established. Both the major source
baseline date and the trigger date are fixed dates, although different dates
apply to (1) S02 and particulate matter, and (2) N02, as follows:
Pollutant Ma.ior Source Baseline Date Trigger Date
PM January 6, 1975 August 7, 1977
S02 January 6, 1975 August 7, 1977
N02 February 8, 1988 February 8, 1988
The minor source baseline date is the earliest date after the trigger
date on which a complete PSD application is received by the permit reviewing
agency. If the application that established the minor source baseline date is
ultimately denied or is voluntarily withdrawn by the applicant, the minor
source baseline date remains in effect nevertheless. Because the date marks
the point in time after which actual emissions changes from all sources affect
the available increment (regardless of whether the emissions changes are a
result of construction), it is often referred to as the "baseline date."
The minor source baseline date for a particular pollutant is triggered
by a PSD applicant only if the proposed increase in emissions of that
pollutant is significant. For instance, a PSD application for a major new
source or modification that proposes to increase its emissions in a
significant amount for S02, but in an insignificant amount for PM, will
establish the minor source baseline date for S02 but not for PM. Thus, the
minor source baseline dates for different pollutants (for which increments
exist) need not be the same in a particular area.
C.8
-------
DRAFT
OCTOBER 1990
II.C ESTABLISHING THE BASELINE AREA
The area in which the minor source baseline date is established by a PSD
permit application is known as the baseline area. The extent of a baseline
area is limited to intrastate areas and may include one or more areas
designated as attainment or unclassified under Section 107 of the Act. The
baseline area established pursuant to a specific PSD application is to include
1) all portions of the attainment or unclassifiable area in which the PSD
applicant would propose to locate, and 2) any attainment or unclassifiable
area in which the proposed emissions would have a significant ambient impact.
2
For this purpose, a significant impact is defined as at least a 1 jug/m annual
increase in the average annual concentration of the applicable pollutant.
Again, a PSD applicant's establishment of a baseline area in one State does
not trigger the minor source baseline date in, or extend the baseline area
into, another State.
II.D REDEFINING BASELINE AREAS (AREA REDESIGNATIONS)
It is possible that the boundaries of a baseline area may not reasonably
reflect the area affected by the PSD source which established the baseline
area. A state may redefine the boundaries of an existing baseline area by
redesignating the section 107 areas contained therein. Section 107(d) of the
Clean Air Act specifically authorizes states to submit redesignations to the
EPA. Consequently, a State may submit redefinitions of the boundaries of
attainment or unclassifiable areas at any time, as long as the following
criteria are met:
2
area redesignations can be no smaller than the 1 ng/m area of
impact of the triggering source-, and
the boundaries of any redesignated area cannot intersect the
1 [ig/m3 area of impact of any major stationary source that
established or would have established a minor source baseline date
for the area proposed for redesignation.
C.9
-------
DRAFT
OCTOBER 1990
II.E INCREMENT CONSUMPTION AND EXPANSION
The amount of PSD increment that has been consumed in a PSD area is
determined from the emissions increases and decreases which have occurred from
sources since the applicable baseline date. It is useful to note, however,
that in order to determine the amount of PSD increment consumed (or the amount
of available increment), no determination of the baseline concentration needs
to be made. Instead, increment consumption calculations must reflect only the
ambient pollutant concentration change attributable to increment-affecting
emissions.
Emissions increases that consume a portion of the applicable increment
are, in general, all those not accounted for in the baseline concentration and
specifically include:
actual emissions increases occurring after the major source
baseline date, which are associated with physical changes or
changes in the method of operation (i.e., construction) at a major
stationary source; and
actual emissions increases at any stationary source, area source,
or mobile source occurring after the minor source baseline date.
The amount of available increment may be added to, or "expanded," in two
ways. The primary way is through the reduction of actual emissions from any
source after the minor source baseline date. Any such emissions reduction
would increase the amount of available increment to the extent that ambient
concentrations would be reduced.
Increment expansion may also result from the reduction of actual
emissions after the major source baseline date, but before the minor source
baseline date, if the reduction results from a physical change or change in
the method of operation (i.e., construction) at a major stationary source.
Moreover, the reduction will add to the available increment only if the
reduction is included in a federally enforceable permit or SIP provision.
Thus, for major stationary sources, actual emissions reductions made prior to
C.10
-------
DRAFT
OCTOBER 1990
the minor source baseline date expand the available increment just as
increases before the minor source baseline date consume increment.
The creditable increase of an existing stack height or the application
of any other creditable dispersion technique may affect increment consumption
or expansion in the same manner as an actual emissions increase or decrease.
That is, the effects that a change in the effective stack height would have on
ground level pollutant concentrations generally should be factored into the
increment analysis. For example, this would apply to a raised stack height
occurring in conjunction with a modification at a major stationary source
prior to the minor source baseline date, or to any changed stack height
occurring after the minor source baseline date. It should be noted, however,
that any increase in a stack height, in order to be creditable, must be
consistent with the EPA's stack height regulations; credit cannot be given for
that portion of the new height which exceeds the height demonstrated to be the
good engineering practice (GEP) stack height.
Increment consumption (and expansion) will generally be based on changes
in actual emissions reflected by the normal source operation for a period of 2
years. However, if little or no operating data are available, as in the case
of permitted emission units not yet in operation at the time of the increment
analysis, the potential to emit must be used instead. Emissions data
requirements for modeling increment consumption are described in
Section IV.D.4. Further guidance for identifying increment-consuming sources
(and emissions) is provided in Section JI/.C.2.
C.ll
-------
DRAFT
OCTOBER 1990
II.F BASELINE DATE AND BASELINE AREA CONCEPTS -- EXAMPLES
An example of how a baseline area is established is illustrated in
Figure C-l. A major new source with the potential to emit significant amounts
of S02 proposes to locate in County C. The applicant submits a complete PSD
application to the appropriate reviewing agency on October 6, 1978. (The
trigger date for S02 is August 7, 1977.) A review of the State's S02
attainment designations reveals that attainment status is listed by individual
counties in the state. Since County C is designated attainment for S02, and
the source proposes to locate there, October 6, 1978 is established as the
minor source baseline date for SOp for the entire county.
Dispersion modeling of proposed S02 emissions in accordance with
approved methods reveals that the proposed source's ambient impact will exceed
1 ug/m (annual average) in Counties A and B. Thus, the same minor source
baseline date is also established throughout Counties A and B. Once it is
triggered, the minor source baseline date for Counties A, B and C establishes
the time after which all emissions changes affect the available increments in
those three counties.
Although S02 impacts due to the proposed emissions are above the
significance level of 1 /ig/m3 (annual average) in the adjoining State, the
proposed source does not establish the minor source baseline date in that
State. This is because, as mentioned in Section II.C of this chapter,
baseline areas are intrastate areas only.
The fact that a PSD source^s emissions cannot trigger the minor source
baseline date across a State's boundary should not be interpreted as
precluding the applicant's emissions from consuming increment in another
State. Such increment-consuming emissions (e.g., S02 emissions increases
resulting from a physical change or a change in the method of operation at a
C.12
-------
DRAFT
OCTOBER 1990
County D
i- S02 Attainment
County E
S02 Unclassified
Baseline Date Triggered 10/6/78
— State line
•••• County line
Figure C-1. Establishing the Baseline Area.
C.13
-------
DRAFT
OCTOBER 1990
major stationary source after January 6, 1975) that affect another State will
consume increment there even though the minor source base!ine date has not
been triggered, but are not considered for increment-consuming purposes until
after the minor source baseline date has been independently established in
that State. A second example, illustrated in Figure C-2, demonstrates how a
baseline area may be redefined. Assume that the State in the first example
decides that it does not want the minor source baseline date to be established
in the western half of County A where the proposed source will not have a
o
significant annual impact (i.e., 1 fig/m , annual average). The State,
therefore, proposes to redesignate the boundaries of the existing section 107
attainment area, comprising all of County A, to create two separate attainment
areas in that county. If EPA agrees that the available data support the
change, the redesignations will be approved. At that time, the October 6,
1978 minor source baseline date will no longer apply to the newly-established
attainment area comprising the western portion of County A.
If the minor source baseline date has not been triggered by another PSD
application having a significant impact in the redesignated western portion of
County A, the S02 emissions changes occurring after October 6, 1978 from minor
point, area, and mobile sources, and from nonconstruction-related activities
at all major stationary sources in this area will be transferred into the
baseline concentration. In accordance with the major source baseline date,
construction-related emissions changes at major point sources continue to
consume or expand increment in the western portion of County A which is no
longer part of the original baseline area.
C.14
-------
DRAFT
OCTOBER 1990
Redesignated Attainment Areas
County E
SOa Unclassified ..-'
County D
r— S02 Attainment
Baseline Date Triggered 10/6/78
State line
County line
Figure C-2. Redefining the Baseline Area.
C.15
-------
DRAFT
OCTOBER 1990
III. AMBIENT DATA REQUIREMENTS
An applicant should be aware of the potential need to establish and
operate a site-specific monitoring network for the collection of certain
ambient data. With respect to air quality data, the PSD regulations contain
provisions requiring an applicant to provide an ambient air quality analysis
which may include pre-application monitoring data, and in some instances post-
construction monitoring data, for any pollutant proposed to be emitted in
significant amounts by the new source or modification. In the absence of
available monitoring data which is representative of the area of concern, this
requirement could involve the operation of a site-specific air quality
monitoring network by the applicant. Also, the need for meteorological data,
for any dispersion modeling that must be performed, could entail the
applicant's operation of a site-specific meteorological network.
Pre-application data generally must be gathered over a period of at
least 1 year and the data are to represent at least the 12-month period
immediately preceding receipt of the PSD application. Consequently, it is
important that the applicant ascertain the need to collect any such data and
proceed with the required monitoring activities as soon as possible in order
to avoid undue delay in submitting a complete PSD application.
III.A PRE-APPLICATION AIR QUALITY MONITORING
For any criteria pollutant that the applicant proposes to emit in
significant amounts, continuous ambient monitoring data may be required as
part of the air quality analysis. If, however, either (1) the predicted
ambient impact, i.e., the highest modeled concentration for the applicable
averaging time, caused by the proposed significant emissions increase (or
significant net emissions increase), or (2) the existing ambient pollutant
concentrations are less than the prescribed significant monitoring value (see
Table C-3), the-permitting agency has discretionary authority to exempt an
applicant from this data requirement.
C.16
-------
DRAFT
OCTOBER 1990
TABLE C-3. SIGNIFICANT MONITORING CONCENTRATIONS
Pollutant
Air Quality Concentration
and Averaging Time
Carbon monoxide
Nitrogen dioxide
Sulfur dioxide
Particulate Matter, TSP
Particulate Matter, PM-10
Ozone
Lead
Asbestos
Beryl 1 i urn
Mercury
Vinyl chloride
Fluorides
Sulfuric acid mist
Total reduced sulfur (including H2S)
Reduced sulfur (including hLS)
Hydrogen sulfide
575
14
13
10
10
a
0.1
b
0.001
0.25
15
0.25
b
b
b
0.2
(8-hour)
(Annual)
(24-hour)
(24-hour)
(24-hour)
(3-month)
(24-hour)
(24-hour)
(24-hour)
(24-hour)
(1-hour)
3 No significant air quality concentration for ozone monitoring has been established. Instead, applicants
with a net emissions increase of 100 tons/year or more of VOC's subject to PSD would be required to perform
an ambient impact analysis, including pre-application monitoring data.
b Acceptable monitoring techniques may not be available at this time. Monitoring requirements for this
pollutant should be discussed with the permitting agency.
C.17
-------
DRAFT
OCTOBER 1990
The determination of the proposed project's effects on air quality (for
comparison with the significant monitoring value) is based on the results of
the dispersion modeling used for establishing the impact area (see Section
IV.B of this chapter). Modeling by itself or in conjunction with available
monitoring data should be used to determine whether the existing ambient
concentrations are equal to or greater than the significant monitoring value.
The applicant may utilize a screening technique for this purpose, or may elect
to use a refined model. Consultation with the permitting agency is advised
before any model is selected. Ambient impacts from existing sources are
estimated using the same model input data as are used for the NAAQS analysis,
as described in section IV.D.4 of this chapter.
If a potential threat to the NAAQS is identified by the modeling
predictions, then continuous ambient monitoring data should be required, even
when the predicted impact of the proposed project is less than the significant
monitoring value. This is especially important when the modeled impacts of
existing sources are uncertain due to factors such as complex terrain and
uncertain emissions estimates.
Also, if the location of the proposed source or modification is not
affected by other major stationary point sources, the assessment of existing
ambient concentrations may be done by evaluating available monitoring data.
It is generally preferable to use data collected within the area of concern;
however, the possibility of using measured concentrations from representative
"regional" sites may be discussed with the permitting agency. The
PSD Monitoring Guideline provides additional guidance on the use of such
regional sites.
Once a determination is made by the permitting agency that ambient
monitoring data must be submitted as part of the PSD application, the
requirement can be satisfied in one of two ways. First, under certain
conditions, the applicant may use existing ambient data. To be acceptable,
such data must be judged by the permitting agency to be representative of the
air quality for the area in which the proposed project would construct and
operate. Although a State or local agency may have monitored air quality for
C.18
-------
DRAFT
OCTOBER 1990
several years, the data collected by such efforts may not necessarily be
adequate for the preconstruction analysis required under PSD. In determining
the representativeness of any existing data, the applicant and the permitting
agency must consider the following critical items (described further in the
PSD Monitoring Guideline):
monitor location;
quality of the data; and
currentness of the data.
If existing data are not available, or they are judged not to be
representative, then the applicant must proceed to establish a site-specific
monitoring network. The EPA strongly recommends that the applicant prepare a
monitoring plan before any actual monitoring begins. Some permitting agencies
may require that such a plan be submitted to them for review and approval. In
any case, the applicant will want to avoid any possibility that the resulting
data are unacceptable because of such things as improperly located monitors,
or an inadequate number of monitors. To assure the accuracy and precision of
the data collected, proper quality assurance procedures pursuant to Appendix B
of 40 CFR Part 58 must also be followed. The recommended minimum contents of
a monitoring plan, and-a discussion of the various considerations to be made
in designing a PSD monitoring network, are contained in the PSD Monitoring
Guideline,
The PSD regulations generally require that the applicant collect 1 year
of ambient data (EPA recommends 80 percent data recovery for PSD purposes).
However, the permitting agency has discretion to accept data collected over a
shorter period of time (but in no case less than 4 months) if a complete and
adequate analysis can be accomplished with the resulting data. Any decision
to approve a monitoring period shorter than 1 year should be based on a
demonstration by the applicant (through historical data or dispersion
modeling) that the required air quality data will be obtained during a time
period, or periods, when maximum ambient concentrations can be expected.
C.19
-------
DRAFT
OCTOBER 1990
For a pollutant for which there is no NAAQS (i.e., a noncriteria
pollutant), EPA's general position is to not require monitoring data, but to
base the air quality analysis on modeled impacts. However, the permitting
agency may elect to require the submittal of air quality monitoring data for
noncriteria pollutants in certain cases, such' as where:
a State has a standard for a non-criteria pollutant;
the reliability of emissions data used as input to modeling
existing sources is highly questionable; and
available models or complex terrain make it difficult to
estimate air quality or the impact of the proposed or
modification.
The applicant will need to confer with the permitting agency to determine
whether any ambient monitoring may be required. Before the agency exercises
its discretion to require such monitoring, there should be an acceptable
measurement method approved by EPA or the appropriate permitting agency.
With regard to particulate matter, where two different indicators of the
pollutant are being regulated, EPA considers the PM-10 indicator to represent
the criteria form of the pollutant (the NAAQS are now expressed in terms of
ambient PM-10 concentrations) and TSP is viewed as the non-criteria form.
Consequently, EPA intends to apply the pre-application monitoring requirements
to PM-10 primarily, while treating TSP on a discretionary basis in light of
its noncriteria status. Although the PSD increments for particulate matter
are still based on the TSP indicator, modeling data, not ambient monitoring
data, are used for increment analyses.
Ambient air quality data collected by the applicant must be presented in
the PSD application as part of the air quality analysis. Monitoring data
collected for a criteria pollutant may be used in conjunction with dispersion
modeling results to demonstrate NAAQS compliance. Each PSD application
involves its own unique set of factors, i.e., the integration of measured
ambient data and modeled projections. Consequently, the amount of data to be
C.20
-------
DRAFT
OCTOBER 1990
used and the manner of presentation are matters that should be discussed with
the permitting agency.
III.B POST-CONSTRUCTION AIR QUALITY MONITORING
The PSD Monitoring Guideline recommends that post-construction
monitoring be done when there is a valid reason, such as (1) when the NAAQS
are threatened, and (2) when there are uncertainties in the data bases for
modeling. Any decision to require post-construction monitoring will generally
be made after the PSD application has been thoroughly reviewed. It should be
noted that the PSD regulations do not require that the significant monitoring
concentrations be considered by the permitting agency in determining the need
for post-construction monitoring.
Existing monitors can be considered for collecting post-construction
ambient data as long as they have been approved for PSD monitoring purposes.
However, the location of the monitors should be checked to ascertain their
appropriateness if other new sources or modifications have subsequently
occurred, because the new emissions from the more recent projects could alter
the location of points of maximum ambient concentrations where ambient
measurements need to be made.
Generally, post-construction monitoring should not begin until the
source is operating near intended capacity. If possible the collection of
data should be delayed until the source is operating at a rate equal to or
greater than 50 percent of design capacity. The PSD Monitoring Guideline
provides, however, that in no case should post-construction monitoring be
delayed later than 2 years after the start-up of the new source or
modification.
Post-approval ozone monitoring is an alternative to pre-application
monitoring for applicants proposing to emit VOC's if they choose to accept
nonattainment preconstruction review requirements, including LAER, emissions
and air quality offsets, and statewide compliance of other sources under the
same ownership. As indicated in Table C-3, pre-application monitoring for
C.21
-------
DRAFT
OCTOBER 1990
ozone is required when the proposed source or modification would emit at least
100 tons per year of volatile organic compounds (VOC). Note that this
emissions rate for VOC emissions is a surrogate for the significant monitoring
concentration for the pollutant ozone (see Table C-3). Under
40 CFR 52.21(m)(l)(vi), post-approval monitoring data for ozone is required
(and cannot be waived) in conjunction with the aforementioned nonaitainment
review requirements when the permitting agency waives the requirement for pre-
application ozone monitoring data. The post-approval period may begin any
time after the source receives its PSD permit. In no case should the post-
approval monitoring be started later than 2 years after the start-up of the
new source or modification.
III.C METEOROLOGICAL MONITORING
Meteorological data is generally needed for model input as part of the
air quality analysis. It is important that such data be representative of the
atmospheric dispersion and climatological conditions at the site of the
proposed source or modification, and at locations where the source may have a
significant impact on air quality. For this reason, site specific data are
preferable to data collected elsewhere. On-site meteorological monitoring may
be required, even when on-site air quality monitoring is not.
The PSD Monitoring Guideline should be used to establish locations for
any meteorological monitoring network that the applicant may be required to
operate and maintain as part of the preconstruction monitoring requirements.
That guidance specifies the meteorological instrumentation to be used in
measuring meteorological parameters such as wind speed, wind direction, and
temperature. The PSD Monitoring Guideline also provides that the retrieval
of valid wind/stability data should not fall below 90 percent on an annual
basis. The type, quantity, and format of the required data will be influenced
by the specific input requirements of the dispersion modeling techniques used
in the air quality analysis. Therefore, the applicant will need to consult
with the permitting agency prior to establishing the required network.
C.22
-------
DRAFT
OCTOBER 1990
Additional guidance for the collection and use of on-site data is
provided in the PSD Monitoring Guideline, Also, the EPA documents entitled
Qn-Site Meteorological Program Guidance for Regulatory Modeling Applications
(Reference 3), and Volume IV of the series of reports entitled Qua!itv
Assurance Handbook for Air Pollution Measurement Systems (Reference 4),
contain information required to ensure the quality of the meteorological
measurements collected.
C.23
-------
DRAFT
OCTOBER 1990
IV. DISPERSION MODELING ANALYSIS
Dispersion models are the primary tools used in the air quality
analysis. These models estimate the ambient concentrations that will result
from the PSD applicant's proposed emissions in combination with emissions from
existing sources. The estimated total concentrations are used to demonstrate
compliance with any applicable NAAQS or PSD increments. The applicant should
consult with the permitting agency to determine the particular requirements
for the modeling analysis to assure acceptability of any air quality modeling
technique(s) used to perform the air quality analysis contained in the PSD
application.
IV.A OVERVIEW OF THE DISPERSION MODELING ANALYSIS
The dispersion modeling analysis usually involves two distinct phases:
(1) a preliminary analysis and (2) a full impact analysis. The preliminary
analysis models only the significant increase in potential emissions of a
pollutant from a proposed new source, or the significant net emissions
increase of a pollutant from a proposed modification. The results of this
preliminary analysis determine whether the applicant must perform a full
impact analysis, involving the estimation of background pollutant
concentrations resulting from existing sources and growth associated with the
proposed source. Specifically, the preliminary analysis:
determines whether the applicant can forego further air quality
analyses for a particular pollutant;
may allow the applicant to be exempted from the ambient monitoring
data requirements (described in section III of this chapter); and
/s used to define the impact area within which a full impact
analysis must be carried out.
The EPA does not require a full impact analysis for a particular
pollutant when emissions of that pollutant from a proposed source or
modification would not increase ambient concentrations by more than prescribed
significant ambient impact levels, including special Class I significance
C.24
-------
DRAFT
OCTOBER 1990
levels. However, the applicant should check any applicable State or local PSD
program requirements in order to determine whether such requirements may
contain any different procedures which may be more stringent. In addition,
the applicant must still address the requirements for additional impacts
required under separate PSD requirements, as described in Chapters D and E
which follow this chapter.
A full impact analysis is required for any pollutant for which the
proposed source's estimated ambient pollutant concentrations exceed prescribed
significant ambient impact levels. This analysis expands the preliminary
analysis in that it considers emissions from:
the proposed source;
existing sources;
residential, commercial, and industrial growth that accompanies
the new activity at the new source or modification (i.e.,
secondary emissions).
For SOp, particulate matter, and NOp, the full impact analysis actually
consists of separate analyses for the NAAQS and PSD increments. As described
later in this section, the selection of background sources (and accompanying
emissions) to be modeled for the NAAQS and increment components of the overall
analysis proceeds under somewhat different sets of criteria. In general,
however, the full impact analysis is used to project ambient pollutant
concentrations against which the applicable NAAQS and PSD increments are
compared, and to assess the ambient impact of non-criteria pollutants.
The reviewer's primary role is to determine whether the applicant
selected the appropriate model(s), used appropriate input data, and followed
recommended procedures to complete the air quality analysis. Appendix C in
the Modeling Guideline provides an example checklist which recommends a
standardized set of data to aid the reviewer in determining the completeness
and correctness of an applicant's air quality analysis.
C.25
-------
DRAFT
OCTOBER 1990
Figure C-3 outlines the basic steps for an applicant to follow for a PSD
dispersion modeling analysis to demonstrate compliance with the NAAQS and PSD
increments. These steps are described in further detail in the sections which
follow.
IV.B DETERMINING THE IMPACT AREA
The proposed project's impact area is the geographical area for which
the required air quality analyses for the NAAQS and PSD increments are carried
out. The impact area includes all locations where the significant increase in
the potential emissions of a criteria pollutant from a new source, or
significant net emissions increase from a modification, will cause a
significant ambient impact (i.e., equal or exceed the applicable significant
ambient impact level, as shown in Table C-4). The highest modeled pollutant
concentration for each averaging time is used to determine whether the source
will have a significant ambient impact for that pollutant. [An impact area is
not defined for noncriteria pollutants in the same way as for criteria
pollutants (see Section IV.C.3 of this chapter for further discussion).]
The impact area is a circular area with a radius extending from the
source to (1) the most distant point where approved dispersion modeling
predicts a significant ambient impact will occur, or (2) a modeling receptor
distance of 50 km, whichever is less. Usually the area of modeled significant
impact does not have a continuous, smooth border. (It may actually be
comprised of pockets of significant impact separated by pockets of
insignificant impact.) Nevertheless, the required air quality analysis is
carried out within the circle that circumscribes the significant ambient
impacts, as shown in Figure C-4.
Initially, for each pollutant subject to review an impact area is
determined for every averaging time. The impact area used for the air quality
analysis of a particular pollutant is the largest of the areas determined for
that pollutant. For example, modeling the proposed S02 emissions from a new
source might show that a significant ambient SOp impact occurs out to a
distance from the source of 2 kilometers for the annual averaging period;
C.26
-------
DRAFT
OCTOBER '990
Meteorological Data
Source Input Data
Sm
.£ S»
I -^
CO
I
CD
(0
Q.
LL
Meteorological Data
Source Input Data
Pollutant Emitted in
Significant Amounts
Model Impact of
Proposed Source
Yes
Ambient
Concentrations
Above Air Quality
Significance
Level
Determine Need for
Pre-application
Monitoring
i
Determine
Impact Area
i
r
Develop Emissions
Inventory
1
!
Model Impact of
Proposed, Existing, and
Secondary Emissions
i
F
Add Monitored
Background Levels
(for NAAQS only)
i
r
Demonstration of
Compliance
No Further NAAQS or
PSD Increment Analysis
Required for Pollutant
Figure C-3. Basic Steps in the Air Quality Analysis
(NAAQS and PSD Increments)
C.27
-------
DRAFT
OCTOBER 1990
TABLE C-4.
SIGNIFICANCE LEVELS FOR AIR QUALITY IMPACTS IN CLASS II AREAS3
Pollutant Annual 24-hour 8-hour 3-hour 1-hour
so2
TSP
PM-10
NOX
CO
°3
15 - 25
15 - -
15 - -
1 _
500 - 2,000
b
a This table does not apply to Class I areas. If a proposed source is
located within 100 kilometers of a Class I area, an impact of 1 fig/m on a
24-hour basis is significant.
- No significant ambient impact concentration has been established. Instead,
any net emissions increase of 100 tons per year of VOC subject to PSD would
be required to perform an ambient impact analysis.
C.28
-------
DRAFT
OCTOBER 1990
Impact Area —T
County B
SO2 Unclassified
County E
S02 Unclassified
State line
County line
County D
SOa Attainment
Figure C-4. Determining the Impact Area.
C.29
-------
DRAFT
OCTOBER 1990
4.3 kilometers for the 24-hour averaging period; and 3.8 kilometers for the 3-
hour period. Therefore, an impact area with a radius of 4.3 kilometers from
the proposed source is selected for the SCL air quality analysis.
