EPA Region 8
Review of EPA's Office of Inspector General
Hotline Complaint No. 2021-0188
Report
July 2022
-------
EPA Region 8 Review of Issues Raised in PEER Complaint
Concerning New Source Review Permitting by the Colorado
Department of Public Health and Environment
Table of Contents
Executive Summary 3
Introduction 5
Legal Requirements; Relevant Policies and Procedures 7
NAAQS and SIPs 7
The NSR Permit Program 8
The Colorado Minor NSR Program 11
Air Quality Modeling in the Colorado Minor NSR Program 12
APCD Modeling Guidance and Policies 13
APCD Organization and Permitting Practices 17
EPA Region 8 Review of Individual Permits and Associated Air Quality Analyses 18
Figure 1. Permitted Sources Identified in Complaint 18
Bighorn Pad in Jackson County, Colorado (Permit Number 16JA1055, issued November 14,
2018) 19
Cripple Creek & Victor Gold Mine in Teller County, Colorado (Permit Number 98TE0545,
issued November 17, 2020) 21
Martin Marietta Materials Highway 34 Facility in Weld County, Colorado (Permit Number
17WE0578, issued December 17, 2018) 23
Williams Willow Creek Gas Plant in Rio Blanco County, Colorado (Permit Number
07RB0616, issued July 7, 2016) 24
Overall Findings 26
Agency Recommendations 28
Appendix A - Memorandum and Attachments from Kevin Collins, Hotline Coordinator, EPA
OIG Office of Counsel and Congressional and Public Affairs, to Deb Thomas, Acting Regional
Administrator EPA Region 8 (May 24, 2021)
Appendix B - Colorado Interim Modeling Guideline for Air Quality Permits (October 2021)
Appendix C - Table, History of Denver Metro North Front Range 2008 Ozone Standard with
Corresponding Ozone Nonattainment New Source Review Requirements for Major Sources
2
-------
Executive Summary
Public Employees for Environmental Responsibility (PEER) filed a complaint with the EPA
Office of Inspector General (OIG) in March 2021 about the implementation of the Clean Air Act
(CAA) minor source permitting program by the Colorado Department of Public Health and
Environment (CDPHE). PEER submitted the complaint as a letter signed by three employees of
CDPHE's Air Pollution Control Division (APCD).
The complaint charges that CDPHE1 had ordered staff not to perform air quality modeling that is
"legally required" to show that new and modified minor stationary sources will not cause or
contribute to a National Ambient Air Quality Standards (NAAQS) exceedance, and that when
modeling had been performed, CDPHE disregarded it. The result, according to PEER, was that
CDPHE issued permits without determining whether emissions increases associated with a
project would violate the NAAQS. The complaint further asserts that this failure to perform air
quality modeling has led to excess emissions that have contributed to the failure to attain the
NAAQS and to "chronic health problems, premature deaths, and severe injury to the
environment."
The EPA OIG referred the PEER complaint to EPA Region 8 for appropriate action. We have
completed our review of the issues raised in the complaint. In conducting this review, we
focused on four of the 11 facilities identified in the complaint and information in the associated
CDPHE permitting files, as well as applicable law and regulations. We also met with CDPHE
personnel, including the signatories to the complaint. This report presents our conclusions and
recommendations.
We have identified important concerns with the state's implementation of the CAA minor source
permitting program. In particular, the CDPHE records for the four facilities did not include
analysis supporting the conclusion that the approved permit actions would not cause a NAAQS
violation, as required by CDPHE's federally approved CAA regulations for minor sources.
Instead, the records showed that CDPHE issued permits even though air quality modeling
analyses predicted NAAQS violations, and that these violations predicted by the model were not
addressed by refined or more representative air quality modeling, use of supplemental air quality
data to support the permit conditions, or additional permit conditions that would address the
predicted violations. These insufficiencies apparently stemmed from CDPHE's practice of
determining when to require air quality modeling for minor source construction permit
applications by using simplified criteria based on total annual emissions, which did not ensure
that permits would be issued consistent with Colorado's applicable regulatory requirements for
NAAQS protection. Additionally, a lack of communication among the various APCD groups
processing permit applications contributed to the failures described above. Further, our review
found that one of these facilities had complex and multi-year projects that were split into smaller,
individual permit actions without a documented supporting basis for treating them as separate
permit actions.
1 CDPHE is the state's environmental agency; APCD is the component of CDPHE that addresses air quality issues
by processing permits, conducting inspections of regulated entities, and proposing new regulations under state laws.
See https://cdphe. Colorado. gov/about. This report primarily refers to "CDHPE," as the more publicly familiar term
and one that encompasses all of the agency's subdivisions, but in doing so no distinction from APCD is intended. In
places, particularly when quoting another source, we refer to "APCD."
3
-------
CDPHE's failure to consider potential NAAQS violations and ensure compliance with the
NAAQS is inconsistent with the requirements of the Colorado State Implementation Plan (SIP)
that implement the CAA Minor New Source Review (Minor NSR) permit program, and could
result in harm to air quality and public health. To address these concerns, we recommend that
CDPHE consider these measures:
1. Ensure that all future Minor NSR permit records are complete and include sufficient
documentation to support permit conditions and contain analyses that demonstrate
that the permit conditions will not cause NAAQS violations. As needed, undertake
additional qualitative or quantitative air quality analyses to demonstrate that the
permit conditions comply with the NAAQS and include these analyses in the permit
record.
2. For the 11 permit records identified in the complaint, amend permit actions as
appropriate by conducting refined modeling, incorporating additional/revised permit
conditions, and/or potentially including post construction ambient air monitoring. For
the Cripple Creek and Victor Gold Mine, this would also include providing an
explanation demonstrating that the various projects were not under aggregated when
determining the projects qualified for Minor NSR permits. In light of the high level of
public interest on these issues, the EPA notes that any revisions to these permits,
including the permit record, would benefit from public notice and comment, even if
state rules would not so require.
3. Maintain complete public records for all NSR permits. Records should be retained for
10 years after the permit expires, is terminated, or withdrawn, or longer if required
under state law.
4. Improve communication and coordination among the APCD groups that work on
NSR permits.
5. Ensure that complex and multi-year projects are covered under the proper major or
minor source program permit requirements based on appropriate aggregation
considerations and have adequate air quality impact analyses (AQIAs), and that any
decisions to permit individual units are justified and adequately documented in the
permit records.
6. During quarterly meetings between APCD and EPA Region 8 permitting managers,
provide status reports on newly issued Minor NSR permits that describe the AQIAs
and permit condition determinations.
We acknowledge that CDPHE has already taken some steps to address the issues raised in the
complaint and that some of recommendations listed above overlap with these efforts. The current
actions being taken by CDPHE to address the allegations in the complaint are detailed on page
29 of this report. We are seeking a written response from CDPHE by October 21, 2022, that
addresses our findings and recommendations; we may revisit some of our above
recommendations after reviewing the response. We are also available to advise and consult with
CDPHE on the actions to address our findings in this report.
4
-------
Introduction
On March 30, 2021, PEER filed a complaint with the EPA OIG requesting an investigation of
CDPHE's implementation of the CAA Minor NSR permitting program.2 The complaint was
submitted in the form of a letter signed by three APCD employees, and raised two overlapping
concerns. First, that CDPHE has directed staff not to perform air quality modeling that is
required to show that new and modified minor stationary sources will not cause or contribute to a
NAAQS violation. Second, that CDPHE issued permits with AQIAs that contained unresolved
predicted NAAQS violations. The result, according to PEER, has been excess emissions that
have contributed to the failure to attain the NAAQS and to "chronic health problems, premature
deaths, and severe injury to the environment."
According to the PEER complaint, these problems have existed at least since September 2010,
when CDPHE published a guidance document known as PS Memo 10-01.3 The letter makes a
number of claims concerning CDPHE's use of PS Memo 10-01, among them that it exempted
sources with NO2 and SO2 emissions of 40 tons per year (tpy) or less from any requirement to
demonstrate compliance with the NAAQS, and alleges "a longstanding informal practice of the
APCD of approving permits at all costs even where modeling has demonstrated NAAQS
violations."4 It also asserts that on March 15, 2021, APCD management issued written
instructions to the air quality modeling staff to refrain from performing modeling for sources
with emissions of criteria pollutants above specified thresholds, including a 40 tpy threshold for
NO2 and SO2.5 The letter points out that NO2 is a precursor pollutant that contributes to the
formation of ozone, for which the Denver Metro / North Front Range area's nonattainment status
has deteriorated.6 The letter charges that using PS Memo 10-01, the March 2021 directives, and
other policies, CDPHE has issued permits "without any attempt to determine whether a proposed
new facility or modification would interfere with the attainment or maintenance of hourly and
daily NAAQS," and also that some permits were issued despite "actual knowledge that the
proposed facilities may have modeled NAAQS violations, or even with knowledge of clear and
conclusive evidence of such violations."7
In response to the PEER complaint, the EPA OIG referred the matter to EPA Region 8 for
review and appropriate action.8 On July 7, 2021, the EPA Region 8 Acting Regional
Administrator sent a letter to the CDPHE Director of Environmental Health and Protection,
outlining the issues presented to the EPA and requested that CDPHE provide EPA Region 8
relevant information by the end of that month. CDPHE submitted responsive documents to EPA
Region 8 on July 26, 2021, and EPA Region 8 completed an initial review of these documents in
2 Letter to EPA Inspector General Sean W. O'Donnell from PEER et al. ("PEER complaint") (March 30, 2021). A
copy is included in Appendix A (Attachment 2) for reference.
3 State officials have informed EPA Region 8 that PS Memo 10-01 is no longer in effect. A copy is included in
Appendix A (Attachment 4) for reference.
4 PEER complaint at 8.
5 PEER complaint at 1, 5.
6 PEER complaint at 7-8.
7 PEER complaint at 10.
8 Memorandum from Kevin Collins, Hotline Coordinator, EPA OIG Office of Counsel and Congressional and
Public Affairs, to Deb Thomas, Acting Regional Administrator EPA Region 8 (May 24, 2021). A copy along with
the attachments is included in Appendix A for reference.
5
-------
October 2021. On November 18, 2021, EPA Region 8 personnel met with CDPHE
representatives to ask follow-up questions and request supplemental information. CDPHE
provided the supplemental information on December 3, 2021. Also, on January 5, 2022, EPA
Region 8 staff held a conference call with PEER and their clients to provide them an opportunity
to discuss their concerns.
EPA Region 8's main goal in this review was to determine whether CDPHE issued any NSR
permits that failed to properly demonstrate compliance with the NAAQS, or otherwise violated
any SIP requirements for their Minor NSR permit program. At various stages during our review,
we provided updates to the OIG and obtained feedback from the OIG as to our approach.
We received several follow-up submittals from PEER after the original complaint was filed,
which, in some cases, would have expanded the scope of the complaint. We explained to the
OIG that, with respect to such documents, we intended to focus only on the issues identified in
the original complaint. The OIG generally agreed that our primary role was to perform a review
focused on the goals described above. The OIG did request that EPA Region 8 refer back to the
OIG any evidence of fraud or misconduct that we encountered, but we did not conduct our
review as an investigation of potential fraud or misconduct; rather, we focused on CAA
requirements and air quality issues.9 In any case, in our review we did not determine that state
employees had committed fraud or other misconduct requiring a further referral to the OIG.10
We did identify potential conflicts of interest in connection with CDPHE decisions on one of the
permit actions, as discussed further below in the discussion of the Cripple Creek and Victor Gold
Mine. We reported these findings of potential conflicts of interest to the OIG. Otherwise, in our
view, the Colorado Attorney General's office would be in the best position to evaluate any
personnel issues raised by these potential conflicts.11
The purpose of this report is to document and explain the results of our review. It is also intended
to inform CDPHE of actions needed to ensure that permits are consistent with attainment of the
NAAQS and to resolve any potential violations of Colorado SIP requirements in their issuance of
9 The PEER complaint does not describe specific instances of fraud or misconduct, except that PEER alleges that
"[a] CDPHE modeler was ordered to falsify data in a modeling report regarding this facility [the Cripple Creek &
Victor Gold Mine] to ensure that no modeled violation would be reported." PEER complaint at 13. We discuss this
allegation below in our detailed review of the permitting actions for the mine.
10 See generally https://www.epa.gOv/office-inspector-general/epa-oig-hotline-information#recognizing (last
accessed on May 25, 2022) ("Understanding Fraud, Waste and Abuse"). The PEER complaint does not describe
specific instances of fraud or misconduct, except that PEER alleges that "[a] CDPHE modeler was ordered to falsify
data in a modeling report regarding this facility [the Cripple Creek & Victor Gold Mine] to ensure that no modeled
violation would be reported." PEER complaint at 13. Separately from the PEER complaint, in August 2021 we and
the OIG received a memorandum from one of the CDPHE employees represented by PEER, containing various
allegations related to CDPHE's handling of air permitting for the Cripple Creek and Victor Gold Mine. In our
review we focused on the legal sufficiency of the permit record for this facility; we did not assess the conduct-
related allegations in the PEER complaint or the August 2021 statement, but highlighted the August 2021
memorandum to the OIG for evaluation. See also our detailed review of the permitting actions for the mine, below.
11 We note in addition that the Colorado Attorney General commissioned an independent investigation of the
allegations outlined in the letter from PEER, and issued a report of the findings from their investigation. Public
Report of Independent Investigation of Alleged Non-Enforcement of National Ambient Air Quality Standards by the
Colorado Department of Public Health and Environment, Troutman Pepper Hamilton Sanders LLP as Special
Assistant Attorneys General for the State of Colorado (Sep. 22, 2021), available at
https://coag.gov/app/uploads/2021/09/CDOL-Report-210922.pdf (last accessed on May 1, 2022).
6
-------
Minor NSR permits. In addition, we expect that the results of this review will assist the EPA
with addressing the concerns outlined in a July 2021 OIG report12 that recommended that the
EPA conduct more oversight of "synthetic minor" source permitting.
Legal Requirements; Relevant Policies and Procedures
This report focuses on whether the permit decisions for the four facilities selected for review are
consistent with the Colorado SIP, and whether the decisions for these permit actions are
supported by the permit record. This report does not attempt to provide a detailed response to the
complaint's characterizations of the law. Instead, this section of the report summarizes the legal
requirements relevant to our review.
NAAQS and SIPs
As required under the CAA, the EPA has established NAAQS for six common pollutants, often
referred to as "criteria" pollutants: ozone, NO2, SO2, particulate matter (PM2.5 and PM10), carbon
monoxide, and lead.13 The NAAQS are set at levels intended to protect public health and
welfare.14 Once the EPA establishes a new or revised NAAQS, states must develop SIPs
containing emission limitations and other control measures to ensure that the relevant NAAQS is
achieved and maintained.15 SIPs must include permit programs (generally referred to as NSR
permit programs) to ensure that the construction and modification of stationary sources are
protective of the NAAQS.16 SIPs, including NSR program requirements, must be submitted to
the EPA for review and approval.17 Once a SIP submission is approved by the EPA, it is
incorporated into federal regulations, and its requirements are federally enforceable. The state
and the EPA share the responsibility to ensure compliance with approved SIPs, and may take
enforcement actions to do so. The EPA must take action on finding that any person has violated a
requirement of an applicable SIP (or CAA Title V operating permit), or on finding that there are
widespread violations of the SIP (or CAA Title V operating permit) that appear to result from a
failure of the state to enforce effectively.18 The EPA must notify the violator and the state of any
such findings, and may enforce any plan or operating permit program requirement by (1) issuing
12 OIG Report No. 21-P-0175 (July 8, 2021) is available at: https://www.epa.gov/office-inspector-general/report-
epa-should-conduct-more-oversight-synthetic-minor-source-permitting (last accessed on Jan. 5, 2022).
13 42 U.S.C. § 7409. See https://www.epa.gov/criteria-air-pollutants for more information on the NAAQS (last
accessed on Apr. 7, 2022). This and many other CAA provisions refer to the EPA Administrator; for simplicity, this
report generally uses "EPA" interchangeably with "EPA Administrator."
14 42 U.S.C. § 7409(b).
15 42 U.S.C. §§ 7407(a), 7410(a)(l)-(2). In some areas, principally in parts of California, local governments are
responsible for developing implementation plans. Tribal governments may choose to develop tribal implementation
plans, Federally recognized tribes may choose to develop Tribal Implementation Plans (TIP) containing NSR permit
programs, which are reviewed and approved by EPA, or may choose to seek delegation of the federal NSR permit
programs, but are not required to do so. 42 U.S.C. §§ 7410(o), 7601(d); 40 C.F.R. part 49, subpart A. If a tribe
chooses not to develop its own NSR programs or seek delegation of the federal programs, EPA implements the
programs and issues the NSR permits. For simplicity, since this report concerns the Colorado plan, it generally
refers to "states" and "SIPs."
16 42 U.S.C. § 7410(a)(2)(C).
17 42 U.S.C. § 7410(a).
18 42 U.S.C. § 7413(a).
7
-------
an order requiring the violator to comply, (2) issuing an administrative penalty order, or (3)
bringing a civil action. For NSR construction permits, the CAA and federal regulations do not
provide the EPA the authority to reopen individual permits.19 Instead, the EPA's oversight of the
NSR permit program occurs principally through the SIP review process, through our issuance of
regulations setting minimum standards for state permitting programs, and through our ability to
provide comments to the state on proposed permits. The EPA may also comment on proposed
state permits, informally in connection with our oversight of state programs, or during a state
comment period.
If the EPA disapproves a SIP submission, it may establish a federal implementation plan (FIP)
instead, unless the state submits a timely replacement and the EPA approves it.20 The EPA may
also request revisions to approved SIPs.21 In particular, the EPA can require the state to revise
the plan when any part of the plan is substantially inadequate to attain or maintain the NAAQS,
address the interstate transport of pollutants, or otherwise comply with the CAA.
In this review, we have not determined that state failures to enforce the SIP have led to
widespread violations within the meaning 42 U.S.C. §7413(a)(2). We do have important
concerns in several areas and are explaining those concerns in this report as a part of EPA
Region 8's regular oversight of SIP implementation by Colorado.
The NSR Permit Program
The NSR permit program is intended to protect human health and the environment from the
adverse effects of pollution increases caused by the construction and modification of stationary
air pollution sources.22 It seeks to achieve these goals in part through a preconstruction
permitting process requiring facilities to install pollution control technology under certain
conditions.
Each state must develop, implement, maintain, and enforce a SIP that includes a program to
regulate the modification and construction of any stationary source to ensure that the NAAQS
are achieved and that sources comply with CAA requirements.23 Also, each SIP must include
procedures that let the state determine whether the construction or modification of a facility will
cause a violation of the control strategy, or will interfere with attainment or maintenance of a
national standard.24 The NSR permit program covers the NAAQS pollutants and additional
pollutants under certain conditions.25 Typically, NSR permits are issued by state or local air
19 We note, however, that as to Prevention of Significant Deterioration permits, the Supreme Court has noted that the
CAA "instructs EPA to 'take such measures, including issuance of an order, or seeking injunctive relief, as
necessary to prevent the construction' of a major pollutant emitting facility that does not conform to the PSD
requirements of the ActAlaska Dep't of Env't Conservation v. E.P.A., 540 U.S. 461, 461 (2004).
20 42 U.S.C. § 7410(c).
21 42 U.S.C. § 7410(k)(5).
22 See 42 U.S.C. §§ 7410(a)(2)(C), 7470(1), 7503(a).
23 42 U.S.C. §§ 7410(a)(1), 7410(a)(2)(C), and 7410(a)(2)(D)(ii)); 40 C.F.R. part 51, subpart I.
24 40 C.F.R. § 51.160.
25 Federal minor source permit requirements extend to NAAQS pollutants, while major source permit requirements
are more expansive. See 40 C.F.R. § 51.160(a) (requiring minor source programs to address "Interference with
attainment or maintenance of a national standard"); 51.165(a)(xxxvii) (defining "regulated NSR pollutant" for major
source permitting).
8
-------
pollution control agencies. States must develop their own NSR permit programs as part of their
EPA-approved SIPs, or the state may be delegated the authority to issue permits on behalf of the
EPA under federal regulations. Once the EPA approves a state's permit program into the SIP, the
program is federally enforceable, and the permits issued by the state under the SIP-approved
program are federally enforceable legal documents.26
The NSR permit program includes the Prevention of Significant Deterioration (PSD) program,
the Nonattainment NSR program, and the Minor NSR program. The legal basis for each of these
is in Title I of the CAA and in related regulations. The PSD program applies to new major
stationary sources and major modifications of existing stationary sources in areas that are
designated as attaining the NAAQS ("attainment areas") or are unclassifiable.27 The
Nonattainment NSR program, like the PSD program, applies to new major sources and to
sources making major modifications, but only in areas that are failing to attain one or more
NAAQS ("nonattainment areas").28 The Minor NSR program applies to new minor sources in
both attainment and nonattainment areas, and to both major and minor sources making minor
modifications.29 The Minor NSR program must address criteria pollutants, and may address
other pollutants, depending on the state.30 A facility may be subject to more than one of these
programs.
Major sources are facilities with the potential to emit pollutants in amounts equal to or greater
than a major source threshold level. These threshold levels vary by pollutant and source
category. Minor sources are facilities with the potential to emit pollutants in amounts less than
the major source thresholds, while synthetic minor sources are facilities that have the potential to
emit pollutants at or above a major source threshold level but have voluntarily accepted
enforceable limits to keep their emissions below the major source thresholds and avoid the major
NSR requirements.
Key to the NSR permit program are procedures and demonstrations to ensure that increased
emissions from new and modified sources will not interfere with attainment of the NAAQS or
with other CAA requirements. In the Nonattainment NSR program, covered sources must offset
emissions increases with greater emission reductions from within the same nonattainment area,
or from another nonattainment area if certain additional requirements are satisfied.31 On the other
hand, the PSD program requires applicants for major source permits to submit an analysis of the
impacts to the existing air quality resulting from the proposed source or modification.32 The
26 42 U.S.C. 7413 (federal enforcement, including enforcement of SIP requirements).
27 See 42 U.S.C. § 7475; 40 C.F.R. §§ 51.166 (requirements for state PSD programs submitted for SIP approval),
52.21 (federal PSD program, which states may request delegation to implement instead of developing their own
program to be submitted for SIP approval).
28 See 42 U.S.C. § 7503; 40 C.F.R. § 51.165.
29 See 42 U.S.C. § 7410(a)(2)(C) (requiring SIPs to contain program for "regulation of the modification and
construction of any stationary source"); 40 C.F.R. § 51.160 (requirements for permit programs in SIPs generally)
(both implicitly including minor sources).
30 Colorado's NSR program requirements are discussed later in this report.
31 See 42 U.S.C. § 7503(a)(1)(A). The CAA includes a limited exception to the offset requirement for sources in
designated economic development zones. 42 U.S.C. § 7503(a)(1)(B). That exception is not relevant here, because no
such zone has been designated in Colorado.
32 42 U.S.C. § 7475; 40 C.F.R. § 51.166(m).
9
-------
applicant must also submit a "source impact analysis" demonstrating that emissions increases
from the project will not cause or contribute to violations of the NAAQS or to exceedances of
regulatory limits on increases in ambient pollution levels ("increments").33 The required source
impact analysis demonstration is often referred to as an "air quality impact analysis," or AQIA.
The source impact analysis primarily involves a model that uses mathematical formulations to
characterize the atmospheric processes that disperse pollutants emitted by a source.34
EPA regulations promulgated to implement the CAA require that each SIP include a Minor NSR
program.35 Under the regulations, each SIP must include "legally enforceable procedures" - that
is, a Minor NSR program - that will inform the state whether the construction or modification of
a minor source will (1) result in a violation of applicable portions of the state's control strategy,
or (2) interfere with attainment or maintenance of any NAAQS in the state or a neighboring
state.36 As a part of these procedures, each SIP must require owners or operators of minor
sources to submit applications that will allow the state to determine whether the construction or
modification of the source will result in a violation of the control strategy or interfere with
attainment of a NAAQS.37 States have significant flexibility in designing and implementing
Minor NSR programs to meet the requirements of the CAA and the regulations. Although EPA
regulations do not require air quality modeling for every minor source permit, the regulations do
require that SIPs discuss the air quality data and modeling that will be used to meet the
requirements of the regulations, and the modeling must be based on EPA's Guideline on Air
Quality Models.38 EPA's preferred air quality model system used to determine compliance with
the NAAQS is the AMS/EPA Regulatory Model (AERMOD). This system also includes a
screening model, AERSCREEN, that produces estimates of "worst-case" air pollution without a
fully developed set of input data. AERSCREEN is typically used for preliminary estimates of air
pollution because it produces estimates that are equal to or greater than the estimates produced
by AERMOD and it does not require substantial resources to use for projects relative to using
AERMOD. If AERSCREEN predicts concentrations greater than applicable thresholds, then
33 42 U.S.C. § 7475; 51.166(k) (source impact analysis); 42 U.S.C. §§ 7473, 7575(a)(3)(A), 40 C.F.R. § 51.166(c)
(increments); see 75 Fed. Reg. 64864, 64868 (Oct. 20, 2010) (explaining PSD increments).
34 Under some circumstances, when emission increases are below specified "Significant Impact Levels," modeling
may not be required to make the required demonstration under 40 C.F.R. § 51.166(k). See Guidance on Significant
Impact Levels for Ozone and Fine Particles in the Prevention of Significant Deterioration Permitting Program, EPA
OAQPS (Apr. 17, 2018) available at https://www.epa.gov/sites/default/files/2018-
04/documents/sils_policy _guidance_document_final_signed_4-17-18.pdf.
35 40 C.F.R. §§ 51.160-51.164. Each SIP must provide for the regulation of the modification and construction of any
stationary source as necessary to assure that the NAAQS are achieved. 42 U.S.C. § 7410(a)(2)(C). Thus, each SIP
must contain a Minor NSR program.
36 40 C.F.R. § 51.160. The SIP must identify the types and sizes of sources that will be subject to Minor NSR, and
must discuss the basis for determining which facilities will be subject to review as minor sources. 40 C.F.R.
51.160(e). It must also identify the state or local agency responsible for implementing the program, and must contain
administrative procedures to be followed in making the determination of whether the construction or modification of
a minor source will violate the state's control strategy or interfere with attainment and maintenance of the NAAQS.
40 C.F.R. §51.163.
37 40 C.F.R. § 51.160(c).
38 40 C.F.R. § 51.160(f) (requiring that the state program "discuss the air quality data and the dispersion or other air
quality modeling used to meet the requirements of this subpart," and requiring that modeling performed must be
based on the Guideline on Air Quality Models or an approved substitute); 40 C.F.R. part 51, App. W (Guideline on
Air Quality Models).
10
-------
AERMOD is used to refine the analysis with more representative input data and to better
understand the potential air quality impacts resulting from a project's emissions.
The PEER complaint primarily focuses on concerns related to the state's Minor NSR program,
and in several ways alleges that CDPHE's implementation of the Minor NSR program is failing
to ensure attainment of the NAAQS and protection of public health. Accordingly, the following
section discusses federal and Colorado Minor NSR requirements in detail.
The Colorado Minor NSR Program
The state's SIP-approved NSR program is contained in "Regulation 3," adopted by the Colorado
Air Quality Control Commission under the authority of the Colorado Air Pollution Prevention
and Control Act.39 As relevant to this report, Regulation 3 Part B includes construction permit
review procedures for all NSR permits. Accordingly, Regulation 3 provides the primary
requirements that are relevant to the state's Minor NSR program, along with requirements
relevant to the state's PSD and Nonattainment NSR programs.40
Under the Colorado program, to be considered complete a minor source permit application must
include all information and data required for the state's permitting group (APCD) to evaluate the
application.41 The APCD's evaluation begins with a required "preliminary analysis" that allows
it to determine whether the new source will comply with (1) all applicable emission control
regulations, (2) applicable regulations for the control of hazardous pollutants, (3) requirements of
the nonattainment and attainment programs, and (4) all applicable ambient air quality standards
and regulations.42 This preliminary analysis must indicate what impact, if any, the proposed new
source will have on all areas (attainment, nonattainment, etc.) within its probable area of
influence. In other words, the analysis must assess the source's potential impact on NAAQS
attainment.43
The program requires public comment on permits for some minor sources, including those with
projected controlled annual emissions of any NAAQS pollutant of more than 25 tpy in
nonattainment areas, or 50 tpy in attainment and attainment/maintenance areas. A public
comment period is also required for sources that would otherwise be subject to the PSD or
Nonattainment NSR programs, but that are seeking a federally enforceable limit on the potential
to emit of the source to avoid these requirements (i.e., synthetic minor sources).44 In addition,
39 Regulation Number 3, Stationary Source Permitting and Air Pollutant Emission Notice Requirements, 5 CCR
1001-5; Colo. Rev. Stat. (C.R.S.) 25-7-114 to 25-7-114.7.
40 The state's regulations allow combined construction and operating permits; when a minor new source permit is
combined with an operating permit, relevant regulations appear in Regulation 3 part C as well as part B. See 5 CCR
1001-5, Part B, III.D.l.; Part C, III. Other potentially applicable parts of Regulation 3 include Part A (general
provisions) and Part D (major source requirements).
41 5 CCR 1001-5, Part B, III.B.4. An applicant can request a pre-application meeting with the APCD concerning
applicable permit requirements (e.g., the information, plans, specifications, and data required to be furnished with
the permit application). Id. at III.A. These application requirements apply to major source construction permits as
well.
42 5 CCR 1001-5, PartB, III.B.5.
43 5 CCR 1001-5, PartB, III.B.5.
44 5 CCR 1001-5, Part B, III.C. 1. Other sources for which public comment is required include those with 200 tpy of
11
-------
APCD can require public comment for other sources when it is "warranted" considering the
source's duration and location, the nature and projected amount of emissions, anticipated public
concern, and other relevant factors.45 When public comment is required, including when the
APCD determines that an application warrants public comment, the APCD must make the
application publicly available and must send a copy of the public notice and the draft permit to
the EPA.46
Under the state's Minor NSR program rules, the APCD must grant a minor source construction
permit if it finds that the proposed project will not cause an exceedance of any NAAQS, will
meet all applicable ambient air quality standards and regulations and requirements for control of
hazardous air pollutants, and has paid the required fees.47 If the APCD determines that a project
cannot comply with any of these requirements, or with any applicable Reasonably Available
Control Technology (RACT) requirements, it must deny the permit application.48
As noted previously, the complaint raises concerns that the state's Minor NSR program is failing
to sufficiently consider whether proposed projects will interfere with attainment of the NAAQS.
In particular, the complaint charges that the state is failing to employ air quality modeling
appropriately. The following section discusses the air quality modeling requirements of the
Colorado Minor NSR program.
Air Quality Modeling in the Colorado Minor NSR Program
The primary concern of the complaint appears to be how APCD has considered modeling in
Minor NSR permit actions. An air quality model uses mathematical and numerical techniques,
along with meteorological data and source information, to simulate the physical and chemical
processes that affect air pollutants as they disperse and react in the atmosphere. Modeling can be
used during the permitting process to verify that a new source or modification will not exceed
ambient air quality standards, and when necessary to determine additional control requirements.
The state's Minor NSR program requires that applications for construction permits include "all
information and data required to evaluate the application."49 As explained previously, the APCD
evaluation begins with a "preliminary analysis" that includes an assessment of the source's
impact on the NAAQS.50 When the preliminary analysis includes modeling, the model used must
lead emissions, regardless of the area's attainment status; those that may violate state requirements concerning
odors; those that are subject to federal National Emission Standards for Hazardous Air Pollutants (NESHAPs), or
federal or state Maximum Achievable Control Technology (MACT) or Generally Available Control Technology
standard; and those that are applying for a Best Available Retrofit Technology (BART) determination or alternative
under the state's regulation implementing the CAA Regional Haze program. Id.
45 5 CCR 1001-5, Part B, III.C.3. One exception to the state's public notice requirements is that sources applying for
construction or modification permits as a part of a combined NSR and operating permit follow only the public
participation requirements of the state's operating permit program. 5 CCR 1001-5, Part B, III.C.2.C.
46 5 CCR 1001-5, PartB, III.C.4.
47 5 CCR 1001-5, PartB, III.D.l. Minor sources may also need to comply with Reasonably Available Control
Technology (RACT) requirements, and additional requirements apply to major source permits. Id. at III.D.2.,
III.D.l.b, III.D.I.e.
48 5 CCR 1001-5, PartB, III.F.l. In addition to the requirements at III.D.l, permits must comply with requirements
49 5 CCR 1001-5, PartB, III.B.4.
50 5 CCR 1001-5, PartB, III.B.5.
12
-------
be appropriate to the topography, meteorology, and other characteristics of the region that the
source will impact, and must be approved in EPA modeling guidelines or otherwise approved by
the EPA.51 To grant the permit, APCD must find, among other things, that the proposed source
will not cause an exceedance of any NAAQS, and that it will "meet any applicable ambient air
quality standards and all applicable regulations."52
The complaint asserts that the SIP requires CDPHE to use air quality modeling for all minor
sources. Specifically, the complaint quotes this SIP-approved provision of the state regulations:
All estimates of ambient concentrations required under this Regulation Number 3 shall be
based on the applicable air quality models, databases, and other requirements generally
approved by U.S. EPA and specifically approved by the Division.53
And from that requirement, the complaint concludes that "CDPHE is required to verify through
air quality modeling that a new major or minor stationary source, or a modification to an existing
source, will not cause, a NAAQS exceedance."54 This is an overstatement. The quoted Colorado
regulations do require that when ambient air quality estimates are used, the estimates must be
based on approved air quality models.55 But the regulations do not require air quality modeling
in every case. It may be possible for a permitting authority to use means other than modeling to
justify a conclusion that a permit will not interfere with attainment of the NAAQS. This option is
within the scope of the state's discretion as the permitting authority under its SIP-approved
Minor NSR program.56
We emphasize that such a conclusion must be justified in the supporting record for the permit.
The complaint alleges, however, that the APCD has issued guidance and unwritten directives
that in effect require an unsupported conclusion that issuance of many permits will not interfere
with the NAAQS. The following section of this report gives an overview of the state's modeling-
related guidance and summarizes the complaint's allegations concerning its use and effect.
APCD Modeling Guidance and Policies
As noted previously, the complaint makes several allegations about the state's use of APCD-
issued guidance, and about instructions reportedly given to APCD staff concerning the use of
modeling for the Minor NSR program. The complaint alleges that CDPHE's application of this
guidance has resulted in failures by CDPHE to comply with the requirements of the SIP and
NSR program. The state guidance under consideration does not establish legal requirements that
CDPHE must follow, and it has not been submitted as a part of the SIP. This report does not
attempt to evaluate all details or the entire history of the modeling-related APCD guidance and
51 5 CCR 1001-5, Part B, III.B.5; Part A, VIII.A.
52 5 CCR 1001-5, Part B, III.D.l.
53 5 CCR 1001-5, Part A, VIII.A.l, quoted inPEER complaint at 5.
54 PEER complaint at 5 (emphasis added).
55 5 CCR 1001-5, PartB, III.B.5.
56 The PEER complaint also incorrectly asserts that the state's regulations impose a cause-or-contribute standard for
evaluating NAAQS in minor source permitting. Instead, the state regulation requires CDPHE to grant the permit if
the new source "will not cause an exceedance of the NAAQS" and will meet other requirements. 5 CCR 1001-5,
Part B, III.D. I.e. Thus, "or contribute" is not a part of the state's SIP-approved requirements for evaluating minor
source impact on the NAAQS.
13
-------
policies, but our review did assess the guidance because the complaint alleges that it influenced
the state's permitting decisions.
The state's primary guidance document related to modeling in the Minor NSR program has
undergone various revisions over the years, but has generally been titled the Colorado Modeling
Guideline for Air Quality Permits.57 The Colorado Modeling Guideline outlines the APCD's
expectations for AQIAs to estimate impacts from minor and major stationary sources, and gives
a starting point for air quality modeling. Versions relevant to our review included the October
2021 "Interim" version, the May 2018 "Draft" version, and various previous versions.58 The
state has frequently revised the Colorado Modeling Guideline over the years, and this report does
not attempt to fully recapitulate the document and its different versions, but rather to note the
provisions of this guidance document that were relevant to our review.
The Colorado Modeling Guideline explains that while the state's regulations require APCD to
perform a preliminary analysis discussing the "impact, if any" from the proposed project, the
guidance does not call for modeling in every case.59 APCD must provide a demonstration of
compliance with all the air quality standards, which for minor sources can be completed using
quantitative (i.e., modeling) or qualitative (i.e., non-modeling) methods, as appropriate.60
According to the Colorado Modeling Guideline, if air quality modeling is not submitted with a
permit application, the APCD will decide whether air quality modeling is warranted to assess the
impact of the facility and demonstrate compliance.61 Table 1 includes the modeling thresholds
and scenarios from the May 2018 version of the Colorado Modeling Guideline. This table is used
to help ACPD staff determine when modeling is warranted for permit actions.
Table 1. [Coloradol Modeling Thresholds
62
If emission rate is less than threshold, a qualitative description of impact
maybe adequate unless a situation warrants modeling.1
Pollutant
Requested Emission Rate from New Source or Facility-Wide Net
Emissions Increase from a Modification
Long-Term (tons per year)
Short-Term (pounds per hour)
Carbon Monoxide
23 pounds per hour
Nitrogen Oxides
40
0.46
Sulfur Dioxide
40
0.46
Particulate Matter < 10 mm
82 pounds per day
57 In this report we refer to this state guidance document as the "Colorado Modeling Guideline," and will identify
any specific reference by its date. In general, the May 2018 version was the one in effect during the permitting
decisions we evaluated.
58 Interim Colorado Modeling Guideline for Air Quality Permits (Oct. 2021). A copy is provided in Appendix B. ;
Colorado Modeling Guideline for Air Quality Permits, APCD, Technical Services Program, Modeling and
Emissions Inventory Unit (Draft, May 2018). A copy is provided in Appendix A (Attachment 5) for reference.
CDPHE has informed Region 8 that the Colorado Modeling Guideline is undergoing another revision.
59 Colorado Modeling Guideline, Draft (May 2018), at 11.
611 Colorado Modeling Guideline, Draft (May 2018), at 11. The Colorado Modeling Guideline does state that new
major sources and major modifications subject to PSD attainment area rules must submit an AQIA that uses air
quality models with the permit application. Id. at 12.
61 Colorado Modeling Guideline, Draft (May 2018), at 12.
62 Reproduced from Colorado Modeling Guideline, Draft (May 2018), at 22.
14
-------
Particulate Matter < 2.5
mm
11 pounds per day
Lead
25 pounds per 3-months
1 Circumstances where source may cause or contribute to a violation of applicable ambient air
quality standards despite being below the thresholds:
(a)
(b)
(c)
(d)
Sources ofS02, N02, PM10, PM2.S, CO, or Pb where a substantial portion of the new or
modified emissions have poor dispersion characteristics (e.g., rain caps, horizontal
stacks, fugitive releases, or building downwash) in close proximity to ambient air at
the site boundary
Sources located in complex terrain (e.g., terrain above stack heights in close proximity
to the source)
Sources located in areas with poor existing air quality
Modification at existing major sources, including grandfathered sources that have
never been modeled before
The state's air quality modelers developed these modeling thresholds through a technical peer
review of APCD's modeling practices, and APCD performed dispersion modeling to help
demonstrate that the thresholds are appropriate.63 The purpose of the thresholds is to identify
new sources and modifications that are expected to have relatively small impacts and thus do not
warrant further air quality analysis. The thresholds were also developed so that the permit
applicants and APCD could avoid situations where the decision to perform modeling takes
longer than actually performing a screening-level modeling analysis would take (screening-level
models can often be run quickly and with minimal cost). According to the Colorado Modeling
Guideline, "modeling will usually be warranted" if the requested emission rate or the facility-
wide net emissions increase is above either threshold for any pollutant. On the other hand, if the
requested rate and emissions increase are below both of the thresholds for each pollutant, then
modeling will not usually be warranted unless one of the situations outlined in the footnotes of
Table 1 applies.64
The APCD may also perform screening-level analyses to demonstrate compliance with the
applicable standards. If the screening-level analysis shows potential predicted violations of the
standards, the applicant may be required to perform refined modeling to demonstrate compliance
with all air quality standards. When air quality modeling is required, the Colorado Modeling
Guideline outlines comprehensive procedures and methods for conducting air quality modeling.
If the air quality model predicts exceedances or violations of standards, the applicant has the
opportunity to adjust the facility's allowable emissions, operating hours, physical barriers to
preclude public access, source parameters, and source configurations in order to demonstrate that
the predicted impact is in compliance with all applicable standards.65
The Colorado Modeling Guideline (May 2018) describes a methodical process for deciding
whether to require air quality modeling in connection with a Minor NSR permit application.
According to the complaint, however, the state has not applied the thresholds or other
considerations in the Colorado Modeling Guideline. Instead, the complaint states that APCD
63 The EPA has not reviewed or approved this analysis and associated thresholds completed by APCD.
64 Colorado Modeling Guideline, Draft (May 2018), at 22.
65 Colorado Modeling Guideline, Draft (May 2018), at 22, 31-32.
15
-------
followed a policy of indiscriminately approving Minor NSR permits without requiring modeling.
Specifically, on September 20, 2010, APCD issued PS Memo 10-01, a guidance document that
purports to relieve "all sources with NO2 or SO2 emissions of 40 tons per year or less.. .from
demonstrating compliance with the corresponding 1-hr [NAAQS]."66 The complaint claims that
APCD has interpreted this policy as "superseding] statutory/regulatory requirements," and has
applied it to issue permits without requiring or considering modeling for a facility with SO2
emissions of "about 22 tpy,"67 for a mine facility application previously rejected for failure to
demonstrate compliance with PM2.5 limits,68 and at least eleven other facilities.69 CDPHE's
Director of Environmental Health and Protection Operations and the Deputy Director of
Regulatory Affairs have informed the EPA that PS Memo 10-01 has been retired and is no longer
implemented by APCD. Nonetheless, according to some CDPHE staff, this document may have
recently been followed in permitting actions.
Further, according to the complaint, on March 15, 2021, APCD management informed air
quality modeling staff that "it was now the policy of the APCD that NAAQS compliance in air
permit applications will no longer be verified" for several NAAQS.70 The complaint explains
this statement by citing an internal APCD email stating that modeling would be performed only
for sources meeting specific emissions thresholds:
• 40 tpy for NO2 and SO2
• 82 lbs/ day for PM10
• 5 tpy for PM2.5
• 23 lbs/hr for CO
• 25 lbs/3-mo for Pb71
Under this approach, the "short-term standards" in the Colorado Modeling Guideline would not
be used, and that document would be removed from the state website.72
The complaint characterizes this 2021 directive as "new policy set by the office which prohibits
employees from conducting legally mandated modelling of air pollution from new stationary
sources of air pollution."73 We note, however, that the complaint's allegation concerning the
2021 policy is broadly similar to its allegations about PS Memo 10-01, and about the state's
66 PEER complaint at 8.
67 PEER complaint at 8 and PEER complaint at 7-9.
68 PEER complaint at 8-9 and PEER complaint at 10-14.
69 PEER complaint at 13-14.
70 PEER complaint at 5.
71 PEER complaint at 5. Apparently, despite the letter's categorical "will no longer be verified" statement, under this
new policy modeling would be performed for sources exceeding the emissions thresholds. Also, the PEER complaint
notes that the March 15, 2021 policy would allow "exceptions" to this approach based on APCD management
approval.
72 The PEER complaint does not identify the version, but this statement apparently refers to the May 2018 draft
version of the Colorado Modeling Guideline.
73 PEER complaint at 1.
16
-------
practices since 2010. In both cases, the complaint charges that APCD has been issuing minor
source permits without conducting legally required modeling to verify NAAQS compliance.74
The following section summarizes the relevant components of APCD's organization and
permitting procedures.
APCD Organization and Permitting Practices
The Colorado Modeling Guideline also outlines a process for conducting an AQIA, and the
APCD staff roles and responsibilities for air quality construction permits. The Colorado
Modeling Guideline says permit applications with air quality modeling should generally be sent
to the Stationary Sources Program (SSP), where a permit engineer will process the permit
application. Any information associated with the air quality modeling should be sent to the
Modeling and Emissions Inventory Unit (MEIU) under the Technical Services Program (TSP)
for review. APCD staff are advised to contact the applicant when oversights, errors, or
questionable assumptions or methods are found during the review process. The applicants are
also encouraged to communicate (e.g., phone conversations, email, etc.) with APCD staff to
address questions or resolve issues. APCD may conduct additional air quality modeling to
address issues found during the review. All unresolved or significant modeling-related issues are
recommended to be documented in writing and added to the permit record.
The Colorado Modeling Guideline does not explain how modeling-related issues should be
resolved within APCD or in the permit conditions to ensure that the proposed project will not
cause violations of applicable standards. The state's Director of Environmental Health and
Protection Operations and its Deputy Director of Regulatory Affairs described the state's typical
process to us. According to them, APCD's practice is to resolve all issues before approving and
issuing the permit, but that the methods for doing so vary. Details of the resolution may be
documented in a final report that is part of the permit record, or may have been discussed among
staff and management without being documented in the permit record. Further, the materials
included in the permit record vary depending on the complexity of the particular permitting
action, but typically include the final permit itself, the permit application (submitted in the form
of an Air Pollution Emission Notice, or APEN), and a preliminary analysis that provides
technical support for the permit issuance. The permit record may also include additional
application materials, including technical materials submitted by the applicant in support of
requested limits, and key written communications (typically emails) between the applicant and
the permit engineer. For permits that undergo public comment, the draft permit, comments
received, and all written responses to public comments should be included in the record. For
permits that require modeling per the Colorado Modeling Guideline, both the modeling
submission from the applicant and the final modeling report should be included as part of the
public record. Finally, the state officials explained that APCD has had some process issues in the
past with getting modeling reports uploaded into the electronic records management system,
though these reports were and should continue to be available to the public.
74 "The March 15, 2021 policy reduces to writing a longstanding informal practice of the APCD of approving
permits at all costs even where modeling has demonstrated NAAQS violations." PEER complaint at 8.
17
-------
EPA Region 8 Review of Individual Permits and Associated Air Quality
Analyses
To assess whether CDPHE is implementing its NSR permit program in a manner that fails to
protect air quality or is contrary to legal requirements, EPA Region 8 reviewed a subset of the 11
permits listed in the complaint: (1) Bighorn Pad, Jackson County; (2) Cripple Creek & Victor
Gold Mine, Teller County; (3) Martin Marietta Materials Highway 34 Facility, Weld County;
and (4) Williams Willow Creek Gas Plant, Rio Blanco County. These permits were selected
based on the size of the permit records, ranging from a complex multi-year project that contained
multiple separate permit actions (e.g., Cripple Creek & Victor Gold Mine) to one that contained
few actions (e.g., Martin Marietta Materials Highway 34 Facility). In addition, the records for
each of these permits included air quality modeling reports that allowed us to assess the validity
and reasonableness of APCD's approach to modeling. All of the selected permits are also minor
source construction permits. Additional permits beyond these four were not reviewed because
our findings remained consistent across these four permits; we believe that evaluating additional
permits would not significantly add to our findings.
Figure 1 displays the locations of each of the 11 facilities referenced in the complaint as holding
inappropriately issued permits, including the four facilities that we have focused on in this report.
The Denver Metro/North Front Range (DMNFR) Ozone Nonattainment Area is also displayed
for context.
Figure 1. Permitted Sources Identified in Complaint
SB
j Fort I
Collins
Columbine National'
t forestt |
Vl v A
Q3
40
Roo^vetl
\ 'National
iMFbrest ^
TS
rn«
5
eeley
Evans
9'
v
Legend
1. Williams Willow Creek Gas Plant
2. ColoWyo Coal Mine
3. Bighorn Pad
4. Martin Marietta Materials - Highway 34
5. JBS Swift Beef Company
6. Asphalt Specialties - Central Plant
7. Asphalt Specialties - 62nd Ave
8. Martin Marietta Materials - Monaghan
9. Aggregate Industries
10. Cripple Creek & Victor Gold Mine
11. McCormick Asphalt
| Ozone Nonattainment Area
^> 'Gunnison ''lI.l'"
^.'National ' \v-
>C-
• X\!' /ore?''
Sj|i
¦ fw® ,
•u"TotesP' r
Longni out
'Boulder
: . ' - \
¦9'
' 1 Vv^Arapahcwoi^dvpltV
''I Natfcnal^U^a
- ' . 1
, Arvada-
I.ik«v{wd t
Eng|eivoo<
J ' , w'
ibmspSr^^ v J & \V '
Ww-
i PikeWaitlj
V
^Forest
V,l-sh.nnSU.51"°""0A g
S Denver
9 Aurora
¦mvood Vm»
Paikci
Col orad pti ngs
10
^Security-Widefield
\ Fountain
COLORADO
11
18
-------
Several months after submitting the original complaint, PEER and their clients submitted
additional information related to ozone attainment planning and to other sources and permits
with which they expressed concerns. This additional information was provided during our review
and was not a part of the OIG's request to EPA Region 8. We did not address this additional
information and will review it outside the context of this report as appropriate.
A summary of the results from our review for each of the four permits follows. Our findings in
each case, including our conclusions concerning the records for these permits, are based on the
permit files provided to us by CDPHE.
Bighorn Pad in Jackson County, Colorado (Permit Number 16JA1055, issued November
14, 2018)
The Bighorn Pad permit application proposed to construct an eight-well oil and gas production
facility that would include crude oil storage tanks, produced water storage tanks, an enclosed
flare, and enclosed combustion devices. Crude oil and produced water would be trucked out of
the facility using tank trucks and submerged fill loading. The application was originally
submitted in December 2016, and the AQIA report was submitted in February 2017. APCD
deemed the application incomplete in February 2017, upon which determination the source re-
submitted the application in March 2017. The application was subsequently deemed complete in
May 2017. The permit record provided to EPA Region 8 contained APENs, permit applications,
issued Construction Permits, inspection reports, an Impact Analysis Review Report written by
the MEIU (dated May 2017), and email communications among APCD staff and the applicant
discussing the air quality modeling analysis.
The applicant submitted modeling analyses for the 1-hour and 8-hour CO NAAQS that used the
EPA's preferred AERMOD modeling system. However, based on the review by the APCD
modeling unit (the MEIU), modeling for other pollutants should have been completed for the
proposed project. In particular, the applicant did not submit a modeling analysis to demonstrate
compliance with the 1-hour NO2 NAAQS, presumably because the proposed facility-wide NO2
emission rate was below the threshold in PS Memo 10-01, which states that modeling for 1-hour
NO2 is only required if the emission rate is greater than 40 tpy. As a result, the MEIU conducted
additional air quality modeling for the 1-hour NO2 NAAQS because the facility's short-term
NO2 emission rate exceeded the 0.46 lbs/hr threshold in the Colorado Modeling Guideline. The
MEIU also prepared an Impact Analysis Review Report based on that modeling. This report
reflects several conservative assumptions - that is, assumptions that would tend to ensure that the
impacts would be over-predicted rather than under-predicted by the model. In particular, the
MEIU analysis assumed a 100% NO to NO2 conversion, instead of incorporating a refined NO2
tiered approach, to treat NO2 chemistry. In addition, the MEIU analysis used the maximum
predicted concentrations for comparison to the NAAQS, instead of using the model results in the
form of the standard. The report also noted updates that the MEIU made to the modeling
configuration to correct potential issues, including changes to the stack parameters for the
enclosed flares and several nearby sources, and the receptor network.
The results of the MEIU analysis showed that the proposed project could potentially exceed the
1-hour NO2 NAAQS. As a result, the MEIU recommended a refined quantitative impact analysis
that uses more representative modeling tools, configurations, and input information to
demonstrate compliance with the 1-hour NO2NAAQS. According to the MEIU, it was unknown
19
-------
whether this facility would cause or contribute to a violation of this standard without such an
analysis. With the exception of NO2, the Impact Analysis Review Report notes that the
compliance demonstration is valid and would support permit conditions that aligned with the
model configuration assumptions, including restricted stack parameters (i.e., emission rates,
stack heights, and stack diameters). The permit conditions in the four permits associated with the
Bighorn Pad project and provided to EPA Region 8 did contain restrictions on the stack
parameters.
The documentation provided to EPA Region 8 did not include information on whether a refined
quantitative impact analysis was completed, as the MEIU had recommended, to demonstrate
compliance with the 1-hour NO2NAAQS for this permit application. Further, the permit record
did not include information that outlines all the modeling details used in the final air quality
modeling. For instance, the permit record did not include a modeling protocol, or an additional or
revised Impact Analysis Review Report with information on the model system and
configuration, input assumptions, supporting data and maps for the stack parameters,
meteorology, and model domain. According to the discussion with the Environmental Health and
Protection Director during the November 18, 2021 meeting with Region 8, the permit engineer
for this permit application did not see that the MEIU had identified potential problems with the
NO2NAAQS, and therefore issued the permit. Due to the lack of information related to the
concerns with the potential 1-hour NO2 impacts, the permit record did not demonstrate that the
proposed project and the allowable emission increases from the proposed source, in conjunction
with all other applicable emission increases or reductions (including secondary emissions),
would not cause violations of the ambient air quality standards.
As noted above, based on the information provided to EPA Region 8 it appears that the MEIU
took a conservative approach to the air quality modeling analyses. While this is not an
inappropriate approach, and it appears to follow EPA guidance, the use of other acceptable
options and assumptions could produce a more refined and representative analysis that could
generate predicted concentrations that are significantly different or in compliance with the
standards. In other words, a more refined analysis might have demonstrated that the proposed
project would not cause a NAAQS violation. But APCD did not provide any such refined air
quality modeling analyses, or any analysis of other supplemental air quality data in the permit
record for the Bighorn Pad that would demonstrate compliance with all air quality standards and
support the permit conditions. Further, documentation included in the permit record did not
explain or resolve the violations that were predicted by MEIU using the conservative approach
described above. The failure to demonstrate compliance and include this demonstration in the
permit record is also inconsistent with the Colorado Modeling Guideline, which states that all
unresolved or significant modeling-related issues should be documented in writing and added to
the permit record. While we offered the opportunity for CDPHE to provide additional
information on the Bighorn Pad permit action during our call with CDPHE management on
November 18, 2021, no additional information was provided beyond the explanation that the
permit writer overlooked identified problems with the modeling. Accordingly, we have
concluded that APCD did not document in the permit record adequate support for their decision
to issue this permit.
20
-------
Cripple Creek & Victor Gold Mine in Teller County, Colorado (Permit Number 98TE0545,
issued November 17, 2020)
The Cripple Creek & Victor Gold Mine involves the mining of ore and overburden; backfilling
and overburden placement; a valley leach facility; and an adsorption, desorption, and recovery
plant. The permit record contains actions and documents from 1984 through 2021. The first
construction permit submitted to the EPA for review was originally issued in 1984. After the
issuance of the first permit, multiple permit modifications occurred, including modifications in
2008, 2011, 2016, and 2020. The permit record contained APENs, permit applications, issued
Construction Permits, inspection reports, public comments and CDPHE's response to those
comments, operation and maintenance plans, stack test results, office memos, Impact Analysis
Review Reports written by the MEIU, timelines outlining the permit air quality modeling
activities, and email communication between APCD staff and the applicant discussing the air
quality modeling analysis.
According to the information APCD provided to EPA Region 8, the applicant submitted initial
air quality modeling analyses in 2008 and worked with APCD on the model configuration and
input assumptions through 2010. APCD then received the official permit application from the
applicant in March 2013. The applicable pollutants that required air quality modeling for this
proposed project included CO, NO2, PM10, PM2.5, and SO2. APCD found several issues with the
original application and attempted to work with the applicant to resolve the issues between
March 2013 and September 2014. In particular, the MEIU found several issues with the air
quality modeling analysis, such as concerns with the model scenarios, meteorological data,
terrain data, background concentrations, the receptor network, stack parameters and project
layout, and the characterization of nearby sources. As a result, the MEIU considered the air
quality modeling analysis flawed and the results invalid for comparison to the NAAQS. The
MEIU also noted that compliance with applicable air quality standards had not been
demonstrated for the proposed project, given the nature of the deficiencies and the magnitude of
predicted concentrations. The original application was withdrawn and replaced with a new
application in March 2015. Again, MEIU found several issues with the air quality modeling
analysis in the new application, including concerns with the ambient air boundary, representation
of the pits and blasting operations in the model, the meteorological dataset, terrain data,
NO2/NOX in-stack ratios, emission rates, and the receptor network. Between 2015 and June
2018, APCD attempted to work with the applicant on addressing the issues, resulting in several
application re-submissions that included updated air quality modeling or other new information.
In June 2018, the MEIU determined that the proposed project would cause modeled violations of
the 1-hr NO2 NAAQS, and that corrections to the air quality modeling analysis would be
required before it could be determined whether the proposed project would cause or contribute to
modeled violations of the remaining applicable air quality standards. Additional information was
provided to APCD to support the assumptions used in the updated air quality modeling
completed by the applicant, but a Modeling Review Report written by MEIU in January 2019
nonetheless stated that the proposed project could cause or contribute to modeled violations of
the NAAQS. This report also noted that the APCD Director did not agree with the MEIU's
concerns related to the updated modeling completed by the applicant, and that the Director
instead accepted the applicant's air quality modeling analysis purporting to show that the
proposed project would be in compliance with all the applicable NAAQS. In APCD's response
to public comments on this permit action, APCD also noted that the facility has agreed to
21
-------
additional monitoring and committed to additional equipment controls (e.g., dust controls).
Based on this company's agreement to install additional monitoring and dust controls, APCD
issued the permit in 2020.
During the permitting of the original project submitted in the 2013 application, APCD
determined to split off a smaller portion of the larger individual project and permit this smaller
activity before finalizing the permit for the larger project. Information submitted by the
complaint indicates that this was done with the understanding that the smaller project would not
by itself contribute to a NAAQS exceedance or violation and that the issued permit would
require Cripple Creek and Victor Gold Mine to undertake modeling as part of a permit condition
in the issued permit. The permit record supplied by CDPHE does not contain the permit
application for this project, or the final issued permit, but does contain several enforcement
discretion approvals indicating that the facility would not be able to meet this permit condition in
the required time period. These enforcement discretion approvals started in 2016 and continued
until 2018 when it was stated that the facility had submitted the required modeling to satisfy the
permit condition. The final permit for the broader project was issued in 2020. The permit record
did not provide documentation to determine whether this approval of the smaller project, or the
decision to split it off from the larger project, was consistent with the requirements of the
Colorado SIP. Further, the permit record did not provide information to indicate whether the
process to split the larger project into smaller projects ensured that the proper permitting and
compliance requirements were applied to this permit action. In particular, failure to adequately
justify splitting a project into more than one permit has the potential to lead to circumvention of
major source permit program requirements. This approach could also result in the air quality
impacts of a larger project not being sufficiently assessed during the permitting process. During
the January 5, 2022 conference call with PEER and their clients, the clients also identified a
concern about CDPHE "breaking up projects," where emitting units that were part of a large
project were being individually permitted.
Due to the lack of information addressing the concerns that MEIU raised with the air quality
analysis and the decisions to split the project into multiple permits, the permit record did not
demonstrate that the proposed project and the allowable emission increases from the proposed
source would not cause or contribute to violations for all the ambient air quality standards. While
we offered the opportunity for CDPHE to provide additional information on this permit action
during our call with CDPHE management on November 18, 2021, no additional information was
provided to address our concerns. As a result, APCD did not sufficiently document their decision
to issue this permit in the permit record.
One of the additional submittals we received from PEER after the original complaint (see
discussion in Introduction, above) was a memorandum with 35 exhibits concerning the Cripple
Creek and Victor Gold Mine, provided in August 2021. As we noted above, we explained to the
OIG that we intended to focus on determining whether CDPHE issued any Minor NSR permits
that failed to properly demonstrate compliance with the NAAQS, or otherwise violated any SIP
requirements for their Minor NSR permit program, and we did not intend to expand our
investigation to include issues newly raised in such additional submissions. Also, the additional
information in the memorandum and exhibits appeared to be consistent with the information we
reviewed in the permit records, and we concluded that a more thorough review would not be
22
-------
likely to change our technical findings outlined in this report. The OIG generally agreed with this
planned approach to the review.
Accordingly, we relied primarily on the permit documents we received from CDPHE in July
2021 for our review, and did not thoroughly review the new information PEER submitted to us
and the OIG in August 2021. Based on our limited review of this information, however, we had
concerns about allegations that CDPHE managers pressured staff to approve analyses which staff
felt were based on false information, and potential conflicts of interest on the part of a CDPHE
manager involved in the permit actions related to this facility. We viewed these concerns
primarily as personnel matters that would best be addressed by the Colorado Attorney General's
office and shared this with the OIG in October 2021.
Martin Marietta Materials Highway 34 Facility in Weld County, Colorado (Permit
Number 17WE0578, issued December 17, 2018)
The Martin Marietta Materials Highway 34 Facility permit application proposed a modification
to add an asphalt plant to the facility. The modification included the addition of an asphalt flow
drum mixer, lime silo, asphalt cement tanks, hot oil heaters, and conveyor transfer points and
material handling. The most recent application was received June 23, 2017, and the MEIU
completed an Impact Analysis Review Report in October 2017. The permit record provided to
EPA Region 8 contained APENs, permit applications, issued Construction Permits, inspection
reports, public comments and CDPHE's response to those comments, operation and maintenance
plans, stack test results, an Impact Analysis Review Report written by the MEIU (dated October
2017), a Modeling Protocol written by the applicant's contractor, and email communication
among APCD staff and the applicant discussing the air quality modeling analysis.
The applicant submitted modeling analyses for some pollutants and averaging periods, but not
the NO2 (1-hour) and SO2 (1-hour, 3-hour, and 24-hour) standards. The applicant did not submit
an analysis to demonstrate compliance with the NO2 and SO2 NAAQS, apparently relying on the
guidelines outlined in PS Memo 10-01. However, the MEIU did perform additional air quality
modeling analyses for the NO2 and SO2 NAAQS. The air quality modeling used EPA's
AERMOD modeling system. The Impact Analysis Review Report written by the MEIU notes
that the MEIU made updates to the modeling configuration to correct potential issues, including
updated emission rates for several sources based on calculations and the inclusion of several
additional nearby sources based on the significant impact area. The results of the analysis
showed that the proposed project could potentially cause or contribute to modeled violations of
the 1-hour NO2 NAAQS. With the exception of this NO2 NAAQS, the Impact Analysis Review
Report notes that the compliance demonstration is valid and would support permit conditions
that aligned with the model configuration assumptions, such as restricted hours of operations, the
maintenance of a continuous fencing along the boundary, and construction layout and road
location restrictions. The permit conditions in the three permits associated with the Martin
Marietta Materials Highway 34 Facility project only contained fence line, construction layout,
and road location restrictions. A permit condition requiring restricted hours of operations was not
outlined in the permit record provided to EPA Region 8.
The documentation provided to EPA Region 8 did not include information that outlines all the
modeling details used in the final air quality modeling. For instance, the permit record did not
include a modeling protocol, or an additional or revised Impact Analysis Review Report with
23
-------
information on the model system and configuration, input assumptions, supporting data and
maps for the stack parameters, meteorology, and model domain. Further, the permit record did
not provide information on whether the concerns with the modeled 1-hour NO2NAAQS
violations were addressed for this permit application, or whether additional air quality modeling
was conducted to resolve the predicted violations. According to the discussion with the
Environmental Health and Protection Director during the November 18, 2021 meeting with
Region 8, the permit incorporated all the recommendations regarding facility layout and
operations from the MEIU, but the permit record does not communicate a final disposition of the
NO2 issue. Due to the lack of information related to the concerns with the air quality modeling,
the permit record did not demonstrate that the proposed project and the allowable emission
increases from the proposed source, in conjunction with all other applicable emission increases
or reductions (including secondary emissions), would not cause or contribute to violations for all
the ambient air quality standards.
Based on the information provided to EPA Region 8, it appears that the MEIU took a
conservative approach to the air quality modeling analyses. While this is not an inappropriate
approach, and it appears to follow EPA guidance, the use of other acceptable options and
assumptions could produce a more refined and representative analysis that could generate
predicted concentrations that are significantly different or in compliance with the standards. In
other words, a more refined analysis might have demonstrated that the proposed project would
not cause a NAAQS violation. But APCD did not provide any such refined air quality modeling
analyses, or any analysis of other supplemental air quality data in the permit record for the
Martin Marietta Materials Highway 34 Facility that would demonstrate compliance with all air
quality standards and support the permit conditions. Further, documentation included in the
permit record did not explain or resolve the violations that were predicted by MEIU using the
conservative approach described above. The failure to demonstrate compliance and include this
demonstration in the permit record is also inconsistent with the Colorado Modeling Guideline,
which states that all unresolved or significant modeling-related issues should be documented in
writing and added to the permit record. While we offered the opportunity for CDPHE to provide
additional information on this permit action during our call with CDPHE management on
November 18, 2021, no additional information was provided to address our concerns.
Accordingly, we have concluded that APCD did not document in the permit record adequate
support for their decision to issue this permit.
Williams Willow Creek Gas Plant in Rio Blanco County, Colorado (Permit Number
07RB0616, issued July 7, 2016)
The Williams Willow Creek Gas Plant is equipped to process, compress, dehydrate, and
transport natural gas; emitting units include turbines, thermal oxidizers, flares, storage tanks,
pipe racks, and miscellaneous gas transmission equipment. The permit record contains actions
and documents dating from 2007 and continuing through 2015. The construction permit was
originally issued in May 2008, construction commenced in 2009, and a permit modification was
requested in 2010 to increase the permit limit for SO2 emissions. The permit record provided to
EPA Region 8 contained APENs, permit applications, issued Construction Permits, inspection
reports, public comments and CDPHE's response to those comments, operation and maintenance
plans, stack test results, Impact Analysis Review Reports written by the MEIU (dated December
24
-------
2007 and February 2012), and email communication among APCD staff and the applicant
discussing the air quality modeling analysis.
According to the Impact Analysis Review Report, MEIU concluded that the air quality impact
modeling analysis completed in 2007 was sufficient to demonstrate that the proposed source
would not cause or contribute to a violation of the ambient air quality standards, including PMio
(annual and 24-hour), NO2 (annual), and CO (1-hour and 8-hour). The air quality modeling used
EPA's AERMOD modeling system. The Impact Analysis Review Report also noted that the
additional criteria pollutant emissions were below the modeling thresholds (presumably referring
to state-defined thresholds in effect at that time) for all pollutants and averaging periods. To
ensure compliance with the standards, the Impact Analysis Review Report noted recommended
permit conditions, including restrictions on the stack heights.