In the event that the maximum ambient impact of a proposed emissions
increase is below the appropriate ambient air quality significance level for
all locations and averaging times, a full impact analysis for that pollutant
is not required by EPA. Consequently, a preliminary analysis which predicts
an insignificant ambient impact everywhere is accepted by EPA as the required
air quality analysis (NAAQS and PSD increments) for that pollutant. [NOTE:
Hhile it may be shown that no impact area exists for a particular pollutant,
the PSD application (assuming it is the first one in the area) still
establishes the PSD baseline area and minor source baseline date in the
section 107 attainment or unclassifiable area where the source will be
located, regardless of its insignificant ambient impact.]
For each applicable pollutant, the determination of an impact area must
include all emissions including quantifiable fugitive emissions, resulting
from the proposed source. For a proposed modification, the determination
includes contemporaneous emissions increases and decreases, with emissions
decreases input as negative emissions in the model. The EPA allows for the
exclusion of temporary emissions (e.g., emissions occurring during the
construction phase of a project) when establishing the impact area and
conducting the subsequent air quality analysis, if it can be shown that such
emissions do not impact a Class I area or an area where a PSD increment for
that pollutant is known to be violated. However, where EPA is not the PSD
permitting authority, the applicant should confer with the appropriate
permitting agency to determine whether it allows for the exclusion of
temporary emissions.
Once defined for the proposed PSD project, the impact area(s) will
determine the scope of the required air quality analysis. That is, the impact
area(s) will be used to:
C.30
-------
DRAFT
OCTOBER 1990
set the boundaries within which ambient air quality monitoring
data may need to be collected,
• • define the area over which a full impact analysis (one that
considers the contribution of all sources) must be undertaken, and
guide the identification of other sources to be included in the
modeling analyses.
Again, if no significant ambient impacts are predicted for a particular
pollutant, EPA does not require further NAAQS or PSD increment analysis of
that pollutant. However, the applicant must still consider any additional
impacts which the proposed source may have concerning impairment on
visibility, soils and vegetation, as well as any adverse impacts on air
quality related values in Class I areas (see Chapters D and E of this part).
IV.C SELECTING SOURCES FOR THE PSD EMISSIONS INVENTORIES
When a full impact analysis is required for any pollutant, the applicant
is responsible for establishing the necessary inventories of existing sources
and their emissions, which will be used to carry out the required NAAQS and
PSD increment analyses. Such special emissions inventories contain the
various source data used as input to an applicable air quality dispersion
model to estimate existing ambient pollutant concentrations. Requirements for
preparing an emissions inventory to support a modeling analysis are described
to a limited extent in the Modeling Guideline. In addition, a number of other
EPA documents (e.g., References 5 through 11) contain guidance on the
fundamentals of compiling emissions inventories. The discussion which follows
pertains primarily to identifying and selecting existing sources to be
included in a PSD emissions inventory as needed for a full impact analysis.
The permitting agency may provide the applicant a list of existing
sources upon request once the extent of the impact area(s) is known. If the
list includes only sources above a certain emissions threshold, the applicant
is responsible for identifying additional sources below that emissions level
which could affect the air quality within the impact area(s). The permitting
agency should review all required inventories for completeness and accuracy.
C.31
-------
DRAFT
OCTOBER 1990
IV.C.I THE NAAQS INVENTORY
Whjle air quality data may be used to help identify existing background
air pollutant concentrations, EPA requires that, at a minimum, all nearby
sources be explicitly modeled as part of the NAAQS analysis. The Modeling
Guideline defines a "nearby" source as any point source expected to cause a
significant concentration gradient in the vicinity of the proposed new source
or modification. For PSD purposes, "vicinity" is defined as the impact area.
However, the location of such nearby sources could be anywhere within the
impact area or an annular area extending 50 kilometers beyond the impact area.
(See Figure C-5.)
In determining which existing point sources constitute nearby sources,
the Modeling Guideline necessarily provides flexibility and requires judgment
to be exercised by the permitting agency. Moreover, the screening method for
identifying a nearby source may vary from one permitting agency to another.
To identify the appropriate method, the applicant should confer with the
permitting agency prior to actually modeling any existing sources.
The Modeling Guideline indicates that the useful distance for guideline
models is 50 kilometers. Occasionally, however, when applying the above
source identification criteria, existing stationary sources located in the
annular area beyond the impact area may be more than 50 kilometers from
portions of the impact area. When this occurs, such sources' modeled impacts
throughout the entire impact area should be calculated. That is, special
steps should not be taken to cut off modeled impacts of existing sources at
receptors within the applicants impact area merely because the receptors are
located beyond 50 kilometers from such sources. Modeled impacts beyond 50
kilometers should be considered as conservative estimates in that they tend to
overestimate the true source impacts. Consequently, if an existing source's
impact includes estimates at distances exceeding the normal 50-kilometer
range, it may be appropriate to consider other techniques, including long-
range transport models. Applicants should consult with the permitting agency
prior to the selection of a model in such cases.
C.32
-------
DRAFT
OCTOBER 1990
Screening Area
Impact Area
County D
SOz Attainment
State line
County line
Figure C-5. Defining the Emissions Inventory Screening Area.
C.33
-------
DRAFT
OCTOBER 1990
It will be necessary to include in the NAAQS inventory those sources
which have received PSD permits but have not yet not begun to operate, as well
as any complete PSD applications for which a permit has not yet been issued.
In the latter case, it is EPA's policy to account for emissions that will
occur at sources whose complete PSD application was submitted as of thirty
days prior to the date the proposed source files its PSD application. Also,
sources from which secondary emissions will occur as a result of the proposed
source should be identified and evaluated for inclusion in the NAAQS
inventory. While existing mobile source emissions are considered in the
determination of background air quality for the NAAQS analysis (typically
using existing air quality data), it should be noted that the applicant need
not model estimates of future mobile source emissions growth that could result
from the proposed project because the definition of "secondary emissions"
specifically excludes any emissions coming directly from mobile sources.
Air quality data may be used to establish background concentrations in
the impact area resulting from existing sources that are not considered as
nearby sources (e.g., area and mobile sources, natural sources, and distant
point sources). If, however, adequate air quality data do not exist (and the
applicant was not required to conduct pre-application monitoring), then these
"other" background sources are also included in the NAAQS inventory so that
their ambient impacts can be estimated by dispersion modeling.
C.34
-------
DRAFT
OCTOBER 1990
IV.C.2 THE INCREMENT INVENTORY
An emissions inventory for the analysis of affected PSD increments must
also be developed. The increment inventory includes all increment-affecting
sources located in the impact area of the proposed new source or modification.
Also, all increment-affecting sources located within 50 kilometers of the
impact area (see Figure C-5) are included in the inventory if they, either
individually or collectively, affect the amount of PSD increment consumed.
The applicant should contact the permitting agency to determine what
particular procedures should be followed to identify sources for the increment
inventory.
In general, the stationary sources of concern for the increment
inventory are those stationary sources with actual emissions changes occurring
since the minor source baseline date. However, it should be remembered that
certain actual emissions changes occurring before the minor source baseline
date (i.e., at major stationary point sources) also affect the increments.
Consequently, the types of stationary point sources that are initially
reviewed to determine the need to include them in the increment inventory fall
under two specific time frames as follows:
After the ma.ior source base! ine date-
existing major stationary sources having undergone a physical
change or change in their method of operation, and
new major stationary sources.
After the minor source baseline date-
existing stationary sources having undergone a physical
change or change in their method of operation;
existing stationary sources having increased hours of
operation or capacity utilization (unless such change was
considered representative of baseline operating conditions); and
new stationary sources.
C.35
-------
DRAFT
OCTOBER 1990
If, in the impact area or surrounding screening area, area or mobile
source emissions will affect increment consumption, then emissions input data
for such minor sources are also included in the increment inventory. The
change in such emissions since the minor source baseline date (rather than the
absolute magnitude of these emissions) is of concern since this change is what
may affect a PSD increment. Specifically, the rate of growth and the amount
of elapsed time since the minor source baseline date was established determine
the extent of the increase in area and mobile source emissions. For example,
in an area where the minor source baseline date was recently established
(e.g., within the past year or so of the proposed PSD project), very little
area and mobile source emissions growth may have occurred. Also, sufficient
data (particularly mobile source data) may not yet be available to reflect the
amount of growth that has taken place. As with the NAAQS analysis, applicants
are not required to estimate future mobile source emissions growth that could
result from the proposed project because they are excluded from the definition
of "secondary emissions."
The applicant should initially consult with the permitting agency to
determine the availability of data for assessing area and mobile source growth
since the minor source baseline date. This information, or the fact that such
data is not available, should be thoroughly documented in the application.
The permitting agency should verify and approve the basis for actual area
source emissions estimates and, especially if these estimates are considered
by the applicant to have an insignificant impact, whether it agrees with the
applicant's assessment.
When area and mobile sources are determined to affect any PSD increment,
their emissions must be reported on a gridded basis. The grid should cover
the entire impact area and any areas outside the impact area where area and
mobile source emissions are included in the analysis. The exact sizing of an
emissions inventory grid cell generally should be based on the emissions
density in the area and any computer constraints that may exist. Techniques
for assigning area source emissions to grid cells are provided in
Reference 11. The grid layout should always be discussed with, and approved
by, the permitting agency in advance of its use.
C.36
-------
DRAFT
OCTOBER 1990
IV.C.3 NONCRITERIA POLLUTANTS INVENTORY
An inventory of all noncriteria pollutants emitted in significant
amounts is required for estimating the resulting ambient concentrations of
those pollutants. Significant ambient impact levels have not been established
for non-criteria pollutants. Thus, an impact area cannot be defined for non-
criteria pollutants in the same way as for criteria pollutants. Therefore, as
a general rule of thumb, EPA believes that an emissions inventory for non-
criteria pollutants should include sources within 50 kilometers of the
proposed source. Some judgment will be exercised in applying this position on
a case-by-case basis.
IV.D MODEL SELECTION
Two levels of model sophistication exist: screening and refined
dispersion modeling. Screening models may be used to eliminate more extensive^
modeling for either the preliminary analysis phase or the full impact analysis
phase, or both. However, the results must demonstrate to the satisfaction of
the permitting agency that all applicable air quality analysis requirements
are met. Screening models produce conservative estimates of ambient impact in
order to reasonably assure that maximum ambient concentrations will not be
underestimated. If the resulting estimates from a screening model indicate a
threat to a NAAQS or PSD increment, the applicant uses a refined model to re-
estimate ambient concentrations (of course, the applicant can select other
options, such as reducing emissions, or to decrease impacts). Guidance on the
use of screening procedures to estimate the air quality impact of stationary
sources is presented in EPA's Screening Procedures for Estimating Air Quality
Impact of Stationary Sources [Reference 12].
A refined dispersion model provides more accurate estimates of a
source's impact and, consequently, requires more detailed and precise input
data than does a screening model. The applicant is referred to Appendix A of
the Modeling Guideline for a list of EPA-preferred models, i.e., guideline
models. The guideline model selected for a particular application should be
the one which most accurately represents atmospheric transport, dispersion,
C.37
-------
DRAFT
OCTOBER 1990
and chemical transformations in the area under analysis. For example, models
have been developed for both simple and complex terrain situations; some are
designed for urban applications, while others are designed for rural
applications.
In many circumstances the guideline models known as Industrial Source
Complex Model Short- and Long-term (ISCST and ISCLT, respectively) are
acceptable for stationary sources and are preferred for use in the dispersion
modeling analysis. A brief discussion of options required for regulatory
applications of the ISC model is contained in the Modeling Guideline. Other
guideline models, such as the Climatological Dispersion Model (COM), may be
needed to estimate the ambient impacts of area and mobile sources.
Under certain circumstances, refined dispersion models that are not
listed in the Modeling Guideline, i.e., non-guideline models, may be
considered for use in the dispersion modeling analysis. The use of a non-
guideline model for a PSD permit application must, however, be pre-approved on
a case-by-case basis by EPA. The applicant should refer to the EPA documents
entitled Interim Procedures for Evaluating Air Quality Models (Revised)
[Reference 13] and Interim Procedures for Evaluating Air Quality Models:
Experience with Implementation [Reference 14]. Close coordination with EPA
and the appropriate State or local permitting agency is essential if a non-
guideline model is to be used successfully.
C.38
-------
DRAFT
OCTOBER 1990
IV.D.I METEOROLOGICAL DATA
Meteorological data used in air quality modeling must be spatially and
climatologically (temporally) representative of the area of interest.
Therefore, an applicant should consult the permitting authority to determine
what data will be most representative of the location of the applicant's
proposed facility.
Use of site-specific meteorological data is preferred for air quality
modeling analyses if 1 or more years of quality-assured data are available.
If at least 1 year of site-specific data is not available, 5 years of
meteorological data from the nearest National Weather Service (NWS) station
can be used in the modeling analysis. Alternatively, data from universities,
the Federal Aviation Administration, military stations, industry, and State or
local air pollution control agencies may be used if such data are equivalent
in accuracy and detail to the NWS data, and are more representative of the
area of concern.
The 5 years of data should be the most recent consecutive 5 years of
meteorological data available. This 5-year period is used to ensure that the
model results adequately reflect meteorological conditions conducive to the
prediction of maximum ambient concentrations. The NWS data may be obtained
from the National Climatic Data Center (Asheville, North Carolina), which
serves as a clearinghouse to collect and distribute meteorological data
collected by the NWS.
IV.D.2 RECEPTOR NETWORK
Polar and Cartesian networks are two types of receptor networks commonly
used in refined air dispersion models. A polar network is comprised of
concentric rings and radial arms extending outward from a center point (e.g.,
the modeled source). Receptors are located where the concentric rings and
radial arms intersect. Particular care should be exercised in using a polar
network to identify maximum estimated pollutant concentrations because of the
inherent problem of increased longitudinal spacing of adjacent receptors as
C.39
-------
DRAFT
OCTOBER 1990
their distance along neighboring radial arms increases. For example, as
illustrated in Figure C-6, while the receptors on individual radials, e.g.,
Al, A2, A3... and Bl, B2,-B3..., may be uniformly spaced at a distance of 1
kilometer apart, at greater distances from the proposed source, the
longitudinal distance between the receptors, e.g., A4 and B4, on neighboring
radials may be several kilometers. As a result of the presence of larger and
larger "blind spots" between the radials as the distance from the modeled
source increases, finding the maximum source impact can be somewhat
problematic. For this reason, using a polar network for anything other than
initial screening is generally discouraged.
A cartesian network (also referred to as a rectangular network) consists
of north-south and east-west oriented lines forming a rectangular grid, as
shown in Figure C-6, with receptors located at each intersection point. In
most refined air quality analyses, a cartesian grid with from 300 to 400
receptors (where the distance from the source to the farthest receptor is 10
kilometers) is usually adequate to identify areas of maximum concentration.
However, the total number of receptors will vary based on the specific air
quality analysis performed.
In order to locate the maximum modeled impact, perform multiple model
runs, starting with a relatively coarse receptor grid (e.g., one or two
kilometer spacing) and proceeding to a relatively fine receptor grid (e.g.,
100 meters). The fine receptor grid should be used to focus on the area(s) of
higher estimated pollutant concentrations identified by the coarse grid model
runs. With such multiple runs the maximum modeled concentration can be
identified. It is the applicant's responsibility to demonstrate that the
final receptor network is sufficiently compact to identify the maximum
estimated pollutant concentration for each applicable averaging period. This
applies both to the PSD increments and to the NAAQS.
C.40
-------
DRAFT
OCTOBER 1990
Polar Grid Network
»
1 km
Jfrl JA2 I A3 I A4 I A5 I A6
.LLLLL
J I 1 1 1 L-
Cartesian Grid Network
Figure C-6. Examples of Polar and Cartesian Grid Networks.
C.41
-------
DRAFT
OCTOBER 1990
Some air quality models allow the user to input discrete receptors at
user-specified locations. The selection of receptor sites should be a case-
by-case determination, taking into consideration the topography, the
climatology, the monitor sites, and the results of the preliminary analysis.
For example, receptors should be located at:
the fence!ine of a proposed facility;
the boundary of the nearest Class I or nonattainment area;
the location(s) of ambient air monitoring sites-, and
locations where potentially high ambient air concentrations are
expected to occur,
In general, model ing receptors for both the NAAQS and the PSD
increment analyses should be placed at ground level points anywhere
except on the applicant's plant property if it is inaccessible to the
general public. Public access to plant property is to be assumed, however,
unless a continuous physical barrier, such as a fence or wall, precludes
entrance onto that property. In cases where the public has access, receptors
should be located on the applicant's property. It is important to note that
ground level points of receptor placement could be over bodies of water,
railroad tracks, roadways, and property owned by other sources. For NAAQS
analyses, modeling receptors may also be placed at elevated locations, such as
on building rooftops. However, for PSD increments, receptors are limited to
locations at ground level.
IV.D.3 GOOD ENGINEERING PRACTICE (GEP) STACK HEIGHT
Section 123 of the Clean Air Act limits the use of dispersion
techniques, such as merged gas streams, intermittent controls, or stack
heights above GEP, to meet the NAAQS or PSD increments. The GEP stack height
is defined under Section 123 as "the height necessary to insure that emissions
from the stack do not result in excessive concentrations of any air pollutant
in the immediate vicinity of the source as a result of atmospheric downwash,
C.42
-------
DRAFT
OCTOBER 1990
eddies or wakes which may be created by the source itself, nearby structures
or nearby terrain obstacles." The EPA has promulgated stack height
regulations under 40 CFR Part 51 which help to determine the GEP stack height
for any stationary source.
Three methods are available for determining "GEP stack height" as
defined in 40 CFR 51.100(11):
use the 65 meter (213.5 feet) de mim'mis height as measured from
the ground-level elevation at the base of the stack;
calculate the refined formula height using the dimensions of
nearby structures (this height equals H + 1.5L, where H is the
height of the nearby structure and L is the lesser dimension of
the height or projected width of the nearby structure); or
demonstrate by a fluid model or field study the equivalent GEP
formula height that is necessary to avoid excessive concentrations
caused by atmospheric downwash, wakes, or eddy effects by the
source, nearby structures, or nearby terrain features.
That portion of a stack height in excess of the GEP height is generally
not creditable when modeling to develop source emissions limitations or to
determine source impacts in a PSD air quality analysis. For a stack height
less than GEP height, screening procedures should be applied to assess
potential air quality impacts associated with building downwash. In some
cases, the aerodynamic turbulence induced by surrounding buildings will cause
stack emissions to be mixed rapidly toward the ground (downwash), resulting in
higher-than-normal ground level concentrations in the vicinity of the source.
Reference 12 contain screening procedures to estimate downwash concentrations
in the building wake region. The Modeling Guideline recommends using the
Industrial Source Complex (ISC) air dispersion model to determine building
wake effects on maximum estimated pollutant concentrations.
For additional guidance on creditable stack height and plume rise
calculations, the applicant should consult with the permitting agency. In
addition, several EPA publications [References 15 through 19] are available
for the applicant's review.
C.43
-------
DRAFT
OCTOBER 1990
IV.D.4 SOURCE DATA
Emissions rates and other source-related data are needed to estimate the
ambient concentrations resulting from (1) the proposed new source or
modification, and (2) existing sources contributing to background pollutant
concentrations (NAAQS and PSD increments). Since the estimated pollutant
concentrations can vary widely depending on the accuracy of such data, the
most appropriate source data available should always be selected for use in a
modeling analysis. Guidance on the identification and selection of existing
sources for which source input data must be obtained for a PSD air quality
analysis is provided in section IV.C. Additional information on the specific
source input data requirements is contained in EPA's Modeling Guideline and in
the users' guide for each dispersion model.
Source input data that must be obtained will depend upon the
categorization of the source(s) to be modeled as either a point, area or line
source. Area sources are often collections of numerous small emissions
sources that are impractical to consider as separate point or line sources.
Line sources most frequently considered are roadways.
For each stationary point source to be modeled, the following minimum
information is generally necessary:
pollutant emission rate (see discussion below)',
stack height (see discussion on GEP stack height)-,
stack gas exit temperature, stack exit inside diameter, and stack
gas exit velocity;
dimensions of all structures in the vicinity of the stack in
question;
the location of topographic features (e.g., large bodies of water,
elevated terrain) relative to emissions points; and
stack coordinates.
C.44
-------
DRAFT
OCTOBER 1990
A source's emissions rate as used in a modeling analysis for any
pollutant is determined from the following source parameters (where MMBtu
means "million Btu's heat input"):
emissions limit (e.g., lb/MMBtu);
operating level (e.g., MMBtu/hour); and
operating factor (e,g., hours/day, hours/year).
Special procedures, as described below, apply to the way that each of these
parameters is used in calculating the emissions rate for either the proposed
new source (or modification) or any existing source considered in the NAAQS
and PSD increment analyses. Table C-5 provides a summary of the point source
emissions input data requirements for the NAAQS inventory.
For both NAAQS and PSD increment compliance demonstrations, the
emissions rate for the proposed new source or modification must reflect the
maximum allowable operating conditions as expressed by the federally
enforceable emissions limit, operating level, and operating factor for each
applicable pollutant and averaging time. The applicant should base the
emissions rates on the results of the BACT analysis (see Chapter B, Part I).
Operating levels less than 100 percent of capacity may also need to be modeled
where differences in stack parameters associated with the lower operating
levels could result in higher ground level concentrations. A value
representing less than continuous operation (8760 hours per year) should be
used for the operating factor only when a federally enforceable operating
limitation is placed upon the proposed source. [NOTE: It is important that
the applicant demonstrate that all modeled emission rates are consistent with
the applicable permit conditions.]
C.45
-------
DRAFT
OCTOBER 1990
-DRAFT-
March 28, 1990
TABLE C-5 POINT SOURCE MODEL INPUT DATA (EMISSIONS) FOR NAAQS COMPLIANCE DEMONSTRATIONS
Averaging Time
Emission Limit
(»/NMBtu)'
Operating Level
(NMBtu/hr)'
Operating Factor
(e.g., hr/yr, hr/day)
Proposed Major New or Modified Source
Annual and quarterly
Short term
(24 hours or less)
Maximum allowable emission
limit or Federally enforceable
permit
Maximum allowable emission
limit or Federally enforceable
permit limit
Design capacity or Federally
enforceable permit condition
Design capacity or Federally
enforceable permit condition3
Continuous operation
(i.e, 8760 hours)'
Continuous operation (i.e.,
all hours of each time
period under consideration)
(for all hours of the
meteorological data base)2
Nearby Background Source(s)'
Annual and quarterly
Short term
Maximum allowable emission
limit or Federally enforceable
permit
Maximum allowable emission
limit or Federally enforceable
permit 1imit
Actual or design capacity
(whichever is greater), or
Federally enforceable permit
condition
Actual or design capacity
(whichever is greater), or
Federally enforceable permit
condition3
Actual operating factor
averaged over the most
recent 2 years5
Continuous operation (i.e.,
all hours of each time
period under consideration)
(for all hours of the
meteorological data base)2
Other Background Source(s)*
Annual and quarterly
Short term
Maximum allowable emission
limit or Federally enforceable
permit limit
Maximum allowable emission
limit or Federally enforceable
permit limit
Annual level when actually
operating, averaged over the
most recent 2 years5
Annual level when actually
operating, averaged over the
most recent 2 years5
Actual operating factor
averaged over the most
recent 2 years5
Continuous operation (i.e.,
all hours of each time
period under consideration)
(for all hours of the
meteorological data base)2
Terminology applicable to fuel burning sources; analogous terminology (e.g., I/throughput) may be used for other types of sources.
If operation does not occur for all hours of the time period of consideration (e.g., 3 or 24 hours) and the source operation is constrained
by a Federally enforceable permit condition, an appropriate adjustment to the modeled emission rate may be made (e.g., if operation is only
8:00 a.m. to 4:00 p.m. each day, only these hours will be modeled with emissions from the source. Modeled emissions should not be averaged
across non-operating time periods).
Operating levels such as 50 percent and 75 percent of capacity should also be modeled to determine the load causing the highest concentration.
Includes existing facility to which modification is proposed if the emissions from the existing facility will not be affected by the
modification. Otherwise use same parameters as for major modification.
Unless it is determined that this period is not representative.
Generally, the ambient impacts from non-nearby background sources can be represented by air quality data unless adequate data do not exist.
C.46
-------
DRAFT
OCTOBER 1990
For those existing point sources that must be explicitly modeled, i.e.,
"nearby" sources (see section IV.C.I of this chapter), the NAAQS inventory
must contain the maximum allowable values for the emissions limit, and
operating level. The operating factor may be adjusted to account for
representative, historical operating conditions.only when modeling for the
annual [or quarterly for lead (Pb)] averaging period. In such cases, the
appropriate input is the actual operating factor averaged over the most recent
2 years (unless the permitting agency determines that another period is more
representative). For short-term averaging periods (24 hours or less), the
applicant generally should assume that nearby sources operate continuously.
However, the operating factor may be adjusted to take into account any
federally enforceable permit condition which limits the allowable hours of
operation. In situations where the actual operating level exceeds the design
capacity (considering any federally enforceable limitations), the actual level
should be used to calculate the emissions rate.
If other background sources need to be modeled (i.e., adequate air
quality data are not available to represent their impact), the input
requirements for the emissions limit and operating factor are identical to
those for "nearby" sources. However, input for the operating level may be
based on the annual level of actual operation averaged over the last 2 years
(unless the permitting agency determines that a more representative period
exists).
The applicant must also include any quantifiable fugitive emissions from
the proposed source or any nearby sources. Fugitive emissions are those
emissions that cannot reasonably be expected to pass through a stack, vent, or
other equivalent opening, such as a chimney or roof vent. Common quantifiable
fugitive emissions sources of particulate matter include coal piles, road
dust, quarry emissions, and aggregate stockpiles. Quantifiable fugitive
^emissions of volatile organic compounds (VOC) often occur at components of
process equipment. An'applicant should consult with the permitting agency to
determine the proper procedures for characterizing and modeling fugitive
emissions.
C.47
-------
DRAFT
OCTOBER 1990
When building downwash affects the air quality impact of the proposed
source or any existing source which is modeled for the NAAQS analysis, those
impacts generally should be considered in the analysis. Consequently, the
appropriate dimensions of all structures around the stack(s) in question also
should be included in the emissions inventory. Information including building
heights and horizontal building dimensions may be available in the permitting
agency's files; otherwise, it is usually the responsibility of the applicant
to obtain this information from the applicable source(s).
Sources should not automatically be excluded from downwash
considerations simply because they are located outside the impact area. Some
sources located just outside the impact area may be located close enough to it
that the immediate downwashing effects directly impact air quality in the
impact area. In addition, the difference in downwind plume concentrations
caused by the downwash phenomenon may warrant consideration within the impact
area even when the immediate downwash effects do not. Therefore, any decision
by the applicant to exclude the effects of downwash for a particular source
should be justified in the application, and approved by the permitting agency.