Another Impact Analysis Review Report was completed by the MEIU in February 2012 to
summarize an air quality analysis that was completed as a result of issues found in a field
inspection. During the field inspection, the facility was found out of compliance with SO2
emissions limits as a result of emissions from the thermal oxidizer. Consequently, a new APEN
was submitted for this unit, requesting an increase in its SO2 emissions limit. In addition, a
permit limit for PM2.5 emissions was established in this action. Because the total emissions for
these pollutants were above the Colorado Modeling Guideline thresholds, CDPHE requested air
quality modeling to demonstrate compliance with the applicable standards. According to the
Impact Analysis Review Report, the applicant submitted modeling analyses for 3-hour, 24-hour,
and annual SO2 standards, but did not complete modeling to demonstrate compliance with the 1-
hour SO2 and the 24-hour and annual PM2.5 NAAQS. The MEIU conducted additional air quality
modeling analyses for the SO2 and PM2.5 NAAQS. The air quality modeling used EPA's
AERMOD modeling system. The Impact Analysis Review Report written by the MEIU noted
updates that the MEIU made to the modeling configuration to correct potential issues, including
updated emission rates and stack parameters, additional operating scenarios, and updated
elevation data. The Impact Analysis Review Report also noted the use of conservative modeling
methods in the analysis. In particular, the analysis used the maximum predicted concentrations,
instead of using the model results in the form of the standard, for comparison to the NAAQS.
The results of the analysis showed that the proposed project could potentially cause or contribute
to modeled violations of the 1-hour SO2 NAAQS. With the exception of 1-hour SO2, the Impact
Analysis Review Report notes that the compliance demonstration is valid and would support
permit conditions that aligned with the model configuration assumptions, such as restricted hours
of operations, the maintenance of a continuous fencing along the boundary, and stack parameters
(i.e., emission rates and stack height) restrictions. The permit conditions in the two permits
associated with the Williams Willow Creek Gas Plant project only contained stack height and
fence line restrictions. Permit conditions for restricted hours of operations and emission rates
were not outlined in the permit record provided to EPA Region 8. The air quality modeling
analysis included these permit conditions, so it is not clear whether the proposed project would
demonstrate compliance without these conditions.
A Construction Permit Application - Preliminary Analysis Summary report (dated August 2015)
was also provided to EPA Region 8. This report references the air quality modeling analysis that
was documented in the February 2012 Impact Analysis Review Report. While MEIU indicated
potential modeled violations of the 1-hour SO2 NAAQS, the modeled violations were not
25
-------
considered by the APCD. According to this report, the modeled violations were not considered
because the application requesting the modification to increase the SO2 emissions was submitted
before the establishment of the 1-hour SO2 NAAQS standard, and the modification submitted in
2015 requested SO2 emissions that were less than the limits requested in the 2010 application.
For these reasons, APCD did not feel a modeling demonstration was required for the 1-hour SO2
NAAQS standard for this application.
Based on the information provided to EPA Region 8, it is not clear whether the changes in SO2
emissions proposed by this project would indeed cause violations of thel-hour SO2 NAAQS.
Additional information that outlines all the modeling details used in the final air quality
modeling analyses was not included in the permit record. For instance, the permit record did not
include a modeling protocol, or an additional or revised Impact Analysis Review Report with
information on the model system and configuration, input assumptions, supporting data and
maps for the stack parameters, meteorology, and model domain. Updated air quality modeling
would also be potentially needed to address this issue. Due to the lack of information related to
the concerns with the air quality modeling, the permit record did not demonstrate that the
proposed project and the allowable emission increases from the proposed source, in conjunction
with all other applicable emission increases or reductions (including secondary emissions),
would not cause violations for all the ambient air quality standards.
As noted above, based on the information provided to EPA Region 8, it appears that the MEIU
took a conservative approach to the air quality modeling analyses. While this is not an
inappropriate approach, and it appears to follow EPA guidance, the use of other acceptable
options and assumptions could produce a more refined and representative analysis that could
generate predicted concentrations that are significantly different or in compliance with the
standards. In other words, a more refined analysis might have demonstrated that the proposed
project would not cause a NAAQS violation. But APCD did not provide any such refined air
quality modeling analyses, or any analysis of other supplemental air quality data in the permit
record for the Williams Willow Creek Gas Plant that would demonstrate compliance with all air
quality standards and support the permit conditions. Further, documentation included in the
permit record did not explain or resolve the violations that were predicted by MEIU using the
conservative approach described above. The failure to demonstrate compliance and include this
demonstration in the permit record is also inconsistent with the Colorado Modeling Guideline,
which states that all unresolved or significant modeling-related issues should be documented in
writing and added to the permit record. While we offered the opportunity for CDPHE to provide
additional information on this permit action during our call with CDPHE management on
November 18, 2021, no additional information was provided to us. Accordingly, we have
concluded that APCD did not document in the permit record adequate support for their decision
to issue this permit.
Overall Findings
Based on our review of the allegations in the complaint and the records for the four permit
actions described above, we have identified important concerns with Colorado's implementation
of its Minor NSR permit program. In particular, we found inconsistencies between CDPHE's
implementation of the Minor NSR permit program and the requirements of the Colorado SIP.
Colorado's SIP-approved regulations require CDPHE to determine that the source will not cause
26
-------
NAAQS violations, and there must be a record to support that determination. But the state
repeatedly failed to include any record supporting the required demonstration that construction
authorized in Minor NSR permit actions would not cause NAAQS violations.
The permit records we reviewed generally included permit application files, issued permits,
engineering notes, APENs, data supplied by the MEIU, AQIA reports, and review notes. These
records did not include information demonstrating that the proposed project would not cause
violations of the NAAQS, such as acceptable air quality modeling showing that the proposed
emissions would not cause NAAQS violations or other documentation (e.g., operating
modifications or limitations) justifying permit decisions for projects that would otherwise cause
adverse impacts to air quality. Further, as discussed above, in several instances CDPHE
apparently disregarded modeling results indicating that the proposed projects could cause a
violation, and instead relied on the thresholds outlined in PS Memo 10-01 to support the basis to
approve the proposed minor source permits.
As discussed above, CDPHE guidelines and policies generated inconsistencies in the
implementation of the state's Minor NSR permit program and the Minor NSR requirements in
the Colorado SIP. The guidelines and policies that caused the inconsistencies included PS Memo
10-01, the March 2021 communications, and the Colorado Modeling Guideline. The lack of a
process for communicating and coordinating among the APCD NSR permit groups may have
also contributed to the issues found in our review. For instance, two of the APCD groups
involved with NSR permits (i.e., the SSP and MEIU groups) implemented separate guidelines
that used different criteria to determine when air quality modeling would be required for minor
source permit applications. This resulted in the APCD modeling unit (i.e., MEIU) completing
additional air quality modeling analyses for pollutants and averaging periods not addressed by
the applicants because the applicants used the PS Memo 10-01 to exclude air quality modeling
analyses for some of the pollutants and averaging periods. The APCD modeling unit conducted
the additional modeling based on the Colorado Modeling Guideline without consideration of PS
Memo 10-01. While the additional modeling provides a more comprehensive AQIA, it illustrates
conflicts within APCD on the validity and implementation of the relevant guidelines and policies
(i.e., PS Memo 10-01 and the Colorado Modeling Guideline). Further, some of the air quality
modeling results provided by the APCD modeling unit that demonstrated potential violations of
the NAAQS and the need for specific permit conditions to avoid violations were dismissed or
overlooked by the Environmental Health and Protection Director and permitting staff (i.e., SSP).
The Colorado SIP-approved regulations (Regulation 3, Part B, III.D.l.c) and the Colorado
Modeling Guideline require APCD to determine that the source will not cause NAAQS
violations before issuing an NSR permit. That determination must have some basis, which must
be included in the record. The permit records from the four facilities selected for our review
included air quality modeling that predicted modeled NAAQS violations for one or more criteria
pollutants, but these predicted violations were not addressed by refined or more representative air
quality modeling, use of supplemental air quality data to support the permit conditions, or permit
conditions that would address the predicted violations and which should have been documented
in the permit records.
27
-------
In most cases, the predicted 1-hour NO2 concentrations exceeded the standard. The APCD
modeling unit also identified issues with the assumptions and configurations used in the
modeling for the selected permit applications, which makes it difficult to determine, without
additional modeling, whether predicted violations are indeed resulting from the proposed
projects. Our review also found that the APCD modeling unit has taken a conservative modeling
approach that may result in higher predicted air quality impacts. While this is not an
inappropriate approach, and it appears to follow EPA guidance, the use of other acceptable and
representative options and assumptions could produce more refined and representative analysis
that could generate predicted concentrations that are significantly different or in compliance with
the standards. In other words, more refined analyses might have demonstrated that the proposed
projects would not cause NAAQS violations. But APCD did not provide any such refined air
quality modeling analyses, or any analyses of other supplemental air quality data in the permit
records to demonstrate compliance with all air quality standards and support the permit
conditions. Further, documentation included in the permit records did not explain or resolve the
predicted violations.
Our review also found that one of these facilities included permit actions that split a complex,
multi-year project into smaller, individual permit actions. In that case, the permit record did not
include information to explain the basis for splitting the proposed project into multiple permit
actions. Further, the permit record did not demonstrate whether the Minor NSR program
requirements were adequately applied in these permit actions instead of the PSD program
requirements. This distinction between PSD and Minor NSR programs is important because the
PSD program requires additional and more extensive air quality analyses.
Overall, the permits reviewed by EPA Region 8 lacked relevant supporting information and did
not provide a sufficient demonstration to ensure that the approved permit actions would not
cause a NAAQS violation. EPA Region 8 did request additional information from CDPHE to
address our findings and demonstrate that these permit actions at the four facilities would not
cause violations of the NAAQS. However, CDPHE was unable to supply pertinent supporting
information. CDPHE management and staff acknowledged that the permit records may lack
documentation demonstrating how potential modeled NAAQS violations were addressed before
issuing the permits, and the analyses or other data to support the permit conditions and decisions
for the final permits.
Agency Recommendations
EPA Region 8 recommends that CDPHE take the following measures to address our findings:
1. Ensure that all future Minor NSR permit records are complete and include sufficient
documentation to support permit conditions and contain analyses that demonstrate that
the permit conditions will not cause NAAQS violations. As needed, undertake additional
qualitative or quantitative air quality analyses to demonstrate that the permit conditions
comply with the NAAQS and include these analyses in the permit record.
2. For the 11 permit records identified in the complaint, amend permit actions as appropriate
by conducting refined modeling, incorporating additional/revised permit conditions,
and/or potentially including post construction ambient air monitoring. For the Cripple
Creek and Victor Gold Mine, this would also include providing an explanation
28
-------
demonstrating that the various projects were not under aggregated when determining the
projects qualified for Minor NSR permits. In light of the high level of public interest on
these issues, the EPA notes that any revisions to these permits, including the permit
record, would benefit from public notice and comment, even if state rules would not so
require.
3. Maintain complete public records for all NSR permits. Records should be retained for 10
years after the permit expires, is terminated, or withdrawn, or longer if required under
state law.
4. Improve communication and coordination among the APCD groups that work on NSR
permits.
5. Ensure that complex and multi-year projects are covered under the proper major or minor
source program permit requirements based on appropriate aggregation considerations and
have adequate air quality impact analyses AQIAs, and that any decisions to permit
individual units are justified and adequately documented in the permit records.
6. During quarterly meetings between APCD and EPA Region 8 permitting managers,
provide status reports on newly issued Minor NSR permits that describe the AQIAs and
permit condition determinations.
We are seeking a written response from CDPHE by October 21, 2022, that addresses our
findings and recommendations; we may revisit some of our above recommendations after
reviewing the response.
We also want to acknowledge that the CDPHE Environmental Health and Protection Director
informed EPA Region 8 that CDPHE has begun enacting improvements, including taking steps
to ensure adequate communication between the APCD NSR permit groups and resolution of
issues raised, among them the creation of a guidance document requiring the groups to work
together in determining when and what modeling is required for minor source permits. CDPHE
is also taking steps to assess and improve the workflow between these groups.
Further, CDPHE recently completed an expert stakeholder group initiative that undertook a
comprehensive assessment of how Colorado should approach modeling of minor sources to
ensure a cohesive, justified, and timely approach to modeling and permitting minor sources that
is consistent with state and federal regulations. CDPHE transmitted the workgroup's final report
to the EPA on May 11, 2022. The report has consensus and majority supported
recommendations. CDPHE scheduled public listening sessions in June and July 2022 and
publicly noticed the recommendations in the report and the Interim Colorado Modeling
Guideline to seek public and stakeholder feedback. As noted in the final report, EPA Region 8
staff was informally involved in the group's discussions.
In addition, EPA Region 8 plans to work with the EPA's Office of Air Quality Planning and
Standards (OAQPS) on methods that could assist APCD in implementing their NSR permit
program more efficiently and effectively. Further, EPA Region 8 plans to work with and provide
technical support to APCD to ensure that the approved permits are protecting air quality and
align with the Colorado SIP. If needed, EPA Region 8 is prepared to work with CDPHE on
29
-------
updating the SIP to ensure that the requirements are properly implemented and followed by the
state.
In July 2021, the OIG issued a report75 directing the EPA to conduct more oversight of synthetic
minor source permitting. Recommendations included:
• Develop and implement an oversight plan for synthetic-minor-source permitting;
• Update EPA's practical enforceability guidance;
• Assess EPA studies and other relevant information on enclosed combustion devices
during its next review of applicable regulations to determine whether revisions to
monitoring, record-keeping, and reporting requirements are needed;
• Develop and issue new guidance that includes key EPA expectations for synthetic-minor-
source permitting; and
• Take steps to assure that all states adhere to public participation requirements for
synthetic-minor permits.
In response to the OIG report, the EPA Office of Air and Radiation (OAR) plans to take the
recommended actions. This work is expected to be completed within the next two years and is
likely to result in improvements to EPA guidance related to the NSR permit program. In
addition, it will provide an opportunity to engage with Colorado and other states that identify
minimum emissions thresholds below which public notice requirements in 40 CFR 51.161 have
not been consistently applied. We expect that this work will also assist CDPHE with improving
their NSR permit program.
Finally, we note that the EPA has recently proposed to reclassify ("bump up") the DMNFR
ozone nonattainment area to severe nonattainment for the 2008 ozone NAAQS. This EPA action
should help address the potential stakeholder concerns related to the CDPHE permitting practices
and the DMNFR ozone nonattainment area. If this action is finalized as proposed, air permitting
emissions thresholds for construction of new major stationary sources will be reduced from 50
tpy to 25 tpy. This means that more sources will need to comply with the more rigorous major
source permitting requirements. Regardless of past permitting practices, new major source
construction will require offsets and, in some cases, more stringent emissions controls. Offsets
are emission reductions, generally obtained from existing sources located in the vicinity of a
proposed source which must (1) offset the emissions increase from the new source or
modification and (2) provide a net air quality benefit. The purpose for requiring offsetting
emissions decreases is to assist an area in moving towards attainment of the NAAQS, while still
allowing some industrial growth. Offsets are just one aspect of a complex suite of strategies that
are required in attainment planning and included in the Colorado SIP. Appendix C includes a
table describing the history of Denver Metro North Front Range 2008 Ozone Standard
designations along with the corresponding ozone Nonattainment New Source Review
requirements for major sources.
75 OIG Report No. 21-P-0175, July 8, 2021, is available at: https://www.epa.gov/office-inspector-general/report-epa-
should-conduct-more-oversight-synthetic-minor-source-permitting (last accessed on Jan. 5, 2022).
30
-------
APPENDIX A
Memorandum and Attachments from Kevin Collins, Hotline Coordinator, EPA OIG Office of
Counsel and Congressional and Public Affairs, to Deb Thomas, Acting Regional Administrator
EPA Region 8 (May 24, 2021)
-------
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
OFFICE OF
INSPECTOR GENERAL
May 24, 2021
MEMORANDUM
SUBJECT: Office of Inspector General Hotline Complaint No. 2021-0188
FROM: Kevin Collins, Hotline Coordinator
Office of Counsel and Congressional and Public Affairs
u
TO:
Deb Thomas, Acting Regional Administrator
EPA Region 8
On March 30, 2021, Public Employees for Environmental Responsibility, commonly referred to as PEER,
publicized a complaint to the Office of Inspector General that alleges violations of law, mismanagement,
and abuse of authority by the Colorado Department of Public Health and Environment, or CDPHE. The
complaint relates to the CDPHE's preliminary analyses of air impacts during air permitting for new and
modified stationary sources. The PEER complaint states that, on March 15, 2021, the CDPHE set a new
policy that prohibits employees from assessing National Ambient Air Quality Standards compliance for
hourly nitrogen dioxide and sulfur dioxide limits, three-hour standards for sulfur dioxide, and daily
standards for particulate matter smaller than 2.5 micrometers. I have included attachments of the
information that the hotline received from PEER. The complaint asks the OIG to investigate PEER'S
claims.
The OIG has established EPA OIG Hotline Complaint No. 2021-0188 to document this complaint. I am
referring this matter to your office for review and appropriate actions. If you find potential instances of
fraud or misconduct on the part of CDPHE managers or staff, please immediately alert me and provide
supporting information. Otherwise, please provide me with your report so that the OIG may close this
complaint. Any further steps by the OIG, potentially including a follow-up audit, will be taken after
consideration of any Region 8 findings or response.
Please inform me within the next five calendar days that this referral was received. Please update us by
June 24, 2021, on the actions taken by your office to review and address as needed the concerns identified
by PEER. Feel free to contact me if you have any questions. I can be reached by email at
collins.kevin@epa.gov or (202) 566-2476.
Attachments
-------
APPENDIX A - Attachment 1
PEER Press Release (March 30, 2021)
-------
5/17/2021
Colorado Orders Staff to Ignore Air Pollution Violations
Colorado Orders Staff to Ignore Air Pollution Violations
£ peer.org/colorado-orders-staff-to-ignore-air-pollution-violations/
March 30, 2021
For Immediate Release: Tuesday, March 30, 2021
Contact: Kevin Bell (240) 247-0298; Kirsten Stade kstade@peer.org
Air Modelers Request Federal Intervention and Whistleblower
Protection
Washington, DC — The entire air modeling staff for the State of Colorado contend that they
are now directed to issue illegal permits, ignore violations, and refrain from verifying pollution
emissions, according to a complaint filed today through Public Employees for Environmental
Responsibility (PEER). The air pollutants at issue, sulfur dioxide, nitrogen dioxide, and
particulates, drive unhealthy levels of ozone long plaguing Colorado's Front Range.
The air permit modeling team within the Colorado Department of Public Health and
Environment (CDPHE) reveals a March 15, 2021 order from the Air Division Director to
cease
checking for these criteria pollutants in existing and pending permits. Under this new policy,
permit applicants no longer have to demonstrate that their facilities comply with air quality
standards. This order also stymied a review of previous permits showing many violations.
Colorado's Front Range, encompassing the nine counties around Denver, has been in
violation of National Ambient Air Quality Standards, set by the U.S. Environmental Protection
Agency (EPA), since 2012. Last year, Governor Jared Polis acknowledged the public health
concerns by reclassifying the region as being in "serious" nonattainment, an action that
should have resulted in more stringent enforcement, not less.
"Colorado's critical air quality safeguards are being monkey wrenched by high level officials,"
stated PEER Rocky Mountain Director Chandra Rosenthal, a former U.S. Department of
Justice attorney. "How can air pollution standards be enforced if emissions are not verified?"
In their complaint to EPA's Office of Inspector General (IG), the air permit modelers seek a
performance review and audit of the CDPHE Air Division. Colorado's air program operates
under a federal delegation with EPA funding. Deviation from EPA requirements can result in
federal sanctions against the state, ranging from loss of funding to Colorado being stripped
of its air permitting authority altogether.
PEER attorneys also wrote, on behalf of the air modelers, to CDPHE Executive Director Jill
Hunsaker Ryan pointing out that these disclosures of air pollution violations are covered by
the Colorado Whistleblower Protection Act which forbids retaliation for reporting violations of
law. PEER requested to meet with Ms. Hunsaker Ryan.
https://www.printfriendly.eom/p/g/ceRuDu
1/2
-------
5/17/2021
Colorado Orders Staff to Ignore Air Pollution Violations
Notably, CDPHE quietly removed the guidance document for air quality permits from the
agency website in mid-March. Nor was there any public notice of this major policy change.
"Colorado law assures state workers that they can report violations of law such as these
without fear of official reprisal," added Kevin Bell, staff counsel for PEER, a national law firm
and organization dedicated to protecting environmental whistleblowers. "This is a case
where speaking truth to power is the essence of public service."
ii ii ii
a a a
Read the complaint to EPA's Inspector General
See letter to CDPHE Director
View emails and records concerning new "no modelling" policy
See the CDPHE policy that was removed after March 15
https://www.printfriendly.eom/p/g/ceRuDu
2/2
-------
APPENDIX A - Attachment 2
PEER Letter to EPA OIG (March 30, 2021)
-------
A PEER
Public Employees for Environmental Responsibility
962 Wayne Ave • Suite 610 ~ Silver Spring, MD 20910
March 30,2021
U.S. Environmental Protection Agency
Office of Inspector General Sean W. O'Donnell
1200 Pennsylvania Avenue, NW. (2410T)
Washington, DC 20460
RE: Request for Review of Intentional Non-Enforcement of National Ambient Air Quality
Standards by Colorado Department of Public Health & Environment Pursuant to State
Implementation Plan Under the Clean Air Act, 42 U.S.C. § 7401 et seq.
Dear Inspector General O'Donnell:
We, the undersigned, are employees of the Colorado Department of Public Health &
Environment (CDPHE), which is responsible for the State Implementation Plan (SIP)
incorporated by the U.S. Environmental Protection Agency (EPA) in its implementation of the
Clean Air Act (CAA).1 We are employed by CDPHE's Air Pollution Control Division (APCD),
which has specific responsibility for the enforcement of Colorado's duties under the National
Ambient Air Quality Standards (NAAQS) program.2
On March 15, 2021, CDPHE issued a blanket prohibition on air quality modeling staff
from reviewing NAAQS compliance for hourly Nitrogen Dioxide (NO2) and Sulfur Dioxide (SO2)
limits, 3-hour standards for SO2, and daily standards for particulate matter smaller than 2.5
micrometers (PM2.5). The undersigned are the staff-level employees with responsibility for
conducting modeling of emissions from new or modified stationary sources of NAAQS criteria
pollutants and have been ordered by our management to not perform certain of those legally
required duties. We request that your office review the following alleged instances of violations
of law, mismanagement, and abuse of authority by CDPHE, particularly a new policy set by the
office which prohibits employees from conducting legally mandated modelling of air pollution
from new stationary sources of air pollution.
Unlawful acts and omissions by CDPHE include -
142 U.S.C. § 7401 et seq.
2 42 U.S.C. § 7410 (also known as "section 110" for the section of the Clean Air Act it
corresponds to)
-------
• Suppressing information demonstrating that pending permits would lead to modeled
violations of NAAQS;
• Approving air quality permits which violate NAAQS; and
• Ordering modeling staff to ignore modeled NAAQS violations which would conflict with
nonbinding agency guidance documents.
I. Applicable Federal Law
The CAA requires states to establish and implement SIPs for "implementation, maintenance, and
enforcement" of primary and secondary National Ambient Air Quality Standards (NAAQS).3
Each SIP must, among other requirements include:
• "enforceable emission limitations" and "appropriate devices, methods, systems, and
procedures necessary to monitor, compile, and analyze data on ambient air quality"4
• "adequate personnel, funding, and authority under State (and, as appropriate, local) law
to carry out such implementation plan"5
• "provide for the performance of such air quality modeling as the Administrator may
prescribe for the purpose of predicting the effect on ambient air quality of any emissions
of any air pollutant for which the Administrator has established a national ambient air
quality standard."6
In addition to the statute, EPA has promulgated a sophisticated regulatory scheme governing the
requirements of SIPs.7 Specifically:
"Each plan must set forth legally enforceable procedures that enable the State or local agency to
determine whether the construction or modification of a facility, building, structure or
installation, or combination of these will result in... interference with attainment or maintenance
of a national standard in the State in which the proposed source (or modification) is located or in
a neighboring State."8
A SIP must include means for a state to prevent the construction or modification of a new
stationary source if it "will interfere with the attainment or maintenance of a national standard."9
Regarding air quality modeling, the procedures in a SIP "must discuss the air quality data and
the dispersion or other air quality modeling used to meet the requirements of this subpart. All
3 42 U.S.C. § 7410(a)(1).
4 § 7410(a)(2)
s § 7410(a)(2)(E)
6 § 7410(a) (2) (K)
7 40 C.F.R. § 51.160
s § 51.160(a)(2)
9 § 51.160(b)(2)
PEER • 962 Wayne Ave • Suite 610 • Silver Spring, MD 20910 • 202-265-7337 • www.peer.org
-------
applications of air quality modeling involved in this subpart shall be based on the applicable
models, data bases, and other requirements specified in appendix W of this part (Guideline on
Air Quality Models)."
EPA has issued NAAQS for six "criteria pollutants" which are measured by averaging the
concentration of a given pollutant over a given period of time to determine whether it exceeds
primary or secondary standards. Primary standards provide public health protection, including
protecting the health of "sensitive" populations such as asthmatics, children, and the elderly.
Secondary standards provide public welfare protection, including protection against decreased
visibility and damage to animals, crops, vegetation, and buildings.
The NAAQS standards are:10
Pollutant
[links to historical tables of
NAAQS reviews]
Primary/
Secondary
Averaging
Time
Level
Form
Carbon Monoxide (CO)
primary
8 hours
9 ppm
Not to be exceeded more than once per year
1 hour
35 ppm
Lead (Pb)
primary
and
secondary
Rolling 3
month
average
0.15 Mg/m3-111
Not to be exceeded
Nitrosen Dioxide (NO?)
primary
1 hour
100 ppb
98th percentile of 1-hour daily maximum
concentrations, averaged over 3 years
primary
and
secondary
1 year
53 ppb
Annual Mean
Ozone (03)
primary
and
secondary
8 hours
0.070 ppm
Annual fourth-highest daily maximum 8-hour
concentration, averaged over 3 years
Particle Pollution
(EMI
PM2.5
primary
1 year
12.0 pg/m3
annual mean, averaged over 3 years
secondary
1 year
15.0 pg/m3
annual mean, averaged over 3 years
primary
and
secondary
24 hours
35 pg/m3
98th percentile, averaged over 3 years
PMi0
primary
and
secondary
24 hours
150 pg/m3
Not to be exceeded more than once per year
on average over 3 years
Sulfur Dioxide (SO?)
primary
1 hour
75 ppb
99th percentile of 1-hour daily maximum
concentrations, averaged over 3 years
secondary
3 hours
0.5 ppm
Not to be exceeded more than once per year
10 Taken from https://www.epa.gov/criteria-air-pollutants/naaqs-table (last visited Mar. 22,
2021).
PEER • 962 Wayne Ave • Suite 610 • Silver Spring, MD 20910 • 202-265-7337 • www.peer.org
-------
I.A. State Law Incorporated into EPA Air Quality Program
In Colorado, authority under the CAA and Colorado's relevant SIP is delegated to the APCD of
CDPHE.11 As part of its SIP, Colorado's permitting program includes Colorado Air Quality
Control Commission (AQCC) Regulation 3.12 Part D of AQCC Regulation 3 addresses the
requirements for Major Stationary Sources within the NSR/PSD permitting program, and Part B
addresses the requirements for construction permits for all sources, major and minor, unless a
specific reference is made to Part D for major sources.
Regulation 3, Part 6 § III.B.5 states:
Except for applications for sources subject to the requirements of Section VI. of
Part D of this regulation (Prevention of Significant Deterioration), the Division
shall prepare its preliminary analysis within sixty calendar days after receipt of a
complete permit application. The preliminary analysis allows the Division to
determine whether the new source will, at date of commencement of operation,
comply with: ... Any applicable ambient air quality standards and all applicable
regulations. . . . The preliminary analysis shall indicate what impact, if any, the
new source will have (as of the projected date of commencement of operation) on
all areas (attainment, attainment/maintenance, nonattainment, unclassifiable),
within the probable area of influence of the proposed source... When the
preliminary analysis includes modeling, the model used shall be an appropriate
one given the topography, meteorology and other characteristics of the region that
the source will impact. Use of any non-guideline model requires U.S. EPA
approval under Section VIII.A. of Part A of this regulation.
Part B § III.D.l states:
The Division shall grant the permit if it finds that: [...]
(c) The proposed source or activity will not cause an exceedance of any National
Ambient Air Quality Standards;
(d) The source or activity will meet any applicable ambient air quality standards
and all applicable regulations;
Part B § III.F.l states:
If the Division determines that a source cannot comply with the provisions of Part
B, Section III.D., of this regulation, the Division shall issue its written denial of the
permit application stating the reasons for such denial.
Finally, regarding modeling, Part A § VIII.A.1 states:
11 See https://www.epa.gov/caa-permitting/caa-permitting-colorado.
12 5 CCR1001-5 (containing the entire text of Regulation 3); see also 40 C.F.R. § 52.320-354
(discussing Colorado SIP).
PEER • 962 Wayne Ave • Suite 610 • Silver Spring, MD 20910 • 202-265-7337 • www.peer.org
-------
All estimates of ambient concentrations required under this Regulation Number 3
shall be based on the applicable air quality models, databases, and other
requirements generally approved by U.S. EPA and specifically approved by the
Division.
In short, CDPHE is required to verify through air quality modeling that a new major or minor
stationary source, or a modification to an existing source, will not cause or contribute to a NAAQS
exceedance. Issuing an air permit for a minor source without verifying NAAQS compliance
violates the requirements of the New Source Review permitting program outlined in the EPA-
approved SIP: Colorado's AQCC Regulation 3.
II. Violations of Law
CDPHE'S PROPOSED POLICY
On March 15, 2021 CDPHE's Air Pollution Control Division (APCD) Director Garry Kaufman
informed the air quality modeling unit that it was now the policy of the APCD that NAAQS
compliance in air permit applications will no longer be verified for the 1-hr NO2,1-hr SO2, 3-hr
SO2, and 24-hr PM2.5 NAAQS.
This policy was set during a meeting that day between Director Kaufman, APCD Chief Strategy
Officer Robyn Wille, APCD Technical Services Program Manager Gordon Pierce, and APCD
Modeling and Emissions Inventory Unit Manager Emmett Malone.13
Instructions were provided via email from Gordon Pierce to Emmett Malone.14 These instructions
were conveyed verbally by Emmett Malone to CDPHE's two active permit modelers, DeVondria
Reynolds and Bradley Rink, on March 16,2021.
The instructions relayed to Malone by Pierce state:
Per the meeting we had today 3/15/21 with Garry and Robyn, Garry specifically
stated that, effective immediately, the short-term thresholds in the Modeling
Guideline will not be used and that modeling only be performed using the
following thresholds:
• 40 tpy for NO2 and SO2
• 82 lbs/day forPMio
• 5 tpy for PM2.5
• 23 lbs/ hr for CO
• 25 lbs/ 3-mo for Pb
13 Emmett Malone is the current supervisor of complainants DeVondria Reynolds and Bradley
Rink and had been the supervisor of Rosendo Majano until his transfer to another unit in 2020.
14 See App'x 1-4.
PEER • 962 Wayne Ave • Suite 610 • Silver Spring, MD 20910 • 202-265-7337 • www.peer.org
-------
He will allow exceptions based on agreement between the modelers and permit
engineers, and his specific approval.
Garry also requested that the Modeling Guideline also be removed from the
website, pending further discussions and revisions. I will ask Ivan to pull it.
Attached is a copy of the current version that is on the website.15
That day, the modeling guidance document on CDPHE's Air QUALITY MODELING GUIDANCE FOR
Permits website was removed, and a note was added next to the removed hyperlink stating
"UNDER REVISION."16
Air Quality
Department of Public Health b Environment
AirQuality Home CDPHE Home Contact Us AirQuality Menu »
Air Quality Modeling Guidance for Permits
*
Queried Sources of Air Pollution Map
This webpage presents current air quality permit modeling guidance for estimating Impacts from stationary sources of air pollution.
For general permitting questions, contact the Stationary Sources Program. Contact monitoring or modeling staff In the Technical Services
Program for monitoring or modeling questions, respectively.
Completeness Checklist for Modeling Submittals
FAQs on Air Quality Modeling Data and Techniques
Colorado Modeling Guideline for Air Quality Permits ^
• Colorado Modeling Guideline for Air Quality Permits UNDER REVISION
• Meteorological Determinations and Application of Dispersion Models
Supplemental Modeling Guidance, Forms, and Information ^
A discrepancy between what was discussed verbally in the March 15 meeting and Pierce's email
was noted by Malone, who requested clarification:
One question that came up and if you or Garry could respond I would appreciate
it. What I took from the meeting was that we are not to address the short term
standards. The way you worded your email is slightly different, are we to address
the short term standards?17
15 App'x 3-4.
16 See https:/ / www.colorado.gov/airqualitv/permits.aspx (last accessed Mar. 22,2021). A
version of the guidelines as they existed prior to March 15,2021 is attached as Appendix B.
17 App'x 3. The line "2 years, 7 months, and 15 days" in Malone's email is believed to refer to the
length of time before he can retire.