For a PSD increment analysis, an estimate of the amount of increment
consumed by existing point sources generally is based on increases in actual
emissions occurring since the minor source baseline date. [Remember that
increment is also consumed by major stationary sources whose actual emissions
have increased (as a result of construction) before the minor source baseline
date but on or before the major source baseline date.] For any increment-
consuming (or increment-expanding) emissions unit, the actual emissions limit,
operating level, and operating factor may all be determined from source
records and other information (e.g., State emissions files), when available,
reflecting actual source operation. For the annual averaging period, the
change in the actual emissions rate should be calculated as the difference
between:
the current average actual emissions rate, and
the average actual emissions rate as of the minor source baseline
date (or major source baseline date for major stationary sources).
C.48
-------
DRAFT
OCTOBER 1990
In each case, the average rate is calculated as the average over the previous
2-year period (unless the permitting agency determines that a different time
period is more representative of normal source operation).
For each short-term averaging period (24 hours and less), the change in
the actual emissions rate for the particular averaging period is calculated as
the difference between:
the current maximum actual emissions rate, and
the maximum actual emissions rate as of the minor source baseline
date (or major source baseline date for applicable major
stationary sources undergoing construction before the minor source
baseline date).
In each case, the maximum rate is the highest occurrence for that averaging
period during the previous 2 years of operation.
Where appropriate, air quality impacts from fugitive emissions and
building downwash are also taken into account for the PSD increment analysis.
Of course, they would only be considered when applicable to increment-
consuming emissions.
If the change in the actual emissions rate at a particular source
involves a change in stack parameters (e.g., stack height, gas exit
temperature, etc.) then the stack parameters and emissions rates associated
with both the baseline case and the current situation must be used as input to
the dispersion model. To determine increment consumption (or expansion) for
such a source, the baseline case emissions are input to the model as negative
emissions, along with the baseline stack parameters. In the same model run,
the current case for the same source is modeled as the total current emissions
associated with the current stack parameters. This procedure effectively
calculates, for each receptor and for each averaging time, the difference
between the baseline concentration and the current concentration (i.e., the
amount of increment consumed by the source).
C.49
-------
DRAFT
OCTOBER 1990
Emissions changes associated with area and mobile source growth
occurring since the minor source baseline date are also accounted for in the
increment analysis by modeling. In many cases state emission files will
contain information on area source emissions or such information may be
available from EPA's AIRS-NEDS emissions data base. In the absence of this
information, the applicant should use procedures adopted for developing state
area source emission inventories. The EPA documents outlining procedures for
area source inventory development should be reviewed.
Mobile source emissions are usually calculated by applying mobile source
emissions factors to transportation data such as vehicle miles travelled
(VMT), trip ends, vehicle fleet characteristics, etc. Data are also required
on the spatial arrangement of the VMT within the area being modeled. Mobile
source emissions factors are available for various vehicle types and
conditions from an EPA emissions factor model entitled MOBILE4. The MOBILE4
users manual [Reference 20] should be used in developing inputs for executing
this model. The permitting agency can be of assistance in obtaining the
needed mobile source emissions data. Oftentimes, these data are compiled by
the permitting agency acting in concert with the local planning agency or
transportation department.
For both area source and mobile source emissions, the applicant will
need to collect data for the minor source baseline date and the current
situation. Data from these two dates will be required to calculate the
increment-affecting emission changes since the minor source baseline date.
C.50
-------
DRAFT
OCTOBER 1990
IV.E THE COMPLIANCE DEMONSTRATION
An applicant for a PSD permit must demonstrate that the proposed source
will not cause or contribute to air pollution in violation of any NAAQS or PSD
increment. This compliance demonstration, for each affected pollutant, must
result in one of the following:
1. The proposed new source or modification will not cause a
significant ambient impact anywhere.
If the significant net emissions increase from a proposed source would
not result in a significant ambient impact anywhere, the applicant is usually
not required to go beyond a preliminary analysis in order to make the
necessary showing of compliance for a particular pollutant. In determining
the significant ambient impact for a pollutant, the highest estimated ambient
concentration of that pollutant for each applicable averaging time is used.
2. The proposed new source or modi'f'ication, in conjunction with
existing sources, will not cause or contribute to a violation of
any NAAQS or PSD increment.
In general, compliance is determined by comparing the predicted ground
level concentrations (based on the full impact analysis and existing air
quality data) at each model receptor to the applicable NAAQS and PSD
increments. If the predicted pollutant concentration increase over the
baseline concentration is below the applicable increment, and the predicted
total ground level concentrations are below the NAAQS, then the applicant has
successfully demonstrated compliance.
The modeled concentrations which should be used to determine compliance
with any NAAQS and PSD increment depend on 1) the type of standard, i.e.,
deterministic or statistical, 2) the available length of record of
meteorological data, and 3) the averaging time of the standard being analyzed.
For example, when the analysis is based on 5 years of National Weather Service
meteorological data, the following estimates should be used:
C.51
-------
DRAFT
OCTOBER 1990
for deterministically based standards (e.g., SCk), the highest,
second-highest short term estimate and the highest annual
estimate; and
for statistically based standards (e.g., PM-10), the highest,
sixth-highest estimate and highest 5-year average estimate.
Further guidance to determine the appropriate estimates to use for the
compliance determination is found in Chapter 8 of the Modeling Guideline for
S02, TSP, lead, N02, and CO; and in EPA's PM-10 SIP Development Guideline
[Reference 21] for PM-10.
When a violation of any NAAQS or increment is predicted at one or more
receptors in the impact area, the applicant can determine whether the net
emissions increase from the proposed source will result in a significant
ambient impact at the point (receptor) of each predicted violation, and at
time the violation is predicted to occur. The source will not be considered
to cause or contribute to the violation if its own impact is not significant
at any violating receptor at the time of each predicted violation. In such a
case, the permitting agency, upon verification of the demonstration, may
approve the permit. However, the agency must also take remedial action
through applicable provisions of the state implementation plan to address the
predicted violation(s).
3. The proposed new source or modification, in conjunction with
existing sources, win cause or contribute to a violation, but
win secure sufficient emissions reductions to offset its adverse
air quality impact.
If the applicant cannot demonstrate that only insignificant ambient
impacts would occur at violating receptors (at the time of the predicted
violation), then other measures are needed before a permit can be issued.
Somewhat different procedures apply to NAAQS violations than to PSD increment
violations. For a NAAQS violation to which an applicant contributes
significantly, a PSD permit may be granted only if sufficient emissions
reductions are obtained to compensate for the adverse ambient impacts caused
by the proposed source. Emissions reductions are considered to compensate for
C.52
-------
DRAFT
OCTOBER 1990
the proposed source's adverse impact when, at a minimum, (1) the modeled net
concentration, resulting from the proposed emissions increase and the
federally enforceable emissions reduction, is less than the applicable
significant ambient impact level at each affected receptor, and (2) no new
violations will occur. Moreover, such emissions reductions must be made
federally enforceable in order to be acceptable for providing the air quality
offset. States may adopt procedures pursuant to federal regulations at
40 CFR 51.165(b) to enable the permitting of sources whose emissions would
cause or contribute to a NAAQS violation anywhere. The applicant should
determine what specific provisions exist within the State program to deal with
this type of situation.
In situations where a proposed source would cause or contribute to a PSD
increment violation, a PSD permit cannot be issued until the increment
violation is entirely corrected. Thus, when the proposed source would cause a
new increment violation, the applicant must obtain emissions reductions that
are sufficient to offset enough of the source's ambient impact to avoid, the
violation. In an area where an increment violation already exists, and the
proposed source would significantly impact that violation, emissions
reductions must not only offset the source's adverse ambient impact, but must
be sufficient to alleviate the PSD increment violation, as well.
C.53
-------
DRAFT
OCTOBER 1990
V. AIR QUALITY ANALYSIS -- EXAMPLE
This section presents a hypothetical example of an air quality analysis
for a proposed new PSD source. In reality, no two analyses are alike, so an
example that covers all modeling scenarios is not possible to present.
However, this example illustrates several significant elements of the air
quality analysis, using the procedures and information set forth in this
chapter.
An applicant is proposing to construct a new coal-fired, steam electric
generating station. Coal will be supplied by railroad from a distant mine.
The coal-fired plant is a new major source which has the potential to emit
significant amounts of S02, PM (particulate matter emissions and PM-10
emissions), NO , and CO. Consequently, an air quality analysis must be
J\
carried out for each of these pollutants. In this analysis, the applicant is
required to demonstrate compliance with respect to -
the NAAQS for S02, PM-10, N02, and CO, and
the PSD increments for S02, TSP, and N02-
V.A DETERMINING THE IMPACT AREA
The first step in the air quality analysis is to estimate the ambient
impacts caused by the proposed new source itself. This preliminary analysis
establishes the impact area for each criteria pollutant emitted in significant
amounts, and for each averaging period. The largest impact area for each
pollutant is then selected as the impact area to be used in the full impact
analysis.
To begin, the applicant prepares a modeling protocol describing the
modeling techniques and data bases that will be applied in the preliminary
analysis. These modeling procedures are reviewed in advance by the permitting
agency and are determined to be in accordance with the procedures described in
the Modeling Guideline and the stack height regulations.
C.54
-------
DRAFT
OCTOBER 1990
Several pollutant-emitting activities (i.e., emissions units) at the
source will emit pollutants subject to the air quality analysis. The two main
boilers emit particulate matter (i.e., particulate matter emissions and PM-10
emissions), SC^, NOX, and CO. A standby auxiliary boiler also emits these
pollutants, but will only be permitted to operate when the main boilers are
not operating.
Particulate matter emissions and PM-10 emissions will also occur at the
coal-handling operations and the limestone preparation process for the flue
gas desulfurization (FGD) system. Emissions units associated with coal and
limestone handling include:
Point sources--the coal car dump, the fly ash silos, and the three
coal baghouse collectors;
Area sources--the active and the inactive coal storage piles and
the limestone storage pile; and
Line sources--the coal and limestone conveying operation.
The emissions from all of the emissions units at the proposed source are
then modeled to estimate the source's area of significant impact (impact area)
for each applicable criteria pollutant. The results of the preliminary
analysis indicate that significant ambient concentrations of NCL and SCL will
occur out to distances of 32 and 50 kilometers, respectively, from the
proposed source. No significant concentrations of CO are predicted at any
location outside the fenced-in property of the proposed source. Thus, an
impact area is not defined for CO, and no further CO analysis is required.
Particulate matter emissions from the coal-handling operations and the
limestone preparation process result in significant ambient TSP concentrations
out to a distance of 2.2 kilometers. However, particulate matter emissions
from the boiler stacks will cause significant TSP concentrations for a
distance of up to 10 kilometers. Since the boiler emissions of particulate
matter are predominantly PM-10 emissions, the same impact area is used for
both TSP and PM-10.
C.55
-------
DRAFT
OCTOBER 1990
This preliminary analysis further indicates that pre-application
monitoring data may be required for two of the criteria pollutants, S02 and
N0~, since the proposed new source will cause ambient concentrations exceeding
the prescribed significant monitoring concentrations for these two pollutants
(see Table C-3). Estimated concentrations of PM-10 are below the significant
monitoring concentration. The permitting agency informs the applicant that
the requirement for pre-application monitoring data will not be imposed with
regard to PM-10. However, due to the fact that existing ambient
concentrations of both S02 and N02 are known to exceed their respective
significant monitoring concentrations, the applicant must address the pre-
application monitoring data requirements for these pollutants.
Before undertaking a site-specific monitoring program, the.applicant
investigates the availability of existing data that is representative of air
quality in the area. The permitting agency indicates that an agency-operated
S02 network exists which it believes would provide representative data for the
applicant's use. It remains for the applicant to demonstrate that the
existing air quality data meet the EPA criteria for data sufficiency,
representativeness, and quality as provided in the PSD Monitoring Guideline.
The applicant proceeds to provide a demonstration which is approved by the
permitting agency. For N02, however, adequate data do not exist, and it is
necessary for the .applicant to take responsibility for collecting such data.
The applicant consults with the permitting agency in order to develop a
monitoring plan and subsequently undertakes a site-specific monitoring program
for N02.
In this example, four intrastate counties are covered by the applicant's
impact area. Each of these counties, shown in Figure C-7, is designated
attainment for all affected pollutants. Consequently, a NAAQS and PSD
C.56
-------
DRAFT
OCTOBER 1990
v.
Figure C- 7. Counties Within 100 Kilometers of Proposed Source.
C.57
-------
DRAFT
OCTOBER 1990
analysis must be completed in each county. With the exception of CO (for
which no further analysis is required) the applicant proceeds with the full
impact analysis for each affected pollutant.
V.B DEVELOPING THE EMISSIONS INVENTORIES
After the impact area has been determined, the applicant proceeds to
develop the required emissions inventories. These inventories contain all of
the source input data that will be used to perform the necessary dispersion
modeling for the required NAAQS and PSD increment analyses. The applicant
contacts the permitting agency and requests a listing of all stationary
sources within a 100-kilometer radius of the proposed new source. This takes
into account the 50-kilometer impact area for S(L (the largest of the defined
impact areas) plus the requisite 50-kilometer annular area beyond that impact
area. For N0« and particulate matter, the applicant needs only to consider
the identified sources which fall within the specific screening areas for
those two pollutants, i.e., the 50-kilometer annular area beyond their
respective impact areas.
Source input data (e.g., location, building dimensions, stack
parameters, emissions factors) for the inventories are extracted from the
permitting agency's air permit and emissions inventory files. Sources to
consider for these inventories also include any that might have recently been
issued a permit to operate, but are not yet in operation. However, in this
case no such "existing" sources are identified. The following point sources
are found to exist within the applicant's impact area and screening area:
Refinery A;
Chemical Plant B;
Petrochemical Complex C;
Rock Crusher D;
Refinery E;
Gas Turbine Cogeneration Facility F; and
Portland Cement Plant G.
C.58
-------
DRAFT
OCTOBER 1990
A diagram of the general location of these sources relative to the
location proposed source is shown in Figure C-8. Because the Portland
Cement Plant G is located 70 kilometers away from the proposed source, its
impact is not considered in the NAAQS or PSD increment analyses for
particulate matter. (The area of concern for particulate matter lies within
60 kilometers of the proposed source.) In this example, the applicant first
develops the NAAQS emissions inventory for S02, particulate matter (PM-10),
and N02.
V.B.I THE NAAQS INVENTORY
For each criteria pollutant undergoing review, the applicant (in
conjunction with the permitting agency) determines which of the identified
sources will be regarded as "nearby" sources and, therefore, must be
explicitly modeled. Accordingly, the applicant classifies the candidate
sources in the following way:
Nearby sources Other Background Sources
Pollutant (explicitly model) (non-modeled background)
S0? Refinery A Port. Cement Plant G
Chemical Plant B
Petro. Complex C
Refinery E
N0? Refinery A, Refinery E
Chemical Plant B
Petro. Complex C
Gas Turbines F
Particulate Refinery A Chemical Plant B
Matter (PM-10) Petro. Complex C Refinery E
Rock Crusher D Gas Turbines F
For each nearby source, the applicant now must obtain emissions input
data for the model to be used. As a conservative approach, emissions input
data reflecting the maximum allowable emissions rate of each nearby source
could be used in the modeling analysis. However, because of the relatively
C.59
-------
DRAFT
OCTOBER 1990
SO2lmpact Area (50 km.)
Portland Cement Plant
Refinery\A •
Chemical Plant B *
Rock Crusher D
Petrochemical
Complex C
Proposed Power Plant
Figure C-8. Point Sources Within 100 Kilometers of Proposed Source.
C.60
-------
DRAFT
OCTOBER 1990
high concentrations anticipated due to the clustering of sources A, B, C and
D, the applicant decides to consider the actual operating factor for each of
these sources for the annual averaging period, in accordance with Table C-5.
For example, for S02, the applicant may determine the actual operating factor
for sources A, B, and C, because they are classified as nearby sources for S02
modeling purposes. On the other hand, the applicant chooses to use the
maximum allowable emissions rate for Source E in order to save the time and
resources involved with determining the actual operating factors for the 45
individual N02 emissions units comprising the source. If a more refined
analysis is ultimately warranted, then the actual hours of operation can be
obtained from Source E for the purposes of the annual averaging period.
i
As another example, for particulate matter (PM-10), the applicant may
determine the actual annual operating factor for sources A, C, and D, because
they are nearby sources for PM-10 modeling purposes. Again, the applicant
chooses to determine the actual hours of annual operation because of the
relatively high concentrations anticipated due to the clustering of these
particular sources.
For each pollutant, the applicant must also determine if emissions from
the sources that were not classified as nearby sources can be adequately
represented by existing air quality data. In the case of S02, for example,
data from the existing State monitoring network will adequately measure
Source G's ambient impact in the impact area. However, for PM-10, the
monitored impacts of Source B cannot be separated from the impacts of the
other sources (A, C, and D) within the proximity of Source B. The applicant
therefore must model this source but is allowed to determine both the actual
operating factor and the actual operating level to model the source's annual
impact, in accordance with Table C-5. For the short-term (24-hour) analysis
the applicant may use the actual operating level, but continuous operation
must be used for the operating factor. The ambient impacts of Source E and
Source F will be represented by ambient monitoring data.
For the N02 NAAQS inventory, the only source not classified as a nearby
source is Refinery E. The applicant would have preferred to use ambient data
C.61
-------
DRAFT
OCTOBER 1990
to represent the ambient impact of this source; however, adequate ambient Nf^
data is not available for the area. In order to avoid modeling this source
with a refined model for N02, the applicant initially agrees to use a
screening technique recommended by the permitting agency to estimate the
impacts of Source E.
Air quality impacts caused by building downwash must be considered
because several nearby sources (A, B, C, and E) have stacks that are less than
GEP stack height. In consultation with the permitting agency, the applicant
is instructed to consider downwash for all four sources in the SC^ NAAQS
analysis, because the sources are all located in the SOp impact area. Also,
after consideration of the expected effect of downwash for other pollutants,
the applicant is told that, for NO^, only Source C must be modeled for its air
quality impacts due to downwash, and no modeling for downwash needs to be done
with respect to particulate matter.
The applicant gathers the necessary building dimension data for the
NAAQS inventory. In this case, these data are available from the permitting
agency through its permit files for sources A, B, and E. However, the
applicant must contact Source C to obtain the data from that source.
Fortunately, the manager of Source C readily provide the applicant this
information for each of the 45 individual emission units.
V.B.2 THE INCREMENT INVENTORY
An increment inventory must be developed for S02, particulate matter
(TSP), and N02. This inventory includes all of the applicable emissions input
data from:
increment-consuming sources within the impact area-, and
increment-consuming sources outside the impact area that affect
increment consumption in the impact area.
In considering emissions changes occurring at any of the major stationary
sources identified earlier (see Figure C-3), the applicant must consider
C.62
-------
DRAFT
OCTOBER 1990
actual emissions changes resulting from a physical change or a change in the
method of operation since the major source baseline date, and any actual
emissions changes since the applicable minor source baseline date. To
identify those sources (and emissions) that consume PSD increment, the
applicant should request information from the permitting agency concerning the
baseline area and all baseline dates (including the existence of any prior
minor source baseline dates) for each applicable pollutant.
A review of previous PSD applications within the total area of concern
reveals that minor source baseline dates for both S02 and TSP have already be
established in Counties A and B. For N02, the minor source baseline date has
already been established in County C. A summary of the relevant baseline
dates for each pollutant in these three counties is shown in Table C-6. The
proposed source will, however, establish the minor source baseline date in
Counties C and D for SCk and TSP, and in Counties A, B and D for NOg.
For SOp, the increment-consuming sources deemed to contribute to
increment consumption in the impact area are sources A, B, C and E. Source B
underwent a major modification which established the minor source baseline
date (April 21, 1984). The actual emissions increase resulting from that
physical change is used in the increment analysis. Source A underwent a major
modification and Source E increased its hours of operation after the minor
source baseline date. The actual emissions increases resulting from both of
these changes are used in the increment analysis, as well. Finally, Source C
received a permit to add a new unit, but the new unit is not yet operational.
Consequently, the applicant must use the potential emissions increase
resulting from that new unit to model the amount of increment consumed. The
existing units at Source C do not affect the increments because no actual
emissions changes have occurred since the April 21, 1984 minor source baseline
C.63
-------
DRAFT
OCTOBER 1990
TABLE C-6. EXISTING BASELINE DATES FOR SOg, TSP,
AND N02 FOR EXAMPLE PSD INCREMENT ANALYSIS
Pollutant
Major Source
Baseline Date
Minor Source
Baseline Date
Affected
Counties
Sulfur dioxide
January 6, 1975
Particulate Matter
(TSP) January 6, 1975
Nitrogen Dioxide February 8, 1988
April 21, 1984
March 14, 1985
June 8, 1988
A and B
A and B
C
C.64
-------
DRAFT
OCTOBER 1990
date. Building dimensions data are needed in the increment inventory for
nearby sources A, B, and E because each has increment-consuming emissions
which are subject to downwash problems. No building dimensions data are
needed for Source C, however, because only the emissions from the newly-
permitted unit consume increment and the stack built for that unit was
designed and constructed at GEP stack height.
For NOp, only the gas turbines located at Cogeneration Station F have
emissions which affect the increment. The PSD permit application for the
construction of these turbines established the minor source baseline date
for NCL (June 8, 1988). Of course, all construction-based actual emissions
changes in NOX occurring after the major source baseline date for N02
(February 8, 1988), at any major stationary source affect increment. However,
no such emissions changes were discovered at the other existing sources in the
area. Thus, only the actual emissions increase resulting from the gas
turbines is included in the NCL increment inventory.
For TSP, sources A, B, C, and E are found to have units whose emissions
may affect the TSP increment in the impact area. Source A established the
minor source baseline date with a PSD permit application to modify its
existing facility. Source B (which established the minor source baseline date
for S02) experienced an insignificant increase in particulate matter emissions
due to a modification prior to the minor source baseline date for particulate
matter (March 14, 1985). Even though the emissions increase did not exceed
the significant emissions rate for particulate matter emissions (i.e., 25 tons
per year), increment is consumed by the actual increase nonetheless, because
the actual emissions increase resulted from construction (i.e., a physical
change or a change in the method of operation) at a major stationary source
occurring after the major source baseline date for particulate matter. The
applicant uses the allowable increase as a conservative estimate of the actual
emissions increase. As mentioned previously, Source C received a permit to
construct, but the newly-permitted unit is not yet in operation. Therefore,
the applicant must use the potential emissions to model the amount of TSP
increment consumed by that new unit.
C.65
-------
DRAFT
OCTOBER 1990
Finally, Source E's actual emissions increase resulting from an increase
in its hours of operation must be considered in the increment analysis. This
source is located far enough outside the impact area that its effects on
increment consumption in the impact area are estimated with a screening
technique. Based on the conservative results, the permitting agency
determines that the source's emissions increase will not affect the amount of
increment consumed in the impact area.
In compiling the increment inventory, increment-consuming TSP and SCu
emissions occurring at minor and area sources located in Counties A and B must
be considered. Also, increment-consuming NO emissions occurring at minor,
j\
area, and mobile sources located in County C must be considered. For this
example, the applicant proposes that because of the low growth in population
and vehicle miles traveled in the affected counties since the applicable minor
source baseline dates, emissions from area and mobile sources will not affect
increment (SC^, TSP, or NC^) consumed within the impact area and, therefore,
do not need to be included in the increment inventory. After reviewing the
documentation submitted by the applicant, the permitting agency approves the
applicant's proposal not to include area and mobile source emissions in the
increment inventory.
V.C The Full Impact Analysis
Using the source input data contained in the emissions inventories, the
next step is to model existing source impacts for both the NAAQS and PSD
increment analyses. The applicant's selection of modelS--ISCST, for short-
term modeling, and ISCLT, for long-term modeling--was made after conferring
with the permitting agency and determining that the area within three
kilometers of the proposed source is rural, the terrain is simple (non-
complex), and there is a potential for building downwash with some of the
nearby sources.
No on-site meteorological data are available. Therefore, the applicant
evaluates the meteorological data collected at the National Weather Service
station located at the regional airport. The applicant proposes the use of
C.66
-------
DRAFT
OCTOBER 1990
5 years of hourly observations from 1984 to 1988 for input to the dispersion
model, and the permitting agency approves their use for the modeling analyses.
The applicant, in consultation with the permitting agency, determines
that terrain in the vicinity is essentially flat, so that it is not necessary
to model with receptor elevations. (Consultation with the reviewing agency
about receptor elevations is important since significantly different
concentration estimates may be obtained between flat terrain and rolling
terrain modes.)
A single-source model run for the auxiliary boiler shows that its
estimated maximum ground-level concentrations of SCL and NCL will be less than
the significant air quality impact levels for these two pollutants (see
Table C-4). This boiler is modeled separately from the two main boilers
because there will be a permit condition which restricts it from operating at
the same time as the main boilers. For particulate matter, the auxiliary
boiler's emissions are modeled together with the fugitive emissions from the
proposed source to estimate maximum ground-level PM-10 concentrations. In
this case, too, the resulting ambient concentrations are less than the
significant ambient impact level for PM-10. Thus, operation of the auxiliary
boiler would not be considered to contribute to violations of any NAAQS or PSD
increment for SOp, particulate matter, or NOp. The auxiliary boiler is
eliminated from further modeling consideration because it will not be
permitted to operate when either of the main boilers is in operation.
V.C.I NAAQS ANALYSIS
The next step is to estimate total ground-level concentrations. For the
SOp NAAQS compliance demonstration, the applicant selects a coarse receptor
grid of one-kilometer grid spacing to identify the area(s) of high impact
caused by the combined impact form the proposed new source and nearby sources.
Through the coarse grid run, the applicant finds that the area of highest
estimated concentrations will occur in the southwest quadrant. In order to
determine the highest total concentrations, the applicant performs a second
model run for the southwest quadrant using a 100-meter receptor fine-grid.
C.67
-------
DRAFT
OCTOBER 1990
The appropriate concentrations from the fine-grid run is added to the
monitored background concentrations (including Source G's impacts) to
establish the total estimated S02 concentrations for comparison against the
NAAQS. The results show maximum S02 concentrations of:
600 ng/m3 , 3-hour average;
•3
155 ng/m , 24- hour average; and
27 ng/m3 , annual average.
Each of the estimated total impacts is within the concentrations allowed by
the NAAQS.
For the N02 NAAQS analysis, the sources identified as "nearby" for N02
are modeled with the proposed new source in two steps, in the same way as for
the S02 analysis: first, using the coarse (1-kilometer) grid network and,
second, using the fine (100-meter) grid network. Appropriate concentration
estimates from these two modeling runs are then combined with the earlier
screening results for Refinery E and the monitored background concentrations.