Colorado Modeling Guideline
Supplemental Guidance, Forms, and Information
Meteorological Data
Maps, Figures, Tables
Federal Modeling Support
On-Llne Training
PEER • 962 Wayne Ave • Suite 610 • Silver Spring MD 20910 • 202-265-7337 • www.peer.org
-------
Pierce responded unequivocally:
No, we are not to address the short-term standards... "the short-term thresholds in
the Modeling Guideline will not be used."18
Reynolds and Rink forwarded these instructions via email to CDPHE's third permit modeler,
Rosendo Majano, who four months earlier transferred to a different unit, but continues to provide
training and guidance to the active permit modelers. The response by the modelers was
unequivocal: this is a violation of the law.19
It is not a great leap to demonstrate that a state which purposefully disregards NAAQS standards
is not complying with those NAAQS standards. This new policy violates Colorado AQCC
Regulation 3, Part B §III.B.520 and §III.D.l,21 and it would also violate Colorado's EPA-approved
SIP and NSR Minor Source Permitting Program established per Title I, Part A, § 110 of the CAA22
and 40 CFR § 51.160, as it exempts new stationary sources and modifications from demonstrating
compliance with short term NAAQS by simply not considering them. This violation of the law is
also contributing directly to chronic health problems, premature deaths, and severe injury to the
environment by permitting ever more dangerous emissions.
Through this policy, CDPHE is attempting a reversal of federalism: voiding in Colorado the three
most recent and most stringent ambient air standards promulgated by EPA, the 1-hr NO2,231-hr
SO2,24 and 24-hr PM2.5 NAAQS;25 all of which were put in place to protect human health and the
environment after a thorough regulatory review and reaffirmed in 2018 and 2020.
In short, under this policy, permit applicants in Colorado no longer have to demonstrate that their
proposed sources or modifications comply with these standards in order to get an air permit from
CDPHE. In fact, numerous permits have already been issued over our objections that the new or
modified sources did not comply with one or more NAAQS.26
It is also worth noting that NO2 is one of the two main precursors of ozone, and that the Denver-
North Front Range ozone nonattainment area has deteriorated from a designation of "Marginal"
nonattainment in 2012 for the 2008 ozone NAAQS, to a designation of "Serious" nonattainment
18 Id. (emphasis and ellipses in original).
19 Both Reynolds and Rink sent separate emails declaring their intent to blow the whistle about
this new policy. App'x 24-28.
20 Requiring that any preliminary analysis demonstrate that the proposed new source "will, at
date of commencement of operation, comply with [... a]ny applicable ambient air quality
standards and all applicable regulations."
21 requiring that permits be granted only upon a finding that the proposed new source would
not cause exceedance of any NAAQS and would comply with applicable air quality standards.
22 42 U.S.C. § 7410.
» 75 Fed. Reg. 6,474 (Feb 9,2010).
24 75 Fed. Reg. 35,520 0un 22,2010).
25 78 Fed. Reg. 3,085 (Mar. 18,2013).
26 Described infra.
PEER • 962 Wayne Ave • Suite 610 • Silver Spring, MD 20910 • 202-265-7337 • www.peer.org
-------
in 2020, and soon to deteriorate to a designation of "Severe." This new policy will enable
unfettered growth of NO2 sources in the nonattainment area and elsewhere in the state at
precisely the time when CDPHE should be making efforts to achieve exactly the opposite.
In addition to violating the aforementioned regulatory provisions, this policy is also being
implemented essentially in secret. This policy was not brought to the Air Quality Control
Commission, has not received public scrutiny, was implemented internally without notice, and
as such violates Colorado's Administrative Procedure Act.27
Pattern of Unlawful Conduct
The March 15, 2021 policy reduces to writing a longstanding informal practice of the APCD of
approving permits at all costs even where modeling has demonstrated NAAQS violations.
Evidence demonstrating this history of unlawful conduct in implementing EPA programs under
the CAA is too plentiful to reproduce in full here and can be provided upon request, but certain
illustrative examples deserve note.
On September 20, 2010 CDPHE instituted a minimum threshold policy under PS Memo 10-0128
in which all sources with NO2 or SO2 emissions of 40 tons per year or less are exempted from
demonstrating compliance with the corresponding 1-hr National Ambient Air Quality Standards
(NAAQS), even if they actually exceed that standard many times over. This exemption enabled
unfettered growth of NO2 emissions in the ozone nonattainment area from sources which have
been able to obtain air pollution permits without any verification or modeling of their impacts on
air quality or public health. The memo, despite being labelled as nonbinding guidance, was
interpreted to "supersede statutory/regulatory requirements" of the CAA in April 2011 and has
been applied as such ever since.29
In December 2011, Memo 10-01 was applied when the modeling unit informed the permitting
unit that a facility that they were preparing to approve a permit for would emit more than ten
times the hourly allowed rate of SO2 emissions. The permitting unit told the modeler that they
did not request SO2 modeling because the facility would emit less than 40 tons per year of SO2 so
the hourly limits would not be considered per the memo. Ultimately management decided that
despite being ten times over the hourly limit, a demonstration of modeling compliance would
not be needed.30
In November 2012, a hard rock mine submitted an application in which it claimed to be exempt
from demonstrating compliance with PM2.5 emissions. The same application had been rejected in
2009 for failure to properly demonstrate modeled compliance with particulate matter emissions
27 C.R.S. § 24-4-106.
28 App'x 5-6, also available at
https://environmentalrecords.colorado.gov/HPRMWebDrawer/RecordView/901901.
29 App'x 50-52 (2011); 53-54 (2013)
so App'x 7-9.
PEER • 962 Wayne Ave • Suite 610 • Silver Spring, MD 20910 • 202-265-7337 • www.peer.org
-------
limits, and this time was approved with no substantial changes based on a management decision
that the facility should not have to demonstrate compliance.31
The next month, APCD Scientist and then-Modeling Unit Supervisor, Chuck Machovec
documented a conversation with APCD Deputy Director Garry Kaufman, who later become
Division Director, his current position. Machovec states that
[Kaufman] also mentioned that at some point he wants to have a meeting to reach
an understanding on "policy" versus "technical" aspects of modeling. I told him
that as long as their [sic] is unwritten policy whereby management asks staff
verbally to not follow language in the rule, there will be issues. Any requests from
management for staff to not follow the requirements of the rule should be in
writing.32
In July 2015, CDPHE employee Doris Jung described the process which led to Memo 10-01's
formation and the motivation for CDPHE's management to suppress inconvenient modeling
reports. In short, the permitting unit did not believe that their applicants could demonstrate that
they complied with the NAAQS but wanted to issue permits regardless:
In addition to discussing PM2.5 in this meeting, we (modelers) were told to raise
the modeling thresholds because compliance with NAAQS could not be
demonstrated or they would just issue permits without considering NAAQS
compliance.
We (modelers) argued that their approach would be inconsistent with regulatory
requirements and raising thresholds when NAAQS are more stringent was not
technically justified but ultimately we were ignored. Consequently, since the
modelers wouldn't cooperate, permitting came out with he [sic] 40 tpy threshold
for 1-hr NO2 and SO2 NAAQS on September 20,2010.
The circumstances and the permitting personnel involved have not changed since
the 2010 meeting/ decision. While I agree concerns for NAAQS attainment remain,
there is nothing that indicates to me that permitting is willing to change their
practice.33
This explanation was given in response to concerns raised by modeling staff that CDPHE was
issuing permits to facilities that had small annual emissions rates but still caused a
disproportionately severe impacts on local air quality, particularly regarding the hourly
NAAQS.34
Multiple times over the years CDPHE made it clear to the modeling unit that Memo 10-01, despite
being issued informally as "guidance," was controlling and, at least for CDPHE, even overrode
si App'x 10-11; 12-14.
32 App'x 15.
33 App'x 16-20.
34 App'x 16.
PEER • 962 Wayne Ave • Suite 610 • Silver Spring, MD 20910 • 202-265-7337 • www.peer.org
-------
state and federal law regarding NAAQS compliance.35 An incomplete list of permits issued
despite modeled violations of one or more NAAQS is included at the end of this section.
This policy has had dangerous consequences for the culture of compliance among industry in
Colorado. In a March 2020 email exchange between a CDPHE permit engineer and an industry
consultant, the consultant explains their reasons for not submitting a NAAQS compliance
demonstration for the 24-hr PM2.5 standard, stating:
I've received guidance from multiple non-oil & gas permit engineers in the last
year or so that the agency only looks at the 5 tpy threshold for PM2.5, unless there
are special circumstances (e.g., high expectation of comments from nearby
residents).36
The response from CDPHE's permit engineer confirms this understanding and paints a chilling
portrait of official complicity with this unlawful conduct:
After consulting with some higher-ups, I agree that modeling for the 24-hour PM2.5
standard will not be required in this case. Unfortunately, we don't have a specific
written policy I can point to that says "you can always ignore the daily PM2.5
modeling threshold," but we do it on more of a case-by-case basis.37
CDPHE is enforcing compliance with the 24-hr PM2.5 NAAQS only when there is a possibility
that the permit will be subject to public scrutiny, and finds it "unfortunate" that the legal
threshold exists at all. Public scrutiny is the only criteria that CDPHE is concerned about when
enforcing, or abstaining from, the NAAQS, not legal obligations or concern for public health and
the environment.
Despite attempts by environmental and conservation groups in Colorado to alert CDPHE
leadership and management of the gravity of the situation, the agency has only buried its head
deeper in the sand to avoid evidence that they are issuing unlawful permits.
A close review of permit applications approved by CDPHE would demonstrate that a very large
number of those permits were issued without any attempt to determine whether a proposed new
facility or modification would interfere with the attainment or maintenance of hourly and daily
NAAQS. In many cases, permits were issued with actual knowledge that the proposed facilities
may have modeled NAAQS violations, or even with knowledge of clear and conclusive evidence
of such violations.
35 App'x 21 ("PS memo 10-01 makes it very clear that we cannot require [an applicant] to do
modeling to the 1 hour NOx standard since emissions of NOx will be less than 40 tons per
year.").
36 App'x 22.
37 Id.
PEER • 962 Wayne Ave • Suite 610 • Silver Spring, MD 20910 • 202-265-7337 • www.peer.org
-------
Visual Example of the Cumulative Impacts of Exempting "Minor" Facilities from
Regulation
The consequences of PS Memo 10-01 and the culture of permitting at all costs at the CDPHE
APCD are best illustrated by example. The Martin Marietta Highway 34 Facility (the "Highway
34 Facility"), an asphalt and concrete plant, was permitted by the Air Pollution Control Division
(APCD) inside the ozone non-attainment area despite exhibiting modeled violations of the 1-hr
NO2 NAAQS.38 It was exempted from demonstrating compliance with this standard based on
the PS Memo 10-01.
About three miles from the Highway 34 Facility, the Extraction Oil & Gas Johnson's Corner
Production Facility (the "Johnson's Corner Facility") started operation in March of 2018 and
applied for a permit based on Colorado Reg. 3 § II.D.l.iii. This facility operated without a permit
for a long time and, like the Highway 34 Facility, has been exempted from demonstrating
compliance with the NAAQS based on the PS Memo 10-01.39 In addition to the violations of the
1-hr NO2 standard already caused by the nearby Highway 34 Facility, it is likely that the Johnson's
Corner facility is amplifying and making worse an already existing violation with negative
implications for air quality and public health.
These are only two small facilities located in an area saturated with hundreds of other facilities,
all inside the ozone non-attainment area, and jointly emitting thousands of tons per year of one
of the main ozone precursors, NO2. Using the CDPHE mapping site we can see the huge number
of facilities operating.40
The graphic below shows that in a radius of 25 kilometers around the Johnson's Corner Facility
and Highway 34 Facility, there are 777 sources emitting a total of 5009.01 tons per year of NO2.
Almost half of that total —48%—comes from facilities with individual emissions below the level
set in PS Memo 10-01 and the general permit's threshold of 40 tons per year. This means that these
facilities have likely been permitted without any assessment of the cumulative impact of their
NO2 emissions.
38 Permit Numbers 16WE0688, 16WE0689. AIRS ID 123-9E8B. NKA Rock & Rail Highway 34
Facility.
39 CDPHE email response of 10/16/19 to PEER public record request. Interoffice email states that
there is no CDPHE air permit modeling report for Johnson's Corner. AIRS ID 123-9FCC.
40 https://www.colorado.gov/airquality/ss map wm.aspx
PEER • 962 Wayne Ave • Suite 610 • Silver Spring, MD 20910 • 202-265-7337 • www.peer.org
-------
Earthstar Geographies i Esri. HERE, Garmin. NGA, USGS
Pointer Longitude:t03.50356
Pointer latitude: 39.98068
Designation AredS (check box}: AMA = Attainment/Maintenance Area NAA = Kon-Attainmant Area
Q PM10 AMA E CO AMA D031hrAMA 12! 03 8hr NAA (0 08 ppm / 0 075 ppm 03 NAAOS)
Other (check box): p] Class I ~ Class II" fj CDPHE Monitoring Sites Clear Graphics
"Class II areas as they existed on August 7 1977 that have Class I SQ2 Increment protection Class II boundaries are approximate
Print Help io print more detailed version of table beiow, dick here> Print Table To pnnt map above, dick here* Print *•
777 sources of NOx total 5009.01 tpy w''in 25 km of 4464207.00 N, 512212.00 m E (UTM Zone 13, NAD 1983)
Had these sources been permitted in compliance with regulatory requirements, the
corresponding facilities would have been required to implement control measures, use better
technology, or downsize their projects. This is turn would have the final effect of reducing the
NO2 emissions to comply with the 1-hr NO2 NAAQS and in turn reducing the formation of ozone.
It is also possible that the area is so saturated that no more NO2 sources would have been
permitted at all, which would have at least slowed down the continuing deterioration of the
ozone problem. Determining the actual status of the air quality in that area is part of CDPHE's
job, but that duty has been neglected for years, leading to the current crisis and the downgrading
of Colorado's NAAQS non-attainment status to "serious" from "moderate."
PEER • 962 Wayne Ave • Suite 610 • Silver Spring, MD 20910 • 202-265-7337 • www.peer.org
-------
Unlawfully Issued Permits
An incomplete list of permits issued with potential or actual modeled NAAQS violations
includes:
• Williams Willow Creek Gas Plant, in Rio Blanco County, CO. AIRS ID 103-0360. Issued
with modeled violations of the 1-hr SO2 NAAQS.41
• ColoWyo Coal Mine, in Moffat and Rio Blanco Counties, CO. AIRS ID 103-0327. Issued
with 1-hr NO2 NAAQS modeled violations.42
• Asphalt Specialties Central Plant in Weld County, CO. AIRS ID 123-A023. Issued with
modeled violations of the 1-hr N02 NAAQS and with suspected violations of the PM10,
and both PM2.5 NAAQS.43
• Martin Marietta Monaghan Facility in Adams County, CO. AIRS 001-2193. Issued with
modeled violations of the 1-hr NO2 NAAQS.44
• Aggregate Industries Oxford Asphalt Plant in Arapahoe County, CO. AIRS ID 005-0116.
Issued with modeled violations of the 1-hr N02 NAAQS.45
• Cripple Creek & Victor Gold Mine in Teller County, CO. AIRS ID 119-001. Issued with
modeled violations of the 1-hr NO2 NAAQS and with suspected violations of the PM10,
and both PM2.5 NAAQS. A CDPHE modeler was ordered to falsify data in a modeling
report regarding this facility to ensure that no modeled violation would be reported.46
The modeler explained as much in writing to his supervisor, Emmett Malone:
"As you know, officially there aren't any modeled violations at the CC&V Mine.
That's because of the 01/14/1947 and 01/28/1948 emails from Gordon Pierce
requesting to remove the concentration exceeding the NAAQS from the report and
to replace them with a value that was lower and that was based on incorrect data.
Therefore officially the highest modeled concentration is of 187.7 ug/m3 (99.77
ppb). The NAAQS is of 100 ppb. Reality however, is very different.
[...]
41 See also App'x 16-17 (discussing how Williams Willow Creek Plant was approved despite
"clear potential" to violate the 1-hr SO2 and NO2 NAAQS).
42 App'x 55-57.
43 App'x 29 ("the facility submitted a NAAQS compliance demonstration only for carbon
monoxide despite having emissions rates of PM10, PM2.5, SO2 and NO*, that exceed the
corresponding short-term modeling thresholds").
44 App'x 30-33 (correspondence between modeling and permitting units regarding this facility).
45 App'x 34-35 (correspondence indicating violations of NAAQS and use of outdated software
by applicant to demonstrate modeling compliance).
46 App'x 36.
47 App'x 40-46
4® App'x 47-49
PEER • 962 Wayne Ave • Suite 610 • Silver Spring, MD 20910 • 202-265-7337 • www.peer.org
-------
Those results above beg the question, if the Division is now acknowledging the
modeled NAAQS violations, why would the permit be issued?
Wouldn't that create the exact same situation as the ColoWyo Mine permit that
was challenged in court for being issued with a NAAQS modeled violation?
• Bighorn Pad in Jackson County, CO. AIRS ID 057-0051. Issued with modeled violations of
the 1-hr NO2 NAAQS.
• Martin Marietta Materials Highway 34 Facility in Weld County, CO. Issued with modeled
violations of the 1-hr NO2 NAAQS.
• JBS Swift Beef Company in Weld County, CO. AIRS ID 123-0018. Issued with modeled
violations of the 1-hr NO2 NAAQS.
• McCormick Asphalt Plant in Yuma County, CO. Issued with modeled violation of the 1-
hr NO2 and 1-hr SO2 NAAQS.49
• Asphalt Specialties Co., Inc. / 62nd Ave. Asphalt Plant in Adams County, CO. Issued with
modeled violation s of the 1-hr NO2 NAAQS and high potential for violation of PM10 and
PM2.5 NAAQS.50
These cases are a small sample demonstrating CDPHE's ongoing failure to enforce its SIP and its
minor source permitting program. The March 15, 2021 policy prohibiting modeling of these
NAAQS is the latest and most concrete instance of a pattern of unlawful conduct which is directly
responsible for Colorado's precipitous decline in air quality in the last decade. Consequently, we
request that EPAOIG take immediate action to investigate the CDPHE NAAQS program and
permits issued under it.
Sincerely,
Rosendo Majano
DeVondria Reynolds
Bradley Rink
March 30,2021
Counsel:
Kevin H. Bell, Staff Counsel
Chandra Rosenthal, Rocky Mountain Director
Public Employees for Environmental Responsibility
49 App'x 37-39 (correspondence).
50 App'x 37-39 (correspondence).
PEER • 962 Wayne Ave • Suite 610 • Silver Spring, MD 20910 • 202-265-7337 • www.peer.org
-------
962 Wayne Ave., Ste 610
Silver Spring, MD 20910
(202) 265-7337
PEER • 962 Wayne Ave • Suite 610 • Silver Spring, MD 20910 • 202-265-7337 • www.peer.org
-------
APPENDIX A - Attachment 3
PEER Letter to CDPHE (March 30, 2021)
-------
A PEER
Public Employees for Environmental Responsibility
962 Wayne Ave • Suite 610 • Silver Spring, MD 20910
March 30, 2021
Dear Director Hunsaker-Ryan,
We are writing to you on behalf of three whistleblowers who currently work at the Colorado
Department of Health and Environment (CDPHE) Air Pollution Control Division: Bradley Rink,
DeVondria Reynolds, and Rosendo Majano. These brave individuals are standing up to
management that has asked them to take actions that would violate the Clean Air Act (CAA) and
the Colorado Air Pollution Prevention and Control Act and approve air permits without critical
review of compliance with National Ambient Air Quality Standards (NAAQS). These three
employees represent the entire air permit modeling team for the state of Colorado. Public
Employees for Environmental Responsibility (PEER) requests that they receive protection under the
Colorado Whistleblower Protection Act and that the state act to bring the air permitting program
into compliance with the Clean Air Act and the Colorado Air Pollution Prevention and Control Act.
On March 15, 2021, CDPHE's Air Division Director banned air quality modeling staff from
conducting any review of NAAQS compliance whatsoever for hourly nitrogen dioxide (NO2) and
Sulfur Dioxide (SO2), 3-hour standards for SO2, and daily standards for particulate matter smaller
than 2.5 micrometers (PM2.5). Review of previous permit applications revealed that a shocking
number of permits have been approved by the Air Division even though air quality modeling
analyses showed NAAQS violations.1 So rather than have a record of these violations in the permit
file, Air Pollution Control Division Director, Garry Kaufman, ordered staff to simply stop modeling
for these standards. See Attachment 1, Email of March 15, 2021. CDPHE's Air Director also
unceremoniously deleted the Air Quality Modeling Guidelines, which stated the legal necessity of
demonstrating compliance with those standards, from their website. See Attachment 2, a 3/17/21
screen shot of the CDPHE website.
1 In many cases, permits were issued despite information indicating that there could be potential modeled NAAQS violations, or even despite clear
and conclusive evidence of such violations. Below is a list with some examples permits issued under those circumstances:
- Williams Willow Creek Gas Plant, in Rio Blanco County, CO. AIRS ID 103-0360. Issued with modeled violations of the 1-hr SO2 NAAQS.
- ColoWyo Coal Mine, in Moffat and Rio Blanco Counties, CO. AIRS ID 103-0327. A lawsuit was filed challenging the issuance of this permit with
l-hrN02 NAAQS modeled violations and then changes were made to reduce emissions which allegedly resolved the violations.
- Asphalt Specialties Central Plant in Weld County, CO. AIRS ID 123-A023. Issued with modeled violations of the 1-hr NO2 NAAQS and with
suspected violations of the PM10, and both PM2.5 NAAQS.
- Martin Marietta Monaghan Facility in Adams County, CO. AIRS 001-2193. Issued with modeled violations of the l-hrN02 NAAQS.
- Aggregate Industries Oxford Asphalt Plant in Arapahoe County, CO. AIRS ID 005-0116. Issued with modeled violations of the l-hrN02 NAAQS.
- Cripple Creek & Victor Gold Mine in Teller County, CO. AIRS ID 119-001. Issued with modeled violations of the l-hrN02 NAAQS and with
suspected violations of the PM10, and both PM2.5 NAAQS.
- Bighorn Pad in Jackson County, CO. AIRS ID 057-0051. Issued with modeled violations of the l-hrN02 NAAQS.
- Martin Marietta Materials Highway 34 Facility in Weld County, CO. Issued with modeled violations of the l-hrN02 NAAQS.
- JBS Swift Beef Company in Weld County, CO. AIRS ID 123-0018. Issued with modeled violations of the l-hrN02 NAAQS.
Some of these cases are recent, others date back to 2012. CDPHE's failure to enforce its minor source permitting program is not an isolated one-time
event, but rather a widespread, chronic situation.
PEER • 962 Wayne Ave • Suite 610 • Silver Spring, MD 20910 • 202-265-7337 • www.peer.org
-------
March 30, 2021
Page 2
The Colorado Front Range, the nine counties around Denver, has a well-known problem with
ground level ozone. The state has-not met federal standards since 2012 due in part to proliferation of
oil and gas facilities in the ozone nonattainment area. Many of these facilities have been permitted
without any NAAQS compliance verification, especially for NO2, which is a known ozone
precursor. In 2020 the Governor's office facilitated a downgrade from "moderate" nonattainment to
"serious" nonattainment. This bold action was a welcome signal to many at the CDPHE that the
state was going to begin the real work of imposing stricter permitting requirements for polluting
entities to finally provide Coloradoans with clear, healthy air.
PEER and a coalition of conservation groups have attempted, multiple times, over the past year and
half, to reach out to you and CDPHE about the illegal lack of enforcement of the Environmental
Policy Act (EPA)-approved minor source Air Permitting Program. See Attachment 3 and 4, Letters
from PEER and the conservation community to CDPHE of 2/23/21 & 9/10/20 requesting that the
state address this issue. Since 2010, minor sources with nitrogen oxides and sulfur dioxides air
pollution emissions below a threshold of 40 tons per year are exempted arbitrarily and illegally from
the requirement of demonstrating compliance with the 1-hr NO2 and 1-hr SO2 NAAQS of the
CAA.2
Despite vigorous insistence by their own employees and conservation groups that this policy
(known as PS Memo 10-01) is illegal, the APCD continued to defend the policy of non-
enforcement. CDPHE has allowed air pollution to proliferate in the state.3 There are entrenched
employees that have been working at CDPHE with a seeming mindset to serve only the polluting
industry and they are rubber-stamping air permits.4
On March 15, 2021 management took it to another level. If this action goes unchallenged, it will be
as if the NAAQS are not in effect in Colorado, an egregious decision with long lasting negative
consequences.
Another critical issue is the fact that this new policy is being implemented completely behind the
scenes, in a way that conceals it from EPA and from the public. The only evidence of this policy
that we have seen is an email exchange between two managers referencing a meeting with Director
Kaufman. The latest unilateral decision was made without input from the permit modelers that will
be asked to implement these illegal changes.
Furthermore, without a formal policy, only verbal direction, Director Kaufman and the managers
are pushing accountability to the permit engineers and permit modelers. They will be the individuals
signing permits and air quality modeling analyses with the implication that they have verified
NAAQS compliance, when in fact they have not. The employees that were asked to implement
these illegal actions courageously came forward and requested whistleblower status. We are
2 See this recent letter in the Daily Camera remarking on the illegal policy, https://www.dailycamera.com/2021/03/22/letters-to-the-editor-cu-south-
federal-judges-covid-vaccine-air-pollution-and-covid-libraiy-funding/
3 Because CDPHE has failed to correct illegal SO2 issues, we have reached out to the EPA. PEER, the Center for Biological Diversity, Colorado
Latino Forum, Colorado Sierra Club, and Colorado Jewish Climate Action on Febuary 2,2021 requested that EPA bring CDPHE into compliance
with SO2 standards. See the joint Press Release with details here: https://www.peer.org/call-for-federal-intervention-into-colorado-air-pollution/
4 See Attachment 5, letter from PEER, the Center for Biological Diversity and the Information Network for Responsible Mining opposing the
proposed air permit for the Newmont Mine- Cripple Creek and Victor. Director Kaufman, former legal representative for the Mine, many times, over
many years granted at least six "enforcement discretion" letters allowing the mine to operate without meeting legal requirements.
PEER • 962 Wayne Ave • Suite 610 • Silver Spring, MD 20910 • 202-265-7337 • www.peer.org
-------
March 30, 2021
Page 3
reaching out to you so that you will be a champion for those employees on the front line and to the
public.5
Under the Colorado Whistleblower Act, CRS 24-50.5-103 "an employee who wishes to disclose
information under protection of this article is obligated make a good-faith effort to provide to his or
her supervisor ... the information to be disclosed prior to the time of its disclosure." These
employees have raised the issue with their supervisors' numerous times in emails and in
conversations over the last ten years, since the adoption of Memo 10-01. These employees that
refuse to violate the CAA and the Colorado Air Pollution Prevention and Control Act and have
disclosed violations of law for the public benefit, deserve protection from retaliation. They have put
their careers at risk. We request that you support them in their stand for compliance.
The minor source air permitting program deserves immediate attention and reform. PEER and these
employees will be filing the attached complaint with the EPA Office of Inspector General
denouncing the State of Colorado's failure to enforce the EPA-approved State Implementation Plan
and the Minor Source Permitting Program. See Attachment 6.
We request that you require the state to meet all Clean Air Act and Colorado Air Pollution
Prevention and Control Act requirements and prohibit the implementation of CDPHE's new policy
that would effectively and illegally revoke several federal air quality standards.
Please be in touch. Let us know if you would like us to forward you any of the many attachments
that correspond to the letters and complaint. We can provide you with independent modeling on
facilities that that correspond to our letters. We can provide you with emails that document the
multiple protected disclosures that the Air Modelers have made to management. We would like to
meet with you to discuss the issues they are raising.
Sincerely,
Chandra Rosenthal
Rocky Mountain Director
Kevin Bell
Staff Counsel
Attachments:
1: Email of 3/15/21 ordering Air Modelers to not model criteria pollutants.
2: Screen shot of Rule Removal from CDPHE website.
3: Letter of 9/10/20 requesting CDPHE stop using illegal Memo 10-01 to justify air modeling
thresholds, with an independently modeled facility demonstrating how the Memo is illegal. Please
see the graphic which shows the number of facilities that are permitted under the minor source
program and demonstrates the cumulative impact.
4: Letter of 2/21/21 requesting CDPHE stop using Memo 10-01, with a second independently
modeled facility that again demonstrates how Memo 10-01 is illegal and also EPA guidance
5 Another whistleblower from the air quality monitoring office made similar allegations in 2019. He left the agency with the perception that the entire
management of the APCD and CDPHE are totally uninterested in actually regulating
emissions: https://www.coloradoindependent.com/2019/04/16/colorado-oil-gas-emissions-regulation/
PEER • 962 Wayne Ave • Suite 610 • Silver Spring, MD 20910 • 202-265-7337 • www.peer.org
-------
March 30, 2021
Page 4
requiring compliance to pollution standards in the face of the cumulative impacts from minor
sources.
5. Letter of 8/19/19 to CDPHE opposing the Cripple Creek and Victor Mine air permit.
6. EPA OIG complaint
PEER • 962 Wayne Ave • Suite 610 • Silver Spring, MD 20910 • 202-265-7337 • www.peer.org
-------
APPENDIX A - Attachment 4
CDPHE Emails (dates vary)
-------
State.co.us Executive Branch Mail - Fwd: Modeling meeting outcome
EPAOIG Appendix 1
STATE OF
COLORADO
Fwd: Modeling meeting outcome
1 message
Rink - CDPHE, Bradley|
To: Rosendo Majanol
Tue, Mar 16.2021 at 7:21 AM
Bradley Rink
Technical Services Program, Air Pollution Control Division
fyi - we've been told to not model short term standards
blr
Forwarded message —
From: Malone - CDPHE, EmmettH
Date: Tue, Mar 16,2021 at 5:18
Subject Fwd: Modeling meeting outcome
To: Bradley Rink -¦
Hi,
We can talk about ttiis at 7
Emmett Malone
Supervisor
Modeling and Emissions Inventory Unit
Technical Services Program
Air Pollution Control Division
Colorado Department of Public Health and Environment
APCD-TS-B1
4300 Cherry Creek Drive South
"Are you curious about ground-level ozone in Colorado? Visit our ozone webpage to team more."
Forwarded message -
From: Pierce - CDPHE, Gordonl
Date: Mon, Mar 15, 2021 at 4:30 T
Subject Modeling meeting outcome
To: Emmett Malone|
Emmett.
Per the meeting we had today 3/15/21 with Garry and Robyn, Garry specifically stated that, effective immediately, the short-term thresholds in the Modeling Guideline will not be
used and that modeling only be performed using he following thresholds:
• 40 tpy for N02 and S02
• 82 lbs/day for PM10
. 5 tpy for PM2.5
• 23 Ibs/hr for CO
• 25 lbs/3-mo for Pb
He will allow exceptions based on agreement between the modelers and permit engineers, and his specific approval.
Garry also requested that the Modeling Guideline also be removed from the website, pending fur her discussions and revisions. I will ask Ivan to pull it. Attached is a copy of he
current version hat is on the website.
Thanks,
Gordon
Gordon Pierce
Program Manager
Technical Services Program
IS
I COLO R ADO
Air Pollution Control Division
I Dtoartmentctf ftilcHe^hiitnrt'on'rKnl
4300 Cherry Creek Drive South, Denver, CO 80246-1530
Aro you curious about ground-levol ozono in Colorado? Visit our ozone wobpag© to loam more.
EPAOIG Appendix 1
II
-------
State.co.us Executive Branch Mail - Fwd: Modeling meeting outcome
EPAOIG Appendix 2
—1 Modeling Guideline_downloaded 20210315.pdf
^ 5097K
EPAOIG Appendix 2
https://mail.google.com/mail/u/0?ik=1144611 beO&view=pt&search=all&permthid=thread-f%3A1694393218126394881 %7Cmsg-f%3A16943949556147.
-------
State.co.us Executive Branch Mail - Fwd: Modeling meeting outcome
EPAOIG Appendix 3
STATE OF
COLORADO
Fwd: Modeling meeting outcome
1 message
Reynolds - CDPHE, DeVondrial
To: Rosendo Majano - CDPHE|
Tue, Mar 16, 2021 at 11:16 AM
- Forwarded message -
From: Malone - CDPHE, Emmettl
Date: Tue, Mar 16,2021 at 1:12 PR
Subject: Fwd: Modeling meeting outcome
To: Bradley Rink - CDPHE |
DeVondria Reynolds - CDPHEl
Gordon response
Emmett Malone
Supervisor
Modeling and Emissions Inventory Unit
Technical Services Program
Air Pollution Control Division
Colorado Department of Public Health and Environment
APCD-TS-B1
4300 Cherry Creek Drive South
Denver, CO 80246-1530
"Are you curious about ground-level ozone in Colorado? Visit our ozone webpage to learn more.'
Forwarded message -
From: Pierce - CDPHE, Gordonl
Date: Tue, Mar 16,2021 at 10:06
Subject: Re: Modeling meeting outcome
To: Malone - CDPHE, Emmettl
No, we are not to address he short-term standards..."the short-term thresholds in the Modeling Guideline will not be used.'
(Based on Brad's email, is he planning something?)
Gordon
On Tue, Mar 16,2021 at 8:04 AM Malone - CDPHE, Emmett|
Hi,
wrote:
Had a discussion with Brad and DeVondria this morning and it went much like I thought it would, badly. I feel like I have stepped back in ime and Doris and Rosendo are
arguing about how this is wrong.
One question that came up and if you or Garry could respond I would appreciate it. What I took from the meeting was that we are not to address the short term standards. The
way you worded your email is slightly different, are we to address the short term standards?
2 years, 7 mon hs, and 15 days.
Emmett Malone
Supervisor
Modeling and Emissions Inventory Unit
Technical Services Program
Air Pollution Control Division
Colorado Department of Public Health and Environment
APCD-TS-B1
4300 Cherry Creek Drive South
Denver, CO 80246-1530
"Are you curious about ground-level ozone in Colorado? Visit our ozone webpage to learn more."