The highest average annual concentration resulting from this approach is 85
//g/m3, which is less than the N02 NAAQS of 100 #g/m3, annual average.
For the PM-10 NAAQS analysis, the same two-step procedure (coarse and
fine receptor grid networks) is used to locate the maximum estimated PM-10
concentration. Recognizing that the PM-10 NAAQS is a statistically-based
standard, the applicant identifies the sixth highest 24-hour concentration
(based on 5 full years of 24-hour concentration estimates) for each receptor
in the network. For the annual averaging time, the applicant averages the
5 years of modeled PM-10 concentrations at each receptor to determine the 5-
year average concentration at each receptor. To these long- and short-term
results the applicant then added the monitored background reflecting the
impacts of sources E and F, as well as surrounding area and mobile source
contributions.
C.68
-------
DRAFT
OCTOBER 1990
- For the receptor network, the highest, sixth-highest 24-hour
2
concentration is 127 /zg/m , and the highest 5-year average concentration is
3
38 M9/m • These concentrations are sufficient to demonstrate compliance with
the PM-10 NAAQS.
V.C.2 PSD Increment Analysis
The applicant starts the increment analysis by modeling the increment-
consuming sources of SC^, including the proposed new source. As a
conservative first attempt, a model run is made using the maximum allowable
SCL emissions changes resulting from each of the increment-consuming
activities identified in the increment inventory. (Note that this is not the
same as modeling the allowable emissions rate for each entire source.) Using
a coarse (1-kilometer) receptor grid, the area downwind of the source
conglomeration in the southwest quadrant was identified as the area where the
maximum concentration increases have occurred. The modeling is repeated for
the southwest quadrant using a fine (100-meter) receptor grid network.
The results of the fine-grid model run show that, in the case of peak
concentrations downwind of the southwest source conglomeration, the allowable
SOp increment will be violated at several receptors during the 24-hour
averaging period. The violations include significant ambient impacts from the
proposed power plant. Further examination reveals that Source A in the
southwest quadrant is the large contributor to the receptors where the
increment violations are predicted. The applicant therefore decides to refine
the analysis by using actual emissions increases rather than allowable
emissions increases where needed.
It is learned, and the permitting agency verifies, that the increment-
consuming boiler at Source A has burned refinery gas rather than residual oil
since start-up. Consequently, the actual emissions increase at Source A's
boiler, based upon the use of refinery gas during the preceding 2 years, is
substantially less than the allowable emissions increase assumed from the use
of residual oil. Thus, the applicant models the actual emissions increase at
Source A and the allowable emissions increase for the other modeled sources.
C.69
-------
DRAFT
OCTOBER 1990
This time the modeling is repeated only for the critical time periods and
receptors.
The maximum predicted S02 concentration increases over the baseline
concentration are as follows:
302 ng/m , 3-hour average;
•j
72 ng/m , 24-hour average; and
12 ng/m , annual average.
The revised modeling demonstrates compliance with the S02 increments. Hence,
no further S02 modeling is required for the increment analysis.
The full impact analysis for the Ntk increment is performed by modeling
Source F--the sole existing NCk increment-consuming source—and the proposed
new source. The modeled estimates yield a maximum concentration increase of
3
21 fig/m , annual average. This increase will not exceed the maximum allowable
o
increase of 25 #g/m for NCL.
With the S02 and NOo increment portions of the analysis complete, the
only remaining part is for the particulate matter (TSP) increments. The
applicant must consider the effects of the four existing increment-consuming
sources (A, B, C, and E) in addition to ambient TSP concentrations caused by
the proposed source (including the fugitive emissions). The total increase
in TSP concentrations resulting from all of these sources is as follows:
28 fiq/m , 24-hour average; and
13 #g/m , annual average.
The results demonstrate that the proposed source will not cause any violations
of the TSP increments.
C.70
-------
DRAFT
OCTOBER 1990
VI. BIBLIOGRAPHY
1. U.S. Environmental Protection Agency. Ambient Monitoring Guidelines for
Prevention of Significant Deterioration (PSD). Research Triangle Park,
NC. EPA Publication No. EPA-450/4-87-007. May 1987.
2. U.S. Environmental Protection Agency. Guideline on Air Quality Models
(Revised). Office of Air Quality Planning and Standards, Research
Triangle Park, NC. EPA Publication No. EPA-450/2-28-027R.
July 1986.
3. U.S. Environmental Protection Agency. On-Site Meteorological Programs
Guidance for Regulatory Modeling Applications. Office of Air Quality
Planning and Standards, Research Triangle Park, North Carolina. EPA
Publication No. EPA-450/4-87-013. June 1987.
4. Finkelstein, P.L., D.A. Mazzarella, T.J. Lockhart, W.J. King and J.H.
White. Quality Assurance Handbook for Air Pollution Measurement
Systems, Volume IV: Meteorological Measurements. U.S. Environmental
Protection Agency, Research Triangle Park, NC. EPA Publication No. EPA-
600/4-82-060. 1983.
5. U.S. Environmental Protection Agency. Procedures for Emission Inventory
Preparation, Volume I: Emission Inventory Fundamentals. Research
Triangle Park, NC. EPA Publication No. EPA-450/4-81-026a.
September 1981.
6. U.S. Environmental Protection Agency. Procedures for Emission Inventory
Preparation, Volume II: Point Sources. Research Triangle Park, NC.
EPA Publication No. EPA-450/4-81-026b. September 1981.
7. U.S. Environmental Protection Agency. Procedures for Emission Inventory
Preparation, Volume III: Area Sources. Research Triangle Park, NC.
EPA Publication No. EPA-450/4-81-026c. September 1981.
8. U.S. Environmental Protection Agency. Procedures for Emissions
Inventory Preparation, Volume IV: Mobile Sources. Research Triangle
Park, NC. EPA Publication No. EPA-450/4-81-026d. September 1981.
9. U.S. Environmental Protection Agency. Procedures for Emissions
Inventory Preparation, Volume V: Bibliography. Research Triangle
Park, NC. EPA Publication No. EPA-450/4-81-026e. September 1981.
10. U.S. Environmental Protection Agency. Example Emission Inventory
Documentation For Post-1987 Ozone State Implementation Plans (SIP's).
Office of Air Quality Planning and Standards, Research Triangle Park,
NC. EPA Publication No. EPA-450/4-89-018. October 1989.
C.71
-------
DRAFT
OCTOBER 1990
11 US Environmental Protection Agency. Procedures for Preparation of
Emission Inventories for Volatile Organic Compounds, Volume I: Emission
Inventory Requirements Photochemical Air Simulation Models. Office or
Air Quality Planning and Standards, NC. EPA Publication No.
EPA-450/4-79-018. September 1979.
12. U.S. Environmental Protection Agency. Screening Procedures for
Estimating Air Quality Impact of Stationary Sources. [Draft for Public
Comment.] Office of Air Quality Planning and Standards, Research
Triangle Park, NC. EPA Publication No. EPA 450/4-88-010. August 1988.
13 U.S. Environmental Protection Agency. Interim Procedures for Evaluation
of Air Quality Models (Revised). Office of Air Quality Planning and
Standards, Research Triangle Park, NC. EPA Publication No.
EPA-450/4-84-023. September 1984.
14 U.S. Environmental Protection Agency. Interim Procedures for Evaluation
of Air Quality Models: Experience with Implementation. Office of Air
Quality Planning and Standards, Research Triangle Park, NC. EPA
Publication No. EPA-450/4-85-006. July 1985.
15. U.S. Environmental Protection Agency. Guideline for Determination of
Good Engineering Practice Stack Height (Technical Support Document for
the Stack Height Regulations), Revised. Office of Air Quality Planning
and Standards, Research Triangle Park, NC. EPA Publication No.
EPA 450/4-80-023R. 1985. (NTIS No. PB 85-225241).
16. U.S. Environmental Protection Agency. Guideline for Use of Fluid
Modeling to Determine Good Engineering Practice (GEP) Stack Height.
Office of Air Quality Planning and Standards, Research Triangle Park,
NC. EPA Publication No. EPA-450/4-81-003. 1981. (NTIS No.
PB 82-145327).
17. Lawson, Jr., R.E. and W.H. Snyder. Determination of Good Engineering
Practice Stack Height: A Demonstration Study for a Power Plant. U.S.
Environmental Protection Agency, Research Triangle Park, NC. EPA
Publication No. EPA 600/3-83-024. 1983. (NTIS No. PB 83-207407).
18. Snyder, W.H., and R.E. Lawson, Jr. Fluid Modeling Demonstration of Good
Engineering-Practice Stack Height in Complex Terrain. U.S.
Environmental Protection Agency, Research Triangle Park, NC. EPA
Publication No. EPA-600/3-85-022. 1985. (NTIS No. PB 85-203107).
19. U.S. Environmental Protection Agency. Workshop on Implementing the
Stack Height Regulations (Revised). U.S. Environmental Protection
Agency, Research Triangle Park, NC. 1985.
20. U.S. Environmental Protection Agency. User's Guide to MOBILE4 (Mobile
Source Emission Factor Model). Office of Mobile Sources, Ann Arbor, MI.
EPA Publication No. EPA-AA-TEB-89-01. February 1989.
C.72
-------
DRAFT
OCTOBER 1990
21 U.S. Environmental Protection Agency. PM-10 SIP Development Guideline.
Office of Air Quality Planning and Standards, Research Triangle Park,
NC. EPA Publication No. EPA-450/2-86-001. June 1987.
C.73
-------
DRAFT
OCTOBER 1990
CHAPTER D
ADDITIONAL IMPACT ANALYSIS
I. INTRODUCTION
All PSD permit applicants must prepare additional impact analyses for
each pollutant subject to regulation under the Act which will be emitted by
this proposed new sources or modifications. This analysis assesses the
impacts of air, ground, and water pollution on soils, vegetation, and
visibility caused by any increase in emissions of any regulated pollutant from
the source or modification under review, and from associated growth.
Other impact analysis requirements may also be imposed on a permit
applicant under local, State or Federal laws which are outside the PSD
permitting process. Receipt of a PSD permit does not relieve an applicant
from the responsibility to comply fully with such requirements. For example,
two Federal laws which may apply on occasion are the Endangered Species Act
and the National Historic Preservation Act. Such legislation may require
additional analyses (although not as part of the PSD permit) if any federally-
listed rare or endangered species, or any sites that are included (or are
eligible to be included) in the National Register of Historic Sites, are
identified in the source's impact area.
Although each applicant for a PSD permit must perform an additional
impact analysis, the depth of the analysis generally will depend on existing
air quality, the quantity of emissions, and the sensitivity of local soils,
vegetation, and visibility in the source's impact area. It is important that
the analysis fully document all sources of information, underlying
assumptions, and any agreements reached with the Agency, the U.S. Forest
Service, etc.
D.I
-------
DRAFT
OCTOBER 1990
Generally, small emissions increases in most areas will not have adverse
impacts on soils, vegetation, or visibility. However, an additional impact
analysis still must be performed. Projected emissions from both the new
source or modification and emissions from associated residential, commercial,
or industrial growth are combined and modeled for the impacts assessment
analysis. While this section offers applicants a general approach to an
additional impact analysis, the analysis does not lend itself to a "cookbook"
approach.
D.2
-------
DRAFT
OCTOBER 1990
II. ELEMENTS OF THE ADDITIONAL IMPACT ANALYSIS
The additional impact analysis generally has four parts, as follows:
(1) growth;
(2) ambient air quality impact analysis;
(3) soils and vegetation impacts; and
(4) visibility impairment.
11. A. GROWTH ANALYSIS
The elements of the growth analysis include:
(1) a projection of the associated1 industrial, commercial, and
residential source growth that will occur in the area due to the
source; and
(2) an estimate of the air emissions generated by the above associated
industrial, commercial, and residential growth.
The purpose of the growth analysis is to quanitfy associated growth;
that is, to predict how much new growth is likely to occur to support the
source or modification under reveiw, and then to estimate the emissions which
will result from that associated growth. First, the applicant needs to assess
the amount of residential growth the proposed source will bring to the area.
The amount of residential growth will depend on the size of the available work
1 Associated growth (and the resultant emissions) are the growth (and
emissions) that come about as the result of the construction or modification
of a source (including secondary emissions), but which are not a part of that
source. It does not include growth which has already occurred, although an
assessment of the air quality impacts of general commercial, industrial, and
other growth which has occurred since 08/07/77 could be required under
40 CFR 51.166(n)(3)(ii) and 40 CFR 52.21(n)(2)(ii).
D.3
-------
DRAFT
OCTOBER 1990
force, the number of new employees, and the availability of housing in the
area. Associated commerical and industrial growth consists of new sources
providing goods and services to the new employees and to the proposed source
itself. Other growth is all growth not covered by the preceding, including
construction-related activities and mobile sources (permanent and temporary).
Having completed this portrait of expected growth, the applicant then
begins developing an estimate of the air pollutant emissions which would
likely result from this associated growth. The applicant should generate
emissions estimates from EPA publication AP-42, vendor emissions rates
guarantees, other PSD applications, and from existing sources.
II.B. AMBIENT AIR QUALITY ANALYSIS
The ambient air quality analysis projects the air quality which will
exist in the area of the proposed source or modification during construction
and after it begins operation. The applicant first combines the air pollutant
emissions estimates for the associated growth with the estimates of emissions
from the proposed source or modification. Next, the projected emissions from
other sources in the area which have been permitted (but are not yet in
operation) are included as inputs to the modeling analysis. The applicant
then models the combined emissions estimate and adds the modeling analysis
results to the background air quality to arrive at an estimate of the total
ground-level concentrations of polluants which can be anticipated as a result
of the construction and operation of the proposed source.
II.C. SOILS AND VEGETATION ANALYSIS
The analysis of soils and vegetation air pollution impacts should be
based on an inventory of the soils and vegetation types found in the impact
area. This inventory should include all vegetation with any commercial or
D.4
-------
DRAFT
OCTOBER 1990
recreational value. This inventory may be available from conservation groups,
State agencies, and universities.
For most types of soils and vegetation, ambient concentrations of
criteria pollutants below the secondary national ambient air quality standards
(NAAQS) will not result in harmful effects. However, there are sensitive
vegetation species (e.g., soybeans and alfalfa) which may be harmed by long-
term exposure to low ambient air concentrations of regulated pollutants for
which there are no NAAQS. For example, exposure of sensitive plant species to
0.5 micrograms per cubic meter of fluorides (a regulated, non-criteria
pollutant) for 30 days has resulted in significant foliar necrosis.
Good references for applicants and reviewers alike include the EPA Air
Quality Criteria Documents; a U.S. Department of the Interior document
entitled Impacts of Coal-Fired Plants on Fish, Uildlife, and Their Habitats;
and the U.S. Forest Service document, A Screening Procedure to Evaluate Air
Pollution Effects on Class I Uilderness Areas. Another source of reference
material is the National Park Service report, Air Quality in the National
Parks, which lists numerous studies on the biological effects of air pollution
on vegetation.
II.D. VISIBILITY IMPAIRMENT ANALYSIS
In the visibility impairment analysis, the applicant is especially
concerned with impacts that occur within the impact area of the proposed new
source or modification. Note that the visibility analysis required here is
distinct from the Class I area visibility analysis requirement. The suggested
components of a good visibility impairment analysis are:
D.5
-------
DRAFT
OCTOBER 1990
a determination of the visual quality of the area,
an initial screening of emission sources to assess the possibility
of visibility impairment, and
if warranted, a more in-depth analysis involving computer models.
The EPA's Horkbook for Plume Visual Impact Screening and Analysis
should be used to conduct a visibility impairments analysis. The workbook
outlines a screening procedure designed to expedite the analysis of emissions
impacts on the visual quality of an area. Although designed for Class I area
impacts, the outlined procedures are also generally applicable to other areas.
The following is a brief synopsis of the screening procedures.
II.D.I. SCREENING PROCEDURES: LEVEL 1
The Level 1 visibility screening analysis is a series of conservative
calculations designed to identify those emission sources that have little
potential for adversely affecting visibility. The VISCREEN model is
recommended for this first level screen. Calculated values relating source
emissions to visibility impacts are compared to a standardized screening
value. Those sources with calculated values greater than the screening
criteria are judged to have potential visibility impairments. If potential
visibility impairments are indicated, then the Level 2 analysis is undertaken.
II.D.2. SCREENING PROCEDURES: LEVEL 2
The Level 2 screening procedure is similar to the Level 1 analysis, but
utilizes more specific information regarding the source, topography, regional
visual range, and meteorological conditions. The VISCREEN model is also
recommended for this second level screening analysis.
D.6
-------
DRAFT
OCTOBER 1990
II.D.3. SCREENING PROCEDURES: LEVEL 3
If the Levels 1 and 2 screening analyses indicate the possibility of
visibility impairment, a still more detailed analysis is undertaken in Level 3
with the aid of the plume visibility model. This analysis may be performed
using models listed in Appendix B of the Guideline on Air Quality Models
(revised) and Supplement A, EPA-450/2-78-0272. The selection of the
appropriate model is done on a case-by-cas.e basis. The models generally
require more site-specific emissions and meteorological and other regional
data. The purpose of the Level 3 analysis is to provide an accurate
description of the magnitude and frequency of occurrence of impact.
II.E. CONCLUSIONS
The additional impact analysis consists of a growth analysis, a soils
and vegetation analysis, and a visibility impairment analysis. After
carefully examining all data on additional impacts, the reviewer must decide
whether the analyses performed by a particular applicant are satisfactory.
General criteria for determining the completeness and adequacy of the analyses
may include the following:
whether the applicant has presented a clear and accurate portrait
of the soils, vegetation, and visibility in the proposed impacted
area;
whether the applicant has provided adequate documentation of the
potential emissions impacts on soils, vegetation, and visibility;
and
whether the data and conclusions are presented in a logical manner
understandable by the affected community and interested public.
D.7
-------
DRAFT
OCTOBER 1990
III. ADDITIONAL IMPACT ANALYSIS EXAMPLE
Sections D.I and D.2 outlined, in general terms, the elements and
considerations found in a successful additional impact analysis. To
demonstrate how this analytic process would be applied to a specific
situation, a hypothetical case has been developed for a mine mouth power
plant. This section will summarize how an additional impact analysis would be
performed on that facility.
III.A. EXAMPLE BACKGROUND INFORMATION
The mine mouth power plant consists of a power plant and an adjoining
lignite mine, which serves as the plant's source of fuel. The plant is
capable of generating 1,200 megawatts of power, which is expected to supply a
utility grid (little is projected to be consumed locally). This project is
located in a sparsely populated agricultural area in the southwestern United
States. The population center closest to the plant is the town of
Clarksville, population 2,500, which is located 20 kilometers from the plant
site. The next significantly larger town is Milton, which is 130 kilometers
away and has a population of 20,000. The nearest Class I area is more than
200 kilometers away from the proposed construction. The applicant has
determined that within the area under consideration there are no National or
State forests, no areas which can be described as scenic vistas, and no points
of special historical interest.
The applicant has estimated that construction of the power plant and
development of the mine would require an average work force of 450 people over
a period of 36 months. After all construction is completed, about 150 workers
will be needed to operate the facilities.
D.8
-------
DRAFT
OCTOBER 1990
III.B. GROWTH ANALYSIS
\
To perform a growth analysis of this project, the applicant began by
projecting the growth associated with the operation of the project.
III.B.I. WORK FORCE
The applicant consulted the State employment office, local contractors,
trade union officers, and other sources for information on labor capability
and availability, then made the following determinations.
Most of the 450 construction jobs available will be filled by workers
commuting to the site, some from as far away as Milton. Some workers and
their families will move to Clarksville for the duration of the construction.
Of the permanent jobs associated with the project, about 100 will be filled by
local workers. The remaining 50 permanent positions will be filled by
nonlocal employees, most of whom are expected to relocate to the vicinity of
Clarksville.
The applicant quantified the temporary mobile source emissions
associated with the construction workers traveling from their homes to the job
site and the permanent increases created by the operating personnel commuting
between the plant and their homes. These emissions estimates were used in the
model ing analysis.
III.B.2. HOUSING
Contacts with local government housing authorities and realtors, and a
survey of the classified advertisements in the local newspaper indicated that
the predominant housing unit in the area is the single family house or mobile
home, and the easy availability of mobile homes and lots provides a local
capacity for quick expansion. Although there will be some emissions
D.9
-------
DRAFT
OCTOBER 1990
associated with the construction of new homes, these emissions will be
temporary and insignificant because of the limited numbers of new homes
expected.
The applicant quantified the temporary increases in area emissions
associated with heating the trailers which would be located in Clarksville
during the construction period and the permanent increase in area emissions
created by the construction and operation of the new homes which would be
built to house the families of operating personnel. These emissions estimates
were also used in the modeling analysis.
III.B.3. INDUSTRY
Although new industrial jobs often lead to new support jobs as well
(i.e., grocers, merchants, cleaners, etc.), the small number of new people
brought into the community through employment at the plant is not expected to
generate any such commercial growth. As a result of the relatively self-
contained nature of mine mouth plant operations, no related industrial growth
is expected to accompany the operation of the plant. Emergency and full
maintenance capacity is contained within the power-generating station. For
example, the proposed source will not require an increase in small support
industries (e.g., small foundries or rock crushing operations).
With no associated commercial or industrial growth projected, it then follows
that there will be no growth-related air pollution impacts.
However, there will be temporary construction-related emissions (such as
fugitive PM) from the plant itself. The applicant quantified these emissions
for the modeling analysis.
D.10
-------
DRAFT
OCTOBER 1990
III.C. AMBIENT AIR QUALITY ANALYSIS
The emissions increases identified in the growth analysis were added to
the projected emissions from the plant for modeling. The modeling analysis
results were then added to the background (monitored) ambient air quality
levels to arrive at total expected air quality loading. The results are as
follows:
Increments NAAQS
S02: 3 hour - 260 fig/m3 1300 fig/m3
24 hour - 130 fig/m3 365 fig/m3
annual - 26 fig/m3 80 fig/m3
N02: annual - 47 fig/m3 100 fig/m3
PM: 24 hour - 100 fig/m3 150 //g/m3
annual - 35 fig/m3 50 fig/m3
III.D. SOILS AND VEGETATION
In preparing a soils and vegetation analysis, the applicant acquired a
list of the soils and vegetation types indigenous to the impact area. The
vegetation is dominated by pine and hardwood trees consisting of loblolly
pine, blackjack oak, southern red oak, and sweet gum. Smaller vegetation
consists of sweetbay and holly. Small farms are found west of the forested
area. The principal commercial crops grown in the area are soybeans, corn,
okra, and peas. The soils range in texture from loamy sands to sandy clays.
The principal soil is sandy loam consisting of 50 percent sand, 15 percent
silt, and 35 percent clay.
The applicant, through a literature search and contacts with the local
universities and experts on local soils and vegetation, determined the
sensitivity of the various soils and vegetation types to each of the
D.ll
-------
DRAFT
OCTOBER 1990
applicable pollutants that will be emitted by the facility in significant
amounts. The applicant then compared this information with the estimates of
pollutant concentrations calculated previously in the air-quality modeling
analysis.
After evaluating the predicted ambient air concentration impacts on
soils and vegetation in the impact area, only soybeans proved to be
potentially sensitive. A more careful examination of soybeans revealed that
no adverse effects were expected at the low concentrations of pollutants
predicted by the modeling analysis. The predicted sulfur dioxide (S02)
ambient air concentration is below the level at which major SCL impacts on
o c
soybeans have been demonstrated (greater than 260 /jg/nr for a 24-hour period).
Fugitive emissions emitted from the mine and from coal pile storage will
be deposited on both the soils and leaves of vegetation in the immediate area
of the plant and mine. Minor leaf necrosis and lower photosynthetic activity
is expected, and over a period of time the vegetation's community structure
may change. However, this impact occurs only in an extremely limited,
nonagricultural area very near the emissions site with no recreational or
commercial value.
The potential impact of limestone preparation and storage also must be
considered. High relative humidity may produce a crusting effect of the
fugitive limestone emissions on nearby vegetation. However, for the same
reasons discussed above, there is no impact on vegetation of commerical or
recreational value.
D.12
-------
DRAFT
OCTOBER 1990
III.I. VISIBILITY ANALYSIS
Next, the applicant performed a visibility analysis, beginning with a
screening procedure similar to that outlined in the EPA document Uorkbook for
Plume Visual Impact Screening and Analysis. The screening procedure is
divided into three levels. Each level represents a screening technique for an
increasing possibility of visibility impairment. The applicant executed a
Level 1 analysis involving a series of conservative tests that permitted the
analyst to eliminate sources having little potential for adverse or
significant visibility impairment. The applicant performed these calculations
for various distances from the power plant taking into consideration the
geometry of the plume-observer relationship. In all cases, the results of the
calculations were numerically below the standardized screening criteria.
Also, in preparing the suggested visual and aesthetic description of the area
under review, the applicant noted the absence of scenic vistas, nearby
airports, or other areas which could be affected by minor reductions in
visibility. Therefore, the applicant concluded that no visibility impairment
was expected to occur within the source impact area and that the Level 2 and
Level 3 analyses were unnecessary.
III.F. EXAMPLE CONCLUSIONS
The applicant completed the additional impact analysis by documenting
every element of the analysis and preparing the report in straightforward,
concise language. This step is important, because a primary intention of the
PSD permit process is to generate public information regarding the potential
impacts of pollutants emitted by proposed new sources or modifications on
their impact areas.
D.13
-------
DRAFT
OCTOBER 1990
NOTE: This example provides only the highlights of an additional impact
analysis for a hypothetical mine mouth power plant. An actual analysis would
contain much more detail, and other types of facilities might produce more
growth and more, or different, kinds of impacts. For example, the
construction of a large manufacturing plant could easily generate air quality-
related growth impacts, such as a large influx of workers into an area and the
growth of associated industries. In addition, the existence of particularly
sensitive forms of vegetation, the presence of Class I areas, and the
existence of particular meteorological conditions would require an analysis of
much greater scope.
D.14
-------
DRAFT
OCTOBER 1990
IV.
1.
2.
3.
4.
5.
BIBLIOGRAPHY
Dvorak, A.J., et al. Impacts of Coal-fired Power Plants on Fish,
Wildlife, and their Habitats. Argonne National Laboratory. Argonne,
Illinois. Fish and Wildlife Service Publication No. FWS/OBS-78/29.
March 1978.
A Screening Procedure to Evaluate Air Pollution Effects on Class I
Wilderness Areas, U.S. Forest Service General Technical Report RM-168.
January, 1989.
Workbook for Plume Visual Impact Screening and Analysis. U.S.
Environmental Protection Agency. Research Triangle Park, N.C.
Publication No. EPA-450/4-88-015. (NTIS PB89-151278).
EPA
Air Quality in the National Parks. National Park Service. Natural
Resources Report 88-1. July, 1988.
Guideline on Air Quality Models (revised) and Supplement A. U.S.
Environmental Protection Agency. Research Triangle Park, N.C. EPA
Publication No. EPA-450/2-78-027R.
D.15
-------
DRAFT
OCTOBER 1990
CHAPTER E
CLASS I AREA IMPACT ANALYSIS
I. INTRODUCTION
Class I areas are areas of special national or regional value from a
natural, scenic, recreational, or historic perspective. The PSD regulations
provide special protection for such areas. This section identifies Class I
areas, describes the protection afforded them under the PSD provisions of the
Clean Air Act (CAA), and discusses the procedures involved in preparing and
reviewing a permit application for a proposed source with potential air
quality impacts on a Class I area.
E.I
-------
DRAFT
OCTOBER 1990
II. CLASS I AREAS AND THEIR PROTECTION
Under the^CAA, three kinds of Class I areas either have been, or may be,
designated. These are:
mandatory Federal Class I areas;
Federal Class I areas-, and
non-Federal Class I areas.
Mandatory Federal Class I areas are those specified as Class I by the CAA on
August 7, 1977, and include the following areas in existence on that date:
international parks;
national wilderness areas (including certain national wildlife
refuges, national monuments and national seashores) which exceed
5,000 acres in size;
national memorial parks which exceed 5,000 acres in size; and
national parks which exceed 6,000 acres in size.
An important feature of mandatory Federal Class I areas is that they may not
be reclassified to Class II or Class III areas. A list of these areas is
provided in Table E-l, by State . As noted in Table E-l, they are managed
either by the Forest Service (FS), National Park Service (NPS), or Fish and
Wildlife Service (FWS). More will be said about the responsibilities of these
Federal land managers (FLM) in Section II.C.
The States and Indian governing bodies have the authority to redesignate
additional areas as Class I areas. States may propose to redesignate either
State or Federal lands as Class I, while Indian governing bodies may propose
to redesignate only lands with the boundaries of their respective
E.2
-------
DRAFT
OCTOBER 1990
TABLE E-l. MANDATORY CLASS I AREAS
State/Tvpe/Area Managing Agency
State/Tvpe/Area Managing Agency
Alabama
National Wilderness Areas
Sipsey FS
Alaska
National Parks
Denali
NPS
National Wilderness Areas
Bering Sea FWS
Simeonof FWS
Tuxedni FWS
Arizona
National Parks
Grand Canyon
Petrified Forest
NPS
NPS
National Wilderness Areas
Chiricahua Nat. Monu. NPS
Chiricahua FS
Galiuro FS
Mazatzal FS
Mt. Baldy FS
Pine Mountain FS
Saguaro Nat. Monu. NPS
Sierra Ancha FS
Superstition FS
Sycamore Canyon FS
Arkansas
National Wilderness Areas
Caney Creek FS
Upper Buffalo FS
California
National Parks
Kings Canyon
Lassen Volcanic
Redwood
Sequoia
Yosemite
NPS
NPS
NPS
NPS
NPS
California - Continued
National Wilderness Areas
Agua Tibia FS
Caribou FS
Cucamonga FS
Desolation FS
Dome Land FS
Emigrant FS
Hoover FS
John Muir FS
Joshua Tree NPS
Kaiser FS
Lava Beds NPS
Marble Mountain FS
Minarets FS
Mokelumne FS
Pinnacles NPS
Point Reyes NPS
San Gabriel FS
San Gorgonio FS
San Jacinto FS
San Rafael FS
South Warner FS
Thousand Lakes FS
Ventana FS
Yolla Bolly-Middle-Eel FS
Colorado
National Parks
Mesa Verde NPS
Rocky Mountain NPS
National Wilderness Areas
Black Canyon of the Gunn. NPS
Eagles Nest FS
Flat Tops FS
Great Sand Dunes NPS
La Garita FS
Maroon Bells Snowmass FS
Mount Zirkel FS
Rawah FS
Weminuche FS
West Elk FS
E.3
-------
TABLE E-l. Continued
DRAFT
OCTOBER 1990
State/Tvpe/Area Managing Agency
State/Type/Area Managing Agency
Florida
National Parks
Everglades NPS
National Wilderness Areas
Bradwell Bay FS
Chassahowitzka FWS
Saint Marks FWS
Georgia
National Wilderness Areas
Cohutta FS
Okefenokee FWS
Wolf Island FWS
Hawaii
National Parks
Haleakala
Hawaii Volcanoes
NPS
NPS
Idaho
National Parks
Yellowstone (See Wyoming)
National Wilderness Areas
Craters of the Moon NPS
Hells Canyon (see Oregon)
Sawtooth FS
Selway-Bitterroot FS
Kentucky
National Parks
Mammoth Cave
Louisiana
National
Breton
Maine
National
Acadia
Wilderness
Parks
NPS
Areas
FWS
NPS
National Wilderness Areas
Moosehorn FWS
Michigan
National Parks
Isle Royale NPS
National Wilderness Areas
Seney FWS
Minnesota
National Parks
Voyageurs NPS
National Wilderness Areas
Boundary Waters Canoe Ar. FS
Missouri
National Wilderness Areas
Hercules-Glades
Mingo
FS
FWS
Montana
National Parks
Glacier NPS
Yellowstone (See Wyoming)
National Wilderness Areas
Anaconda-Pintlar FS
Bob Marshall FS
Cabinet Mountains FS
Gates of the Mountain FS
Medicine Lake FWS
Mission Mountain FS
Red Rock Lakes FWS
Scapegoat FS
Selway-Bitterroot (see Idaho)
U.L. Bend FWS
Nevada
National Wilderness Areas
Jarbridge
FS
New Hampshire
National Wilderness Areas
Great Gulf FS
Presidential Range-Dry R. FS
E.4
-------
DRAFT
OCTOBER 1990
TABLE E-l. Continued
st.ate/Tvpe/Area Managing Agency
State/Tvpe/Area Managing Agency
New Jersey
National Wilderness Areas
Brigantine FWS
New Mexico
National Parks
Carlsbad Caverns
NPS
National Wilderness Areas
Bandelier NPS
Bosque del Apache FWS
Gila FS
Pecos FS
Salt Creek FWS
San Pedro Parks FS
Wheeler Peak FS
White Mountain FS
North Carolina
National Parks
Great Smoky Mountains (see Tennessee)
National Wilderness Areas
Joyce Kilmer-Siickrock FS
Linville Gorge FS
Shining Rock FS
Swanquarter FWS
North Dakota
National Parks
Theodore Roosevelt
NPS
National Wilderness Areas
Lostwood FWS
Oklahoma
National Wilderness Areas
Wichita Mountains FWS
Oregon
National Parks
Crater Lake
Oregon - Continued
National Wilderness Areas
Diamond Peak FS
Eagle Cap FS
Gearhart Mountain FS
Hells Canyon FS
Kalmiopsis FS
Mountain Lakes FS
Mount Hood FS
Mount Jefferson FS
Mount Washington FS
Strawberry Mountain FS
Three Sisters FS
South Carolina
National Wilderness Areas
Cape Remain FWS
South Dakota
National Parks
Wind Cave
NPS
National Wilderness Areas
Badlands NPS
Tennessee
National Parks
Great Smoky Mountains NPS
National Wilderness Areas
Joyce Kilmer-Siickrock
(see North Carolina)
Texas
National Parks
Big Bend NPS
Guadalupe Mountain NPS
NPS
E.5
-------
DRAFT
OCTOBER 1990
TABLE E-l.* Continued
State/Tvpe/Area Managing Agency
State/Tvoe/Area Managing Agency
Utah
National Parks
Arches
Bryce Canyon
Canyon!ands
Capitol Reef
NPS
NPS
NPS
NPS
Vermont
National Wilderness Areas
Lye Brook FS
Virgin Islands
National Parks
Virgin Islands
Virginia
National Parks
Shenandoah
NPS
NPS
National Wilderness Areas
James River Face FS
Washington
National Parks
Mount Rainier
North Cascades
Olypmic
NPS
NPS
NPS
National Wilderness Areas
Alpine Lakes FS
Glacier Peak FS
Goat Rocks FS
Mount Adams FS
Pasayten FS
West Virginia
National Wilderness Areas
Dolly Sods FS
Otter Creek FS
Wisconsin
National Wilderness Area
Rainbow Lake FWS
Wyoming
National Parks
Grand Teton NPS
Yellowstone NPS
National Wilderness Areas
Bridger FS
Fitzpatrick FS
North Absaroka FS
Teton FS
Washakie FS
International Parks
Roosevelt-Campobello n/a
* For reference, all mandatory Federal Class I areas except two (Rainbow Lake
in Wisconsin and Bradwell Bay in Florida) are listed at 40 CFR 81, Subpart D -
Mandatory Class I Federal Areas Where Visibility is an Important Value.
E.6
-------
DRAFT
OCTOBER 1990
reservations. Any Federal lands which a State so redesignates are considered
Federal Class I areas. In so far as these areas are not mandatory Federal
Class II areas, these areas may be again reclassified at some later date.
(There are, as of the date of this manual, no State-designated Federal Class I
areas.) However, in accordance with the CAA the following areas may be
redesignated only as Class I or II:
an area which as of August 7, 1977, exceeded 10,000 acres in size
and was a national monument, a national primitive area, a national
preserve, a national recreational area, a national wild and scenic
river, a national wildlife refuge, a national lakeshore or seashore;
and
a national park or national wilderness area established after
August 7, 1977, which exceeds 10,000 acres in size.
Federal Class I areas are managed by the Forest Service (FS), the National
Park Service (NPS), or the Fish and Wildlife Service (FWS).
State or Indian lands reclassified as Class I are considered non-Federal
Class I areas. Four Indian Reservations which are non-Federal Class I areas
are the Northern Cheyenne, Fort Peck, and Flathead Indian Reservations in
Montana, and the Spokane Indian Reservation in Washington.
One way in which air quality degradation is limited in all Class I areas
is by stringent limits defined by the Class I increments for sulfur dioxides
(S02), particulate matter [measured as total suspended particulate (TSP)], and
nitrogen dioxide (N02). As explained previously in Chapter C, Section II.A,
PSD increments are the maximum increases in ambient pollutant concentrations
E.7
-------
DRAFT
OCTOBER 1990
allowed over baseline concentrations. In addition, the FLM of each Class I
area is charged with the affirmative responsibility to protect that area's
unique attributes, expressed generically as air quality related values
(AQRV's). The FLM, including the State or Indian governing body, where
applicable, is responsible for defining specific AQRV's for an area and for
establishing the criteria to determine an adverse impact on the AQRV's.
Congress intended the Class I increments to serve a special function in
protecting the air quality and other unique attributes in Class I areas. In
Class I areas, increments are a means of determining which party, i.e., the
permit applicant or the FLM, has the burden of proof for demonstrating whether
the proposed source would not cause or contribute to a Class I increment
violation, the FLM may demonstrate to EPA, or the appropriate permitting
authority, that the emissions from a proposed source would have an adverse
impact on any AQRV's established for a particular Class I area.
If, on the other hand, the proposed source would cause or contribute to a
Class I increment violation, the burden of proof is on the applicant to
demonstrate to the FLM that the emissions from the source would have no
adverse impact on the AQRV's. These concepts are further described in
Section III.D of this chapter.
II.A. CLASS I INCREMENTS
The Class I increments for total suspended particulate (TSP), SC^, and
N02 are listed in Table E-2. Increments are the maximum increases in ambient
pollutant concentrations allowed over baseline concentrations. Thus, these
increments should limit increases in ambient pollutant concentrations caused
by sources near Class I areas. Increment consumption analyses for Class I
E.8
-------
DRAFT
OCTOBER 1990
TABLE E-2. CLASS I INCREMENTS (ug/m3)
Pollutant Annual 24-hour 3-hour
Sulfur dioxide 2 5 25
Particulate matter (TSP) 5 10 N/A
Nitrogen dioxide 2.5 N/A N/A
E.9
-------
DRAFT
OCTOBER 1990
areas should include not only emissions from the proposed source, but also
increment-consuming emissions from other sources.
II.B. AIR QUALITY-RELATED VALUES (AQRV's)
The AQRV's are those special attributes of a Class I area that
deterioration of air quality may adversely affect. For example, the Forest
Service defines AQRV's as "features or properties of a Class I area that made
it worthy of designation as a wilderness and that could be adversely affected
by air pollution." Table E-3 presents an extensive (though not exhaustive)
list of example AQRV's and the parameters that may be used to detect air
pollution-caused changes in them. Adverse impacts on AQRV's in Class I areas
may occur even if pollutant concentrations do not exceed the Class I
increments.
Air quality-related values generally are expressed in broad terms. The
impacts of increased pollutant levels on some AQRV's are assessed by measuring
specific parameters that reflect the AQRV's status. For instance, the
projected impact on the presence and vitality of certain species of animals or
plants may indicate the impact of pollutants on AQRV's associated with species
diversity or with the preservation of certain endangered species. Similarly,
an AQRV associated with water quality may be measured by the pH of a water
body or by the level of certain nutrients in the water. The AQRV's of various
Class I areas differ, depending on the purpose and characteristics of a
particular area and on assessments by the area's FLM. Also, the concentration
at which a pollutant adversely impacts an AQRV can vary between Class I areas
because the sensitivity of the same AQRV often varies between areas.
E.10
-------
DRAFT
OCTOBER 1990
TABLE E-3. EXAMPLES OF AIR QUALITY-RELATED VALUES AND POTENTIAL
AIR POLLUTION-CAUSED CHANGES
Air Quality Related Value
Potential Air Pollution-Caused Changes
Flora and Fauna
Growth, Mortality, Reproduction, Diversity,
Visible Injury, Succession, Productivity,
Abundance
Water
Visibility
Total Alkalinity, Metals Concentration,
Anion and Cation Concentration, pH,
Dissolved Oxygen
Contrast, Visual Range, Coloration
Cultural-Archeol og ical
and Paleontological
Decomposition Rate
Odor
Odor
E.ll
-------
DRAFT
OCTOBER 1990
When a proposed major source's or major modification's modeled emissions
may affect a Class I area, the applicant analyzes the source's anticipated
impact on visibility and provides the information needed to determine its
effect on the area's other AQRV's. The FLM's have established criteria for
determining what constitutes an "adverse" impact. For example, the NPS
defines an "adverse impact" as "any impact that: (1) diminishes the area's
national significance; (2) impairs the structure or functioning of ecosystems;
or (3) impairs the quality of the visitor experience." If a FLM determines,
based on any information available, that a source will adversely impact AQRV's
in a Class I area, the FLM may recommend that the reviewing agency deny
issuance of the permit, even in cases where no applicable increments would be
exceeded.
II.C. FEDERAL LAND MANAGER
The FLM of a Class I area has an affirmative responsibility to protect
AQRV's for that area which may be adversely affected by cumulative ambient
pollutant concentrations. The FLM is responsible for evaluating a source's
projected impact on the AQRV's and recommending that the reviewing agency
either approve or disapprove the source's permit application based on
anticipated impacts. The FLM also may suggest changes or conditions on a
permit. However, the reviewing agency makes the final decisions on permit
issuance. The FLM also advises reviewing agencies and permit applicants about
other FLM concerns, identifies AQRV's and assessment parameters for permit
applicants, and makes ambient monitoring recommendations.
The U.S. Departments of Interior (DOI) and Agriculture (USDA) are the
FLM's responsible for protecting and enhancing AQRV's in Federal Class I
areas. Those areas in which the DOI has authority are managed by the NPS and
the FWS, while the USDA Forest Service separately reviews impacts on Federal
E.12
-------
DRAFT
OCTOBER 1990
Class I national wildernesses under its jurisdiction. The Federal PSD
regulations specify that the Administrator furnish written notice of any
permit application for a proposed major stationary source or major
modification, the emissions from which may affect a Class I area, to the FLM
and the official charged with direct responsibility for management of any
lands within the area. Although the Secretaries of Interior and Agriculture
are the FLM's for Federal Class I areas, they have delegated permit review to
specific elements within each department. In the DOI, the NPS Air Quality
Division reviews PSD permits for both the NPS and FWS. Hence, for sources
that may affect wildlife refuges, applicants and reviewing agencies should
contact and send correspondence to both the NPS and the wildlife refuge
manager located at the refuge. Table E-4 summarizes the types of Federal
Class I areas managed by each FLM. In the USDA, the Forest Service has
delegated to its regional offices (listed in Table E-5) the responsibility for
PSD permit application review.
E.13
-------
DRAFT
OCTOBER 1990
TABLE E-4. FEDERAL LAND MANAGERS
Federal Land
Manager
Federal Class I Areas
Managed
Address
National Park
Service (DOI)
National Memorial Parks
National Monuments
National Parks
National Seashores
1
Air Quality Division
National Park Service - Air
P.O. Box 25287
Denver, CO 80225-0287
Fish and Wildlife
Service (DOI)
National Wildlife
Refuges1
Send to NPS, above, and
to Wildlife Refuge
Manager.
Forest Service
(USDA)
National Wildernesses
Send to Forest Service
Regional Office
(See Table E-5)
those national monuments, seashores, and wildlife refuges which also
were designated wilderness areas as of August 7, 1977 are included as .
mandatory Federal Class I areas.
2The Wildlife Refuge Manager is located at or near each refuge.
E.14
-------
DRAFT
OCTOBER 1990
TABLE E-5. USDA FOREST SERVICE REGIONAL OFFICES
AND STATES THEY SERVE*
USDA Forest Service
Northern Region
Federal Building
P.O. Box 7669
Missoula, MT 59807
[ID, ND, SD, MT]
USDA Forest Service
Rocky Mountain Region
11177 West 8th Avenue
P.O. Box 25127
Lakewood, CO 80225
[CO, KS, NE, SD, WY]
USDA Forest Service
Southwestern Region
Federal Building
517 Gold Avenue, SW
Albuquerque, NM 87102
[AZ, NM]
USDA Forest Service
Intermountain Region
Federal Building
324 25th Street
Ogden, UT 84401
[ID, UT, NV, WY]
USDA Forest Service
Pacific Southwest Region
630 Sansome Street
San Francisco, CA 94111
[CA, HI, GUAM, Trust Terr, of Pacific]
USDA Forest Service
Pacific Northwest Region
P.O. Box 3623
Portland, OR 97208
[WA, OR]
USDA Forest Service
Southern Region
1720 Peachtree Road, NW
Atlanta, GA 30367
[AL, AR, FL, GA, KY, LA, MS, NC, OK,
PR, SC, TN, TX, VI, VA]
USDA Forest Service
Alaska Region
P.O. Box 21628
Juneau, AK 99802-1628
[AK]
USDA Forest Service
Eastern Region
310 W. Wisconsin Avenue, Room 500
Milwaukee, WI 53203
[CT, DE, IL, IN, IA, ME, MD, MA, MI,
MN, MO, NH, NY, NJ, OH, PA, RI, VT,
WV, WI]
* Some Regions serve only part of a State.
E.15
-------
DRAFT
OCTOBER 1990
III. CLASS I AREA IMPACT ANALYSIS AND REVIEW
This section presents the procedures an applicant should follow in
preparing an analysis of a proposed source's impact on air quality and AQRV's
in Class I areas, including recommended informal steps. For each participant
in the analysis - the permit applicant, the FLM, and the permit reviewing
agency - the section summarizes their role and responsibilities.
III.A. SOURCE APPLICABILITY
If a proposed major source or major modification may affect a Class I
area, the Federal PSD regulations require the reviewing authority to provide
written notification of any such proposed source to the FLM (and the DOI and
USDA officials delegated permit review responsibility). The meaning of the
term "may affect" is interpreted by EPA policy to include all major sources or
major modifications which propose to locate within 100 kilometers (km) of a
Class I area. Also, if a major source proposing to locate at a distance
greater than 100 km is of such size that the reviewing agency or FLM is
concerned about potential emission impacts on a Class I area, the reviewing
agency can require the applicant to perform an analysis of the source's
potential emissions impacts on the Class I area. This is because certain
meteorological conditions, or the quantity or type of air emissions from large
sources locating further than 100 km, may cause adverse impacts on a Class I
area. A reviewing agency should not exclude a major new source or major
modification from performing an analysis of the potential impacts if the FLM
identifies some reason to believe that the source would affect a Class I area.
The EPA requires a NAAQS and increment analysis of any PSD source the
emissions from which increase pollutant concentration by 1 #g/m3 or more (24-
E.16
-------
DRAFT
OCTOBER 1990
hour average) in a Class I area. However, certain AQRV's may be sensitive to
pollutant increases less than 1 #g/m3. Some Class I areas may be approaching
the threshold for effects by a particular pollutant on certain resources and
consequently may be sensitive to even small increases in pollutant
concentrations. For example, in some cases increases in sulfate concentration
of 1 /zg/m3 or less may adversely impact visibility. Thus, a 24-hour average
increase of 1 /jg/m3 should not absolutely determine whether an AQRV impact
analysis is needed. The reviewing agency should consult the FLM to determine
whether to require all the information necessary for a complete AQRV impact
analysis of a proposed source.
III.B. PRE-APPLICATION STAGE
A pre-application meeting between the applicant, the FLM, and the
reviewing agency to discuss the information required of the source is highly
recommended. The applicant should contact the appropriate FLM as soon as
plans are begun for a major new source or modification near a Class I area
(i.e., generally within 100 km of the Class I area). A preapplication
meeting, while not required by regulation, helps the permit applicant
understand the data and analyses needed by the FLM. At this point, given
preliminary information such as the source's location and the type and
quantity of projected air emissions, the FLM can:
agree on which Class I areas are potentially affected by the source;
discuss AQRV's for each of the areas(s) and the indicators that may
be used to measure the source's, impact on those AQRV's;
advise the source about the scope of the analysis for determining
whether the source potentially impacts the Class I area(s);
discuss which Class I area impact analyses the applicant should
include in the permit application; and
E.17
-------
DRAFT
OCTOBER 1990
discuss all pre-application monitoring in the Class I area that may
be necessary to assess the current status of, and effects on, AQRV's
(this monitoring usually is done by the applicant).
III.C. PREPARATION OF PERMIT APPLICATION
For each proposed major new source or major modification that may affect
a Class I area, the applicant is responsible for:
identifying all Class I areas within 100 km of the proposed source
and any other Class I areas potentially affected;
performing for each Class I area any preliminary analysis required
by a reviewing agency to find whether the source may increase the
ambient concentration of any pollutant by 1 #g/m3 (24-hour average)
or more;
performing all necessary Class I increment analyses (including any
necessary cumulative impact analyses) when a significant ambient
impact is predicted;
providing the information necessary to conduct the AQRV impact
analyses;
performing any monitoring within the Class I area required by the
reviewing agency; and
providing the reviewing agency with any additional relevant
information the agency requests to "complete" the Class I area
impacts analysis.
By involving the FLM early in preparation of the Class I area analysis, the
applicant can identify and address FLM concerns, avoiding delays later during
permit review.
The FLM is the AQRV expert for Class I areas. As such, the FLM can
recommend to the applicant:
E.18
-------
DRAFT
OCTOBER 1990
the AQRV's the applicant should address in the PSD permit
application's Class I area impact analysis;
techniques for analyzing pollutant effects on AQRV's;
the criteria the FLM will use to determine whether the emissions
from the proposed source would have an adverse impact on any AQRV;
the pre-construction and post-construction AQRV monitoring the FLM
will request that the reviewing agency require of the applicant; and
the monitoring, analysis, and quality assurance/quality control
techniques the permit applicant should use in conducting the AQRV
monitoring.
The permit applicant and the FLM also should keep the reviewing agency
apprised of all discussions concerning a proposed source.
III.D. PERMIT APPLICATION REVIEW
Where a reviewing agency anticipates that a proposed source may affect a
Class I area, the reviewing agency is responsible for:
sending the FLM a copy of any advance notification that an applicant
submits within 30 days of receiving such notification;
sending EPA a copy of each permit application and a copy of any
action relating to the source;
sending the FLM a complete copy of all information relevant to the
permit application, including the Class I visibility impacts
analysis, within 30 days of receiving it and at least 60 days before
any public hearing on the proposed source (the reviewing agency may
wish to request that the applicant furnish 2 copies of the permit
applicat ion);
providing the FLM a copy of the preliminary determination document;
and
E.19
-------
DRAFT
OCTOBER 1990
making a final determination whether construction should be
approved, approved with conditions, or disapproved.
A reviewing agency's policy regarding Class I area impact analyses can
ensure FLM involvement as well as aid permit applicants. Some recommended
policies for reviewing agencies are:
not considering a permit application complete until the FLM
certifies that it is "complete" in the sense that it contains
adequate information to assess adverse impacts on AQRV's;
recommending that the applicant agree with the FLM (usually well
before the application is received) on the type and scope of AQRV
analyses to be done;
deferring to the FLM's adverse impact determination, i.e., denying
permits based on FLM adverse impact certifications; and
where appropriate, incorporating permit conditions (e.g., monitoring
program) which will assure protection of AQRV's. Such conditions
may be most appropriate when the full extent of the AQRV impacts is
uncertain.
In addition, the reviewing agency can serve as an arbitrator and advisor in
FLM/applicant agreements, especially at meetings and in drafting any written
agreements.
While the FLM's review of a permit application focuses on emissions
impacts on visibility and other AQRV's, the FLM may comment on all other
aspects of the permit application. The FLM should be given sufficient time
(at least 30 days) to thoroughly perform or review a Class I area impact
analysis and should receive a copy of the permit application either at the
same time as the reviewing agency or as soon after the reviewing agency as
possible.
E.20
-------
DRAFT
OCTOBER 1990
The FLM can make one of two decisions on a permit application: (1) no
adverse impacts; or (2) adverse impact based on any available information.
Where a proposed major source or major modification adversely impacts a
Class I area's AQRV's, the FLM can recommend that the reviewing agency deny
the permit request based on the source's projected adverse impact on the
area's AQRV's. However, rather than recommending denial at this point, the
FLM may work with the reviewing agency to identify possible permit conditions
that, if agreed to by the applicant, would make the source's effect on AQRV's
acceptable. In cases where the permit application contains insufficient
information for the FLM to determine AQRV impacts, the FLM should notify the
reviewing agency that the application is incomplete.
During the public comment period, the FLM can have two roles: 1) final
determination on the source's impact on AQRV's with a formal recommendation to
the reviewing agency; and 2) a commenter on other aspects of the permit
application (best available control technology, modeling, etc.). Even for PSD
permit applications where a proposed source's emissions clearly would not
cause or contribute to exceedances of any Class I increment, the FLM may
demonstrate to the reviewing agency that emissions from the proposed source or
modification would adversely impact AQRV's of a Class I area and recommend
denial. Conversely, a permit applicant may demonstrate to the FLM that a
proposed source's emissions do not adversely affect a Class I area's AQRV's
even though the modeled emissions would cause an exceedance of a Class I
increment. Where a Class I increment is exceeded, the burden of proving no
adverse impact on AQRV's is on the applicant. If the FLM concurs with this
demonstration, the FLM may recommend approval of the permit to the reviewing
agency and such a permit may be issued despite projected Class I increment
violations. However, in those cases where the permitted source would cause or
contribute to a Class I increment violation, such pollutant increases must not
be allowed to cause or contribute to ambient concentration increases which
would violate the Class II increments (see Table C-2, Chapter C).