On Mon, Mar 15,2021 at 4:30 PM Pierce - CDPHE, Gordon |
Emmett,
wrote:
Per the meeting we had today 3/15/21 with Garry and Robyn, Garry specifically stated that, effective immediately, the short-term thresholds in the Modeling Guideline will
not be used and that modeling only be performed using the following thresholds:
• 40 tpy for N02 and S02
• 82 lbs/day for PM 10
EPAOIG Appendix 3
h ttps://mail.google.com/mail/u/0?ik=1144611 beO&view=pt&search=all&permthid=thread-f%3A1694393218126394881 %7Cmsg-f%3A16944096843982... 1/2
-------
State.co.us Executive Branch Mail - Fwd: Modeling meeting outcome
. 5tpyfor PM2.5 EPAOIG Appendix 4
• 23 Ibs/hr for CO
• 25 lbs/3-mo for Pb
He will allow exceptions based on agreement between he modelers and permit engineers, and his specific approval.
Garry also requested that the Modeling Guideline also be removed from the website, pending further discussions and revisions. I will ask Ivan to pull it. Attached is a copy of
the current version that is on the website.
Thanks,
Gordon
Gordon Pierce
Program Manager
Technical Services Program
COLORADO
Air Pollution Control Division
I Department of Public Health 6 Environment
4300 Cherry Creek Drive South, Denver, CO 80246-1530
Are you curious about ground-level ozone in Colorado? Visit our ozone webpage to learn more.
Gordon Pierce
Program Manager
Technical Services Program
IS
COLORADO
Air Pollution Control Division
I Department of Public Health & Environment
4300 Cherry Creek Drive South, Denver, CO 80246-1530
Are you curious about ground-level ozone in Colorado? Visit our ozone webpage to learn more.
DeVondria Reynolds, MS
Air Quality Modeler
Modeling and Emissions Inventory Unit
Technical Services Program
IS
COLORADO
Air Pollution Control Division
Department of Public Health & Environment
P 303.692.3249 I F 303.782.5493
4300 Cherry Creek Drive South, Denver, CO 80246-1530
Learn more about COVID-19 https://www.colorado.gov/pacific/cdphe/apcd-covid-19 .
Are you curious about ground-level ozone in Colorado? Visit our ozone webpage to learn more.
EPAOIG Appendix 4
https://mail.google.com/mail/u/0?ik=1144611be0&view=pt&search=all&permthid=thread-f%3A1694393218126394881%7Cmsg-f%3A16944096843982.
-------
EPAOIG Appendix 5
COLORADO DEPARTMENT OF PUBLIC HEALTH AND ENVIRONMENT
Stationary Sources Program / Air Pollution Control Division
INTER-OFFICE COMMUNICATION
PS Memo 10-01
TO: Stationary Sources Staff, Local Agencies, Regulated Community
FROM: Kirsten King and Roland C. Hea
DATE: September 20, 2010
RE: Permit Modeling Requirements for the 1 -Hour NO2 and SO2 NAAQS
The Division is establishing this guidance for use by minor stationary sources of
nitrogen dioxide (NO2) and sulfur dioxide (SO2) in evaluating whether modeling is
necessary for permitting purposes to determine whether a permit applicant's emissions
will comply with the new 1-hour NO2 and/or the new 1-hour SO2 National Ambient Air
Quality Standard (NAAQS). The United States Environmental Protection Agency (EPA)
published implementation guidance on June 28, 2010 and August 23, 2010 regarding
demonstrating compliance with the new standards for Prevention of Significant
Deterioration (PSD) sources.1 The Division finds it useful to publish this supplemental
state guidance to ensure that minor sources are addressed in a manner consistent with the
EPA guidance for PSD sources.
Under federal rules, an ambient air quality impact analysis is required for each
pollutant that a PSD source has the potential to emit in significant amounts. Such
analysis includes modeling. The metric used by EPA to measure significant amounts is
the significant emissions rate (SER). Federal rules currently define the SER for NOx and
SO2 as 40 tons per year (tpy). (40 CFR 52.21(b)(23)(i); 40 CFR 51.166(b)(23)(i)). EPA
recently evaluated and decided to apply on an interim basis the 40 tpy SER to major
source permitting compliance demonstrations for the hourly NO2 and SO2
standards. EPA concludes and states that an ambient air quality impact analysis is not
necessary for PSD sources with projected NO2 or SO2 emissions rates below the SER.
(Wood Memoranda at p. 11 and p.4)
1 See June 28, 2010, Anna Marie Wood, Acting Director, Air Quality Policy Division,
Office of Air Quality Planning and Standards Memorandum "General Guidance for
Implementing the 1-hour NO2 National Ambient Air Quality Standard in Prevention of
Significant Deterioration Permits, Including an Interim 1-hour NO2 Significant Impact Level"
and August 23, 2010 Memorandum "General Guidance for Implementing the 1-hour SO2
National Ambient Air Quality Standard in Prevention of Significant Deterioration Permits,
Including an Interim 1-hour SO2 Significant Impact Level" ("Wood Memoranda").
EPAOIG Appendix 5
-------
EPAOIG Appendix 6
The Division has evaluated EPA's rationale for establishing NO2 and SO2 SERs
for modeling the 1-hour NO2 and SO2 standards. The Wood Memoranda guidance set
forth EPA's reasoning that its SER for SO2 (a pollutant with shorter-term 3-hour and 24-
hour averaging times) is 40 tpy, and, for this pollutant, ambient air quality impact
analyses have not been necessary at levels below the SER. EPA has concluded that this
reasoning applies to the one-hour NO2 and SO2 standards on an interim basis. EPA states
it intends to conduct an evaluation of screening tools available to permitting agencies. In
the interim, it recommends the continued use of the existing SER for NOx and SO2
emissions with respect to the 1-hour NO2 and SO2 standards, and thus ambient air quality
impact analyses are not necessary for either NO2 or SO2 emissions below the 40 tpy SER.
EPA's Wood Memoranda guidance address PSD sources. The Division believes
that the same principles apply to minor sources, in part, to ensure consistency of
treatment in permitting and to ensure that it is not imposing different requirements on
minor sources than those to which PSD sources are subject. The Division is aware of no
factual basis to impose more stringent requirements on minor sources than EPA would
impose on the largest air pollution sources. Therefore, the Division will apply EPA's
SERs for NOx and SCMo the 1-hour NO2 and 1-hour SO2 standards for all stationary
source permitting activities, including determining when ambient air quality impact
analyses are necessary for permitting, pending the consideration of any further guidance
issued by EPA on this subject.
EPAOIG Appendix 6
-------
EPAOIG Appendix 7
Majano, Rosendo
From:
Sent:
To:
Subject:
Attachments:
Majano, Rosendo
Thursday, December 08, 2011 3:02 PM
Money, Carissa D.
RE: Enterprise CTF - S02 emissions
ENTERPRISE CTF - NE TERRAIN ELEVATION PROFILE.jpg
Thanks Carissa for this information. I just reviewed the memo you are referring to in your email, and I have a couple of
questions/comments:
1- The memo seems to apply only to 1-hr N02 and 1-hr S02, but the S02 NAAQS/CAAQS also include the 24-hr/3-
hr averaging periods.
2- Where does this memo leave the short-term modeling thresholds indicated in the CO Modeling Guideline?
Should I assume that the memo supersedes the Modeling Guideline in this matter?
Anyhow, regardless of these questions and comments, there is a strong technical reason why, despite the guidance
provided by the memo, I recommend that the S02 modeling be required in this particular case: There are complex
terrain situations all around the CTF facility, which means that the elevation of the surrounding terrain is higher than the
elevation of the highest stack at CTF. The consequence of this is that the plume of gases will impact the higher terrain
causing higher concentrations than what you would normally see with simple terrain situations. How much higher these
concentrations will be will depend on the distance at which the elevated terrain is located since this will in turn
determine how much the plume of gases will have traveled and how much dispersion of the pollutants will have
occurred. You can better visualize this situation by looking at the attached file. This is showing the terrain elevation
profile in the north-east direction from the CTF site, and as you can see, the elevation of the surrounding terrain starts
being higher than the highest stack at approximately 5 km and from there you can expect higher concentrations. This is
only one direction, and the CTF facility is surrounded by higher terrain in almost every direction. The yellow dots
represent the receptors where the concentrations are being modeled and a big portion of them will be located in
complex terrain.
For the specific case of the CTF facility, you have to consider also the fact that the 1-hr N02 modeling is already showing
violations of the corresponding NAAQS. As I explained to you yesterday, these violations are not necessarily being
caused by CTF, but that is something that is still pending to be determined. And although the S02 emissions are much
lower, the dispersion conditions and terrain characteristics are the same, so if there are high concentrations for 1-hr
N02, the possibility of getting high 1-hr S02 concentrations is real and shouldn't be ignored.
So, I'm not saying that the short term S02 NAAQS will be exceeded, but the fact that the S02 emissions are below 40 tpy
is not by itself sufficient technical evidence to prove the contrary, at least not in the particular case of the CTF facility. In
this case I think that the Division should ask the applicant to submit modeling to demonstrate compliance with the S02
NAAQS.
Let me know your thoughts about this.
Thanks,
Rosendo.
From: Money, Carissa D.
Sent: Thursday, December 08, 2011 12:02 PM
To: Majano, Rosendo
Subject: RE: Enterprise CTF - S02 emissions
l
EPAOIG Appendix 7
-------
Rosendo,
EPAOIG Appendix 8
I did not request modeling of S02 since per PS Memo 10-01 modeling is only required if S02 emissions exceed 40 tpy.
Please let me know if you have additional questions about this determination. (Here is the link to PS memo in case you
have not seen it before)
http://www.cdphe.state.co.us/ap/down/PS10-01.pdf
Thank you!
Carissa
From: Majano, Rosendo
Sent: Thursday, December 08, 2011 11:34 AM
To: Money, Carissa D.
Subject: Enterprise CTF - S02 emissions
Carissa,
While looking at the APENs I noticed that the facility has about 22 tpy of S02 emissions (approximately 5 Ib/hr). This
would exceed the short term modeling threshold of 0.46 Ib/hr and they would need to submit modeling to verify
compliance with the short-term S02 NAAQS, specially with the 1-hr standard.
Do you know if this issue has been previously discussed with the applicant?
Thanks,
Rosendo.
2
EPAOIG Appendix 8
-------
EPAOIG Appendix 9
Majano, Rosendo
From:
Sent:
To:
Subject:
Money, Carissa D.
Tuesday, December 13, 2011 9:35 AM
Majano, Rosendo
RE: Enterprise - CTF and Meeker
Rosendo,
I have not yet heard back about the flares so I will follow up with Enterprise today.
Also, I talked with my supervisor yesterday about the S02 issue (my guidance so no modeling required and your
guidance says that S02 should be modeled). We then followed up with Kirsten and I think she has already talked with
Gordon this morning about the issue. Ultimately I am leaving it up to management to decide but please let me know if
you have any questions/concerns for me.
I also agree with getting back to Enterprise quickly about the modeling issues since it is quite a bit of work for them so
thank you for moving everything along!
Thanks,
Carissa
From: Majano, Rosendo
Sent: Monday, December 12, 2011 5:53 PM
To: Sebesta, Jacob C.; Money, Carissa D.
Subject: Enterprise - CTF and Meeker
Hi Carissa and Jacob,
I just wanted to check if you had a chance to discuss the flare emissions for both facilities with the applicant. I don't
mean to rush you, I just want to know where is the situation at for the purpose of planning my workload.
For CTF, I'm done reviewing the modeling analysis and I'm ready to send Enterprise an email with all the modeling issues
they need to address, so once I get confirmation if the flare's emissions are OK or will change, then I'll send the email.
For Meeker, it looks like the consulting company is already working on the new modeling, so my concern here is that if
the flare's emissions change, they will have to re-do whatever work they have completed, and it will be worse if the re-
submit the modeling and then we have to tell them the repeat it. So I was thinking that it might be a good idea to
contact the consulting company and applicant and tell them to hold the re-submittal until this issue is solved. Jacob,
please let me know if you are OK with doing that.
Also, since both facilities are about 12 kilometers apart, they have to include each other in their respective off-site
emissions inventory. Therefore, I will ask them to do that taking into account all the corrections that are currently being
made. This means that CTF will have to model Meeker as an off-site source with all the recent corrections, and Meeker
will have to do the same with CTF. This request shouldn't be a problem for them considering that we are talking about
the same applicant and same consulting company. The only problem is that I need to notify them of the corrections for
both plants, and I haven't done that yet for CTF. So for that reason it also makes sense to tell them to hold the re-
submittal for Meeker.
Thanks,
Rosendo.-
l
EPAOIG Appendix 9
-------
State.co.us Executive Branch Mail - Re: Bowie Mine
State
of
Colorado
EPAOIG Appendix 10
Re: Bowie Mine
1 message
Jung, Doris
To: Rosendo Majano
Cc: Chuc
Tue, Nov 27, 2012 at 6:46 AM
Rosendo,
I completely disagree with permitting on how they are handling Bowie with regards to PM2.5 and, on a more
basic level, how they are handling impact analyses and NAAQS compliance demonstrations.
I had conveyed to permitting in March of this year that Bowie's modeling should include PM2.5. See attached
PDF.
At this point, permitting has already told the consultant to ignore PM2.5 (per Chuck Pray) and it's not the first
time they have done so. The modeling/impact analysis review report can discuss how PM2.5 has not been
addressed and include any data/results/concerns relating to PM2.5 attainment.
Doris
On Mon, Nov 26, 2012 at 3:24 PM, Rosendo Majano |
Doris,
wrote:
I was reviewing the comments that Jon Torizzo made to the consulting company related to the modeling
submitted prior to 2009, and there were all sort of issues ranging from incorrect emission rates to incorrect
source characterization and definition of ambient air. I don't think this would fall under the explanation
described in the memo of "applications that exceeded the normal permit processing timeframes."
This one did not exceed the normal time frame. The applicant just failed to demonstrate compliance with the
NAAQS through no fault of the Division. In my opinion it is completely wrong to exempt them from PM2.5
modeling based on that 2008 application.
How many years can they keep resubmitting the modeling analysis and still consider it the same application?
Are you completely ruling out the possibility of requesting the applicant to submit PM2.5 modeling?
Rosendo
On Mon, Nov26, 2012 at 12:58 PM, Jung, Doris |
Rosendo,
wrote:
After CC&V lobbied up the management chain in 2010 to relieve themselves of submitting a PM2.5 impact
analysisfbr their current permit, the previous Division director (Paul Tourangeau) decided that applications in-
house prior to 2009 would not need to submit PM2.5 modeling (as reflected in the memo, although his name
does not appear anywhere).
The application originally submitted by 136611 completed due to delays on the
https://mail.google.com/mail/?ui=2&ik=1144611be0&view=pt&search=inbox&th=13b421f8d3d3b9a8
1/2
-------
State.co.us Executive Branch Mail - Re: Bowie Mine
applicant/consultant side. Thus, such uncooperative behavior is rewarded by nof asseslinjpi$ai nlrieht: with
PM2.5 standards.
I agree that EPA's surrogate policy does not apply but, according to permitting, they are relying on the
attached memo. The memo does not reference the surrogate policy. By the date of the memo, case law had
ruled/clarified the limited use of the surrogate policy (a demonstration needs to be made). At this time, there
is no technical reason for not assessing PM2.5.
As with the 1-hr N02 and S02 standards, the modeling/impact analysis review report can discuss how
PM2.5 has not been addressed and include any data/results/concerns relating to PM2.5 attainment.
I believe Bowie is the last of the pre-2009 applications.
Doris
Forwarded message
From: Rosendo Majano|
Date: Wed, Nov 21, 2012 at 3:03 PM
Subject: Bowie Mine
To: Doris Jung |
Cc: Chuck Machovec I
Doris,
I started reviewing the Bowie Mine application - received at the SSP on 02/29/2012 - and noticed that the
applicant is claiming to be exempt from conducting PM2.5 modeling (please see the attached file). I talked
to Chuck Pray about this and he confirmed it. According to him, this is exactly the same application that
had been submitted several years ago, with no changes, and therefore the PM10 surrogate policy would still
apply.
However, from what I have seen, it is not that the application has been sitting on someone's desk waiting for
review, it looks like the modeling was rejected and resubmitted several times. I dont know the details but I
dont think this could be considered as the same application submitted prior to 2009.
Given that you have more knowledge of what happened with the previous submittals, you might be in a better
position to determine if the PM10 surrogate policy would still apply. I don't think it does and I would request
PM2.5 modeling to be submitted.
Let me know your thoughts about this.
Thanks,
Rosendo.
RE_ Bowie Mining.pdf
J 67K
EPAOIG Appendix 11
https://mail.google.com/mail/?ui=2&ik=1144611be0&view=pt&search=inbox&th=13b421f8d3d3b9a8
2/2
-------
State.co.us Executixe Branch Mail - Re: Bowie Mine
EPAOIG Appendix 12
State
of
Colorado
Re: Bowie Mine
1 message
Pray - CDPHE, Charles
To: Rosendo Majano
Cc: "Jung, Doris"
Tue, Feb 19, 2013 at 12:56 PM
, Chuck Machovec - CDPHE
Rosendo:
Yes, the decision to allow Bowie, and three other sources, to not be required to model the PM2.5 was made by
upper management. I haven't been told it has been rescinded. I think they get a pass until this permit action is
completed.
The APEN the Division received in January was not for this purpose. It was to expand the existing GOB pile to
allow the mine to continue to operate, again until this permit process is complete for the newly located GOB pile
and associated haul. The expansion of GOB pile #2 has been allowed under "enforcement discretion" to keep the
mine open, but at only half the volume requested to cmply with the DRMS permitting.
BLM's lease arrangements do allow a lessee to prevent access under certain circumstances, while not requiring
fencing to allow for animal migration. How a company enforces this is probably up to them, but I'm certain that
the mine would be trying to prevent access to any operations which could potentially create a lawsuit situation.
It's easier in open pit mining, you fence everything you can and let terrain take care of the rest. The deer, elk and
antelope (and occasionally black bear) always seemed to find a way in and out.
I didn't receive a copy of the diagram mentioned in Aaron's e-mail, so I cant review that.
Chuck
On Tue, Feb 19, 2013 at 12:33 PM, Rosendo Majano|
Chuck,
wrote:
FYI, please see the email below. Specifically, see the last paragraph. It is saying that they will submit a
revised modeling analysis to account for changes in several sources. An APEN related to these changes was
apparently submitted in January, 2013 and if I understand correctly, you will also receive revised emissions
calculations.
Under these circumstances I can't do absolutely anything with the modeling review until I receive the new files,
so I will put this application on hold in the meantime.
Also, in this new resubmittal, will the SSP continue to exempt this facility from demonstrating compliance with
the PM2.5 NAAQS?
The email below is describing violations of the 24-hr PM10 NAAQS, so there is reason for concern about the
PM2.5 NAAQS.
Let me know what you decide.
Thanks,
Rosendo
EPAOIG Appendix 12
https ://mai I .g oog I e.corrVmai l/?ui=2&ik=1144611be0&view= pt&search= i nbc»&th= 13cf407e0cfd71 a5
1/3
-------
State.co.us Executixe Branch Mail - Re: Bowie Mine
EPAOIG Appendix 13
Forwarded message
From: Aaron Martinkus^^^^^H
Date: Fri, Feb 15, 2013 at 12:37 PM
Subject: Re: Bowie Resources
To: Rosendo Majano^^^^^^^H
Rosendo,
I have conducted a modeling analysis of the Bowie operations without excluding any receptors. The model
results show some predicted exceedances of the 24-hour and annual PM-10 NAAQS, but they only occur in
areas that are physically controlled by Bowie. I have attached a diagram showing the receptor locations
where the model predicted an exceedance. As you can see, these receptors are located near the facility wash
plant, and at the top of the mine. There is one predicted exceedance at the Terror Creek Site, but that area is
surrounded by fencing, and not publically accessible. This analysis includes the assumed 24-hour and annual
background concentrations of 29 ug/m3 and 16 ug/m3, respectively.
I have also attached a diagram with the locations of the gates that keep the public from accessing the roads
north of the mine. Installation of fencing around the entire mine property would not be possible; the lease
agreement between Bowie and the Bureau of Land Management would not allow for this type of activity.
Consideration must be made when there are competing interests amongst regulatory agencies; installation of
fencing would cause land disturbances that are not permitted, and there are also considerations regarding
wildlife migration and habitat protection.
While Bowie does not have a complete physical barrier around the entire mine, public access to the areas
where an exceedance is predicted to occur is controlled due to these gates, the topography, and because the
general public will not be able to reach these active mine areas without the permission of Bowie. With
permission to be on-site, any visitors (e.g. contractors) are no longer considered by EPA to be the general
public. MSHA regulations also require that people coming onsite must register with the mine upon arrival, they
must be accompanied by mine personnel, and that they do not allow general public access to the mine.
However, this requirement does not insist on fencing or physical barriers; it is understood that the actions of
mine personnel is sufficient to preclude public access to active mining areas. Bowie personnel have been
trained to approach any unfamiliar persons onsite, and to escort them off-site as necessary. Again, EPA does
not define preclude as making public access absolutely impossible; rather that the likelihood of such access is
small.
In addition, APCD has approved this approach for other permitting/modeling actions for other coal mines in
the North Fork Valley, such as Oxbow Mining, LLC and Mountain Coal Company, LLC. APCD has previously
understood the unique aspects of the siting of these mines requires different rationale to be applied in the
permitting process, and has used discretion and judgement when reviewing and approving permit applications
for these sources. This is not the first time this approach has been proposed, nor the first time it has been
approved by the Division for permitting. Bowie is just asking for a fair and consistent application of regulatory
interpretations.
Please note, the dispersion modeling has been updated to account for an APEN submitted in January to
expand Gob Pile #2; additional "area-poly sources" have been added (NEWGOB2 and NEWGOB2DOZ), a
volume source for the dropping of gob onto the pile (GOBPILENEW2), and an on-site haul road route has been
modified (CYN001-062) to account for where this expansion of the stockpile will occur. In addition, the size of
the area-poly source Gob Pile #3 (NEWGOB) was expanded to more accurately reflect the size of the
proposed pile, and as such the haul road (GRN001-029) associated with this stockpile was lengthened. I will
send you the revised input files in a separate email, and will forward the associated emission inventory to
Chuck Pray.
Thanks for your help. EpA0|G 13
https ://mai I .g oog I e.corrVmai l/?ui=2&ik=1144611be0&view= pt&search= i nbc»&th= 13cf407e0cfd71 a5 2/3
-------
Regards,
Aaron
State.co.us Executixe Branch Mail - Re: Bowie Mine
EPAOIG Appendix 14
Chuck Pray, P.E.-P.L.S.
Permit Engineer, Air Pollution Control Division
Colorado Department of Public Health and Environment
4300 Cherry Creek Drive South
Denver, CO 80246-I530
www.colorado.gov/cdphe
My office hours are 7:30 AM to 4:00 PM.
Please be aware that any information submitted to the Air Division regarding emissions of pollutants is potentially
subject to the Colorado Open Records Act.
EPAOIG Appendix 14
https ://mai I .g oog I e.corrVmai l/?ui=2&i k= 1144611beO&view= pt&search= i nbc»&th= 13cf407e0cfd71 a5
-------
State.co.us Executive Branch Mail - LEAN meeting
EPAOIG Appendix 15
State
of
Colorado
LEAN meeting
Garry wants to schedule a status update LEAN meeting on Dec 17, 18, or 19th.
I don't get the impression he is expecting that we are "done," rather, he wants to just make sure the process
continues mowng to implementation.
He seemed fine with the progress so far.
He also discussed metrics. For example, a year from now he/we need to be able to show how LEAN has or
hasnt helped. I told him that we'll need to careful with metrics because other variables are at work such as
decreased workloads (i.e., although I didn't quite say this... I suggested that a finding of an association between
better "turnaround time" of modeling reviews and LEAN implementation does not necessarily establish a "causal"
relationship because of the other variables at play).
He also mentioned that at some point he wants to have a meeting to reach an understanding on "policy" versus
"technical" aspects of modeling. I told him that as long as their is unwritten policy whereby management asks
staff N^rbally to not follow language in the rule, there will be issues. Any requests from management for staff to
not follow the requirements of the rule should be in writing. He agrees that there should be more discussion.
On a side discussion, he mentioned that he isn't con\inced that the modeling completeness language should be
included in the letter sent by the engineer (i.e., he seemed to be suggesting that the modeling completeness
letter and permit engineer completeness letter are separate things... but that said, he said that he is open to
discussion). My two cents... which I didn't share with Garry... is that if we are issuing a completeness letter on
modeling alone, then it would need a disclaimer that another letter may or may not be sent by the engineer...
and that a "complete" modeling analysis does not indicate a complete application, etc.
Chuck Machovec
Scientist/Supervisor
Modeling, Meteorology, and Emission Inventory Unit
Technical Services Program, Air Pollution Control Division
Colorado Department of Public Health and Environment
APCD-TS-B1
4300 Cherry Creek Drive South
Denver, CO 80246-1530
I «4orai" t V|Mrti»rrt
I'h Wn I IraltH
ji*J t jmrDftmtvrt
EPAOIG Appendix 15
tittps://rnai!googlecorn/mait/?ui=2&ik=1144611be0&v iew=pt&search=inbox&th=13b6c 1483741 a28b
1/1
-------
State.co.us Executive Branch Mail - Re: Brief Survey - Minor Source Modeling
EPAOIG Appendix 16
STATE OF
COLORADO
Re: Brief Survey - Minor Source Modeling
1 message
Jung -CDPHE, Doris^ Mon, Jul 6, 2015 at 10:43 AM
To: Rosendo Majano -
Cc: "gordon.pierce@state.co.us
Rosendo,
The issues you have brought to Roland's attention in the email below are among the many we (Chuck, Jon, and I) had informed permitting about in the August 18, 2010 meeting
hat was organized to discuss PM2.5 modeling.
Attendees included Chuck, Jon, me, Gordon, Roland Hea, Jim King (operating permit supervisor at the time), Chip Hancock, Mark McMillan, and Chris Laplante.
In addition to discussing PM2.5 in this meeting, we (modelers) were told to raise the modeling thresholds because compliance with NAAQS could not be demonstrated or they
would just issue permits without considering NAAQS compliance.
We (modelers) argued that their approach would be inconsistent with regulatory requirements and raising hresholds when NAAQS are more stringent was not technically
justified but ultimately we were ignored. Consequently, since the modelers wouldn't cooperate, permitting came out with he 40 tpy hreshold for 1-hr N02 and S02 NAAQS on
September 20, 2010.
The circumstances and the permitting personnel involved have not changed since the 2010 meeting/decision. While I agree concerns for NAAQS attainment remain, there is
nothing that indicates to me that permitting is willing to change their practice.
At this time, dialogue on this subject with permitting is beating a dead horse. Our resources are better spent elsewhere.
Doris
Forwarded message
From: Rosendo Majano
Date: Mon, Jul 6,2015 at
Subject: Fwd: Brief Survey
To: Doris Jung - CDPHE
FYI. I should have CCd you. I apologize for not doing it.
Forwarded message -
From: Hea - CDPHE, Rolandl
Date: Mon, Jul 6,2015 at 8:53
Subject: Re: Brief Survey - Minor Source Modeling
To: Rosendo Majano|
Cc: Gordon Pierce -
Rosendo,
Thanks for your additional thoughts. I do believe that we will need to continue these discussions over time to find the right balance between workloads, resources, outcomes, etc.
From a permitting perspective, we need to find solutions that balance environmental and public health protection, available staff resources, our permit timeliness requirements,
applicants' capacities and abilities to provide detailed technical information (especially small businesses), etc. I look forward to continuing the discussion and have given Mark
McMillan a heads-up as well.
Roland
On Tue, Jun 30, 2015 at 1:10 PM, Rosendo Majano^^^^^^^^^^^^^^^^fwrote:
Thanks Roland for your response.
My opinion is that the broader internal discussion is long overdue. The workable set of thresholds/criteria that you talk about establishing, already exist in the CO Modeling
Guideline. Table 1 of his document establishes long-term and short-term thresholds (0.46 Ib/hr for N02 and S02) and also a list of criteria (footnotes of Table 1) for when
modeling could be warranted even when the emissions rates are below the short and/or long-term hresholds.
Those thresholds and criteria to determine when modeling is required were developed by the APCD modeling experts and I never understood why the SSP has chosen to
ignore them all these years.
I can't take credit for any content of the CO Modeling Guideline since it was developed before I started working here at the APCD, but I can perfectly see and understand the
logic in he content of Table 1 and its footnotes, considering the clear requirement in AQCC Regulation 3 to verify compliance with the NAAQS as condition to issue permits,
and the unique situation of Colorado with most of its emission sources located in complex terrain (i.e. surrounding terrain elevations above the height of the stack).
Under these circumstances it is perfectly feasible to have sources with small annual emission rates cause very high impacts on air quality, specially on the short-term NAAQS,
because of the unique meteorological conditions created by complex terrain. I think that looking at the annual emissions alone to determine when modeling is warranted is a
big mistake, as it is ignoring the expert advise of those who have a better understanding of the dispersion modeling process, and the corresponding regulations and guidance
issued by the EPA.
By doing this, the Division might be issuing permits to facilities that are causing or contributing to modeled violations of the NAAQS, which I think is something we should all be
concerned about.
I agree with you that this discussion is way beyond the scope of the NACAA survey, and I never intended to suggest otherwise, but the questions in the survey reminded me of
this unresolved issue at the APCD: the requirement in AQCC Regulation 3 to verify compliance with the NAAQS before issuing a permit, and the question of what kind of
demonstration of compliance with the NAAQS has been performed for all those applications with emissions below the annual thresholds (e.g. 40 tpy for N02 and S02) that
were never required to submit modeling, but hat have emissions above the short-term thresholds or that meet other criteria that would warrant a modeling analysis.
I have made this same question to several people at the APCD before, including Will, and just like in your case now, the question has always been left unanswered. And I'm
not referring to those hundreds of applications with VOC-only I'm referring to applications like Williams Willow Creek or Williams
https://mail.google.com/mail/u/0?ik=1144611 beO&view=pt&search=all&permthid=thread-f%3A1505960242271872266%7Cmsg-f%3A15059659534209... 1/5
-------
State.co.us Executive Branch Mail - Re: Brief Survey - Minor Source Modeling
Crawford Trail, and many other that despite the clear paten ial to cause modeled violations of the 1-hr N02 or S02 NAAQS, all Tative
demonstration of compliance with those NAAQS.
Anyway, I can see that under the current circumstances my questions are mostly rhetorical, and I admit that all this discussion is way above my pay grade. But as technical
staff constantly dealing with this issue, I thought the NACAA survey was a good opportunity to, once again, bring this problem to the attention of decision makers like you, who
have steered fie Division all these years and are ultimately responsible for defining how hings work at he APCD.
I appreciate your taking the time to reply to my email and address my concerns.
Rosendo
Rosendo Majano
Air Quality Scientist
Modeling, Meteorology and Emission Inventory Unit
Air Pollution Control Division
Colorado Department of Public Health and Environment
On Mon, Jun 29,2015 at 4:57 PM, Hea - CDPHE, Roland wrote:
Rosendo,
Thanks for your thoughts. You raise some very good questions that warrant a broader internal discussion.
For purposes of responding to the New Mexico NACAA survey request, I tried to keep my responses as brief as possible. On Questions 1 and 2, they were simply asking for
a yes/no response. The only reason I included some additional discussion was to convey hat in Colorado we require modeling analysis for some of our minor source permit
applications based on a set of hresholds/criteria that we have established over ime.
The 40 tpy NOx and S02 threshold I cited was simply an example of one of the criteria we use. The New Mexico questions appear to be geared to how other agencies are
addressing modeling (or not modeling) for he 1-hour N02 and S02 standards. I responded N/A for Ques ion 3 because for some applications we do model to demonstrate
compliance with the 1-hour standards. We typically receive several housand minor source permit applications per year and the vast majority do not trigger our modeling
thresholds/criteria based either on their level of requested emissions or because of he pollutant involved, e.g., we gets lots of oil and gas E&P site applications that only
have reportable VOC emissions.
For me, the questions regarding which of these applications we do require quantitative modeling analysis for and how we establish a workable set of thresholds/criteria were
way beyond the scope of the response back to New Mexico. We should be discussing these questions internally between SSP, TSP, P&P, etc. if we feel there are issues
with our approach.
Roland
On Mon, Jun 29, 2015 at 9:53 AM, Rosendo Majano^^^^^^^^^^^^^^^^J wrote:
Roland,
This email was forwarded to me and I wanted to ask you a couple of things given that your response doesn't seem to completely address the requirements in AQCC
Regulation 3.
1- The requirement in Reg 3 is not to do modeling, but to verify compliance with the NAAQS (AQCC Regulation 3, Part B, III.D.1.C and d).
2- My understanding is that dispersion modeling is the only mechanism approved by the EPA to quantify impacts on air quality. Per AQCC Regulation 3, Part B, III.B.5 the
Division needs to prepare a preliminary analysis that will indicate what impact, if any, the new source will have. Such preliminary analysis will allow the Division to
determine, among other requirements, whether the new source will comply with he NAAQS.
3- So according to your response to questions 1 and 2 below the SSP does not require modeling for sources with proposed permitted emissions below the 40 tpy
threshold for N02 and S02. So in those cases, how does he SSP verify compliance wi h the 1-hr NQ2 and 1-hr SQ2 NAAQS? That's question 3 below that was left
unanswered.
According to your email that's the case for the vast majority of minor source applications, so I'm curious to know for those applications with N02 and/or S02 emissions
below the 40 tpy threshold, if the impacts on ambient air are not quantified through dispersion modeling, what kind of demonstration of compliance with the NAAQS has
been performed that has allowed he SSP to issue these permits all hese years?