E.21
-------
DRAFT
OCTOBER 1990
IV. VISIBILITY IMPACT ANALYSIS AND REVIEW
Visibility is singled out in the regulations for special protection and
enhancement in accordance with the national goal of preventing any future, and
remedying any existing, impairment of visibility in Class I areas caused by
man-made air pollution. The visibility regulations for new source review
(40 CFR 51.307 and 52.27) require visibility impact analysis in PSD areas for
major new sources or major modifications that have the potential to impair
visibility in any Class I area. Information on screening models available for
visibility analysis can be found in the manual "Workbook for Plume Visual
Impact Screening and Analysis," EPA-450/4-88-015 (9/88).
IV.A VISIBILITY ANALYSIS
An "adverse impact on visibility" means visibility impairment which
interferes with the management, protection, preservation, or enjoyment of a
visitor's visual experience of the Class I area. The FLM makes the
determination of an adverse impact on a case-by-case basis taking into account
the geographic extent, duration, intensity, frequency and time of visibility
impairment, and how these factors correlate with (1) times of visitor use of
the Class I area, and (2) the frequency and timing of natural conditions that
reduce visibility. Visibility perception research indicates that the visual
effects of a change in air quality requires consideration of the features of
the particular vista as well as what is in the air, and that measurement of
visibility usually reflects the change in color, texture, and form of a scene.
The reviewing agency may require visibility monitoring in any Class I area
near a proposed new major source or modification as the agency deems
appropriate.
E.22
-------
DRAFT
OCTOBER 1990
An integral vista is a view perceived from within a Class I area of a
specific landmark or panorama located outside of the Class I area. A
visibility impact analysis is required for the integral vistas identified at
40 CFR 81, Subpart D, and for any other integral vista identified in a SIP.
IV.B PROCEDURAL REQUIREMENTS
When the reviewing agency receives advance notification (e.g., early
consultation with the source prior to submission of the application) of a
permit application for a source that may affect visibility in a Class I area,
the agency must notify the appropriate FLM within 30 days of receiving the
notification. The reviewing agency must, upon receiving a permit application
for a source that may affect Class I area visibility, notify the FLM in
writing within 30 days of receiving it and at least 60 days prior to the
public hearing on the permit application. This written notification must
include an analysis of the source's anticipated impact on visibility in any
Class I area and all other information relevant to the permit application.
The FLM has 30 days after receipt of the visibility impact analysis and other
relevant information to submit to the reviewing agency a finding that the
source will adversely impact visibility in a Class I area.
If the FLM determines that a proposed source will adversely impact
visibility in a Class I area and the reviewing agency concurs, the permit may
not be issued. Where the reviewing agency does not agree with the FLM's
finding of an adverse impact on visibility the agency must, in the notice of
public hearing, either explain its decision or indicate where the explanation
can be obtained.
E.23
-------
DRAFT
OCTOBER 1990
V. BIBLIOGRAPHY
1. Workbook for Plume Visual Impact Screening and Analysis. U.S.
Environmental Protection Agency, Research Triangle Park, NC.
EPA-450/4-88-015. September 1988.
2. Workbook for Estimating Visibility Impairment. U.S. Environmental
Protection Agency. Research Triangle Park, NC. EPA-450/4-80-031.
November 1980. (NTIS No. PB 81-157885).
3. USDA Forest Service (1987a) Air Resource Management Handbook, FSH
2509.19.
4. USDA Forest Service (1987b) Protocols for Establishing Current Physical.
Chemical, and Biological Conditions of Remote Alpine and Subalpine
Ecosystems. Rocky Mountain Forest and Range Experiment Station General
Technical Report No. 46, Fort Collins, Colorado.
5. USDA Forest Service (1987c). A Screening Procedure to Evaluate Air
Pollution Effects on Class I Wilderness Areas. Rocky Mountain Forest and
Range Experiment Station GTR 168. Fort Collins, Colorado.
6. USDI (1982) "Internal Procedures for Determinations of Adverse Impact
Under Section 165(d)(2)(C)(ii) and (iii) of the Clean Air Act" 47 FR
30226, July 12, 1982.
7. DOI National Park Service (1985) Permit Application Guidance for New Air
Pollution Sources. Natural Resources Report Series 85-2, National Park
Service, Air Quality Division, Permit Review and Technical Support
Branch, Denver, Colorado.
8. DOI National Park Service, Air Resource Management Manual. National Park
Service, Air Quality Division, Permit Review and Technical Support
Branch, Denver, Colorado.
E.24
-------
PART II
NONATTAINMENJ AREAS
Chapter F - Nonattainment Area Applicability
Chapter G - Nonattainment Area Requirements
-------
DRAFT
OCTOBER 1990
CHAPTER F
NONATTAINMENT AREA APPLICABILITY
I. INTRODUCTION
Many of the elements and procedures for source applicability under the
nonattainment area NSR applicability provisions are similar to those of PSD
applicability. The reader is therefore encouraged to become familiar with the
terms, definitions and procedures from Part I.A., "PSD Applicability," in this
manual. Important differences occur, however, in three key elements that are
common to applicability determinations for new sources or modifications of
existing sources located in attainment (PSD) and nonattainment areas. Those
elements are:
• Definition of "source,"
• Pollutants that must be evaluated (geographic effects); and
• Applicability thresholds
Consequently, this section will focus on these three elements in the context
of a nonattaiment area NSR program. Note that the two latter elements,
pollutants that must be evaluated for nonattainment NSR due to the location of
the source in designated nonattainment areas (geographic effects) and
applicability thresholds, are not independent. They will, therefore, be
discussed in section III.
F.I
-------
DRAFT
OCTOBER 1990
II. DEFINITION OF SOURCE
The original NSR regulations required that a source be evaluated
according to a dual source definition. On October 14, 1981, however, the EPA
revised the new source review regulations to give a State the option of
adopting a plantwide definition of stationary source in nonattainment areas,
if the State's SIP did not rely on the more stringent "dual" definition in its
attainment demonstration. Consequently, there are two stationary source
definitions for nonattainment major source permitting: a "plantwide"
definition and a "dual" source definition. The permit application must use.
and be reviewed according to. whichever of the two definitions is used to
define a stationary source in the applicable SIP.
II.A. "PLANTWIDE" STATIONARY SOURCE DEFINITION
The EPA definition of stationary source for nonattainment major source
permitting uses the "plantwide" definition, which is the same as that used in
PSD. A complete discussion of the concepts associated with the plantwide
definition of source are presented in the PSD part of this manual (see
section II). In essence, this definition provides that only physical or
operation changes that result in a significant net emissions increase at the
entire plant are considered a major modification to an existing major source
(see sections II and III).
For example, if an existing major source proposes to increase
emissions by constructing a new emissions unit but plans to reduce
actual emissions by the same amount at another emissions unit at
the plant (assuming the reduction is federally enforceable and is
the only contemporaneous and creditable emissions change at the
source), then there would be no net increase in emissions at the
plant and therefore no "major" modification to the stationary
source.
F.2
-------
DRAFT
OCTOBER 1990
II.B. "DUAL SOURCE" DEFINITION OF STATIONARY SOURCE
The "dual" definition of stationary source defines the term stationary
source as ". . . any building, structure, facility, or installation which
emits or has the potential to emit any air pollutant subject to regulation
under the Clean Air Act." Under this definition, the three terms building,
structure, or facility are defined as a single term meaning all of the
pollutant-emitting activities which belong to the same industrial grouping
(i.e.,same two-digit SIC code), are located on one or more adjacent
properties, and are under the control of the same owner or operator. The
fourth term, installation, means an identifiable piece of process equipment.
Therefore, a stationary source is both:
a building, structure, or facility (plantwide); and
an installation (individual piece of equipment).
In other words, the "dual source" definition of stationary source treats
each emissions unit as (1) a separate, independent stationary source, and (2)
a component of the entire stationary source.
For example., in the case of a power plant with three large boilers
each emitting major amounts (i.e., >100 tpy) of NO , each of the
three boilers is an individual stationary source ami all three
boilers together constitute a stationary source. [Note that the
power plant would be seen only as a single stationary source under
the plantwide definition (all three boilers together as one
stationary source)].
Consequently, under the dual source definition, the emissions from each
physical or operational change at a plant are reviewed both with and without
regard to reductions elsewhere at the plant.
For example, a power plant is an existing major SO- source in an
SO, nonattainment area. The power plant proposes to 1) install
SOJ scrubbers on an existing boiler and 2) construct a new boiler
F.3
-------
DRAFT
OCTOBER 1990
at the same facility. Under the "piantwide" definition, the S02
reductions from the scrubber installation could be considered,
along with other contemporaneous emissions changes at the plant
and the new emissions increase of the new boiler to arrive at the
source's net emission increase. This might result in a net
emissions change which would be below the SO- significance level
and the new boiler would "net" out of revietras major
modification. Under the dual source definition, however, the new
boiler would be regarded as a individual source and would be
subject to nonattainment NSR requirements if its potential
emissions exceed the 100 tpy threshold. The emissions reduction
from the scrubber could not be used to reduce net source
emissions, but would instead be regarded as an S02 emissions
reduction from a separate source.
The following examples are provided to further clarify the application
of the dual source definition to determine if a modification to an existing
major source is major and, therefore, subject to major source NSR permitting
requirements.
Example 1 An existing major stationary source is located in a
nonattainment area for NO where the "dual source"
definition applies, and has the following emissions units:
Unit #1 with a potential to emit of 120 tpy of NOX
Unit #2 with a potential to emit of 80 tpy of NOX
Unit #3 with a potential to emit of 120 tpy of NO
A
Unit #4 with a potential to emit of 130 tpy of NO
A
Case 1 A modification planned for Unit #1 will result in an emissions
increase of 45 tpy of NO The following emissions changes are
contemporaneous with the proposed modification (all case examples
assume that increases and decreases are creditable and will be
made federally enforceable by the reviewing authority when the
modification is permitted and will occur before construction of
the modification):
Unit #3 had an actual decrease of 10 tpy NO
A
Unit #4 had an actual decrease of 10 tpy NO
A.
F.4
-------
DRAFT
OCTOBER 1990
Only contemporaneous emissions changes at Unit #1 are considered
because Unit #1 is a major source of NO by itself (i.e.,
potential emissions of NO are greater than 100 tpy). The
proposed increase at unit #1 of 45 tpy is greater than the 40 tpy
NO significant emissions rate since the emissions changes at the other
units are not considered. Consequently, the proposed modification to
Unit #1 is major under the dual source definition.
Case 2 A modification to unit #2 is planned which will result in an emissions
increase of 45 tpy of NO . The following emissions changes are
contemporaneous with the proposed modification:
Unit #1 had an actual decrease of 10 tpy
Unit #3 had an actual decrease of 10 tpy
Unit #2 7's not a major stationary source in and of itself (i.e.,
its potential to emission of 80 tpy NO is less than the 100 tpy
major source threshold). Therefore, tne major stationary source
being modified is the whole plant and the emissions decreases at
units #1 and #3 are considered in calculating the net emissions
change at the source. The net emissions change of 25 tpy (the sum
of +45, -10, and -10) at the source is less than the applicable 40
tpy NO significant emissions rate. Consequently, the proposed
modification is not major.
x
Case 3 A brand new unit #5 with a potential to emission of 45 tpy of NO
(note that potential emissions are less than the 100 tpy major
source cutoff) is being added to the plant. The following
emissions changes are contemporaneous with the proposed
modification:
Unit #1 had an actual decrease of 15 tpy
Unit #2 had an actual increase of 25 tpy
Unit #3 had an actual decrease of 20 tpy
The new unit #5 is not a major stationary source in and of itself.
Therefore, the major stationary source being modified is the whole
plant and the emissions decreases at units #1, #2 and #3 are
considered in calculating the net emissions change at the source.
The net emissions change of 35 tpy (the sum of + 45, -15, +25, and
-20) at the source is less than the applicable 40 tpy NO
significance level. Therefore, the proposed unit #5 TS wot a
major modification.
F.5
-------
DRAFT
OCTOBER 1990
Case 4 A brand new unit #6 with a potential to emit of NOX of 120 tpy is
being added to the plant. Because the new unit is, by itself, a
new major source (i.e., potential NO emissions are greater than
the 100 tpy major source cutoff), it cannot net out of review
(using emissions reductions achieved at other emissions units at
the plant) under the dual source definition.
Example 2 An existing plant has only two emissions units. The units have a
potential to emit of 25 tpy and 40 tpy. Here, any modification to
the plant would have to have a potential to emit greater than 100
tpy before the modification is major and subject to review. This
is because neither of the two existing emissions units (at 25 tpy
and 40 tpy), nor the total plant (at 65 tpy) are considered to be
a major source (i.e., existing potential emissions do not exceed
100 tpy). If, however, a third unit with potential emissions of
110 tpy were added, that unit would be subject to review
regardless of any emissions reductions from the two existing
units.
F.6
-------
DRAFT
OCTOBER 1990
III. POLLUTANTS ELIGIBLE FOR REVIEW AND APPLICABILITY THRESHOLDS
III.A. POLLUTANTS ELIGIBLE FOR REVIEW (GEOGRAPHIC CONSIDERATIONS)
A new source will be subject to nonattainment area preconstruction
review requirements only if it will emit, or will have the potential to emit,
in major amounts any criteria pollutant for which the area has been designated
nonattainment. Similarly, only if a modification results in a significant
increase (and significant net emissions increase under the plantwide source
definition) of a pollutant, for which the source is major and for which the
area is designated nonattainment, do nonattainment requirements apply.
III.B. MAJOR SOURCE THRESHOLD
For the purposes of nonattainment NSR, a major stationary source is
any stationary source which emits or has the potential
to emit 100 tpy of any [criteria] pollutant subject to
regulation under the CAA, or
any physical change or change in method of operation at an
existing non-major source that constitutes a major
stationary source by itself.
Note that the 100 tpy threshold applies to all sources. The alternate
250 tpy major source threshold [for PSD sources not classified under one of
the 28 regulated source categories identified in Section 169 of the CAA (See
Section I.A.2.3 and Table I-A-1) as being subject to a 100 tpy threshold] does
not exist for nonattainment area sources.
F.7
-------
DRAFT
OCTOBER 1990
III.C. MAJOR MODIFICATION THRESHOLDS
Major modification thresholds for nonattainment areas are those same
significant emissions values used to determine if a modification is major for
PSD. Remember, however, that only criteria pollutants for which the location
of the source has been designated nonattainment are eligible for evaluation.
F.8
-------
DRAFT
OCTOBER 1990
IV. NONATTAINMENT APPLICABILITY EXAMPLE
The following example illustrates the criteria presented in sections II
and III above.
Construction of a new plant with potential emissions of 500 tpy S0-, 50
tpy VOC and 30 tpy NO is proposed for an area designated nonattainment
for SOy and ozone amT attainment for NO . (Recall that VOC is the
regulated surrogate pollutant for ozone.) The new plant is major for
S0y and therefore would be subject to nonattainment requirements for S02
only. Even though the VOC emissions are significant, the source is
minor for VOC, and according to nonattainment regulations, is not
subject to major source review. For purposes of PSD, the NO emissions
are neither major nor significant and are, therefore, not subject to PSD
review.
Two years after construction on the new plant commences, a modification
of this plant is proposed that will result in an emissions increase of
60 tpy VOC and 35 tpy NOV without any creditable contemporaneous
emissions reductions. Again, the VOC emissions increase would not be
subject, because the existing source is not major for VOC. The
emissions increase of 35 tpy N0y is not significant and again, is not
subject to PSD review. Note, however, that the plant would be
considered a major source of VOC in subsequent applicability
determinations.
One year later, the plant proposes another increase in VOC emissions by
75 tpy and NOV by another 45 tpy, again with no contemporaneous
emissions redactions. Because the existing plant is now major for VOL
and will experience a significant net emissions '"crease of that
pollutant, it will be subject to nonattainment HSR for VOC. Because tht
source is major for a regulated pollutant (VOC) and will experience a
significant net emissions increase of an attainment pollutant (NOJ, it
will also be subject to PSD review.
F.9
-------
DRAFT
OCTOBER 1990
CHAPTER G
NONATTAINMENT AREA REQUIREMENTS
I. INTRODUCTION
The preconstruction review requirements for major new sources or major
modifications locating in areas designated nonattainment pursuant to section
107 of the Act differ from prevention of significant deterioration (PSD)
requirements. First, the emissions control requirement for nonattainment
areas, lowest achievable emission rate (LAER), is defined differently than the
best available control technology (BACT) emissions control requirement.
Second, the source must obtain any required emissions reductions (offsets) of
the nonattainment pollutant from other sources which impact the same area as
the proposed source. Third, the applicant must certify that all other sources
owned by the applicant in the State are complying with all applicable
requirements of the CAA, including all applicable requirements in the State
implementation plan (SIP). Fourth, such sources impacting visibility in
mandatory class I Federal areas must be reviewed by the appropriate Federal
land manager (FLM).
G.I
-------
DRAFT
OCTOBER 1990
II. LOWEST ACHIEVABLE EMISSION RATE (LAER)
For major new sources and major modifications in nonattainment areas,
LAER is the most stringent emission limitation derived from either of the
following:
the most stringent emission limitation contained in the
implementation plan of any State for such class or category of
source; or
the most stringent emission limitation achieved in practice by
such class or category of source.
The most stringent emissions limitation contained in a SIP for a class or
category of source must be considered LAER, unless (1) a more stringent
emissions limitation has been achieved in practice, or (2) the SIP limitation
is demonstrated by the applicant to be unachievable. By definition LAER can
not be less stringent than any applicable new source performance standard
(NSPS).
There is, of course, a range of certainty in such a definition. The
greatest certainty for a proposed LAER limit exists when that limit is
actually being achieved by a source. However, a SIP limit, even if it has not
yet been applied to a source, should be considered initially to be the product
of careful investigation and, therefore, achievable. A SIP limit's
credibility diminishes if a) no sources exist to which it applies; b) it is
generally acknowledged that sources are unable to comply with the limit and
the State is in the process of changing the limit; or c) the State has relaxed
the original SIP limit. Case-by-case evaluations need to be made in these
situations to determine the SIP limit's achievability.
The same logic applies to SIP limits to which sources are subject but
with which they are not in compliance. Noncompliance by a source with a SIP
G.2
-------
DRAFT
OCTOBER 1990
limit, even if it is the only source subject to that specific limit, does not
automatically constitute a demonstration that the limit is unachievable. The
specific reasons for noncompliance must be determined, and the ability of the
source to comply assessed. However, such noncompliance may prove to be an
indication of nonachievability, so the achievability of such a SIP limitation
should be carefully studied before it is used as the basis of a LAER
determination. Some recommended sources of information for determining LAER
are:
SIP limits for that particular class or category of sources;
preconstruction or operating permits issued in other
nonattainment areas; and
the BACT/LAER Clearinghouse.
Several technological considerations are involved in selecting LAER.
The LAER is an emissions rate specific to each emissions unit including
fugitive emissions sources. The emissions rate may result from a combination
of emissions-limiting measures such as (1) a change in the raw material
processed, (2) a process modification, and (3) add-on controls. The
reviewing agency determines for each new source whether a single control
measure is appropriate for LAER or whether a combination of emissions-limiting
techniques should be considered.
The reviewing agency also can require consideration of technology
transfer. There are two types of potentially transferable control
technologies: (1) gas stream controls, and (2) process controls and
modifications. For the first type of transfer, classes or categories of
sources to consider are those producing similar gas streams that could be
controlled by the same or similar technology. For the second type of
transfer, process similarity governs the decision.
G.3
-------
DRAFT
OCTOBER 1990
Unlike BACT, the LAER requirement does not consider economic, energy, or
other environmental factors. An emissions limit should not be considered for
LAER (achievable) if the cost of maintaining the associated level of control
is so great that a major new source could not be built or operated. This
applies generically, i.e., no new plants could be built in a particular
industry due to economic constraints incurred by a particular control
technology. If another plant in the same (or comparable) industry already
uses that control technology, then such use constitutes evidence that the cost
to the industry of that control is not prohibitive. Thus, for a new source,
LAER costs are considered only to the degree that they reflect unusual
circumstances which in some manner differentiate the cost of control for that
source from control costs for the rest of the industry. Therefore, no
discussion of costs is necessary or appropriate if sources in the industry are
already using that control.
Where technically feasible, LAER generally is specified as both a
numerical emissions limit (e.g., Ib/MMBtu) and an emissions rate (e.g.,
Ib/hr). Where numerical levels reflect assumptions about the performance of a
control technology, the permit should specify both the numerical emissions
rate and limitation and the control technology. In some cases where
enforcement of a numerical limitation is judged to be technically infeasible,
the permit may specify a design, operational, or equipment standard; however,
such standards must be clearly enforceable, and the reviewing agency must
still make an estimate of the resulting emissions for offset purposes.
G.4
-------
DRAFT
OCTOBER 1990
III. EMISSIONS REDUCTIONS "OFFSETS"
Unless the area of concern is covered by an EPA-approved growth
allowance, a major source or major modification planned in a nonattainment
area must obtain emissions offsets as a condition for approval. These
offsets, generally obtained from existing sources located in the vicinity of a
proposed source, must (1) offset the emissions increase from the new source or
modification and (2) provide a net air quality benefit. The obvious purpose
of acquiring offsetting emissions decreases is to allow an area to move
towards attainment of the NAAQS while still allowing some industrial growth.
Air quality improvement may not be realized if all emissions increases are not
accounted for and if emissions offsets are not real.
In evaluating a nonattainment NSR permit, the reviewing agency ensures
that offsets are developed in accordance with the provisions of the applicable
State or local nonattainment NSR rules. The following factors are considered
in reviewing offsets :
the pollutants requiring offsets and amount of offset required;
the location of offsets relative to the proposed source;
the allowable sources for offsets;
the "baseline" for calculating emissions reduction credits; and
the enforceability of proposed offsets.
Each of these factors should be discussed with the reviewing agency to ensure
that the specific requirements of that agency are met.
The offset requirement applies to each pollutant which triggered
nonattainment NSR applicability. For example, a permit for a proposed
G.5
-------
DRAFT
OCTOBER 1990
petroleum refinery which will emit more than 100 tpy of sulfur dioxide (S02)
and particulate matter in an area designated, pursuant to § 107 of the Act as
nonattainment for both pollutants, is required to obtain emissions offsets of
S02 and particulate matter.
III.A. CRITERIA FOR EVALUATING EMISSIONS OFFSETS
Emissions reductions obtained to offset new source emissions in a
nonattainment area must meet two important objectives:
• ensure reasonable progress toward attainment of the NAAQS; and
• provide a positive net air quality benefit in the area affected by
the proposed source.
States have latitude in determining what requirements offsets must meet to
achieve these NAA program objectives. The EPA has set forth minimum
considerations under the Interpretative Ruling (40 CFR 51, Appendix S).
Acceptable offsets also must be creditable, quantifiable, federally
enforceable, and permanent.
While an emissions offset must always result in reasonable progress
toward attainment of the NAAQS, it need not show that the area will attain the
NAAQS. Therefore, the ratio of required emissions offset to the proposed
source's emissions must be greater than one. The State determines what offset
ratio is appropriate for a proposed source, taking into account the location
of the offsets, i.e., how close the offsets are to the proposed source.
To satisfy the criterion of a net air quality benefit does not mean that
the applicant must show an air quality improvement at every location affected
by the proposed source. Sources involved in an offset situation should impact
air quality in the same general area as the proposed source, but the net air
G.6
-------
DRAFT
OCTOBER 1990
quality benefit test should be made "on balance" for the area affected by the
new source. Generally, offsets for VOC's are acceptable if obtained from
within the same air quality control region as the new source or from other
nearby areas which may be contributing to an ozone nonattainment problem. For
all pollutants, offsets should be located as close to the proposed site as
possible. Applicants should always discuss the location of potential offsets
with the reviewing agency to determine whether the offsets are acceptable.
III.B. AVAILABLE SOURCES OF OFFSETS
In general, emissions reductions which have resulted from some other
regulatory action are not available as offsets. For example, emissions
reductions already required by a SIP cannot be counted as offsets. Also,
sources subject to an NSPS in an area with less stringent SIP limits cannot
use the difference between the SIP and NSPS limits as an offset. In addition,
any emissions reductions already counted in major modification "netting" may
not be used as offsets. However, emissions reductions validly "banked" under
an approved SIP may be used as offsets.
III.C. CALCULATION OF OFFSET BASELINE
A critical element in the development or review of nonattainment area
new source permits is to determine the appropriate baseline emissions level
for the source from which offsetting emissions reductions are being sought.
In most cases the SIP emissions limit in effect at the time that the permit
application is filed may be used. This means that offsets will be based on
emissions reductions below these SIP limits. Where there is no meaningful or
applicable SIP requirement, the applicant is required to use actual emissions
as the offset baseline.
G.7
-------
DRAFT
OCTOBER 1990
III.D. ENFORCEABILITY OF PROPOSED OFFSETS
The reviewing agency ensures that all offsets are the result of a new,
federally enforceable emissions limitations. Such emissions limitations
should be specifically stated in the permit, regulation or other document
which establishes a Federal enforceability requirement. Offsets must be
established by conditions in the federally enforceable operating permit of the
plant from which the emissions reduction is obtained or in a SIP revision
which establishes the new emissions limitation for that source.
G.8
-------
DRAFT
OCTOBER 1990
IV. OTHER REQUIREMENTS
An applicant proposing a major new source or major modification in a
nonattainment area must certify that all major stationary sources owned or
operated by the applicant (or by any entity controlling, controlled by, or
under common control with the applicant) in that State are in compliance with
all applicable emissions limitations and standards under the CAA. This
includes all regulations in an EPA-approved SIP, including those more
stringent than Federal requirements.
In accordance with requirements set forth under 40 CFR 52.28, any major
new source or major modification proposed for a nonattainment area that may
impact visibility in a mandatory Class I Federal area is subject to review by
the appropriate Federal land manager (FLM). The reviewing agency for any
nonattainment area should ensure that the FLM of such mandatory Class I
Federal area receives appropriate notification and copies of all documents
relating to the permit application received by the agency.