I will appreciate your response.
Rosendo
Rosendo Majano
Air Quality Scientist
Modeling, Meteorology and Emission Inventory Unit
Air Pollution Control Division
Colorado Department of Public Health and Environment
Forwarded message -
From: Hea - CDPHE, Roland[
Date: Fri, Jun 26, 2015 at4:4i
Subject: Fwd: Brief Survey - Minor Source Modeling
To: William Allison - CDPHE
, Gordon Pierce - CDPHE
Christopher Laplante -
All,
FYI - Here is a response I sent to Rita Bates with the State of New Mexico as a result of a request from NACAA. It was straightforward, so I just went ahead and replied.
Thanks and have a nice weekend,
Roland
EPAOIG Appendix 17
https://mail.google.com/mail/u/0?ik=1144611 beO&view=pt&search=all&permthid=thread-f%3A1505960242271872266%7Cmsg-f%3A15059659534209... 2/5
-------
State.co.us Executive Branch Mail - Re: Brief Survey - Minor Source Modeling
¦ Forwarded message EPAOIG AppSflCliX 1 8
From: Hea - CDPHE, Rolandl
Date: Fri, Jun 26, 2015 at 4:29
Subject: Re: Brief Survey - Minor Source Modeling
Karen
Rita,
Here are Colorado's responses to your questions:
Does your state require facility modeling analysis for minor source permitting?
Yes, for some of our minor source permit applica ions. We have developed a set of hresholds and criteria for when modeling is required for minor source applications. For
example, new sources whose proposed permitted emissions are equal to or greater than 40 tons per year of NOx or S02, or existing sources who are proposing to increase
their permit limits by 40 tons per year or more of NOx or S02 are required to model for that pollutant. As a matter of practice, the vast majority of our minor source
applications do not require modeling analysis because they don't trigger the modeling hresholds or criteria.
If you require modeling analysis for minor sources do you require modeling demonstration of compliance with 1-hr N02 and 1-hr S02 NAAQS?
Yes, if they trigger our modeling thresholds or criteria for NOx or S02.
If you do not require a modeling analysis, do you show compliance with the N02 and S02 1-hour standards in another way? If yes, how?
Not applicable based on our responses to Questions 1 and 2.
Please let me know if you have any questions or would like to discuss in greater detail.
Best regards,
Roland
On Fri, Jun 26, 2015 at 10 00 AM, Karen Mongoven^^^^^^^^^^^^^^J wrote:
TO: NACAA NEW SOURCE REVIEW COMMITTEE
NACAA EMISSIONS AND MODELING COMMITTEE
Rita Bates (New Mexico) asked me to circulate he following brief survey questions to committee members and would appreciate it if you could take a few moments to
respond. This email is formatted so that responses will be addressed to both Rita and me. Thank you in advance for your assistance.
Does your state require facility modeling analysis for minor source permitting?
Yes or No
If you require modeling analysis for minor sources do you require modeling demonstration of compliance with 1-hr N02 and 1-hr S02 NAAQS?
Yes or No
If you do not require a modeling analysis, do you show compliance with the NQ2 and SQ2 1-hour standards in another way? If yes, how?
Rita Bates
Planning Section Chief
Ail Quality Bureau — New Mexico Environment Department
525 Camino de los Marque^ Suite I
Santa Fe, New Mexico 87505
* * *
Karen K. Mongoven
EPAOIG Appendix 18
https://mail.google.com/mail/u/0?ik=1144611 beO&view=pt&search=all&permthid=thread-f%3A1505960242271872266%7Cmsg-f%3A15059659534209...
-------
- Re: Brief Survey - Minor Source Modeling
EPAOIG Appendix 19
National Association of Clean Air Agencies
444 North Capitol Street, NW, Suite 307
Washington, DC 20001
Roland C. Hea, P.E.
Permitting Section Supervisor
Roland C. Hea, P.E.
Permitting Section Supervisor
Roland C. Hea, P.E.
Permitting Section Supervisor
Roland C. Hea, P.E.
Permitting Section Supervisor
EPAOIG Appendix 19
https://mail.google.com/mail/u/0?ik=1144611be0&view=pt&search=all&permthid=thread-f%3A1505960242271872266%7Cmsg-f%3A15059659534209.
-------
State.co.us Executive Branch Mail - Re: Brief Survey - Minor Source Modeling
EPAOIG Appendix 20
EPAOIG Appendix 20
https://mail.google.com/mail/u/0?ik=1144611be0&view=pt&search=all&permthid=thread-f%3A1505960242271872266%7Cmsg-f%3A15059659534209
-------
State.co.us Executive Branch Mail - Excel Denver Steam Plant
EPAOIG Appendix 21
STATE OF
COLORADO
Excel Denver Steam Plant
1 message
Rosendo,
In your e-mail to Xcel you made some excellent arguments why impact modeling to the 1 hour NOx standard would be a good idea for he proposed new boiler at the existing
steam plant. However, PS memo 10-01 makes it very clear that we cannot require Xcel to do modeling to the 1 hour NOx standard since emissions of NOx will be less than 40
tons per year.
I will let Gary Magno know that we will not be requiring modeling to the 1 hour NOx standard for the new boiler so long as emissions are limited to less than 40 tons per year.
Chip
R K "Chip" Hancock III, P.E.
Construction Permit Unit Supervisor
Stationary Sources Program
Air Pollution Control Division
E
NOTE: As of January 1,2014, the Colorado Air Pollution Control Division no longer accepts blank or incomplete APENs. Additional fees may apply If an APEN is submitted
without the necessary information. An application with missing information may result in longer processing times. Please note that all APEN submissions should be completed
using forms currently supplied by the Division (See Reg. 3, Part A, Section IIA) Current APEN forms can be found at bttps //www Colorado gov/cdphe/APENforms.
EPAOIG Appendix 21
https://mail.google.com/mail/u/0?ik=1144611be0&view=pt&search=all&permthia=thread-f%3A1544742496627340042%7Cmsg-f%3A15447424966273... 1/1
-------
State.co.us Executive Branch Mai) - Fwd: MMM Monaghan - air modeling status
EPAOIG Appendix 22
STATE OF
COLORADO
Fwd: MMM Monaghan - air modeling status
1 message
Rink - CDPHE, Bradley|
To: Rosendo Majanol
, Emmett Malonel
Wed. Mar 11.2020 at 1:51 PM
I "Reynolds - CDPHE, DeVondria"!
fyi
Bradley Rink
Technical Services Program, Air Pollution Control Division
vjemailsig.png
Forwarded message -
From: Brickey - CDPHE, Jonathan]
Date: Wed, Mar 11, 2020 at 1:45 F
Subject Re: MMM Monaghan - air modeling status
To: Anna Unruh I
Cc: Hanna Warl"
, Erin Kunkell
I, Bradley Rink - CDPHEl
Anna,
After consul ing with son* higher-ups, I agree that modeling for the 24-hour PM2.5 standard will not be required in this case.
Unfortunately, we don't have a specific written policy I can point to that says "you can always ignore the daily PM2.5 modeling hreshold", but we do it on more of a case-by-case
basis. In the case of Monaghan, its nearly a halt mile from the nearest residential area (and predominant wind patterns in the area are in the opposite direction), the topography
of the area is very flat and I havent heard any community objections, so I dont have an issue with overlooking this particular PM2.5 modeling.
Thanks,
Jonathan Brickey, P£
Permit Engineer
Construction Permits Unit
10
COLORADO
Air Pollution Control Division
Department of Public Health & Environment
P 303.691.4093 | F 303.782.0278
Are you curious about tfround-level ozone in Colorado? Visit our ozone webpage to team more.
On Wed, Mar 11, 2020 at 12:09 PM Anna Unruh|
Hi Jonathan,
I wrote:
I've received guidance from multiple non-oil & gas permit engineers in the last year or so that the agency only looks at the S tpy threshold for PM2.5, unless there are special
circumstances (e.g., high expectation of comments from nearby residents]. The first time 1 heard it I was surprised so I've brought it up with other permit engineers (basically
anytime I've had the chance) and was told the same thing Based on this guidance, we did not submit daily PM 2.S modeling
Thanks,
Anna
Anna Unruh
Senior Consultant
Trinity Consultants
1391 N Speer Blvd Suite 350 | Denver, CO 80204
From: Brickey - CDPHE, Jona han
Sent: Wednesday, March 11. 2020
To: Anna Unruh
Subject: MMM
EPAOIG Appendix 22
https://mail.google.com/mail/u/0?ik=1144611 beO&view=pt&search=all&permthia=thread-f%3A1660898564694092280%7Cmsg-f%3A16608985646940...
1/2
-------
Anna,
State.co.us Executive Branch Mail - Fwd: MMM Monaghan - air modeling status
EPAOIG Appendix 23
I was going over the Monaghan modeling with Bradley, and I wanted to give you a quick update.
All the emission factors/calculations look good. Bradley was going to make some slight changes to how the emission types were binned, but he said any changes to the final
outcome will be minimal, nothing to worry about.
However, your analysis doesn't seem to address he 24-hour PM2.5 standard. Table 7-1 notes that the annual PM2.5 emissions are below the modeling threshold, but makes
no mention of the short term PM2.5 NAAQS (35 ug/m3). Considering the facility-wide daily PM10 emissions in Table D-1 show 321.63 ppd of emissions, it's unlikely that
facility-wide daily PM2 5 emissions will be below the modeling threshold of 11 ppd, but I don't see any daily total calculated. What was your reasoning for leaving out daily
PM2.5 from he modeling?
Thanks,
Jonathan Brickey, PE
Permit Engineer
Construction Permits Unit
COLORADO
ifc-- ,'^i Air Pollution Control Division
Department of Public Health 6 Environment
P 303.691.4093 I F 303.782.0278
Are you curious about ground-level ozone in Colorado? Visit our ozone webpage to learn more.
EPAOIG Appendix 23
https://mail.google.com/mail/u/0?ik=1144611be0&view=pt&search=all&permthid=thread-f%3A1660898564694092280%7Cmsg-f%3A16608985646940.
-------
State.co.us Executive Branch Mail - Fwd: Modeling meetinn
STATE OF
COLORADO
Fwd: Modeling meeting outcome
1 message
Reynolds - CDPHE, DeVondrial
To: Matt Burgett - CDPHE|
tyi
DeVondria Reynolds, MS
Air Quality Modeler
Modeling and Emissions Inventory Unit
Technical Services Program
Tue, Mar 16.2021 at 1:39 PM
Tajano - CDPHEl
COLORADO
Air Pollution Control Division
Department of Public Health 6 Environment
Learn more about COVID 19 https://www.colorado.gov/padflc/cdphe/apcd-covid-19 .
Are you curious about ground-level ozone in Colorado? Visit our ozone webpage to leam more.
Forwarded message -
From: Reynolds - CDPHE, DeVondria|
Date: Tue, Mar 16,2021 at 3:38 PM
Subject Re: Modeling meeting outcome
To: Kaufman - CDPHE, Garrison |
Cc: Gordon Pierce - CDPHEl
FEmmeti Malonel
I Bradley Rink - CDPHEl
Hello Gary,
Based on the above communication that took place during the uncompromising meeting Emrnett encountered yesterday with Gordon, Robyn and yourself, (t is unclear how
Regulation 3, Part B, Sections III .B.5 d and IILD.1 c will be satisfied.
As stated in the Clean Air Act and per my job description it is my duty to determine compliance or noncompliance with the National Ambient Air Quality Standards, the Colorado
Ambient Air Quality Standards, PSD Increments, and Air Quality Related Values, for major and minor sources of air pollution It is my understanding that you are instructing me
to turn a blind eye to the work that we are supposed to do in order to protect and improve the health of Colorado's citizens and the quality of its environment
I refuse to sign my name on any fur her modeling reports under your mandate without a written documented memo (or likewise) that has your signature of approval that is
dictating this direct violation of law and regulations. Considering the lawsuits that have been presented to our division I do not want litigation brought upon myself seeing that it is
your direction that has led us down this path (you are more than welcome to sign off on all modeling reports going forward, not something I would recommend seeing that you
did not do the work but that choice is yours).
I do want to make you aware that if this email remains unresponsive by you I will forward his correspondence and your lack of reply to the Environmental Program Manager, a
general assembly, EPA and environmental groups if need be.
I would like for it to also be noted that I do not appreciate being bullied and forced into a situation that is in direct violation of federal regulations. As much as you and your higher
official peers assume modeling is a meaningless unit resources should not be taken away nor should our job responsibilities be reduced. Your push should be to only add
resources to this unit not take away
DeVondria Reynolds, MS
Air Quality Modeler
Modeling and Emissions Inventory Unit
Technical Services Program
COLORADO
Air Pollution Control Division
Department of Public Health b Environment
4300 Cherry Creek Drive South, Denver, CO 80246-1530
Leam more about COVID-19 https://www.colorado.gov/pacific/cdphe/apcd-covid-19 .
Are you curious about ground-level ozone in Colorado? Visit our ozone webpage to leam more.
On Tue, Mar 16,2021 at 2:43 PM Rink - CDPHE, Bradley|
yep
Bradley Rink
Technical Services Program, Air Pollution Control Division
I wrote:
From: Malone - CDPHE, Emmett
EPAOIG Appendix 24
https ://maiLgooglecom/mail/u/0?ik=1144611 beO&view=pt&search=all&permthia=thread-f%3AI 694393218126394881 %7Cmsg-f%3AI 6944187073217.
1/3
-------
Date: Tue, Mar 16, 2021 at 11:12 AM
Subject: Fwd: Modeling mee ing outcome
To: Bradley Rink - CDPHEl
State.co.us Executive Branch Mail - Fwd: Modeling meeting outcome
EPAOIG Appendix 25
, DeVondria Reynolds - CDPHEl
Gordon response
Emmett Malone
Supervisor
Modeling and Emissions Inventory Unit
Technical Services Program
Air Pollution Control Division
Colorado Department of Public Health and Environment
APCD-TS-B1
4300 Cherry Creek Drive South
Denver, CO 80246-1530
"Are you curious about ground-level ozone in Colorado? Visit our ozone webpage to learn more.'
- Forwarded message -
From: Pierce - CDPHE, Gordonl
Date: Tue, Mar 16, 2021 at 10:06
Subject: Re: Modeling meeting outcome
To: Malone - CDPHE, Emmettl
No, we are not to address the short-term standards..."the short-term thresholds in the Modeling Guideline will not be used."
(Based on Brad's email, is he planning something?)
Gordon
On Tue, Mar 16, 2021 at 8:04 AM Malone - CDPHE, Emmett|
Hi,
wrote:
Had a discussion with Brad and DeVondria this morning and it went much like I thought it would, badly. I feel like I have stepped back in time and Doris and Rosendo are
arguing about how this is wrong.
One question that came up and if you or Garry could respond I would appreciate it. What I took from the mee ing was hat we are not to address the short term standards.
The way you worded your email is slightly different, are we to address the short term standards?
2 years, 7 months, and 15 days.
Emmett Malone
Supervisor
Modeling and Emissions Inventory Unit
Technical Services Program
Air Pollution Control Division
Colorado Department of Public Heal h and Environment
APCD-TS-B1
4300 Cherry Creek Drive South
Denver, CO 80246-1530
"Are you curious about ground-level ozone in Colorado? Visit our ozone webpage to learn more."
On Mon, Mar 15, 2021 at 4:30 PM Pierce - CDPHE, Gordon|
Emmett,
wrote:
Per the meeting we had today 3/15/21 with Garry and Robyn, Garry specifically stated that, effec ive immediately, the short-term thresholds in the Modeling Guideline will
not be used and that modeling only be performed using the following thresholds:
• 40 tpy for N02 and S02
• 82 lbs/day for PM10
• 5 tpy for PM2.5
• 23 Ibs/hr for CO
• 25 lbs/3-mo for Pb
He will allow exceptions based on agreement between the modelers and permit engineers, and his specific approval.
Garry also requested that the Modeling Guideline also be removed from the website, pending further discussions and revisions. I will ask Ivan to pull it. Attached is a copy
of the current version that is on the website.
Thanks,
Gordon
Gordon Pierce
Program Manager
EPAOIG Appendix 25
https://mail.google.com/mail/u/0?ik=1144611 beO&view=pt&search=all&permthid=thread-f%3A1694393218126394881 %7Cmsg-f%3A16944187073217...
-------
Technical Services Program
State.co.us Executive Branch Mail - Fwd: Modeling meeting outcome
EPAOIG Appendix 26
42
COLORADO
Air Pollution Control Division
Department of Public Health 6 Environment
4300 Cherry Creek Drive South, Denver, CO 80246-1530
Are you curious about ground-level ozone in Colorado? Visit our ozone webpage to learn more.
Gordon Pierce
Program Manager
Technical Services Program
^AICOLDRAD0
Air Pollution Control Division
| Department of Public Health 6 Environment
4300 Cherry Creek Drive South, Denver, CO 80246-1530
Are you curious about ground-level ozone in Colorado? Visit our ozone webpage to learn more.
EPAOIG Appendix 26
https://mail.google.com/mail/u/0?ik=1144611be0&view=pt&search=all&permthid=thread-f%3A1694393218126394881%7Cmsg-f%3A16944187073217.
-------
State.CO-US Executive Branch Mail -
Re: Modeling meeting outcome
EPAOIG Appendix 27
STATE OF
COLORADO
Re: Modeling meeting outcome
1 message
Rink - CDPHE, Bradley|
To: Gordon Pierce - CDI
Emmett Malone]
"Garrison W Kaufman"!
, "Reynolds - CDPHE, DeVondria"
, "Burgett- CDPHE, Matt"!
Tue, Mar 16.2021 at 1:31 PM
, Rosendo
FYl - these actions undertaken by APCD management, et al., are in viola ion of the law defined by the Clean Air Act, and rules defined by APCD and he AQCC. I object to
hese directions by my management as they are in violation of Federal Law and State Rule; i am requesting whis leblower status.
regards
blr
Bradley Rink
Technical Services Program, Air Pollution Control Division
On Tue, Mar 16,2021 at 12:42 PM Rink - CDPHE, Bradley|
yep
Bradley Rink
Technical Services Program. Air Pollution Control Division
I wrote:
Forwarded message -
From: Malone - CDPHE, Emmettl
Date: Tue, Mar 16, 2021 at 11:12 /
Subject Fwd: Modeling mee ing outcome
To: Bradley Rink - CDPHEl
DeVondria Reynolds - CDPHEl
Gordon response
Emmett Malone
Supervisor
Modeling and Emissions Inventory Unit
Technical Services Program
Air Pollution Control Division
Colorado Department of Public Health and Environment
APCD-TS-B1
4300 Cherry Creek Drive South
Denver. CO 80246-1530
"Are you curious about ground-level ozone in Colorado? Visit our ozone webpage to learn more.'
Forwarded message —
From: Pierce - CDPHE, GordonM
Date: Tue, Mar 16,2021 at
Subject Re: Modeling meeting outcome
To: Malone - CDPHE, Emmett^
No, we are not to address the short-term standards ."the short-term thresholds in the Modeling Guideline will not be used."
(Based on Brad's email, is he planning something?)
Gordon
On Tue, Mar 16,2021 at 8:04 AM Malone - CDPHE, Emmett^^^^^^^^^^^^^^H wrote:
Hi,
Had a discussion with Brad and DeVondria this morning and it went much like I thought it would, badly. I feel like I have stepped back in time and Doris and Rosendo are
arguing about how this is wrong.
One question that came up and if you or Garry could respond I would appreciate it What I took from the mee Ing was hat we are not to address the short term standards.
The way you worded your email is slightly different, are we to address the short term standards?
2 years, 7 months, and 15 days.
Emmett Malone
Supervisor
Modeling and Emissions Inventory Unit
EPAOIG Appendix 27
https://mail.google.com/mail/u/0?ik=1144611 beO&view=pt&search=all&permthid=thread-f%3A1694393218126394881 %7Cmsg-f%3A16944182210504... 1/2
-------
Technical Services Program
Air Pollution Control Division
Colorado Department of Public Heal h and Environment
APCD-TS-B1
4300 Cherry Creek Drive South
Denver, CO 80246-1530
"Are you curious about ground-level ozone in Colorado? Visit our ozone webpage to learn more."
State.co.us Executive Branch Mail - Re: Modeling meeting outcome
EPAOIG Appendix 28
On Mon, Mar 15, 2021 at 4:30 PM Pierce - CDPHE, Gordonwrote:
Emmett,
Per the meeting we had today 3/15/21 with Garry and Robyn, Garry specifically stated that, effec ive immediately, the short-term thresholds in the Modeling Guideline will
not be used and that modeling only be performed using the following thresholds:
• 40 tpy for N02 and S02
• 82 lbs/day for PM10
• 5 tpy for PM2.5
• 23 Ibs/hr for CO
• 25 lbs/3-mo for Pb
He will allow exceptions based on agreement between the modelers and permit engineers, and his specific approval.
Garry also requested that the Modeling Guideline also be removed from the website, pending further discussions and revisions. I will ask Ivan to pull it. Attached is a copy
of the current version that is on the website.
Thanks,
Gordon
Gordon Pierce
Program Manager
Technical Services Program
colorad°
V ~-f Air Pollution Control Division
I Department of Public Health 6 Environment
4300 Cherry Creek Drive South, Denver, CO 80246-1530
Are you curious about ground-level ozone in Colorado? Visit our ozone webpage to learn more.
Gordon Pierce
Program Manager
Technical Services Program
COLORADO
Air Pollution Control Division
Department of Public Health 6 Environment
4300 Cherry Creek Drive South, Denver, CO 80246-1530
Are you curious about ground-level ozone in Colorado? Visit our ozone webpage to learn more.
EPAOIG Appendix 28
https://mail.google.com/mail/u/0?ik=1144611be0&view=pt&search=all&permthid=thread-f%3A1694393218126394881%7Cmsg-f%3A16944182210504.
-------
State.co.us Executive Branch Mail - Re: Asphalt Specialties -
STATE OF
COLORADO
Re: Asphalt Specialties - Central Plant
1 message
Moseley - CDPHE, Aaron
To: Rosendo Majano
Cc: "Reynolds - CDPHt, DeVondria
, Emmett Malone^^^^^^^^^^^^^^^J, Chip Hancock- CDPHE|
Rosendo,
Per Chip's guidance, for this facility, SSP is only asking for a demonstration of compliance with the carbon monoxide 1-hr and 8-hr NAAQS.
Thanks, stay warm!
Aaron Moseley
Permit Engineer
Thu, Oct 10,2019 at 4:53 PM
Stationary Sources Program
IS
COLORADO
Air Pollution Control Division
Department of Public Health b Environment
Are you curious about ground-level ozone in Colorado? Visit our ozone webpage to learn more
On Thu, Oct 10, 2019 at 2:12 PM Rosendo Majano|
Hi Aaron,
wrote:
We have started the review of the subject application and found that the facility submitted a NAAQS compliance demonstration only for carbon monoxide despite having
emissions rates of PM10, PM2.5, S02 and NOx, that exceed the corresponding short-term modeling thresholds.
Could you please confirm for which pollutants-averaging periods did the Stationary Sources Program request a NAAQS compliance demonstration?
Thanks!
Rosendo
Rosendo Majano
Appendix 29
https://mail.google.com/mail/u/0?ik=1144611be0&view=pt&sea^^^^^fithid=thread-a%3Ar7176976387311761846%7Cmsg-f%3A1647048714537... 1/1
-------
State.co.us Executive Branch Mail - Fwd: MMM Monaghan - air modeling status
EPAOIG Appendix 30
STATE OF
COLORADO
Fwd: MMM Monaghan - air modeling status
1 message
Rosendo Majano
To: "Brickey, Jonathan
Cc: Emmett Malone|
Bcc: Bradley Rink -
Jonathan,
Thu, Mar 12,2020 at 8:28 AM
, "Reynolds - CDPHE, DeVondria"!
Your email below was forwarded to me. I know that you are following instructions from your supervisors when not requesting a 24-hr PM2.5 NAAQS compliance demonstration,
but I would fail to do my job if I don't inform you that the Martin Marietta Monaghan facility submitted a screening modeling analysis for CO, PM10, and PM2.5 last year (see the
attached modeling review report) and that such analysis resulted in a modeled viola ion of the 24-hr PM2.5 NAAQS.
Because the previous results are from a screening analysis, the next recommended step should be to request a refined modeling analysis to verify NAAQS compliance per CO
Regulation 3, Part B §III.B.5.d and §III.D.1 .c. That's what I would suggest.
Once again, I understand the constraints of your situation, but my role in this permit application is to verify NAAQS compliance, so I feel compelled to inform you of the pre-
existing PM2.5 NAAQS compliance issue at this facility.
Thanks,
Rosendo Majano
Permit Modeling Work Lead
Modeling and Emissions Inventory Unit
IS
COLORADO
Air Pollution Control Division
Department of Public Health & Environment
- Forwarded message -
From: Rink - CDPHE, Bradleyl
Date: Wed, Mar 11,2020 at 1:5
Subject: Fwd: MMM Monaghan - air modeling status
To: Rosendo Majanol
, Emmett Malonel
Reynolds - CDPHE, DeVondrial
fyi
Bradley Rink
Technical Services Program, Air Pollution Control Division
emailsig.png
From: Brickey - CDPHE, Jonathan|
Date: Wed, Mar 11,2020 at 1:45 Pl\*
Subject: Re: MMM Monaghan - air modeling status
To: Anna Unruh[~~
Cc: Hanna WarlicE
, Erin Kunkell
I, Bradley Rink-CDPHE I
Anna,
After consul ing with some higher-ups, I agree that modeling for the 24-hour PM2.5 standard will not be required in this case.
Unfortunately, we don't have a specific written policy I can point to that says "you can always ignore the daily PM2.5 modeling hreshold", but we do it on more of a case-by-case
basis. In the case of Monaghan, it's nearly a half mile from the nearest residential area (and predominant wind patterns in the area are in the opposite direction), the topography
of the area is very flat, and I haven't heard any community objections, so I don't have an issue with overlooking this particular PM2.5 modeling.
Thanks,
Jonathan Brickey, PE
Permit Engineer
Construction Permits Unit
COLORADO
Air Pollution Control Division
Department of Public Health & Environment
Are you curious about ground-level ozone in Colorado? Visit our ozone webpage to learn more.
EPAOIG Appendix 30
h ttps://mail.google.com/mail/u/0?ik=1144611 be0&view=pt&search=all&permthid=thread-f%3A1660898564694092280%7Cmsg-a%3Ar2903233432121... 1/2
-------
State.co.us Executive Branch Mail - Fwd: MMM Monaghan
- air modeling status
EPAOIG Appendix 31
On Wed, Mar 11, 2020 at 12:09 PM Anna Unruh
Hi Jonathan,
I've received guidance from multiple non-oil & gas permit engineers in the last year or so that the agency only looks at the 5 tpy threshold for PM2.5, unless there are special
circumstances (e.g., high expectation of comments from nearby residents). The first time I heard it; I was surprised, so I've brought it up with other permit engineers (basically
anytime I've had die chance) and was told the same thing Based on this guidance, we did not submit daily PM2.5 modeling
Thanks,
Anna
Anna Unruh
Senior Consultant
Trinity Consultants
1391 N Speer Blvd Suite 350 | Denver, CO 80204
wrote:
- CDPHE, Jona
Sent: Wednesday, March 11, 2020
Anna
Subject: MMM Monagna^^iHrioaelin^tatu^^
Anna,
I was going over the Monaghan modeling with Bradley, and I wanted to give you a quick update.
All the emission factors/calculations look good. Bradley was going to make some slight changes to how the emission types were binned, but he said any changes to the final
outcome will be minimal, nothing to worry about.
However, your analysis doesn't seem to address he 24-hour PM2.5 standard. Table 7-1 notes that the annual PM2.5 emissions are below the modeling threshold, but makes
no mention of the short term PM2.5 NAAQS (35 ug/m3). Considering the facility-wide daily PM10 emissions in Table D-1 show 321.63 ppd of emissions, it's unlikely that
facility-wide daily PM2 5 emissions will be below the modeling threshold of 11 ppd, but I don't see any daily total calculated. What was your reasoning for leaving out daily
PM2.5 from he modeling?
Thanks,
Jonathan Brickey, PE
Permit Engineer
Construction Permits Unit
IS
COLORADO
Air Pollution Control Division
Department of Public Health 6 Environment
jonathan.brickey@state.co.us I www.colorado.gov/cdphe/apcd
Are you curious about ground-level ozone in Colorado? Visit our ozone webpage to learn more.
—i Martin Marietta Monaghan Facility - Modeling Review Report 03082019.pdf
^ 807K
EPAOIG Appendix 31
https://mail.google.com/mail/u/0?ik=1144611be0&view=pt&search=all&permthid=thread-f%3A1660898564694092280%7Cmsg-a%3Ar2903233432121.
-------
STATE OF
COLORADO
State.co.us Executive Branch Mail - Re: MMM Monaghan - air modeling status
EPAOIG Appendix 32
Re: MMM Monaghan - air modeling status
1 message
Rosendo Majano
To: "Rink - CDPHt, Bradley'
Cc: "Reynolds - CDPHE, DeVondria
Regarding the email hread below:
Thu, Mar 12,2020 at 7:47 AM
, Emmett Malonel
From CO Regulation 3, Part B §III.D.1.c
"... the Division shall grant the permit if it finds that: The proposed source or activity will not cause an exceedance of any National Ambient Air Quality
Standards;"
From CDPHE's 03/06/2020 press release distributed among all employees:
"Our rules and regulations exist for a reason: they protect Coloradans' health and the environment that we all cherish, so compliance is not an option; it's an
imperative," said Garry Kaufman, director of the Colorado Air Pollution Control Division.
The modeling analysis submitted by the Monaghan facility for this same permit application already resulted in a 24-hr PM2.5 NAAQS modeled viola ion.
Rosendo Majano
Forwarded message -
From: Brickey - CDPHE, Jonathanl
Date: Wed, Mar 11, 2020 at 1:45 PIV
Subject: Re: MMM Monaghan - air modeling status
To: Anna Unruhl""
Cc: Hanna WarlicE
, Erin Kunkell
I, Bradley Rink-CDPHEl
Anna,
After consulting with some higher-ups, I agree that modeling for the 24-hour PM2.5 standard will not be required in this case.
Unfortunately, we don't have a specific written policy I can point to that says "you can always ignore the daily PM2.5 modeling threshold", but we do it on more of a case-by-
case basis. In the case of Monaghan, it's nearly a half mile from the nearest residential area (and predominant wind patterns in the area are in the opposite direction), the
topography of the area is very flat, and I haven't heard any community objections, so I don't have an issue with overlooking this particular PM2.5 modeling.
Thanks,
Jonathan Brickey, PE
Permit Engineer
Construction Permits Unit
IS
COLORADO
Air Pollution Control Division
Department of Public Health & Environment
Are you curious about ground-level ozone in Colorado? Visit our ozone webpage to learn more.
On Wed, Mar 11, 2020 at 12:09 PM Anna Unruh|
Hi Jonathan,
wrote:
I've received guidance from multiple non-oil & gas permit engineers in the last year or so that the agency only looks at the 5 tpy threshold forPM2.5, unless there are special
circumstances (e.g., high expectation of comments from nearby residents). The first time I heard it; I was surprised, so I've brought it up with other permit engineers (basically
anytime I've had the chance) and was told the same thing Based on this guidance, we did not submit daily PM2.5 modeling
Thanks,
Anna
Anna Unruh
Senior Consultant
Trinity Consultants
1391 N Speer Blvd Suite 350 | Denver, CO 80204
EPAOIG Appendix 32
https://mail.google.com/mail/u/0?ik=1144611be0&view=pt&search=all&permthid=thread-f%3A1660898564694092280%7Cmsg-a%3Ar-397035900210... 1/2
-------
State.co.us Executive Branch Mail - Re: MMM Monaghan - air modeling status
EPAOIG Appendix 33
- CDPHE, Jonathan
Sent: Wednesday, March 11, 2020
Anna UnruhJH^^J^^f^im^^J
Subject: MMM Monagna^^inrioaelin^tatu^^
Anna,
I was going over the Monaghan modeling with Bradley, and I wanted to give you a quick update.
All the emission factors/calculations look good. Bradley was going to make some slight changes to how the emission types were binned, but he said any changes to the final
outcome will be minimal, nothing to worry about.
However, your analysis doesn't seem to address the 24-hour PM2.5 standard. Table 7-1 notes that the annual PM2.5 emissions are below the modeling threshold, but
makes no mention of he short term PM2.5 NAAQS (35 ug/m3). Considering the facility-wide daily PM10 emissions in Table D-1 show 321.63 ppd of emissions, it's unlikely
that facility-wide daily PM2.5 emissions will be below the modeling threshold of 11 ppd, but I don't see any daily total calculated. What was your reasoning for leaving out
daily PM2.5 from the modeling?
Thanks,
Jonathan Brickey, PE
Permit Engineer
Construction Permits Unit
IS
COLORADO
Air Pollution Control Division
Department of Public Health b Environment
Are you curious about ground-level ozone in Colorado? Visit our ozone webpage to learn more.
EPAOIG Appendix 33
https://mail.google.com/mail/u/0?ik=1144611 beO&view=pt&search=all&permthid=thread-f%3A1660898564694092280%7Cmsg-a%3Ar-397035900210... 2/2
-------
EPAOIG Appendix 34
STATE OF
COLORADO
Oxford Asphalt Plant
Rosendo Majano
To: Aaron Moseley
Cc: Emmett Malori
^Reynolds - CDPHE, DeVondria"
Wed, Feb 19,2020 at 8:23 AM
Hi Aaron,
I was discussing with DeVondria yesterday the Oxford Asphalt Plant modeling analysis, and there are some issues that we need to bring to your attention.
1 - Particulate matter modeling.
The modeling report that was submitted earlier this month has the following language:
In an email dated May 2, 2019, Mr. Moseleyofthe CDPHE indicated that, based on his calculations, CO and particulate matter (PM) must be modeled. He also indicated in this email communication that he would perform the
SCREEN3 modeling for the particulate sources, and that Aggregate Industries would only be responsible for the CO modeling.
SCREEN3 is an outdated program that is no longer an EPA regulatory model for permitting purposes. If I'm not mistaken it was replaced by AERSCREEN in early 2011, so I would not recommend to use SCREEN3 results to
support a permitting action. Moreover, both SCREEN3 and AERSCREEN are one-source models, meaning that you can only model once source at a time, so there is no way to represent at the same time all the sources of
fugitive dust, which typically are responsible for a much larger fraction of the total particulate matter emissions. Hence our long standing recommendation to use AERMOD when modeling this type of facilities.
Currently the consultants have not provided the necessary information that would allow us to model the fugitive dust sources, so our recommendation is to request the applicant to submit the particulate matter modeling in
AERMOD for us to review. However, this is your decision. Please let us know how you would like to proceed.
2- Stack parameters of the drum mixer.
The exhaust velocity of the drum mixer used in the modeling analysis is of 440.9 m/s, which seems quite high. Has the consultant sent you specification sheets for this unit or do you know from experience if this range of exhaust
velocity is normal for a drum mixer?
This facility is reporting Pb emissions of 5.2 lb/day and indicating that the requested operating limit is 5000 hours per year. At this emission rate the facility will exceed EPA's 0.5 tpy Pb emissions threshold for monitoring, so we
would need to notify the Division's particulate monitoring group. This daily emission rate also exceeds by far the modeling threshold in our CO Modeling Guideline, so we would also recommend that Pb modeling be submitted as
well. Alternatively, we could do that modeling analysis in house, but in any case, we would ask you for confirmation if that daily emission rate is correct and if this facility will have the 5000 hrs/year restriction in the permit. Also,
could you please confirm if there will be any restriction in the daily hours of operation?
4-APENs and PA.
Could you please provide us with a copy of the APENs and the preliminary analysis for this facility?
Thanks in advance for your help!
3- Pb emissions.
Rosendo Majano
EPAOIG
Appendix
34
-------
EPAOIG Appendix 35
STATE OF
COLORADO
Oxford Asphalt Plant
Rosendo Majanc
To: "Reynolds - C
Thu, Feb 27,2020 at 11:35 AM
Hi DeVondria,
I talked to Aaron about the issues with the Oxford Asphalt Plant, and he said he would get back to us with the information: APENs, PA, confirmation of the Pb emissions, and high velocity of the exhaust from the drum mixer.
The one issue for which he gave me a definitive answer is the modeling of the PM emissions with SCREEN3. I explained that this is no longer a regulatory model and he told me that he was instructed to do use that model, so
that's what he is going to do.
I would recommend to include a statement indicating that SCREEN3 is no longer a regulatory model and that those results can't be used for a NAAQS compliance determination. I would also recommend to include an
explanation about that being a single-source model and that fugitive PM emissions can't be included in the analysis.
Finally, I recommend that you place this application on hold until you get all the necessary information from Aaron.
We can discuss this in more detail but I just wanted to let you know about these answers.
Thanks,
Rosendo Majano
EPAOIG Appendix 35
-------
State.co.us Executive Branch Mail - CC&V Modeling Results
EPAOIG Appendix 36
STATE OF
COLORADO
CC&V Modeling Results
1 message
Rosendo Thu, Mar 12, 2020 at 2:33 PM
To: Emmett
Emrnett,
You requested the maximum modeled concentrations on bo h the sou h and north portion of the CC&V Mine property boundary shown in the figure below and listed as 1-hr N02
Modeled NAAQS violations.
As you know, officially there aren't any modeled violations at the CC&V Mine. That's because of the 01/14/19 and 01/28/19 emails from Gordon Pierce requesting to remove the
concentration exceeding the NAAQS from the report and to replace them with a value that was lower and that was based on incorrect data. Therefore officially the highest
modeled concentra ion is of 187.7 ug/m3 (99.77 ppb). The NAAQS is of 100 ppb.
Reality however, is very different. The actual highest modeled design concentration in the southern area of the mine is of 229.34 ug/m3 (121 ppb). In the northern area of the
mine this concentration is of 225.78 ug/m3 (120 ppb). This is the information you are now requesting.
Those results above beg the question, if ttie Division is now acknowledging the modeled NAAQS violations, why would the permit be issued?
Wouldn't that create the exact same situation as the ColoWyo Mine permit that was challenged in court for being issued with a NAAQS modeled violation?
Rosendo Majano
Permit Modeling Work Lead
Modeling and Emissions Inventory Unit
IS
COLORADO
Air Pollution Control Division
Department of Public Health & Environment
11530
EPAOIG Appendix 36
https://mail.google.com/mail/u/0?ik=1144611be0&view=pt&search=all&permthid=thread-a%3Ar-7026110936250503657%7Cmsg-a%3Ar-58742828847... 1/1
-------
STATE OF
COLORADO
EPAOIG Appendix 37
H1H and McCormick
4 messages
Malone - CDPHE, Emmett|
To: Rosendo Majano - CDP
Cc: "Gordon.Pierce@state.co.us"!
I Bradley Rink-CDPHE|
¦ DeVondria Reynolds - CDPHEi
Tue, Sep 15,2020 at 1:26 PM
Hi,
Today's meeting with management is over. It was decided to use option 3 for the H1H question. The variation of this option chosen is , when there is not representative meteorological data the two most representative
meteorological data sets available for the site will be provided to the applicant/consultant. Both data sets will be modeled. The data set with the highest impacts will be used for the design concentration. The design concentration
will be the form of the standard.
I will let the industry workgroup know what was decided Sep 30. If there are no major concerns raised by industry we will implement the new policy Oct 1.
On the McCormick Asphalt Plant there was no conclusive answer on how to handle this type of situation other than we will continue to talk about it in the quarterly meetings I have with management. In McCormicks case since
the owner was open to raising the stacks I was asked to ask him to do so. But before doing so it was thought that we should do some quick model runs to get an idea how high the stacks would need to be raised to make sure
we are not asking something that is not practical. I have asked DeVondria to do some runs to get an idea of how high the stacks need to be.
I called Steve McCormick and let him know what was going on and he was OK with the approach.
Let me know if you have any questions.
Emmett Malone
Supervisor
Modeling and Emissions Inventory Unit
Technical Services Program
Air Pollution Control Division
Colorado Department of Public Health and Environment
APCD-TS-B1
4300 Cherry Creek Drive South
Denver, CO 80246-1530
"Are you curious about ground-level ozone in Colorado? Visit our ozone webpage to learn more."
Rosendo Majanol
To: "Malone - CDHHt, tmmett'
Cc: Bradley Rink - CDPHEi
Tue, Sep 15,2020 at 3:15 PM
DeVondria Reynolds - CDPHEi
"G ordo n. P ierce@state .co. us" I
I was taking a look at all the information and it seems to me that the issues with the McCormick plant are not only with its design, but also with its location. There is an ethanol plant located about a kilometer away that is causing
modeled violations of the 24-hr PM2.5 NAAQS and that might also be causing high 1-hr N02 impacts. If that's the case raising the stacks at McCormick might not solve the problem.
My recollection is that no nearby sources were included in the 1-hr N02, and if that's the case the impacts from McCormick don't have to be that high to contribute to modeled violations of these standards.
There is also the problem that the modeling for PM did not include any fugitive emissions. McCormick submitted modeling for Carbon Monoxide only, which means that they only modeled point sources, so they did not provide
any information that would allow us to model fugitives (e.g. haul road traffic, materials handling, etc.).
What has been modeled so far is the PM emissions from the stacks, and the 24-hr PM2.5 impacts from those emissions alone were fairly high, although still below the corresponding NAAQS. That's why the draft table of results
includes the following conclusion: Although a complete analysis could not be completed for PM fugitive emissions it is possible that the facility will contribute to a model violation of the PM2.5 NAAQS standard if additional
fugitive emissions are added to the modeling including background.
In my opinion, raising the stacks might be a solution only for the 1-hr and 3-hr S02 NAAQS modeled violations (unless the ethanol plant is burning coal or other fuel with high sulfur content), but for 1-hr N02 and 24-hr PM2.5 I
think that full cumulative analyses would be needed for to determine if the McCormick plant would contribute to modeled violations.
The test runs with different stack heights will give us only partial answers and will leave us with the uncertainty of what the cumulative impacts will be, so it might be better if we request from the applicant to submit full modeling
analyses for all the troubled pollutants and averaging periods. That way we will have certainty of where this facility stands in terms of NAAQS compliance.
Rosendo Majano
[Quoted text hidden]
Reynolds - CDPHE, DeVbndria[
To: Rosendo Majanoj
Cc: "Malone - CDPHE, tmmett
All,
Tue, Sep 15,2020 at 5:16 PM
, Bradley Rink - CDPHEi
Please see the attached document with my findings.
Please let me know if you have any questions or comments.
Rassndfii
I requested nearby source inventory from David and both the previous modeling results and these include nearby sources for all pollutants, you will see that if the 'HMA' stack is raised to 20m then cumulative modeling is no
longer required for CO (the only pollutant they were instructed to model, but all other pollutants that they were not instructed to model will require cumulative modeling *ironic I know...)
I do agree that a complete analysis should be done especially for PM emissions but I do not think we should be doing it for them, at all. The applicant should resubmit their own modeling for testing out raising stacks, additional
PM emissions, etc.
Thanks.
DeVondria Reynolds, MS
Air Quality Modeler
Modeling and Emissions Inventory Unit
Technical Services Program
&
COLORADO
Air Pollution Control Division
Department of Public Health b Environment
Learn more about COVID-19 https //www Colorado gov/pacific/cdphe/apcd-covid-19
Are you curious about ground-level ozone in Colorado? Visit our ozone webpage to learn more
[Quoted text hidden]
McCormick Asphalt NAAQS Analysis Summary-raised stack, do cx
EPAOIG Appendix 37
-------
EPAOIG Appendix 38
Reynolds - CDPHE, DeVondrial
To: Rosendo MajanoJ
Cc: "Malone - CDPHE, tmmeu
Tue, Sep 15,2020 at 5:45 PM
.Bradley Rink-CDPHEi
" Gordo n. P ierce@state .co. us" I
DeVondria Reynolds, MS
Air Quality Modeler
Modeling and Emissions Inventory Unit
Technical Services Program
Learn more about COVID-19 https //www Colorado gov/pacific/cdphe/apcd-covid-19
Are you curious about ground-level ozone in Colorado? Visit our ozone webpage to learn more
[Quoted text hidden]
EPAOIG Appendix 38
-------
EPAOIG Appendix 39
STATE OF
COLORADO
McCormick Asphalt Plant
Bradley Rink - CDPHEl
Rosendo Majano - CDPHE
Wed, Aug 26,2020 at 2:57 PM
Hi Matt,
I wanted to give you a heads up. McCormick Asphalt Plant was told to only model CO. When the permit modelers reviewed the application they became concerned about other pollutants because of the nearby sources, low
stack heights, and the dirty fuel the plant plans on using.
As I understand it the plant causes or contributes to exceedances of the NAAQS for:
They are using representative meteorological data so these exceedances are using the form of the standard not the H1H. I have attached tables DeVondria has created showing the results. PM was not modeled with fugitive
emissions therefore a refined analysis may have even higher impacts. The tables shows that the exceedances are significantly over the NAAQS.
My question becomes do we (Air Division) want to have the source address these impacts or have DeVondria to write her report saying the source was only required to model CO but the source contributes to or causes modeled
exceedances of the NAAQS for
PM2.5 24hr
S02 8hr
S02 1 hr
N02 1 hr
Attached are some tables showing the modeling results.
Let me know what you think.
Emmett Malone
Supervisor
Modeling and Emissions Inventory Unit
Technical Services Program
Air Pollution Control Division
Colorado Department of Public Health and Environment
APCD-TS-B1
4300 Cherry Creek Drive South
Denver, CO 80246-1530
"Are you curious about ground-level ozone in Colorado? Visit our ozone webpage to learn more."
rgpi McCormick Asphalt NAAQS Analysis Summary-2.docx
^ am m
PM2.5 24hr
S02 8hr
S02 1hr
N02 1 hr
EPAOIG
Appendix
39
-------
State.co.us Executive Branch Mail - CC&V
EPAOIG Appendix 40
STATE OF
COLORADO
CC&V
Pierce - CDPHE, Gordon
To: Rosendo Majano
Cc: Emmett Malonef
Rosendo,
Mon, Jan 14,2019 at 2:24 PM
Please see below for my requested edits in red. The primary one involves the 1-hour N02. While I agree that the use of OLMGROUP ALL is incorrect (per the June Modeler's
Workshop), that was an error on our part as we told CC&V to model it that way (as you mention in the writeup). As we have done for other applicants when there is an error
on our part, we do not request it be fixed unless there is a need to re-model. Thus, the report summary table should reflect the values without that error being corrected. I do
believe that this error should be discussed in the writeup following the summary table, as you have done. If you could please send a revised version for review, I would
appreciate it.
Thanks,
Gordon
The Division's Modeling and Emissions Inventory Unit (MEIU) received on April 23, 2018 a dispersion modeling analysis submitted by Cripple Creek & Victor Gold Mining
Company (CC&V) as part of the application to modify the Construction Permit NO.98TE0545 and implement the Mine Life Extension #2 Cresson Project in Teller County,
CO.
Review of that modeling analysis was finalized on June 2018 and a final report was written on 06/26/2018 concluding that the proposed project at the Cripple Creek & Victor
facility will cause modeled violations of the 1-hr NO2 NAAQS. In addition, it was concluded that there were several critical errors in the modeling analysis that should require
correction before it could be determined whether the proposed project will or will not cause and/or contribute to modeled violations of the rest of applicable NAAQS.
After fulfilling administrative requirements of gathering all the communications records for the project, entering all the application information into the Modeling Unit's
database, generating a review checklist report, and creating a billing report, the final modeling review report was provided to the permit engineer, Jonathan Brickey, on
07/09/2018.
On 08/13/2018 the Modeling Unit MEIU manager, Emmett Malone, forwarded two emails from the Air Pollution Control Division (APCD) director, Garry Kaufman, in which he
provided arguments to reject the main findings and conclusions of the CC&V final modeling review report in what pertains to the engine loads and N02/NOx in-stack ratios
for the non-road diesel engines, and also provided instructions to review the modeling analysis with the originally submitted emission rates and to accept the originally
submitted N02/NOx in-stack ratios for the non-road diesel engines. All the aforementioned emails are included in Appendix 1 of this document.
In compliance with the APCD director's instructions, CC&V's 04/23/18 modeling submittal has been reviewed for a second time leaving intact the original emission rates and
in-stack ratios of the non-road diesel engines. In addition, an analysis of the APCD director's arguments to approve CC&V's data has also been conducted to determine if
they are supported by sound scientific principles and applicable regulations and guidance. Such analysis is included in a subsequent section of this document. The results of
this new review are provided below.
There are NAAQS and CAAQS currently in effect for the following pollutants and averaging periods: CO (1-hrand 8-hr), Pb (3-month rolling average), N02(1-hrand annual),
PM10 (24-hr), PM2.5 (24-hr and annual), SO2 (1-hr and 3-hr), and ozone (8-hr). For this permit application, quantitative impact analyses are warranted to demonstrate
compliance with the NAAQS/CAAQS for the following pollutants and averaging periods: NO2 (1-hr and Annual), PM2.5 (24-hr and Annual), PM10; CO (1-hr and 8-hr); and
S02(1-hr and 3-hr).
The applicant submitted modeling analyses for all required pollutants and averaging periods, and the results of the modeling analyses are summarized below:
Pollutant
Averaging Period
Maximum Modeled
Concentration (|jg/m3)
Background Concentration
(|jg/m3)
Total Impact
(|jg/m3)
NAAQS
(|jg/m3)
PM2.5
24-hr
19.72
12.7
32.42
35
PM2.5
Annual
5.36
6.0
11.36
12
PM10
24-hr
124.03
24.0
148.03
150
S02
1-hr
34.5
86.4
120.9
196.5
S02
3-hr
19.6
62.8
82.4
700
no2
1-hr
220.34 (use values
with OLMGROUP
ALL) with background
Seasonal-hourly profile
220.34 (use
values with
OLMGROUP
ALL)
188.34
N02
Annual
37.4
28.22
65.62
100
CO
1-hr
8,509.3
4580
13,089.3
40,000
CO
8-hr
628.3
2290
2918.3
10,000
The meteorological data used was adequately representative of the dispersion conditions at the mine and therefore the design concentrations of the results listed above
match the original form of the corresponding NAAQS.
For 24-hr PM2.5 the modeled design concentrations is the highest eighth high 24-hr concentration averaged across the 3 years of meteorological data.
For annual PM2.5 the modeled design concentration is the maximum annual concentration averaged across the 3 years of meteorological data.
For 24-hr PM10 the modeled design concentration is the highest fourth high 24-hr concentration of the 3-year period of meteorological data.
For 1 -hr SO2 the modeled design concentration is the highest fourth high maximum daily 1 -hr concentration averaged across the 3 years of meteorological data.
EPAOIG Appendix 40
https://mail.google.com/mail/u/0?ui=2&ik=1144611be0&view=lg&permmsgid=msg-f:1622672522560002904 1/7
-------
State.co.us Executive Branch Mail - CC&V
For 3-hr SO2 the modeled design concentration is the highest second high 3-hr concentration of the 3-year period of meteoro^icc^Sl^ App©ndiX 41
For 1 -hr NO2 the modeled design concentration is the highest eighth high maximum daily 1 -hr concentration averaged across the 3 years of meteorological data.
For annual NO2 the modeled design concentration is the highest annual concentration of the 3-year period of meteorological data.
For 1 -hr CO the modeled design concentration is the highest second high 1 -hr concentration of the 3-year period of meteorological data.
For 8-hr CO the modeled design concentration is the highest second high 8-hr concentration of the 3-year period of meteorological data.
It is noted that the meteorological data was reprocessed with AERMET v18081, and AERMOD v18081 was used to conduct all the final runs. This version of AERMET
includes a correction in the Bulk Richardson algorithm when using on-site meteorological data, a situation that applies directly to the CC&V case. CC&V submitted this
modeling analysis using meteorological data processed with the previous version, V16216r, which was correctly the latest version available at the time of submittal.
After review of the information submitted by the applicant with revisions by the MEIU, it is concluded that the CC&V gold mine will cause modeled violations of the 1-hr N02
NAAQO. After review of the information submitted by the applicant, and requested by the Division Director, with revisions by the MEIU, it is concluded that the CC&V gold
mine will/will not cause modeled violations of the 1-hr N02 NAAQS/any pollutants.
The 1-hr NO2 analysis that CC&V submitted was performed with the Ozone Limiting Method and the "OLMGROUP ALL" option by grouping the representation of multiple
operating scenarios into single AERMOD runs through the use of source groups.
The applicant has indicated that blasts do not occur in more than one pit at the same time, consequently, CC&V was instructed by the MEIU to model this activity as taking
[11
place in each pit at a time. MEIU further indicated that this could be accomplished with separate model runs per scenario or by using source groups in AERMOD
However, during the June, 2018 EPA Regional/State/Local Modelers Workshop, James Thurman from EPA's OAQPS office did a presentation on the use of source groups
in AERMOD and explicitly indicated that source groups should not be used when modeling NO2 with the Ozone Limiting Method using the OLMGROUP ALL option because
the model would underestimate impacts.
After this presentation, AERMOD's documentation was reviewed and it was verified that in fact, the Ozone Limiting Method when using the OLMGROUP ALL option will
distribute the available ambient ozone equally among all of the emission sources declared in the AERMOD input file, without any regard for which sources are included or
not included in the different source groups. The consequence of this situation is that the background ozone concentration provided in the input file would in practice be
diluted, thus resulting in less conversion of NOx to NO2 and therefore in lower modeled NO2 concentrations.
Consequently, If this error was corrected by the MEIU by separating the different operating scenarios and reviewing them by conducting a separate, individual AERMOD run
per each individual scenario. This allowed , then the available ozone to would be distributed only among those sources that are active on each operating scenario. Applying
this correction, modeled 1-hour N02 concentrations would be in excess of 225 |jg/m , well above the NAAQS.
In addition to this problem, the design concentration for the 1-hr NO2 NAAQS involves doing a receptor-specific average across the different years of meteorological data
used. This average is generally done as post-processing after the AERMOD runs, but CC&V did not provide any evidence of doing such average, so it was performed by the
MEIU on all the final 1-hr NO2 runs during the review process.
The modeled results for 1-hr NO^ are above the corresponding standard, but it is noted that they are expected to be even higher if the issues discussed below were to be
corrected.
Although the engine loads used by CC&V have been approved by the APCD director, they were calculated using statistical techniques that are inadequate and that have
skewed the results. If these calculations were corrected, it is expected that the emission rates will increase and so would the resulting modeled concentrations. More details
on this topic are provided in a subsequent section of this document.
Also, the NO2/NOX in-stack ratios used for some of the non-road diesel engines are of 0.2 and 0.01; values that are not technically supported for the type of engines in
question. While these values were approved by the APCD director, the Division's subject matter experts have not been able to find any scientific literature or technical
documentation to support them. Values as low as 0.22 have been documented for on-road engines only, and the 0.01 value simply defies logic is highly suspect for any type
of mobile source diesel engine, with or without emission control devices.
EPA recommends the use of a NCyNOx in-stack ratio value of 0.5 for any case for which a different value cannot be adequately justified.
For the rest of the pollutants and averaging periods the following conclusions have been reached:
- 24-hr and annual PM2.5.
While the results are numerically below the corresponding NAAQS, compliance with such standards cannot be determined.
The emission rates used by CC&V are based on emission factors that have not been adjusted with a deterioration factor to account for the increase in emissions resulting
from usage as the engine ages, and also with a transient adjustment factor to account for the change in emissions due to transient demands of the engine.
Both adjustment factors are included in the data base of EPA's mobile sources emissions model MOVES, and the resulting emission rates are higher after they are applied.
However, CC&V did not use MOVES and instead used what appear to be only zero-hour steady state factors, thus adding another layer of error in the calculation of mobile
engine emissions. More details on this topic can be found in Appendix 2 of this document.
Considering that the modeled concentrations for 24-hr and annual PM2.5 are at 92.6% and 94.6% respectively of their corresponding NAAQS, it is very feasible that once
these adjustments are applied, the resulting concentrations could reach or exceed the standards.
- 24-hr PM10.
While the results are numerically below the corresponding NAAQS, compliance with such standards cannot be determined.
Similar to the case of PM2.5, the emission rates used by CC&V are based on emission factors that have not been adjusted with a deterioration factor to account for the
increase in emissions resulting from usage as the engine ages, and also with a transient adjustment factor to account for the change in emissions due to transient demands
of the engine.
EPAOIG Appendix 41
https://mail.google.com/mail/u/0?ui=2&ik=1144611be0&view=lg&permmsgid=msg-f:1622672522560002904
2/7
-------
3/28/2021 State.co.us Executive Branch Mail - CC&V
Both adjustment factors are included in the data base of EPA's mobile sources emissions model MOVES, and the resulting applied.
However, CC&V did not use MOVES and instead used what appear to be only zero-hour steady state factors, thus adding another layer of error in the calculation of mobile
engine emissions. More details on this topic can be found in Appendix 2 of this document.
Considering that the modeled concentration is at 98.7% of the NAAQS, it is likely that once these adjustments are applied, the resulting concentration could reach or exceed
the standards.
- Annual N02.
While the results are numerically below the corresponding NAAQS, compliance with such standards cannot be determined.
Although the engine loads used by CC&V have been approved by the APCD director, they were calculated using statistical techniques that are inadequate and that have
skewed the results. If these calculations were corrected it is expected that the emission rates will increase and so would the resulting modeled concentrations. More details
on this topic are provided in a subsequent section of this document.
In addition, the NO2/NOX in-stack ratios used for some of the non-road diesel engines are of 0.2 and 0.01, values that are not technically supported for the type of engines in
question. While these values were approved by the APCD director, the Division's subject matter experts have not been able to find any scientific literature or technical
documentation to support them. Values as low as 0.22 have been documented for on-road engines only, and the 0.01 value simply defies logic is highly suspect for any type
of mobile source diesel engine, with or without emission control devices.
EPA recommends the use of a N02/NOx in-stack ratio value of 0.5 for any case for which a different value cannot be adequately justified.
Currently the modeled concentration is at 65.8% of the NAAQS, and it is impossible to predict how much it would increase after applying these corrections.
- 1-hrand 8-hr CO.
While the results are numerically below the corresponding NAAQS, compliance with such standards cannot be determined is likely demonstrated.
Similar to the case of N02, although the engine loads used by CC&V have been approved by the APCD director, they were calculated using statistical techniques that are
inadequate and that have skewed the results. If these calculations were corrected it is expected that the emission rates will increase and so would the resulting modeled
concentrations. However, considering that the current results are at 29.2% and 32.7% for the 8-hr and 1-hr standards respectively, it is unlikely that the corrections would
cause increases in modeled concentrations such that these NAAQS would be reached or exceeded.
-1 -hr and 3-hr S02.
Compliance with these NAAQS has been demonstrated.
While the engine loads were calculated incorrectly as described above for the N02 and CO cases, S02 emissions are primarily a function of the sulfur content in the fuel,
and the influence of the engine load is insignificant. Considering that the current results are at 11.8% and 61.5% of the 3-hr and 1-hr NAAQS respectively, applying the
corrections will not raise the emissions enough to cause modeled violations of these standards.
Non-road engine loads and emissions.
With regards to the non-road engine loads and emissions, the APCD director's emails (See Appendix 1) require some clarifications.
When discussing non-road engine loads and emissions the director's emails described the estimation of actual worst-case emissions from non-road engines as an
extremely complex challenge, referring to CC&V's methodology as a fairly simplified approach and to the corrections applied by the MEIU as a more sophisticated approach.
CC&V's approach consisted in estimating engine loads based on fuel consumption and engine usage data for a period of 3 years. This methodology is in principle
scientifically sound, but CC&V failed to use basic statistical techniques in the handling of the raw data thus skewing the resulting calculations of the engine loads.
The MEIU on the other hand, did not propose a different methodology for estimating emissions, but only proposed to use the correct statistical techniques to organize and
analyze the raw data. So there has never been a simplified approach or a sophisticated approach. The methodology for estimating engine loads and emission rates based
on fuel consumption and engine usage is essentially the same, and the only difference has been in the statistical processing of the raw data.
Barring honest mistakes, it is a reasonable expectation that the technical analyses submitted by permit applicants comply with basic scientific and mathematical principles,
and that expectation has not been met in CC&Vs case when processing the raw data in the calculation of engine loads and emission rates.
NOg/NOx in-stack ratios.
On the APCD director's emails (See Appendix 1) he indicates that CC&V submitted information from the representative for the engine manufacturer stating that in-stack ratio
of between 0.15 and 0.20 would be reasonable for four of the engines and that an in-stack ratio of less than 0.01 would be reasonable for the fifth engine. This information
consists resides in an email from a Caterpillar dealership, Wagner Equipment (www.wagnerequipment.com), which included a table in which these N02/NOx in-stack ratios
were listed for the specific Tier 4 engines owned by CC&V.
In general, the approval process for the N02/NOx in-stack ratios starts with the permit applicant proposing source-specific values for their emission units, and providing
supporting documentation to justify such values. Such supporting documentation is generally a reference to EPA's in-stack ratios database, copy of documentation prepared
by federal or State regulatory agencies, reference to or copy of scientific or technical literature citing results of studies conducted on the topic, manufacturer's specification
sheets, or stack test reports (tailpipe emission testing in this case).
The first and the last of this list are commonly the most reliable sources of documentation, and they are essentially the same type of data, as EPA has in the last few years
undertaken the effort of collecting and validating stack testing results for different types of sources and compiling them in a database. Unfortunately for non-road diesel
engines there is no such information available.
Stack tests, in the manner defined by EPA, are not performed on non-road engines or in general on mobile engines. Instead, for mobile sources emissions testing is
conducted in a laboratory where the entire vehicle is placed in a dynamometer to apply load to the engine and emissions are collected and analyzed during specific driving
cycles. For heavy duty engines, on-road and non-road, the engine alone, removed from the vehicle, is placed on an engine test bench where load is applied directly to the
engine axle while operated at specific cycles as emissions are collected and analyzed.
Thus, as opposed to stack testing on stationary engines, emission testing on non-road engines is not something that is done by the owner or operator of the actual permitted
emission unit. This type of tests are performed only by the engine manufacturer (published in specification sheets or equipment manuals), by large regulatory agencies like
EPA or CARB (California Air Resources Board), or by research institutions on some emission units representative of specific engine types. And while emission factors and
EPAOIG Appendix 42
https://mail.google.com/mail/u/0?ui=2&ik=1144611be0&view=lg&permmsgid=msg-f:1622672522560002904
3/7
-------
3/28/2021 State.co.us Executive Branch Mail - CC&V
limits have been developed for non-road engines with this type of tests, specific data to derive the N02/N0X in-stack ratios^js*either
because NOx measurements were not conducted during the tests or because the data was not of interest at the time and not published.
The remaining source of data to support specific N02/N0X in-stack ratios are the manufacturers or scientific literature. MEIU staff researched the manuals and specification
sheets available to the public for the type of engines used by CC&V. The research also included EPA and CARB documentation, and a fair amount of scientific literature
discussing this topic; and none of these documents contained any N02/N0X in-stack ratio information applicable to CC&Vs engines.
However, there were several of these documents discussing NOx and NO2 emissions from mobile diesel engines, and all of them consistently explained that the operating
principle of the emissions control devices that make Tier 4 diesel engines much cleaner, would also result in increased N02/N0X in-stack ratios.
As explained in the 06/26/2018 final modeling review report, Diesel engines without these control devices (i.e. Tier 3 engines and below) typically have an N02/N0X in-stack
ratio of about 0.1, but the oxidation catalyst and the regeneration mechanism of the particulate filter will have the side effect of oxidizing a large fraction of the NOx emissions
into NO2 therefore increasing substantially the ratio. As it also was explained in that report, there is no mechanism by which that ratio would be reversed to its original value
or lower. The reduction catalyst will reduce both NO2 and NOx into elemental nitrogen, and there is nothing in the available literature to suggest that one chemical reaction
occurs at a faster rate than the other one to alter the final N02/N0X ratio significantly. On the contrary, all the available literature points to an ideal ratio of 0.5 for the final
reduction reaction to be optimal and to a final ratio well above the original 0.1 that existed before the exhaust went through the emissions control devices.
Sound scientific principles and judgement support a ratio much higher than the ones proposed by CC&V, and accepting their proposed values as recommended by the
APCD director is not scientifically defensible unless testing data is provided to support them. This is particularly true for the 0.01 ratio, which is not defensible even for the
older Tier 1 through 3 engines.
[1]
09/19/2014 Modeling Review Report-List of Outstanding Issues - Cripple Creek & Victor Gold Mining Company - Mine Life Extension #2 Cresson Project
On Thu, Jan 10,2019 at 10:22 AM Rosendo wrote:
Emmett - As you requested, below is the full contenunamilre^ncluae^^ii^C&V report.
Rosendo Majano
The Division's Modeling and Emissions Inventory Unit (MEIU) received on April 23, 2018 a dispersion modeling analysis submitted by Cripple Creek & Victor Gold Mining
Company (CC&V) as part of the application to modify the Construction Permit NO.98TE0545 and implement the Mine Life Extension #2 Cresson Project in Teller County,
CO.
Review of that modeling analysis was finalized on June 2018 and a final report was written on 06/26/2018 concluding that the proposed project at the Cripple Creek &
Victor facility will cause modeled violations of the 1-hr NO2 NAAQS. In addition, it was concluded that there were several critical errors in the modeling analysis that
require correction before it could be determined whether the proposed project will or will not cause and/or contribute to modeled violations of the rest of applicable
NAAQS.
After fulfilling administrative requirements of gathering all the communications records for the project, entering all the application information into the Modeling Unit's
database, generating a review checklist report, and creating a billing report, the final modeling review report was provided to the permit engineer, Jonathan Brickey, on
07/09/2018.
On 08/13/2018 the Modeling Unit manager, Emmett Malone, forwarded two emails from the Air Pollution Control Division (APCD) director, Garry Kaufman, in which he
provided arguments to reject the main findings and conclusions of the CC&V final modeling review report in what pertains to the engine loads and N02/NOx in-stack ratios
for the non-road diesel engines, and also provided instructions to review the modeling analysis with the originally submitted emission rates and to accept the originally
submitted N02/NOx in-stack ratios for the non-road diesel engines. All the aforementioned emails are included in Appendix 1 of this document.
In compliance with the APCD director's instructions, CC&Vs 04/23/18 modeling submittal has been reviewed for a second time leaving intact the original emission rates
and in-stack ratios of the non-road diesel engines. In addition, an analysis of the APCD director's arguments to approve CC&V's data has also been conducted to
determine if they are supported by sound scientific principles and applicable regulations and guidance. Such analysis is included in a subsequent section of this
document. The results of this new review are provided below.