G.9
-------
APPENDICES
A - Definition of Selected Terms
B - Estimating Control Costs
C - Potential to Emit
-------
DRAFT
OCTOBER 1990
APPENDIX A
DEFINITION OF SELECTED TERMS
-------
APPENDIX ft - DEFINITION OF SELECTED NSR TERMS
BRCT
Best Available Control Technology is the control level required for sources subject to PSD. From the regulation
(reference 40 CFR 52.21(b)) BACT means "an emissions limitation (including a visible emission standard) based on the maximum
degree of reduction for each pollutant subject to regulation under the Clean Air Act which would be emitted from any proposed
major stationary source or major modification which the Administrator, on a case-by-case basis, taking into account energy,
environmental, and economic impacts and other costs, determines is achievable for such source or modification through
application of production processes or available methods, systems, and techniques, including fuel cleaning or treatment or
innovative fuel combustion techniques for control of such pollutant. In no event shall application of best available control
technology result in emissions of any pollutant which would exceed the emissions allowed by any applicable standard under •
40 CFR Parts 60 and 61. If the Administrator determines that technological or economic limitations on the application of
measurement methodology to a particular emissions unit would make the imposition of an emissions standard infeasible, a
design, equipment, work practice, operational standard, or combination thereof, may be prescribed instead to satisfy the
requirement for the application of best available control technology. Such standard shall, to the degree possible, set forth
the emissions reduction achievable by implementation of such design, equipment, work practice or operation, and shall provide
for compliance by means which achieve equivalent results."
Emission Units
Increments
The individual emitting facilities at a location that together make up the source. From the regulation (reference
40 CFR 52.21(b)), it means "any part of a stationary source which emits or would have the potential to emit any pollutant
subject to regulation under the Act."
The maximum permissible level of air quality deterioration that may occur beyond the baseline air quality level. Increments
were defined statutorily by Congress for S02 and PM. Recently EPA also has promulgated increments for NO . Increment is
consumed or expanded by actual emissions changes occurring after the baseline date and by construction related actual
emissions changes occurring after January 6, 1975, and February 8, 1988 for PM/S02 and NOX, respectively.
a.l
-------
APPENDIX ft - DEFINITION OF SELECTED NSR TERMS (Continued)
Innovative Control
Technology From the regulation (reference 40 CFR 52.21(b)(19)) "Innovative control technology" means any system of air pollution control
that has not been adequately demonstrated in practice, but would have a substantial likelihood of achieving greater
continuous emissions reduction than any control system in current practice or of achieving at least comparable reductions at
lower cost in terms of energy, economics, or nonair quality environmental impacts. Special delayed compliance provisions
exist that may be applied when applicants propose innovative control techniques.
LBER Lowest Achievable Emissions Rate is the control level required of a source subject to nonattainment review. From the
regulations (reference 40 CFR 51.165(a)), it means for any source "the more stringent rate of emissions based on the
following:
(a) The most stringent emissions limitation which is contained in the implementation plan of any State for such class or
category of stationary source, unless the owner or operator of the proposed stationary source demonstrates that such
limitations are not achievable; or
(b) The most stringent emissions limitation which is achieved in practice by such class or category of stationary sources.
This limitation, when applied to a modification, means the lowest achievable emissions rate of the new or modified emissions
units within a stationary source. In no event shall the application of the term permit a proposed new or modified stationary
source to emit any pollutant in excess of the amount allowable under an applicable new source standard of performance."
a.2
-------
APPENDIX A - DEFINITION OF SELECTED NSR TERMS (Continued)
Major Modification A major modification is a modification to an existing major stationary source resulting in a significant net emissions
increase (defined elsewhere in this table) that, therefore, is subject to PSD review. From the regulation (reference
40 CFR 52.21(b)(2)):
"(i) 'Major modification' means any physical change in or change in the method of operation of a major stationary source that
would result in a significant net emissions increase of any pollutant subject to regulation under the Act.
(ii) Any net emissions increase that is significant for volatile organic compounds shall be considered significant for ozone.
(iii) A physical change or change in the method of operation shall not include:
(a) routine maintenance, repair and replacement;
(c) use of an alternative fuel by reason of an order or rule under Section 125 of the Act;
(d) Use of an alternative fuel at a steam generating unit to the extent that the fuel is generated from municipal solid
waste;
(e) Use of an alternative fuel or raw material by a stationary source which:
(1) The source was capable of accommodating before January 6, 1975, unless such change would be prohibited under any
Federally enforceable permit condition which was established after January 6, 1975, pursuant to 40 CFR 52.21 or under
regulations approved pursuant to 40 CFR Subpart I or 40 CFR 51.166; or
(2) The source is approved to use under any permit issued under 40 CFR 52.21 or under regulations approved pursuant to
40 CFR 51.166;
(f) an increase in the hours of operation or in the production rate, unless such change would be prohibited under any
federally enforceable permit condition which was established after January 6, 1975, pursuant to 40 CFR 52.21 or under
regulations approved pursuant to 40 CFR Subpart I or 40 CFR 51.166; or
(g) any change in ownership at a stationary source."
a.3
-------
APPENDIX A - DEFINITION OF SELECTED NSR TERMS (Continued)
Major Stationary Source A major stationary source is an emissions source of sufficient size to warrant PSD review. Major modification to major
stationary sources are also subject to PSD review. From the regulation (reference 40 CFR 52.21(b)(l)), (i) "Major stationary
source" means:
"(a) Any of the following stationary sources of air pollutant which emits, or has the potential to emit, 100 tons per year or
more of any pollutant subject to regulation under the Act: Fossil fuel-fired steam electric plants of more than 250 million
British thermal units per hour heat input, coal cleaning plants (with thermal dryers), Kraft pulp mills, Portland cement
plants, primary zinc smelters, iron and steel mill plants, primary aluminum ore reduction plants, primary aluminum ore
reduction plants, primary copper smelters, municipal incinerators capable of charging more than 250 tons of refuse per day,
hydrofluoric, sulfuric, and nitric acid plants, petroleum refineries, lime plants, phosphate rock processing plants, coke
oven batteries, sulfur recovery plants, carbon black plants (furnace process), primary lead smelters, fuel conversion plants,
sintering plants, secondary metal production plants, chemical process plants, fossil fuel boilers (or combinations thereof)
totaling more than 250 million British thermal units per hour heat input, petroleum storage and transfer units with a total
storage capacity exceeding 300,000 barrels, taconite ore processing plants, glass fiber processing plants, and charcoal
production plants;
(b) Notwithstanding the stationary source size specified in paragraph (b)(l)(i) of this section, any stationary source which
emits, or has the potential to emit, 250 tons per year or more of any air pollutant subject to regulation under the Act; or
(c) Any physical change that would occur at a stationary source not otherwise qualifying under paragraph (b)(l) as a major
stationary source not otherwise qualifying under paragraph (b)(l) as a major stationary source, if the changes would
constitute a major stationary source by itself.
(ii) A major stationary source that is major for volatile organic compounds shall be considered major for ozone."
NMQS National Ambient Air Quality Standards are Federal standards for the minimum ambient air quality needed to protect public
health and welfare. They have been set for six criteria pollutants including S02, PM/PM10, NOX, CO, 03 (VOC), and Pb.
a.4
-------
APPENDIX A - DEFINITION OF SELECTED NSR TERMS (Continued)
NESHflP, or National Emission Standard for Hazardous flir Pollutants, is a technology-based standard of performance prescribed
for hazardous air pollutants from certain stationary source categories under Section 112 of the Clean Air Act. Where they
apply, NESHAP represent absolute minimum requirements for BACT.
NSPS NSPS, or New Source Performance Standard, is an emission standard prescribed for criteria pollutants from certain stationary
source categories under Section 111 of the Clean Air Act. Where they apply, NSPS represent absolute minimum requirements for
BACT.
PSD Prevention of significant deterioration is a construction air pollution permitting program designed to ensure air quality
does not degrade beyond the NAAQS levels or beyond specified incremental amounts above a prescribed baseline level. PSD also
ensures application of BACT to major stationary sources and major modifications for regulated pollutants and consideration of
soils, vegetation, and visibility impacts in the permitting process.
Regulated Pollutants1 Refers to pollutants that have been regulated under the authority of the Clean Air Act (NAAQS, NSPS, NESHAP):
03 (VOC) - Ozone, regulated through volatile organic compounds as precursors
Nitrogen oxides
Sulfur dioxide
Total suspended particulate matter
- Particulate matter with <10 micron aerometric diameter
Carbon monoxide
NO,
S02"
PM (TSP)-
PM (PM10)
CO
Pb
As
Be
Hg
vc
F
H2S04 -
H2S
Lead
Asbestos
Beryllium
Mercury
Vinyl chloride
Fluorides
Sulfuric acid mist
Hydrogen sulfide
TRS - Total reduced sulfur (including H2S)
RDS - Reduced Sulfur Compounds (including H2S)
Bz - Benzene
Rd - Radionuclides
As - Arsenic
CFC's - Chlorofluorocarbons
Rn-222 - Radon-222
Halons
1 The referenced list of regulated pollutants is current as of November 1989. Presently, additional pollutants may also be subject to regulation
under the Clean Air Act.
a.5
-------
APPENDIX A - DEFINITION OF SELECTED NSR TERMS (Continued)
Significant Emissions Increase For new major stationary sources and major modifications, a significant emissions increase triggers PSD review.
Review requirements must be met for each pollutant undergoing a significant net emissions increase. From the
regulation (reference 40 CFR 52.21(b)(23)).
(i) "Significant" means, in reference to a net emissions increase from a modified major source or the potential of a new
major source to emit any of the following pollutants, a rate of emissions that would equal or exceed any of the following *
rates:
Carbon monoxide: 100 tons per year (tpy)
Nitrogen oxides: 40 tpy
Sulfur dioxide: 40 tpy
Particulate matter: 25 tpy
PH10: 15 tpy
Ozone: 40 tpy of volatile organic compounds
Lead: 0.6 tpy
Asbestos: 0.007 tpy
Beryllium: 0.0004 tpy
Mercury: 0.1 tpy
Vinyl chloride: 1 tpy
Fluorides: 3 tpy
Sulfuric acid mist: 7 tpy
Hydrogen Sulfide (H2S): 10 tpy
Total reduced sulfur (including H2S): 10 tpy
Reduced sulfur compounds (including H2S): 10 tpy
(ii) "Significant" means, in reference to a net emissions increase or the potential of a source to emit a pollutant subject
to regulation under the Act, that (i) above does not list, any emissions rate.
(For example, benzene and radionuclides are pollutants falling into the "any emissions rate" category.)
(iii) Notwithstanding, paragraph (b)(23)(i) of this section, "significant means any emissions rate or any net emissions
increase associated with a major stationary source or major modification which would construct within 10 kilometers of a
Class I area, and have an impact on such an area equal to or greater than 1 ug/m3, (24-hour average).
a.6
-------
APPENDIX A - DEFINITION OF SELECTED NSR TERMS (Continued)
SIP State Implementation Plan is the federally approved State (or local) air quality management authority's statutory plan for
attaining and maintaining the HAftQS. Generally, this refers to the State/local air quality rules and permitting requirements
that have been accepted by EPft as evidence of an acceptable control strategy.
Stationary Source For PSD purposes, refers to all emissions units at one location under common ownership or control. From the regulation
(reference 40 CFR 52.21(b)(5) and 51.166(b)(5)), it means "any building, structure, facility, or installation which emits or
may emit any air pollutant subject to regulation under the Act."
"Building, structure, facility, or installation" means all of the pollutant-emitting activities which belong to the same
industrial grouping, are located on one or more contiguous or adjacent properties, and are under the control of the same
person (or person under common control). Pollutant-emitting activities shall be considered as part of the same industrial
grouping if they belong to the same "Major Group" (i.e., which have the same first two digit code) as described in the
Standard Industrial Classification Manual, 1972, as amended by the 1977 Supplement (U.S. Government Printing Office stock
numbers 4101-0066 and 003-005-00176-0, respectively).
a.7
-------
DRAFT
OCTOBER 1990
APPENDIX B
ESTIMATING CONTROL COSTS
-------
DRAFT
OCTOBER 1990
APPENDIX B
ESTIMATING CONTROL COSTS
I. CAPITAL COSTS
Capital costs include equipment costs, installation costs, indirect
costs, and working capital (if appropriate). Figure B-4 presents the
elements of total capital cost and represents a building block approach that
focuses on the control device as the basic unit of analysis for estimating
total capital investment. The total capital investment has a role in the
determination of total annual costs and cost effectiveness.
One of the most common problems which occurs when comparing costs at
different facilities is that the battery limits are different. For example,
the battery limit of the cost of a electrostatic precipitation might be the
precipitator itself (housing, plates, voltage regulators, transformers, etc.),
ducting from the source to the precipitator, and the solids handling system.
The stack would not be included because a stack will be required regardless of
whether or not controls are applied. Therefore, it should be outside the
battery 1imits of the control system.
Direct installation costs are the costs for the labor and materials to
install the equipment and includes site preparation, foundations, supports,
erection and handling of equipment, electrical work, piping, insulation and
painting. The equipment vendor can usually supply direct installation costs.
The equipment vendor should be able to supply direct installation costs
estimates or general installation costs factors. In addition, typical
installation cost factors for various types of equipment are available in the
following references.
b.l
-------
o Primary Control Device
o Auxiliary Equipment
(including ductwork)
o Modification to Other Equipment
o Instrumentation (a)
o Sales Taxes (a)
o Freight (a)
o Foundation and Supports
o Handling and Erection
o Electrical
o Piping
o Insulation
o Painting
o Engineering
o Construction and Field Expenses
o Contractor Fees
o Start-up
o Performance Tests
o Contingencies
Direct
Installation
Costs (b)
Site Preparation (c,d)
Buildings (d)
Land (e)
Working Capital (e)
Total
Direct
Costs
Indirect
Installation «
Costs (b)
Total
Indirect
Costs
Total
Nondepreciable
Investment
"Battery
Limits"
Costs
Off-site
Facilities (e)
Total
Depreciable
Investment
Total
Capital
Investment
(a) These costs are factored from the sum of the control device and auxiliary equipment costs.
(b) These costs are factored from the purchased control equipment.
(c) Usually required only at "grass roots" installations.
(d) Unlike the other direct and indirect costs, costs for these items are not factored from the
purchased equipment cost. Rather, they are sized and costed separately.
(e) Normally not required with add-on control systems.
°
FIGURE B-4. Elements of Total Capital Costs
-------
DRAFT
OCTOBER 1990
OAQPS Control Cost Manual (Fourth Edition), January 1990,
EPA 450/3-90-006
Control Technology for Hazardous Air Pollutants (HAPS) Manual.
September 1986, EPA 625/6-86-014
Standards Support Documents
Background Information Documents
Control Techniques Guidelines Documents
Other EPA sponsored costing studies
Engineering Cost and Economics Textbooks
Other engineering cost publications
These references should also be used to validate any installation cost factors
supplied from equipment vendors.
If standard costing factors are used, they may need to be adjusted due to
site specific conditions. For example, in Alaska installation costs are on
the order of 40-50 percent higher than in the contiguous 48 states due to
higher labor prices, shipping costs, and climate.
Indirect installation costs include (but are not limited to) engineering,
construction, start-up, performance tests, and contingency. Estimates of
these costs may be developed by the applicant for the specific project under
evaluation. However, if site-specific values are not available, typical
estimates for these costs or cost factors are available in:
OAQPS Control Cost Manual (Fourth Edition), EPA 450/3-90-006
Cost Analysis Manual for Standards Support Documents, April 1979
These references can be used by applicants if they do not have
site-specific estimates already prepared, and should also be used by the
reviewing agency to determine if the applicant's estimates are reasonable.
Where an applicant uses different procedures or assumptions for estimating
b.3
-------
DRAFT
OCTOBER 1990
control costs other than those contained in the referenced material or
outlined in this document, the nature and reason for the differences are to be
documented in the BACT analysis.
Working capital is a fund set aside to cover initial costs of fuel,
chemicals, and other materials and other contingencies. Working capital costs
for add on control systems are usually relatively small and, therefore, are
usually not included in cost estimates.
Table B-ll presents an illustrative example of a capital cost estimate
developed for an ESP applied to a spreader-stoker coal-fired boiler. This
estimate shows the minimum level of detail required for these types of
estimates. If bid costs are available, these can be used rather than study
cost estimates.
II. TOTAL ANNUAL COST
The permit applicant should use the levelized annual cost approach for
consistency in BACT cost analysis. This approach is also called the
"Equivalent Uniform Annual Cost" method, or simply "Total Annual Cost" (TAC).
The components of total annual costs and their relationships are shown in
Figure B-5. The total annual costs for control systems is comprised of three
elements: "direct" costs (DC), "indirect costs" (1C), and "recovery credit"
(RC), which are related by the following equation:
TAC = DC + 1C - RC
b.4
-------
DRAFT
OCTOBER 1990
TABLE B-ll.
EXAMPLE OF A CAPITAL COST ESTIMATE FOR AN
ELECTROSTATIC PRECIPITATOR
Capital
cost
($)
Direct Investment
Equipment cost
ESP unit
Ducting
Ash hand!ing system
Total equipment cost
Installation costs
ESP unit
Ducting
Ash hand!ing system
Total installation costs
Total direct investment (TDI)
(equipment + installation)
Indirect Investment
Engineering (10% of TDI)
Construction and field expenses (10% of TDI)
Construction fees (10% of TDI)
Start-up (2% of TDI)
Performance tests (minimum $2000)
Total indirect investment (Til)
Contingencies (20% of TDI + Til)
TOTAL TURNKEY COSTS (TDI + Til)
Working Capital (25% of total direct operating costs)a
GRAND TOTAL
175,800
64,100
97,200
337,100
175,800
102,600
97,200
375,600
712,700
71,300
71,300
71,300
71,300
14,300
3,000
231,200
188,800
1,132,700
21,100
1,153,800
b.5
-------
DRAFT
OCTOBER 1990
o Raw Materials
o Utilities
- Electricity
- Steam
- Water
- Others
o Labor
- Operating
- Supervisory
- Maintenance
o Maintenance materials
o Replacement parts
Variable
Semivariable
Direct
Annual
Costs
o Overhead
o Property Taxes
o Insurance
o Capital Recovery
o Recovered Product
o Recovered Energy
o Useful byproduct
o Energy Gain
Indirect
Annual
Costs
Recovery
Credits
Total
= Annual
Costs
FIGURE B-5. Elements of Total Annual Costs
b.6
-------
DRAFT
OCTOBER 1990
Direct costs are those which tend to be proportional or partially
proportional to the quantity of exhaust gas processed by the control system
or, in the case of inherently lower polluting processes, the amount of
material processed or product manufactured per unit time. These include costs
for raw materials, utilities (steam, electricity, process and cooling water,
etc.), and waste treatment and disposal. Semivariable direct costs are only
partly dependent upon the exhaust or material flowrate. These include all
associated labor, maintenance materials, and replacement parts. Although
these costs are a function of the operating rate, they are not linear
functions. Even while the control system is not operating, some of the
semivariable costs continue to be incurred.
Indirect, or "fixed", annual costs are those whose values are relatively
independent of the exhaust or material flowrate and, in fact, would be
incurred even if the control system were shut down. They include such
categories as overhead, property taxes, insurance, and capital recovery.
Direct and indirect annual costs are offset by recovery credits, taken
for materials or energy recovered by the control system, which may be sold,
recycled to the process, or reused elsewhere at the site. These credits, in
turn, may be offset by the costs necessary for their purification, storage,
transportation, and any associated costs required to make then reusable or
resalable. For example, in auto refinishing, a source through the use of
certain control technologies can save on raw materials (i.e., paint) in
addition to recovered solvents. A common oversight in BACT analyses is the
omission of recovery credits where the pollutant itself has some product or
process value. Examples of control techniques which may produce recovery
credits are equipment leak detection and repair programs, carbon absorption
systems, baghouse and electrostatic precipitators for recovery of reusable or
saleable solids and many inherently lower polluting processes.
b.7
-------
DRAFT
OCTOBER 1990
Table B-12 presents an example of total annual costs for the control
system previously discussed. Direct annual costs are estimated based on
system design power requirements, energy balances, labor requirements, etc.,
and raw materials and fuel costs. Raw materials and other consumable costs
should be carefully reviewed. The applicant generally should have documented
delivered costs for most consumables or will be able to provide documented
estimates. The direct costs should be checked to be sure they are based on
the same number of hours as the emission estimates and the proposed operating
schedule.
Maintenance costs in some cases are estimated as a percentage of
the total capital investment. Maintenance costs include actual costs to
repair equipment and also other costs potentially incurred due to any
increased system downtime which occurs as a result of pollution control system
maintenance.
Fixed annual costs include plant overhead, taxes, insurance, and capital
recovery charges. In the example shown, total plant overhead is calculated as
the sum of 30 percent of direct labor plus 26 percent of all labor and
maintenance materials. The OAQPS Control Cost Manual combines payroll and
plant overhead into a single indirect cost. Consequently, for "study"
estimates, it is sufficiently accurate to combine payroll and plant overhead
into a single indirect cost. Total overhead is then calculated as 60 percent
of the sum of all labor (operating, supervisory, and maintenance) plus
maintenance materials.
Property taxes are a percentage of the fixed capital investment. Note
that some jurisdictions exempt pollution control systems from property taxes.
Ad valorem tax data are available from local governments. Annual insurance
charges can be calculated by multiplying the insurance rate for the facility
by the total capital costs. The typical values used to calculate taxes and
b.8
-------
DRAFT
OCTOBER 1990
TABLE B-12. EXAMPLE OF A ANNUAL COST ESTIMATE FOR AN ELECTROSTATIC
PRECIPITATOR APPLIED TO A COAL-FIRED BOILER
Annual costs
(S/yr)
Direct Costs
Direct labor at $12.02/man-hour 26,300
Supervision at $15.63/man-hour 0
Maintenance labor at $14.63/man-hour 16,000
Replacement parts " 5,200
Electricity at $0.0258/kWh 3,700
Water at $0.18/1000 gal 300
Waste disposal at $15/ton (dry basis) 33,000
Total direct costs 84,500
Indirect Costs
Overhead
Payroll (30% of direct labor) 7,900
Plant (26% of all labor and replacement parts) 12,400
Total overhead costs 20,300
Capital charges
G&A taxes and insurance 45,300
(4% of total turnkey costs)
1 oo i r\(\
Capital recovery factor iw.iuu
(11.75% of total turnkey costs)
Interest on working capital 2'100
(10% of working capital)
Total capital charges 180,500
TOTAL ANNUALIZED COSTS 285,300
b.9
-------
DRAFT
OCTOBER 1990
insurance is four percent of the total capital investment if specific facility
data are not readily available.
The annual costs previously discussed do not account for recovery of the
capital cost incurred. The capital cost shown in Table B-2 is annualized
using a capital recovery factor of 11.75 percent. When the capital recovery
factor is multiplied by the total capital investment the resulting product
represents the uniform end of year payment necessary to repay the investment
in "n" years with an interest rate "i".
The formula for the capital recovery factor is:
CRF = i (1 -t- i)n
(1 + D-l
where:
CPF = capital recovery factor
n = economic life of equipment
i = real interest rate
The economic life of a control system typically varies between 10 to 20
years and longer and should be determined consistent with data from EPA cost
support documents and the IRS Class Life Asset Depreciation Range System.
From the example shown in Table B-12 the interest rate is 10 percent and
the equipment life is 20 years. The resulting capital recovery factor is
11.75 percent. Also shown is interest on working capital, calculated as the
product of interest rate and the working capital.
It is important to insure that the labor and materials costs of parts of
the control system (such as catalyst beds, etc.) that must be replaced before
the end of the useful life are subtracted from the total capital investment
before it is multiplied by the capital recovery factor. Costs of these parts
b.10
-------
DRAFT
OCTOBER 1990
should be accounted for in the maintenance costs. To include the cost of
those parts in the capital charges would be double counting. The interest
rate used is a real interest rate (i.e., it does not consider inflation). The
value used in most control costs analyses is 10 percent in keeping with
current EPA guidelines and Office of Management and Budget recommendations for
regulatory analyses.
It is also recommended that income tax considerations be excluded from
cost analyses. This simplifies the analysis. Income taxes generally
represent transfer payments from one segment of society to another and as such
are not properly part of economic costs.
III. OTHER COST ITEMS
Lost production costs are not included in the cost estimate for a new or
modified source. Other economic parameters (equipment life, cost of capital,
etc.) should be consistent with estimates for other parts of the project.
b.ll
-------
PART III
EFFECTIVE PERMIT HRITING
Chapter H - Elements of an Effective Permit
Chapter I - Permit Drafting
-------
DRAFT
OCTOBER 1990
CHAPTER H
ELEMENTS OF AN EFFECTIVE CONSTRUCTION PERMIT
I. INTRODUCTION
A construction permit is the legal tool used to establish all the source
limitations deemed necessary by the reviewing agency during review of the
permit application, as described in Parts I and II of this manual, and is the
primary basis for enforcement of NSR requirements. In essence, the
construction permit may be viewed as an extension of the regulations. It
defines as clearly as possible what is expected of the source and reflects the
outcome of the permit review process. A construction permit may limit the
emissions rate from various emissions units or limit operating parameters such
as hours of operation and amount or type of materials processed, stored, or
combusted. Operational limitations frequently are used to establish a new
potential to emit or to implement a desired emissions rate. The construction
permit must be a "stand-alone" document that:
• identifies the emissions units to be regulated;
• establishes emissions standards or other operational limits to be met;
• specifies methods for determining compliance and/or excess emissions,
including reporting and recordkeeping requirements; and
• outlines the procedures necessary to maintain continuous compliance
with the emission limits.
To achieve these goals, the permit, which is in effect a contract between the
source and the regulatory agency, must contain specific, clear, concise, and
enforceable conditions.
This part of the manual gives a brief overview of the development of a
construction permit, which ensures that major new sources and modifications
H.I
-------
DRAFT
OCTOBER 1990
will be constructed and operated in compliance with the applicable new source
review (NSR) regulations [including prevention of significant deterioration
(PSD) and nonattainment area (NAA) review], new source performance standards
(NSPS), national emissions standards for hazardous air pollutants (NESHAP),
and applicable state implementation plan (SIP) requirements. In particular, a
construction permit contains the specific conditions and limitations which
ensure that:
• an otherwise major source will remain minor;
• all contemporaneous emissions increases and decreases are creditable
and federally-enforceable; and
• where appropriate, emissions offset transactions are documented
clearly and offsets are real, creditable, quantifiable,
permanent and federally-enforceable.
For a more in-depth study, refer to the Air Pollution Training Institute
(APTI) course SI 454 (or Workshop course 454 given by APTI) entitled
"Effective Permit Writing." This course is highly recommended for all permit
writers and reviewers.
H.2
-------
DRAFT
OCTOBER 1990
II. TYPICAL CONSTRUCTION PERMIT ELEMENTS
While each final construction permit is unique to a particular source due
to varying emission limits and specific special terms and conditions, every
such permit must also contain certain basic elements:
legal authority;
technical specifications;
emissions compliance demonstration;
definition of excess emissions;
administrative procedures; and
other specific conditions.
Although many of these elements are inherent in the authority to issue permits
under the SIP, they must be explicit within the construction of a NSR permit.
Table H-l lists a few typical subelements found in each of the above. Some
permit conditions included in each of these elements can be considered
standard permit conditions, i.e., they would be included in nearly every
permit. Others are more specific and vary depending on the individual source.
II.A. LEGAL AUTHORITY
In general, the first provision of a permit is the specification of the
legal authority to issue the permit. This should include a reference to the
enabling legislation and to the legal authority to issue and enforce the
conditions contained in the permit and should specify that the application is,
in essence, a part of the permit. These provisions are common to nearly all
permits and usually are expressed in standard language included in every
permit issued by an agency. These provisions articulate the contract-1 ike
nature of a permit in that the permit allows a source to emit air pollution
only if certain conditions are met. A)specific citation of any applicable
H.3
-------
DRAFT
OCTOBER 1990
TABLE H.I. SUGGESTED MINIMUM CONTENTS OF AIR EMISSION PERMITS
Permit Category
Typical Elements
Legal Authority
Technical Specifications
Emission Compliance Demonstration
Definition of Excess Emissions
Administrative
Other Conditions
Basis—statute, regulation, etc.
Conditional Provisions
Effective and expiration dates
Unit operations covered
Identification of emission units
Control equipment efficiency
Design/operation parameters
Equipment design
Process specifications
Operating/maintenance procedures
Emission limits
Initial performance test and methods
Continuous emission monitoring and
methods
Surrogate compliance measures
- process monitoring
- equipment design/operations
- work practice
Emission limit and averaging time
Surrogate measures
Malfunctions and upsets
Follow-up requirements
Recordkeeping and reporting procedures
Commence/delay construction
Entry and inspections
Transfer and severability
Post construction monitoring
Emissions offset
H.4
-------
DRAFT
OCTOBER 1990
permit effective date and/or expiration date is usually included under the
legal authority as well.
II.B. TECHNICAL SPECIFICATIONS
Overall, the technical specifications may be considered the core of the
construction permit, in that they specifically identify the emissions unit(s)
covered by the permit and the corresponding emission limits with which the
source must comply. Properly identifying each emissions unit is important so
that (1) inspectors can easily identify the unit in the field and (2) the
permit leaves no question as to which unit the various permit limitations and
conditions apply. Identification usually includes a brief description of the
source or type of equipment, size or capacity, model number or serial number,
and the source's identification of the unit.
Emissions and operational limitations are included in the technical
specifications and must be clearly expressed, easily measurable, and allow no
subjectivity in their compliance determinations. All limits also must be
indicated precisely for each emissions point or operation. For clarity, these
limits are often best expressed in tabular rather than textual form. In
general, it is best to express the emission limits in two different ways, with
one value serving as an emissions cap (e.g., Ibs/hr.) and the other ensuring
continuous compliance at any operating capacity (e.g., Ibs/MMBtu). The permit
writer should keep in mind that the source must comply with both values to
demonstrate compliance. Such limits should be of a short term nature,
continuous and enforceable. In addition, the limits should be consistent with
the averaging times used for dispersion modeling and the averaging times for
compliance testing. Since emissions limitation values incorporated into a
permit are based on a regulation (SIP, NSPS, NESHAP) or resulting from new
source review, (i.e., BACT or LAER requirements), a reference to the
applicable portion of the regulation should be included.
H.5
-------
DRAFT
OCTOBER 1990
II.C. EMISSIONS COMPLIANCE DEMONSTRATION
The construction permit should state how compliance with each limitation
will be determined, and include, but not be limited to, the test method(s)
approved for demonstrating compliance. These permit compliance conditions
must be very clear and enforceable as a practical matter (see Appendix C).
The conditions must specify:
when and what tests should be performed;
under what conditions tests should be performed;
the frequency of testing;
the responsibility for performing the test;
that the source be constructed to accommodate such testing;
procedures for establishing exact testing protocol; and
requirements for regulatory personnel to witness the testing.
Where continuous, quantitative measurements are infeasible, surrogate
parameters must be expressed in the permit. Examples of surrogate parameters
include: mass emissions/opacity correlations, maintaining pressure drop
across a control (e.g., venturi throat of a scrubber), raw material input/mass
emissions output ratios, and engineering correlations associated with specific
work practices. These alternate compliance parameters may be used in
conjunction with measured test data to monitor continuous compliance or may be
independent compliance measures where source testing is not an option and work
practice or equipment parameters are specified. Only those parameters that
exhibit a correlation with source emissions should be used. Identifying and
quantifying surrogate process or control equipment parameters (such as
pressure drop) may require initial source testing or may be extracted from
confirmed design characteristics contained in the permit application.
H.6
-------
DRAFT
OCTOBER 1990
Parameters that must be monitored either continuously or periodically
should be specified in the permit, including averaging time for continuously
monitored data, and data recording frequency for periodically (continually)
monitored data. The averaging times should be of a short term nature
consistent with the time periods for which dispersion modeling of the
respective emissions rate demonstrated compliance with air quality standards,
and consistent with averaging times used in compliance testing. This
requirement also applies to surrogate parameters where compliance may be time-
based, such as weekly or monthly leak detection and repair programs (also see
Appendix C). Whenever possible, "never to be exceeded" values should be
specified for surrogate compliance parameters. Also, operating and
maintenance (O&M) procedures should be specified for the monitoring
instruments (such as zero, span, and other periodic checks) to ensure that
valid data are obtained. Parameters which must be monitored continuously or
continually are those used by inspectors to determine compliance on a real-
time basis and by source personnel to maintain process operations in
compliance with source emissions limits.
II.D. DEFINITION OF EXCESS EMISSIONS
The purpose of defining excess emissions is to prevent a malfunction
condition from becoming a standard operating condition by requiring the source
to report and remedy the malfunction. Conditions in this part of the
construction permit:
precisely define excess emissions;
outline reporting requirements;
specify actions the source must take; and
indicate time limits for correction by the source.
H.7
-------
DRAFT
OCTOBER 1990
Permit conditions defining excess emissions may include alternate conditions
for startup, shutdown, and malfunctions such as maximum emission limits and
operational practices and limits. These must be as specific as possible since
such exemptions can be misused. Every effort should be made to include
adequate definitions of both preventable and nonpreventable malfunctions.
Preventable malfunctions usually are those which cause excess emissions due to
negligent maintenance practices. Examples of preventable malfunctions may
include: leakage or breakage of fabric filter bags; baghouse seal ruptures;
fires in electrostatic precipitators due to excessive build up of oils or
other flammable materials; and failure to monitor and replace spent activated
carbon beds in carbon absorption units. These examples reinforce the need for
good O&M plans and keeping records of all repairs. Permit requirements
concerning malfunctions may include: timely reporting of the malfunction
duration, severity, and cause; taking interim and corrective actions; and
taking actions to prevent recurrence.
II.E. ADMINISTRATIVE PROCEDURES
The administrative elements of construction permits are usually standard
conditions informing the source of certain responsibilities. These
administrative procedures may include:
recordkeeping and reporting requirements, including all continuous
monitoring data, excess emission reports, malfunctions, and
surrogate compliance data;
notification requirements for performance tests, malfunctions,
commencing or delay of construction;
entry and inspection procedures;
the need to obtain a permit to operate; and
specification of procedures to revoke, suspend, or modify the
permit.
H.8
-------
DRAFT
OCTOBER 1990
Though many of these conditions will be entered into the permit via standard
permit conditions, the reviewer must ensure the language is adequate to
establish precisely what is expected or needed from the source, particularly
the recordkeeping requirements.
II.F. OTHER CONDITIONS
In some cases, specific permit conditions which do not fit into the above
elements may need to be outlined. Examples of these are conditions requiring:
the permanent shutdown of (or reduced emissions rates for) other emissions
units to create offsets or netting credits; post-construction monitoring;
continued Statewide compliance; and a water truck to be dedicated solely to a
haul road. In the case of a portable source, a condition may be included to
require a copy of the effective permit to be on-site at all times. Some O&M
procedures, such as requiring a 10 minute warmup for an incinerator, would be
included in this category, as well as conditions requiring that replacement
fabric filters and baghouse seals be kept available at all times. Any source-
specific condition which needs to be included in the permit to ensure
compliance should be listed here.
III. SUMMARY
Assuming a comprehensive review, a construction permit is only as clear,
specific, and effective as the conditions it contains. As such, Table H-2 on
the following page lists guidelines for drafting actual construction permit
conditions. The listing specifies how typical permit elements should be
written. For further discussion on drafting "federally enforceable" permit
conditions as a practical matter, please refer to Appendix C - "Potential to
Emit."
H.9
-------
DRAFT
OCTOBER 1990
TABLE H.2. GUIDELINES FOR WRITING EFFECTIVE SPECIFIC CONDITIONS IN NSR
PERMITS
1. Make each permit condition simple, clear, and specific such that it
"stands alone."
2. Make certain 7ega7 authority exists to specify conditions.
3. Permit conditions should be objective and meaningful,
4. Provide description of processes, emissions units and control equipment
covered by the permit, including operating rates and periods.
5. Clearly identify each permitted emissions unit such that it can be
located in the field.
6. Specify allowable emissions (or concentration, etc.) rates for each
pollutant and emissions unit permitted, and specify each applicable
emissions standard by name in the permit.
7. Allowable emissions rates should reflect the conditions of BACJ/LAER and
Air Quality Analyses (e.g., specify limits two ways: maximum mass/unit
of process and maximum mass/unit time)
8. Specify for all emissions units (especially fugitive sources) permit
conditions that require continuous application of BACJ/LAER to achieve
maximum degree of emissions reduction.
9. Initial and subsequent performance tests should be conducted at worst
case operating (non-malfunction) conditions for all emissions units.
Performance tests should determine both emissions and control equipment
efficiency.
10. Continual and continuous emissions performance monitoring and
recordkeeping (direct and/or surrogate) should be specified where
feasible.
11. Specify test method (citation) and averaging period by which all
compliance demonstrations (initial and continuous) are to be made.
12. Specify what conditions constitute "excess emissions," and what is to be
done in those cases.
H.10
-------
DRAFT
OCTOBER 1990
CHAPTER I
PERMIT DRAFTING
I. RECOMMENDED PERMIT DRAFTING STEPS
This section outlines a recommended five-step permit drafting process
(see Table 1-1). These steps can assist the writer in the orderly preparation
of air emissions permits following technical review.
Step 1 concerns the emissions units and requires the listing and
specification of three things. First, list each new or modified emissions
unit. Second, specify each associated emissions point. This includes
fugitive emissions points (e.g., seals, open containers, inefficient capture
areas, etc.) and fugitive emissions units (e.g., storage piles, materials
handling, etc.). Be sure also to note emissions units with more than one
ultimate exhaust and units sharing common exhausts. Third, the writer must
describe each emissions unit as it may appear in the permit and identify, as
well as describe, each emissions control unit. Each new or modified emissions
unit identified in Step 1 that will emit or increase emissions of any
pollutant is considered in Step 2.
Step 2 requires the writer to specify each pollutant that will be emitted
from the new or modified source. Some pollutants may not be subject to
regulation or are of de minimis amounts such that they do not require major
source review. All pollutants should be identified in this step and reviewed
for applicability. Federally enforceable conditions must be identified for
de minimis pollutants to ensure they do not become significant (see
Appendix C - Potential to Emit). An understanding of "potential to emit" is
pertinent to permit review and especially to the drafting process.
I.I
-------
DRAFT
OCTOBER 1990
TABLE 1-1. FIVE STEPS TO PERMIT DRAFTING
STEP 1. SPECIFY EMISSIONS UNITS
Identify each new (or modified) emissions unit that will emit (or
increase) any pollutant.
Identify any pollutant and emissions units involved in a netting or
emissions reduction proposal (i.e., all contemporaneous emissions
increases and decreases).
Include point and fugitive emissions units.
Identify and describe emissions unit and emissions control
equipment.
STEP 2. SPECIFY POLLUTANTS
Pollutants subject to NSR/PSD.
Pollutants not subject to NSR/PSD but could reasonably be expected
to exceed significant emissions levels. Identify conditions that
ensure de minimis (e.g., shutdowns, operating modes, etc..).
STEP 3. SPECIFY ALLOWABLE EMISSION RATES AND BACT/LAER REQUIREMENTS
Minimum number of allowable emissions rates specified is equal to at
least two limits per pollutant per emissions unit.
One of two allowable limits is unit mass per unit time (Ibs/hr)
which reflects application of emissions controls at maximum
capacity.
Maximum hourly emissions rate must correspond to that used in air
qua!ity analysis.
Specify BACT/LAER emissions control requirements for each
pollutant/emissions unit pair.
1.2
-------
DRAFT
OCTOBER 1990
TABLE 1-1. - Continued
STEP 4. SPECIFY COMPLIANCE DEMONSTRATION METHODS
Continuous, direct emission measurement is preferable.
Specify initial and periodic emissions testing where necessary.
Specify surrogate (indirect) parameter monitoring and recordkeeping
where direct monitoring is impractical or in conjunction with tested
data.
Equipment and work practice standards should complement other
compliance monitoring.
STEP 5. OTHER PERMIT CONDITIONS
Establish the basis upon which permit is granted (legal authority).
Should be used to minimize "paper" allowable emissions.
Federally enforceable permit conditions limiting potential to emit.
1.3
-------
DRAFT
OCTOBER 1990
Step 3 pools the data collected in the two previous steps. The writer
should specify the pollutants that will be emitted from each emission unit and
identify associated emission controls for each pollutant and/or emission unit.
(Indicate if the control has been determined to be BACT.) The writer also
must assess the minimum number of allowable emissions rates to be specified in
the permit. Each emissions unit should have at least two allowable emissions
rates for each pollutant to be emitted. This is the most concise manner in
which to present permit allowables and should be consistent with the averaging
times and emissions ratio used in the air quality analysis. As discussed
earlier in Section H, the applicable regulation should also be cited as well
as whether BACT, LAER, or other SIP requirements apply to each pollutant to be
regulated.
Step 4 essentially mirrors the items discussed in the previous Chapter H,
Section IV., Emissions Compliance Demonstration. At this point the writer
enters into the permit any performance testing required of the source. The
conditions should specify what emissions test is to be performed and the
frequency of testing. Any surrogate parameter monitoring must be specified.
Recordkeeping requirements and any equipment and work practice standards
needed to monitor the source's compliance should be written into the permit
in Step 4. Any remaining or additional permit conditions, such as legal
authority and conditions limiting potential to emit can be identified in
StepS, (Other Permit Conditions, see Table 1-1.) At this point, the permit
should be complete. The writer should review the draft to ensure that the
resultant permit is an effective tool to monitor and enforce source
compliance. Also, the compliance inspector should review the permit to ensure
that the permit conditions are enforceable as a practical matter.
1.4
-------
DRAFT
OCTOBER 1990
II. PERMIT WORKSHEETS AND FILE DOCUMENTATION
Some agencies use permit drafting worksheets to store all the required
information that will be incorporated into the permit. The worksheets may be
helpful and are available at various agencies and in other EPA guidance
documents. The worksheets serve as a summary of the review process, though
this summation should appear in the permit file with or without a worksheet.
Documenting the permit review process in the file cannot be overemphasized.
The decision-making process which leads to the final permit for a source must
be clearly traceable through the file. When filing documentation, the
reviewer must also be aware of any confidential materials. Many agencies have
special procedures for including confidential information in the permit file.
The permit reviewer should follow any special procedures and ensure the permit
file is documented appropriately.
III. SUMMARY
Listed below are summary "helpful hints" for the permit writer, which
should be kept in mind when reviewing and drafting the permit. Many of these
have been touched on throughout Part III, but are summarized here to help
ensure that they are not overlooked:
Document the review process throughout the file.
Be aware of confidentiality items, procedures, and the consequences
of the release of such information.
Ensure the application includes all pertinent review information
(e.g., has the applicant identified solvents used in some coatings;
are solvents used, then later recovered; ultimate disposal of
collected wastes identified; and applicable monitoring and modeling
results included).
Address secondary pollutant formation.
1.5
-------
DRAFT
OCTOBER 1990
Ensure that all applicable regulations and concerns have been
addressed (e.g., BACT, LAER, NSPS, NESHAP, non-regulated toxics,
SIP, and visibility).
Ensure the permit is organized well, e.g., conditions are
independent of one another, and conditions are grouped so as not be
cover more than one area at a time.
Surrogate parameters listed are clear and obtainable.
Emissions limits are clear. In cases of multiple or common exhaust,
limits should specify if per emissions unit or per exhaust.
Every permit condition is 1) reasonable, 2) meaningful,
3) monitorable, and 4) always enforceable as a practical matter.
1.6
-------
DRAFT
OCTOBER 1990
APPENDIX C
POTENTIAL TO EMIT
-------
DRAFT
OCTOBER 1990
APPENDIX C1
POTENTIAL TO EMIT
Upon commencing review of a permit application, a reviewer must define
the source and then determine how much of each regulated pollutant the source
potentially can emit and whether the source is major or minor (nonmajor). A
new source is major if its potential to emit exceeds the appropriate major
emissions threshold, and a change at an existing major source is a major
modification if the source's net emissions increase is "significant." This
determination not only quantifies the source's emissions but dictates the
level of review and applicability of various regulations and new source review
requirements. The federal regulations, 40 CFR 52.21(b)(4), 51.165(a)(l)(iii),
and 51.166(b)(4), define the "potential to emit" as:
"the maximum capacity of a stationary source to emit a pollutant under its
physical and operational design. Any physical or operational limitation on the
capacity of the source to emit a pollutant, including air pollution control
equipment and restrictions on hours of operation or on the type or amount of
material combusted, stored or processed, shall be treated as part of its design
if the limitation or the effect it would have on emissions is federally
enforceable."
In the absence of federally enforceable restrictions, the potential to emit
calculations should be based on uncontrolled emissions at maximum design or
achievable capacity (whichever is higher) and year-round continuous operation
(8760 hours per year).
1 This Appendix is based largely on an EPA memorandum "Guidance on
Limiting Potential to Emit in New Source Permitting," from Terrell E. Hunt,
Office of Enforcement and Compliance Monitoring, and John S. Seitz, Office or
Air Quality Planning and Standards, June 13, 1989.
c.l
-------
DRAFT
OCTOBER 1990
When determining the potential to emit for a source, emissions should be
estimated for individual emissions units using an engineering approach. These
individual values should then be summed to arrive at the potential emissions
for the source. For each emissions unit, the estimate should be based on the
most representative data available. Methods of estimating potential to emit
may include:
Federally enforceable operational limits, including the effect of
pollution control equipment;
performance test data on similar units;
equipment vendor emissions data and guarantees;
test data from EPA documents, including background information
documents for new source performance standards, national emissions
standards for hazardous air pollutants, and Section lll(d) standards
for designated pollutants;
AP-42 emission factors;
emission factors from technical literature; and
State emission inventory questionnaires for comparable sources.
NOTE: Potential to emit values reflecting the use of pollution control
equipment or operational restrictions are usable only to the extent that the
unit/process under review utilizes the same control equipment or operational
constraints and makes them federally enforceable In the permit.
Calculated emissions will embrace all potential, not actual, emissions
expected to occur from a source on a continuous or regular basis, including
fugitive emissions where quantifiable. Where raw materials or fuel vary in
their pollutant-generating capacity, the most pollutant-generating substance
must be used in the potential-to-emit calculations unless such materials are
c.2
-------
DRAFT
OCTOBER 1990
restricted by federally enforceable operational or usage limits. Historic
usage rates alone are not sufficient to establish potential-to-emit.
Permit limitations are significant in determining a source's potential to
emit and, therefore, whether the source is "major" and subject to new source
review. Permit limitations are the easiest and most common way for a source
to restrict its potential to emit. A source considered major, based on
emission calculations assuming 8760 hours per year of operation, can often be
considered minor simply by accepting a federally enforceable limitation
restricting hours of operation to an actual schedule of, for example, 8 hours
per day. A permit does not have to be a major source permit to legally
restrict potential emissions. Minor source construction permits are often
federally enforceable. Any limitation can legally restrict potential to emit
if it meets three criteria: 1) it is federally enforceable as defined by
40 CFR 52.21(b)(17), 52.24(f)(12), 51.165(a)(l)(xiv), or 51.166(b)(17), i.e.,
contained in a permit issued pursuant to an EPA-approved permitting program or
a permit directly issued by EPA, or has been submitted to EPA as a revision to
a State Implementation Plan and approved as such by EPA; 2) it is enforceable
as a practical matter; and (3) it meets the specific criteria in the
definition of "potential to emit," (i.e., any physical or operational
limitation on capacity, including control equipment and restrictions on hours
of operation or type or amount of material combusted, stored, or processed).
The second criterion is an implied requirement of the first. A requirement
may purport to be federally enforceable, but in reality cannot be federally
enforceable if it cannot be enforced as a practical matter.
In the absence of dissecting the legal aspects of "federal
enforceability," the permit writer should always assess the enforceability of
a permit restriction based upon its practicability. Compliance with any
limitation must be able to be established at any given time. When drafting
c.3
-------
DRAFT
OCTOBER 1990
permit limitations, the writer must-always ensure that restrictions are
written in such a manner that an inspector could verify instantly whether the
source is or was complying with the permit conditions. Therefore, short-term
averaging times on limitations are essential. If the writer does this, he or
she can feel comfortable that limitations incorporated into a permit will be
federally enforceable, both legally and practically.
The types of limitations that restrict potential to emit are emission
limits, production limits, and operational limits. Emissions limits should
reflect operation of the control equipment, be short term, and, where
feasible, the permit should require a continuous emissions monitor. Blanket
emissions limits alone (e.g., tons/yr, Ib/hr) are virtually impossible to
verify or enforce, and are therefore not enforceable as a practical matter.
Production limits restrict the amount of final product which can be
manufactured or produced at a source. Operational limits include all
restrictions on the manner in which a source is run, e.g., hours of operation,
amount of raw material consumed, fuel combusted or stored, or specifications
for the installation, maintenance and operation of add-on controls operating
at a specific emission rate or efficiency. All production and operational
limits except for hours of operation are limits on a source's capacity
utilization. To appropriately limit potential to emit consistent with a
previous Court decision [United States v. Louisiana-Pacific Corporation.
682 F. Supp. 1122 (D. Colo. Oct. 30, 1987) and 682 F. Supp. 1141 (D. Colo.
March 22, 1988)], all permits issued must contain a production or operational
limitation in addition to the emissions limitation and emissions averaging
time in cases where the emission limitation does not reflect the maximum
emissions of the source operating at full design capacity without pollution
control equipment. In the permit, these limits must be stated as conditions
that can be enforced independently of one another. This emphasizes the idea
of good organization when drafting permit conditions and is discussed in more
c.4
-------
DRAFT
OCTOBER 1990
detail in the Part III text. The permit conditions must be clear, concise,
and independent of one another such that enforceability is never questionable.
When permits contain production or operational limits, they must also
have requirements that allow a permitting agency to verify a source's
compliance with its limits. These additional conditions dictate
enforceability and usually take the form of recordkeeping requirements. For
example, permits that contain limits on hours of operation or amount of final
product should require use of an operating log for recording the hours of
operation and the amount of final product produced. For organizational
purposes, these limitations would be listed in the permit separately and
records should be kept on a frequency consistent with that of the emission
limits. It should be specified that these logs be available for inspection
should a permitting agency wish to check a source's compliance with the terms
of its permit.
When permits require add-on controls operated at a specified efficiency
level, the writer should include those operating parameters and assumptions
upon which the permitting agency depended to determine that controls would
achieve a given efficiency. To be enforceable, the permit must also specify
that the controls be equipped with monitors and/or recorders measuring the
specific parameters cited in the permit or those which ensure the efficiency
of the unit as required in the permit. Only through these monitors could an
inspector instantaneously measure whether a control was operating within its
permit requirements and thus determine an emissions unit's compliance. It is
these types of additional permit conditions that render other permit
limitations practically and federally enforceable.
Every permit also should contain emissions limits, but production and
operational limits are used to ensure that emissions limits expressed in the
c.5
-------
DRAFT
OCTOBER 1990
permit are not exceeded. Production limits are most appropriately expressed
in the shortest time periods as possible and generally should not exceed
1 month (i.e., pounds per hour or tons per day), because compliance with
emission limits is most easily established on a short term basis. An
inspector, for example, could not verify compliance for an emissions unit with
only monthly and annual production, operational or emission limits if the
inspection occurred anytime except at the end of a month. In some rare
situations a 1-month averaging time may not be reasonable. In these cases, a
limit spanning a longer period is appropriate if it is a rolling average
limit. However, the limit should not exceed an annual limit rolled on a
monthly basis. Note also that production and operational recordkeeping
requirements should be written consistent with the emissions limits. Thus, if
an emissions unit was limited to a particular tons per day emissions rate,
then production records which monitor compliance with this limit should be
kept on a daily basis rather than weekly.
One final matter to be aware of when calculating potential to emit
involves identifying "sham" permits. A sham permit is a federally enforceable
permit with operating restrictions limiting a source's potential to emit such
that potential emissions do not exceed the major or de minimis levels for the
purpose of allowing construction to commence prior to applying for a major
source permit. Permits with conditions that do not reflect a source's planned
mode of operation may be considered void and cannot shield the source from the
requirement to undergo major source preconstruction review. In other words,
if a source accepts operational limits to obtain a minor source construction
permit but intends to operate the source in excess of those limitations once
the unit is built, the permit is considered a sham. If the source originally
intended or planned to operate at a production level that would make it a
major source, and if this can be proven, EPA will seek enforcement action and
the application of BACT and other requirements of the PSD program.
c.6
-------
DRAFT
OCTOBER 1990
Additionally, a permit may be considered a sham permit if it is issued for a
number of pollution-emitting modules that keep the source minor, but within a
short period of time an application is submitted for additional modules which
will make the total source major. The permit writer must be aware of such
sham permits. If an application for a source is suspected to be a sham, EPA
enforcement and source personnel should be alerted so details may be worked
out in the initial review steps such that a sham permit is not issued. The
possibility of sham permits emphasizes the need, as discussed in the Part III
text, to organize and document the review process throughout the file. This
documentation may later prove to be evidence that a sham permit was issued, or
may serve to refute the notion that a source was seeking a sham permit.
Overall, the permit writer should understand the extreme importance of
potential to emit calculations. It must be considered in the initial review
and continually throughout the review process to ensure accurate emission
limits that are consistent with federally enforceable production and
operational restrictions.
c.7
------- |