There are NAAQS and CAAQS currently in effect for the following pollutants and averaging periods: CO (1-hr and 8-hr), Pb (3-month rolling average), NO2 (1-hr and
annual), PM10 (24-hr), PM2.5 (24-hr and annual), SO2 (1-hr and 3-hr), and ozone (8-hr). For this permit application, quantitative impact analyses are warranted to
demonstrate compliance with the NAAQS/CAAQS for the following pollutants and averaging periods: NO2 (1-hr and Annual), PM2.5 (24-hr and Annual), PM10; CO (1-hr
and 8-hr); and SO2 (1-hr and 3-hr).
The applicant submitted modeling analyses for all required pollutants and averaging periods, and the results of the modeling analyses are summarized below:
Pollutant
Averaging Period
Maximum Modeled
Background Concentration
Total Impact
NAAQS
Concentration (|jg/m3)
(|jg/m3)
(|jg/m3)
(|jg/m3)
PM2.5
24-hr
19.72
12.7
32.42
35
PM2.5
Annual
5.36
6.0
11.36
12
PM10
24-hr
124.03
24.0
148.03
150
S02
1-hr
34.5
86.4
120.9
196.5
S02
3-hr
19.6
62.8
82.4
700
N02
1-hr
229.34 with background
Seasonal-hourly profile
229.34
188.34
N02
Annual
37.4
28.22
65.62
100
CO
1-hr
8,509.3
4580
13,089.3
40,000
E
PAOIG Appendix 4
3
https://mail.google.com/mail/u/0?ui=2&ik=1144611be0&view=lg&permmsgid=msg-f:1622672522560002904
4/7
-------
State.co.us Executive Branch Mail - CC&V
| CO | 8-hr | 628.3 | 2290 | 2918.3 [
^PpAOIG (Appendix 44
The meteorological data used was adequately representative of the dispersion conditions at the mine and therefore the design concentrations of the results listed above
match the original form of the corresponding NAAQS.
For 24-hr PM2.5 the modeled design concentrations is the highest eighth high 24-hr concentration averaged across the 3 years of meteorological data.
For annual PM2.5 the modeled design concentration is the maximum annual concentration averaged across the 3 years of meteorological data.
For 24-hr PM10 the modeled design concentration is the highest fourth high 24-hr concentration of the 3-year period of meteorological data.
For 1-hr SO2 the modeled design concentration is the highest fourth high maximum daily 1-hr concentration averaged across the 3 years of meteorological data.
For 3-hr SO2 the modeled design concentration is the highest second high 3-hr concentration of the 3-year period of meteorological data.
For 1-hr NO2 the modeled design concentration is the highest eighth high maximum daily 1-hr concentration averaged across the 3 years of meteorological data.
For annual NO2 the modeled design concentration is the highest annual concentration of the 3-year period of meteorological data.
For 1-hr CO the modeled design concentration is the highest second high 1-hr concentration of the 3-year period of meteorological data.
For 8-hr CO the modeled design concentration is the highest second high 8-hr concentration of the 3-year period of meteorological data.
It is noted that the meteorological data was reprocessed with AERMET v18081, and AERMOD v18081 was used to conduct all the final runs. This version of AERMET
includes a correction in the Bulk Richardson algorithm when using on-site meteorological data, a situation that applies directly to the CC&V case. CC&V submitted this
modeling analysis using meteorological data processed with the previous version, V16216r, which was correctly the latest version available at the time of submittal.
After review of the information submitted by the applicant with revisions by the MEIU, it is concluded that the CC&V gold mine will cause modeled violations of the 1-hr
N02 NAAQS.
The 1-hr NO2 analysis that CC&V submitted was performed with the Ozone Limiting Method and the "OLMGROUP ALL" option by grouping the representation of multiple
operating scenarios into single AERMOD runs through the use of source groups.
The applicant has indicated that blasts do not occur in more than one pit at the same time, consequently, CC&V was instructed by the MEIU to model this activity as taking
[11
place in each pit at a time. MEIU further indicated that this could be accomplished with separate model runs per scenario or by using source groups in AERMOD
However, during the June, 2018 EPA Regional/State/Local Modelers Workshop, James Thurman from EPA's OAQPS office did a presentation on the use of source
groups in AERMOD and explicitly indicated that source groups should not be used when modeling NO2 with the Ozone Limiting Method using the OLMGROUP ALL
option because the model would underestimate impacts.
After this presentation, AERMOD's documentation was reviewed and it was verified that in fact, the Ozone Limiting Method when using the OLMGROUP ALL option will
distribute the available ambient ozone equally among all of the emission sources declared in the AERMOD input file, without any regard for which sources are included or
not included in the different source groups. The consequence of this situation is that the background ozone concentration provided in the input file would in practice be
diluted, thus resulting in less conversion of NOx to NO2 and therefore in lower modeled NO2 concentrations.
Consequently, this error was corrected by the MEIU by separating the different operating scenarios and reviewing them by conducting a separate, individual AERMOD run
per each individual scenario. This allowed the available ozone to be distributed only among those sources that are active on each operating scenario.
In addition to this problem, the design concentration for the 1-hr NO2 NAAQS involves doing a receptor-specific average across the different years of meteorological data
used. This average is generally done as post-processing after the AERMOD runs, but CC&V did not provide any evidence of doing such average, so it was performed on
all the final 1-hr NO2 runs during the review process.
The modeled results for 1 -hr N02 are above the corresponding standard, but it is noted that they are expected to be even higher if the issues discussed below were to be
corrected.
Although the engine loads used by CC&V have been approved by the APCD director, they were calculated using statistical techniques that are inadequate and that have
skewed the results. If these calculations were corrected, it is expected that the emission rates will increase and so would the resulting modeled concentrations. More
details on this topic are provided in a subsequent section of this document.
Also, the NO2/NOX in-stack ratios used for some of the non-road diesel engines are of 0.2 and 0.01; values that are not technically supported for the type of engines in
question. While these values were approved by the APCD director, the Division's subject matter experts have not been able to find any scientific literature or technical
documentation to support them. Values as low as 0.22 have been documented for on-road engines only, and the 0.01 value simply defies logic for any type of mobile
source diesel engine, with or without emission control devices.
EPA recommends the use of a NO2/NOX in-stack ratio value of 0.5 for any case for which a different value cannot be adequately justified.
For the rest of the pollutants and averaging periods the following conclusions have been reached:
- 24-hr and annual PM2.5.
While the results are numerically below the corresponding NAAQS, compliance with such standards cannot be determined.
The emission rates used by CC&V are based on emission factors that have not been adjusted with a deterioration factor to account for the increase in emissions resulting
from usage as the engine ages, and also with a transient adjustment factor to account for the change in emissions due to transient demands of the engine.
Both adjustment factors are included in the data base of EPA's mobile sources emissions model MOVES, and the resulting emission rates are higher after they are
applied. However, CC&V did not use MOVES and instead used what appear to be only zero-hour steady state factors, thus adding another layer of error in the calculation
of mobile engine emissions. More details on this topic can be found in Appendix 2 of this document.
EPAOIG Appendix 44
https://mail.google.com/mail/u/0?ui=2&ik=1144611be0&view=lg&permmsgid=msg-f:1622672522560002904 5/7
-------
State.co.us Executive Branch Mail - CC&V
EPAOIG Appendix 45
Considering that the modeled concentrations for 24-hr and annual PM2.5 are at 92.6% and 94.6% respectively of their corresponding NAAQS, it is very feasible that once
these adjustments are applied the resulting concentrations could reach or exceed the standards.
-24-hr PM10.
While the results are numerically below the corresponding NAAQS, compliance with such standards cannot be determined.
Similar to the case of PM2.5, the emission rates used by CC&V are based on emission factors that have not been adjusted with a deterioration factor to account for the
increase in emissions resulting from usage as the engine ages, and also with a transient adjustment factor to account for the change in emissions due to transient
demands of the engine.
Both adjustment factors are included in the data base of EPA's mobile sources emissions model MOVES, and the resulting emission rates are higher after they are
applied. However, CC&V did not use MOVES and instead used what appear to be only zero-hour steady state factors, thus adding another layer of error in the calculation
of mobile engine emissions. More details on this topic can be found in Appendix 2 of this document.
Considering that the modeled concentration is at 98.7% of the NAAQS, it is likely that once these adjustments are applied the resulting concentration could reach or
exceed the standards.
-Annual N02.
While the results are numerically below the corresponding NAAQS, compliance with such standards cannot be determined.
Although the engine loads used by CC&V have been approved by the APCD director, they were calculated using statistical techniques that are inadequate and that have
skewed the results. If these calculations were corrected it is expected that the emission rates will increase and so would the resulting modeled concentrations. More
details on this topic are provided in a subsequent section of this document.
In addition, the NO2/NOX in-stack ratios used for some of the non-road diesel engines are of 0.2 and 0.01, values that are not technically supported for the type of engines
in question. While these values were approved by the APCD director, the Division's subject matter experts have not been able to find any scientific literature or technical
documentation to support them. Values as low as 0.22 have been documented for on-road engines only, and the 0.01 value simply defies logic for any type of mobile
source diesel engine, with or without emission control devices.
EPA recommends the use of a N02/NOx in-stack ratio value of 0.5 for any case for which a different value cannot be adequately justified.
Currently the modeled concentration is at 65.8% of the NAAQS, and it is impossible to predict how much it would increase after applying these corrections.
1 -hr and 8-hr CO.
While the results are numerically below the corresponding NAAQS, compliance with such standards cannot be determined.
Similar to the case of N02, although the engine loads used by CC&V have been approved by the APCD director, they were calculated using statistical techniques that are
inadequate and that have skewed the results. If these calculations were corrected it is expected that the emission rates will increase and so would the resulting modeled
concentrations. However, considering that the current results are at 29.2% and 32.7% for the 8-hr and 1-hr standards respectively, it is unlikely that the corrections would
cause increases in modeled concentrations such that these NAAQS would be reached or exceeded.
1 -hr and 3-hr S02.
Compliance with these NAAQS has been demonstrated.
While the engine loads were calculated incorrectly as described above for the N02 and CO cases, S02 emissions are primarily a function of the sulfur content in the fuel,
and the influence of the engine load is insignificant. Considering that the current results are at 11.8% and 61.5% of the 3-hr and 1-hr NAAQS respectively, applying the
corrections will not raise the emissions enough to cause modeled violations of these standards.
Non-road engine loads and emissions.
With regards to the non-road engine loads and emissions, the APCD director's emails (See Appendix 1) require some clarifications.
When discussing non-road engine loads and emissions the director's emails described the estimation of actual worst-case emissions from non-road engines as an
extremely complex challenge, referring to CC&Vs methodology as a fairly simplified approach and to the corrections applied by the MEIU as a more sophisticated
approach.
CC&Vs approach consisted in estimating engine loads based on fuel consumption and engine usage data for a period of 3 years. This methodology is in principle
scientifically sound, but CC&V failed to use basic statistical techniques in the handling of the raw data thus skewing the resulting calculations of the engine loads.
The MEIU on the other hand, did not propose a different methodology for estimating emission, but only proposed to use the correct statistical techniques to organize and
analyze the raw data. So there has never been a simplified approach or a sophisticated approach. The methodology for estimating engine loads and emission rates based
on fuel consumption and engine usage is essentially the same, and the only difference has been in the statistical processing of the raw data.
Barring honest mistakes, it is a reasonable expectation that the technical analyses submitted by permit applicants comply with basic scientific and mathematical principles,
and that expectation has not been met in CC&Vs case when processing the raw data in the calculation of engine loads and emission rates.
NO-./NOX in-stack ratios.
On the APCD director's emails (See Appendix 1) he indicates that CC&V submitted information from the representative for the engine manufacturer stating that in-stack
ratio of between 0.15 and 0.20 would be reasonable for four of the engines and that an in-stack ratio of less than 0.01 would be reasonable for the fifth engine. This
information consists in an email from a Caterpillar dealership, Wagner Equipment (www.wagnerequipment.com), which included a table in which these N02/NOx in-stack
ratios were listed for the specific Tier 4 engines owned by CC&V.
In general, the approval process for the N02/NOx in-stack ratios starts with the permit applicant proposing source-specific values for their emission units, and providing
supporting documentation to justify such values. Such supporting documentation is generally a reference to EPA's in-stack ratios database, copy of documentation
prepared by federal or State regulatory agencies, reference to or copy of scientific or technical literature citing results of studies conducted on the topic, manufacturer's
specification sheets, or stack test reports (tailpipe emission testing in this case).
The first and the last of this list are commonly the most reliable sources of documentation, and they are essentially the same type of data, as EPA has in the last few years
undertaken the effort of collecting and validating stack testing results for different types of sources and compiling them in a database. Unfortunately for non-road diesel
engines there is no such information available.
EPAOIG Appendix 45
https://mail.google.com/mail/u/0?ui=2&ik=1144611be0&view=lg&permmsgid=msg-f:1622672522560002904 6/7
-------
State.co.us Executive Branch Mail - CC&V
EPAOIG Appendix 46
Stack tests, in the manner defined by EPA, are not performed on non-road engines or in general on mobile engines. Instead, for mobile sources emissions testing is
conducted in a laboratory where the entire vehicle is placed in a dynamometer to apply load to the engine and emissions are collected and analyzed during specific
driving cycles. For heavy duty engines, on-road and non-road, the engine alone, removed from the vehicle, is placed on an engine test bench where load is applied
directly to the engine axle while operated at specific cycles as emissions are collected and analyzed.
Thus, as opposed to stack testing on stationary engines, emission testing on non-road engines is not something that is done by the owner or operator of the actual
permitted emission unit. This type of tests are performed only by the engine manufacturer (published in specification sheets or equipment manuals), by large regulatory
agencies like EPA or CARB (California Air Resources Board), or by research institutions on some emission units representative of specific engine types. And while
emission factors and limits have been developed for non-road engines with this type of tests, specific data to derive the N02/N0X in-stack ratios has not been found from
EPA or CARB, either because NOx measurements were not conducted during the tests or because the data was not of interest at the time and not published.
The remaining source of data to support specific N02/N0X in-stack ratios are the manufacturers or scientific literature. MEIU staff researched the manuals and
specification sheets available to the public for the type of engines used by CC&V. The research also included EPA and CARB documentation, and a fair amount of
scientific literature discussing this topic; and none of these documents contained any N02/N0X in-stack ratio information applicable to CC&Vs engines.
However, there were several of these documents discussing NOx and NO2 emissions from mobile diesel engines, and all of them consistently explained that the operating
principle of the emissions control devices that make Tier 4 diesel engines much cleaner, would also result in increased N02/N0X in-stack ratios.
As explained in the 06/26/2018 final modeling review report, Diesel engines without these control devices (i.e. Tier 3 engines and below) typically have an N02/N0X in-
stack ratio of about 0.1, but the oxidation catalyst and the regeneration mechanism of the particulate filter will have the side effect of oxidizing a large fraction of the NOx
emissions into NO2 therefore increasing substantially the ratio. As it also was explained in that report, there is no mechanism by which that ratio would be reversed to its
original value or lower. The reduction catalyst will reduce both NO2 and NOx into elemental nitrogen, and there is nothing in the available literature to suggest that one
chemical reaction occurs at a faster rate than the other one to alter the final N02/N0X ratio significantly. On the contrary, all the available literature points to an ideal ratio
of 0.5 for the final reduction reaction to be optimal and to a final ratio well above the original 0.1 that existed before the exhaust went through the emissions control
devices.
Sound scientific principles and judgement support a ratio much higher than the ones proposed by CC&V, and accepting their proposed values as recommended by the
APCD director is not scientifically defensible unless testing data is provided to support them. This is particularly true for the 0.01 ratio, which is not defensible even for the
older Tier 1 through 3 engines.
[1]
09/19/2014 Modeling Review Report - List of Outstanding Issues - Cripple Creek & Victor Gold Mining Company - Mine Life Extension #2 Cresson Project
Gordon Pierce
Program Manager
Technical Services Program
/VBg^ COLORADO
Air Pollution Control Division
* I Department of Public Health & Environment
4300 Cherry Creek Drive South, Denver, CO 80246-1530
Are you curious about ground-level ozone in Colorado? Visit our ozone webpage to learn more.
EPAOIG Appendix 46
https://mail.google.com/mail/u/0?ui=2&ik=1144611be0&view=lg&permmsgid=msg-f:1622672522560002904
7/7
-------
State.co.us Executive Branch Mail - Re: CC&V
EPAOIG Appendix 47
STATE OF
COLORADO
Re: CC&V
1 message
Rosendo Majano^ Tue, Jan 29, 2019 at 11:08 AM
To: "Pierce - CDPI
Cc: Emmett Malonel
Bcc: Marie Bernardo
Gordon,
Just so it's clear in the record, your statement below about not catching the source group error in the June 2018 review is not accurate. I did catch that error in the June 2018
review. There is an explanation of this issue on Page 5 of the 06/26/18 report. Garry also addressed this issue on his 08/13/18 email to you.
However, per your request, I have reported in the table the results prior to the correction.
Rosendo Majano
On Mon, Jan 28, 2019 at 9:34 AM Pierce - CDPHE, Gordon wrote:
Rosendo,
Thanks for your response and clarifications. You are correct that my understanding and statement on the OLMGROUP ALL was not accurate and that the way the source
groups were combined into a single run is really the issue.
I will stand by my statement that since we did not catch the error in our June 2018 review, it should be allowed to stand in this review. This also follows Garry's request. I do
believe there is precedent based on what you listed (and I believe from GCC modeling in the past as well). As such, I would like the results reported without that correction in
the table. And, as I requested (and you mention), please make sure that the results with the correction are part of the write-up in the report. As you say, and I agree, it is not
correct to conceal or downplay it.
For my request on the CO conclusions, it was based on the last sentence you had written in that paragraph, which I was seeing as contradictory. If the paragraph is amended,
as you suggest, that would be fine.
Thanks,
Gordon
On Wed, Jan 16, 2019 at 7:24 AM Rosendo Majano wrote:
Gordon,
I'm afraid you have not understood the error with the 1-hr N02 modeling because what you describe in your email is inaccurate and what you have requested in the table of
results is already there. So I want to clarify the following:
1- The use of OLMGROUP ALL is not incorrect, at all. I did not make any statement to the contrary, nor did EPA on the Modeler's Workshop.
2- The 1-hr N02 results that I provided on the table were already obtained using the OLMGROUP ALL option in AERMOD.
The error in CC&V's N02 modeling consists in representing different operating scenarios in single AERMOD runs with source groups while at the same time using the
Ozone Limiting Method with the OLMGROUP ALL option. That combination of the use of source groups and OLMGROUP ALL is what has the effect of diluting the
background ozone concentration thus causing lower N02 modeled concentrations. But using either source groups or OLMGROUP ALL separately is perfectly acceptable.
So the correction that I applied to CC&V's modeling was to not use source groups to represent different operating scenarios when modeling N02 with the Ozone Limiting
Method and OLMGROUP ALL.
Also, the use of both source groups and OLMGROUP ALL are optional, not mandatory, and I did not tell CC&V that they had to model using either of those options,
separate or combined.
Moreover, I did not tell CC&V to do the modeling that way, combining source groups with OLMGROUP ALL. That was CC&V's choice. My mistake was to not catch this error
in CC&V's modeling before. But as I explained, I became of aware of this issue until it was presented in EPA's Modeler's Workshop.
My instructions to CC&V were related to how to represent blasting activities in the model depending on the pollutant and averaging period being modeled, and specifically to
represent blasting as taking place in each pit at a time through different operating scenarios. In that context I indicated that such representation could be accomplished with
separate model runs or by using source groups, one or the other.
Those instructions are correct because I was referring to modeling blasting activities in general for all pollutants and averaging periods without any reference whatsoever to
N02, to the Ozone Limiting Method or OLMGROUP ALL. I do admit that I should have included a caveat warning CC&V of the possible conflict with source groups in case
that they decided to use the Ozone Limiting Method with OLMGROUP ALL. But once again, I became of aware of this issue until it was presented in EPA's Modeler's
Workshop.
The full language of my instructions to CC&V are in the attached document, in Item 10 of Appendix A.
So characterizing the situation as an error on our part because we told CC&V to model it that way is inaccurate, as it is also inaccurate to say that I mentioned that in my
write up. The error on our part (my error to be specific) was to not catch CC&V's mistake and to not include a caveat in my instructions warning of the conflict in the two
options in AERMOD.
On the other hand, I don't recall any application undergoing modeling review in which we found an error on our part that was not required to be fixed, so I think it's
inaccurate to say that we have done that with other applicants. I could be wrong though, because I don't remember the details of every application.
I recall two applications with an error on our part: The Gypsum Ranch Pit and the Gypsum Biomass Project, for which the same meteorological data set was processed and
given to the applicants by the MEIU containing large amounts of calms in excess of the 10% threshold allowed by EPA for permit modeling. When this error came to light
both modeling reviews had been completed, the Biomass permit issued and the Ranch Pit application had been withdrawn because the facility was sold.
The ColoWyo Mine is another case, in which the emissions were calculated with Mobile v6.2 at a time when MOVES was already available as EPA's preferred emissions
model for mobile sources. This error was overlooked completely by the MEIU for the South Taylor Pit and when the modeling was submitted for the Collom Pit the applicant
was asked to fix this issue and use MOVES. Note that this the only outstanding issue at the moment so there was no need to remodel other than to fix this problem.
EPAOIG Appendix 47
https://mail.google.com/mail/u/0?ik=1144611be0&view=pt&search=all&permthid=thread-a%3Ar-4405689205444361434%7Cmsg-a%3Ar51658973748... 1/3
-------
¦ State.co.us Executive Branch Mail - Re: CC&V
My point is that I don't think there is really a precedent for not fixing an error, specially an error that when corrected would ^iRJSffyiiyrli&c^Siing a
NAAQS. However, if you still want me to report the results without correcting the error, that is your prerogative, I just want the facts to be clear. Please let me know in writing
if this is what you want to do. In that case I will state in my report that I am presenting those erroneous results following your instructions and I will attach your email to the
report, just like I did with Garry's instructions.
I will also explain in the report that the N02 modeling has been conducted with the correction described above and consequently I will provide the actual results (although
not in the table per your request) as opposed to explaining this in a conditional tense, as if it hasn't been done. This is information that I do have and that I relied upon to
reach my conclusion on N02 NAAQS compliance, so I don't think ifs correct to conceal it or downplay it.
I would also ask you for similar type of instructions in writing for including your conclusion that 1-hr and 8-hr CO is likely demonstrated. The emission rates for all mobile
sources were not calculated correctly, and I simply have no idea how much they would increase after the correction. Rebecca Simpson's analysis focused mostly on N02
and to a lesser extent on particulate matter, but did not provide information on CO. We do know that those emissions will increase because this pollutant is highly influenced
by combustion efficiency which in turn is highly influenced by engine load, but I haven't seen any data quantifying the increase. I suspect, intuitively based on the fact that
these are diesel engines and some of them have controls, that the increase might not be enough to reach the NAAQS, but I won't gamble to say or imply that NAAQS
compliance has been demonstrated or likely demonstrated. Just look at the CO results for the ColoWyo case with similar type of engines and how close they are to the
NAAQS.
So my conclusion as a dispersion modeling expert is that with the information available I cannot determine compliance with the CO NAAQS. If you disagree and have
reached a different conclusion and want your conclusion reflected in the report, once again, that is your prerogative, but in that case I would ask you to include that
conclusion under your name and not under mine.
I now realize that my write up on that topic might be contradictory by saying that I cannot determine NAAQS compliance and then saying that the corrections are unlikely to
lead to a modeled violation of the NAAQS, so I will amend that paragraph and remove the latter statement.
I will await your instructions and if you have any questions please let me know.
Rosendo Majano
On Mon, Jan 14, 2019 at 2:25 PM Pierce - CDPHE, Gordon|
Rosendo,
wrote:
Please see below for my requested edits in red. The primary one involves the 1-hour N02. While I agree that the use of OLMGROUP ALL is incorrect (per the June
Modeler's Workshop), that was an error on our part as we told CC&V to model it that way (as you mention in the writeup). As we have done for other applicants when
there is an error on our part, we do not request it be fixed unless there is a need to re-model. Thus, the report summary table should reflect the values without that
error being corrected. I do believe that this error should be discussed in the writeup following the summary table, as you have done. If you could please send a
revised version for review, I would appreciate it.
Thanks,
Gordon
The Division's Modeling and Emissions Inventory Unit (MEIU) received on April 23, 2018 a dispersion modeling analysis submitted by Cripple Creek & Victor Gold
Mining Company (CC&V) as part of the application to modify the Construction Permit NO.98TE0545 and implement the Mine Life Extension #2 Cresson Project in
Teller County, CO.
Review of that modeling analysis was finalized on June 2018 and a final report was written on 06/26/2018 concluding that the proposed project at the Cripple Creek &
Victor facility will cause modeled violations of the 1-hr NO2 NAAQS. In addition, it was concluded that there were several critical errors in the modeling analysis that
should require correction before it could be determined whether the proposed project will or will not cause and/or contribute to modeled violations of the rest of
applicable NAAQS.
After fulfilling administrative requirements of gathering all the communications records for the project, entering all the application information into the Modeling Unit's
database, generating a review checklist report, and creating a billing report, the final modeling review report was provided to the permit engineer, Jonathan Brickey,
on 07/09/2018.
On 08/13/2018 the Modeling Unit MEIU manager, Emmett Malone, forwarded two emails from the Air Pollution Control Division (APCD) director, Garry Kaufman, in
which he provided arguments to reject the main findings and conclusions of the CC&V final modeling review report in what pertains to the engine loads
and N02/NOx in-stack ratios for the non-road diesel engines, and also provided instructions to review the modeling analysis with the originally submitted emission
rates and to accept the originally submitted N02/NOx in-stack ratios for the non-road diesel engines. All the aforementioned emails are included in Appendix 1 of this
document.
In compliance with the APCD director's instructions, CC&V's 04/23/18 modeling submittal has been reviewed for a second time leaving intact the original emission
rates and in-stack ratios of the non-road diesel engines. In addition, an analysis of the APCD director's arguments to approve CC&V's data has also been conducted
to determine if they are supported by sound scientific principles and applicable regulations and guidance. Such analysis is included in a subsequent section of this
document. The results of this new review are provided below.
There are NAAQS and CAAQS currently in effect for the following pollutants and averaging periods: CO (1-hr and 8-hr), Pb (3-month rolling average), N02(1-hr and
annual), PM10 (24-hr), PM2.5 (24-hr and annual), SO2 (1-hr and 3-hr), and ozone (8-hr). For this permit application, quantitative impact analyses are warranted to
demonstrate compliance with the NAAQS/CAAQS for the following pollutants and averaging periods: NO2 (1-hr and Annual), PM2.5 (24-hr and Annual), PM10; CO
(1-hrand 8-hr); and SO2 (1-hr and 3-hr).
The applicant submitted modeling analyses for all required pollutants and averaging periods, and the results of the modeling analyses are summarized below:
Pollutant
PM2.5
PM2.5
PM10
S02
Averaging Period
24-hr
Annual
24-hr
-------
3/28/2021
State.co.us Executive Branch Mail - Re: CC&V
family:Arial, sans-
serif'
EPAOIG Appendix 49
EPAOIG Appendix 49
https://mail.google.com/mail/u/0?ik=1144611be0&view=pt&search=all&permthid=thread-a%3Ar-4405689205444361434%7Cmsg-a%3Ar51658973748.
-------
EPAOIG Appendix 50
Machovec, Chuck M.
From: Hancock, Chip
Sent: Monday, April 18, 2011 3:49 PM
To: Jung, Doris W.
Cc: Fadeyi, Sunday; Machovec, Chuck M.; Hea, Roland C.; King, Kirsten L.
Subject: RE: UCD - permit 96AD234
Per Kirsten and Roland it does.
Chip
From: Jung, Doris W.
Sent: Monday, April 18, 2011 3:43 PM
To: Hancock, Chip
Cc: Fadeyi, Sunday; Machovec, Chuck M.; Hea, Roland C.; King, Kirsten L.
Subject: RE: UCD - permit 96AD234
PS memo 10-01 is only a guidance document. It does not supersede statutory/regulatory requirements.
From: Hancock, Chip
Sent: Monday, April 18, 2011 3:38 PM
To: Jung, Doris W.
Cc: Fadeyi, Sunday; Machovec, Chuck M.; Hea, Roland C.; King, Kirsten L.
Subject: RE: UCD - permit 96AD234
Doris,
Does not matter - PS memo 10-01 over-rides.
Chip
From: Jung, Doris W.
Sent: Monday, April 18, 2011 3:22 PM
To: Hancock, Chip
Cc: Fadeyi, Sunday; Machovec, Chuck M.; Hea, Roland C.; King, Kirsten L.
Subject: RE: UCD - permit 96AD234
Chip,
I think you missed it. Impacts of 1-hr N02 and S02 exceed the NAAQS.
Doris
From: Hancock, Chip
Sent: Monday, April 18, 2011 3:16 PM
To: Jung, Doris W.
Cc: Fadeyi, Sunday; Machovec, Chuck M.; Hea, Roland C.; King, Kirsten L.
Subject: RE: UCD - permit 96AD234
Doris,
l
EPAOIG Appendix 50
-------
EPAOIG Appendix 51
Per PS memo 10-01, modeling for NOx and SOx are not required. The increases in PM2.5 and PM10 are very small (and
the totals are small).
I reviewed the report you attached prior to our making the decision. The previous PM10 had a huge cushion in the
annual impact - ratio up their impact to the new requested level gives a total impact of well less than 110 ng/m3. Doing
the same for the annual number gives well under the 50 ng/m3 standard - and this was using the high 1st high. Total
requested PM10 (not just the increase) is 7 tons per year. Therefore, PM10 modeling not necessary.
The increase in PM2.5 is less than 1.5 tons per year - way below the significance level. Short term increase also below
the 11 pounds per day. So, PM2.5 modeling not necessary.
Chip
From: Jung, Doris W.
Sent: Monday, April 18, 2011 2:43 PM
To: Hancock, Chip
Cc: Fadeyi, Sunday; Machovec, Chuck M.; Hea, Roland C.; King, Kirsten L.
Subject: RE: UCD - permit 96AD234
Chip,
We evaluated the impacts associated with this facility last year (see attached e-mail). At that time it was determined that
demonstrating compliance with the S02, N02, PM10, and PM2.5 NAAQS is necessary for this modification.
Doris
From: Hancock, Chip
Sent: Monday, April 18, 2011 2:11 PM
To: Jung, Doris W.
Cc: Fadeyi, Sunday; Machovec, Chuck M.; Hea, Roland C.; King, Kirsten L.
Subject: UCD - permit 96AD234
Doris,
After a recent meeting Roland and I had with the source, we took another look at this one. After further review the
Construction Permits Unit has determined that impact modeling is not required for the current application and we will
proceed with processing the permit.
This is based on the requested emission levels. The NOx increase is well below 40 tons per year. So, per PS memo 10-01
impact modeling for the NOx one hour standard is not required. All of the other pollutant increases are well below
significance levels. In fact total facility emissions are below significance levels for all of the other pollutants except PM2.S.
The PM2.s level is fairly low and the increase is small (~1 tpy).
I also took a look at the modeling done for the 2009 application. From that it does not appear that the proposed
increases will result in exceedences of the NAAQS.
Therefore, please remove this application (96AD234) from your queue of impact modeling submittals to review. If you
have spent anytime on this application, please send your hourly time summary to Sunday (or me) so it can be added to
PTS and included in the invoice.
2
EPAOIG Appendix 51
-------
EPAOIG Appendix 52
Thanks,
Chip
3
EPAOIG Appendix 52
-------
Updated: Sun Valley Steam Plant Permitting Effort Kick off Me..
State
of
Colorado
EPAOIG Appendix 53
Fwd: Updated: Sun Valley Steam Plant Permitting Effort Kick off Meeting
1 message
Jung - CDPHE, Doris
To: emmett.malon
public health first - NOT!
Forwarded message
Mon, Jan 7, 2013 at 2:57 PM
Rosendo Majano - CDPHE
III
Date: Mon, Jan 7, 2013 at 2:50 PM
Subject: Re: Updated: Sun Valley Steam Plant Permitting Effort Kick off Meeting
"Jung - CDPHE, Doris"
Cc: Chuck Machovec - Kirsten King - CDPHE
Doris,
I hear what you are saying. However, per PS memo 10-02 the Division does not require impact modeling to the 1
hour standard for sources of S02 or NOx with a net change in emissions of less than 40 ton per year. Per the
last paragraph (emphasis added):"..., the Division will apply EPA's SERs for NOx and SO2 to the 1-hour NO2
and 1-hour SO2 standards for all stationary source permitting activities, including determining when ambient air
quality impact analyses are necessary for permitting,...".
As for the reasoning, it is explained better in the memo than I can.
Chip
On Mon, Jan 7, 2013 at 2:20 PM, Jung - CDPHE, Doris wrote:
Chip,
It is our understanding that the Sun Valley Steam Center project is a modification at Zuni Station per the
testimony before the Colorado PUC on 12/12/12: "In 2013, the Company plans to undertake much-needed
repairs and improvements at Zuni Electric Generating Station ("Zuni Station") to continue steam production
operations until it can be replaced late in the third quarter of 2015 by the proposed new Sun Valley Steam
Center ("SVSC"). This new steam production facility will consist of two 300 Mlb per hour package boilers and
is expected to cost $29 million (excluding allowance for funds used during construction and escalations)."
Near-field (including flagpole receptors at the stadium) ambient air impacts from this new steam center should
be assessed for the following reasons regardless of the outcome of any netting exercise (i.e., even if it is a
minor modification):
- The S02 impact analysis on file indicates that the existing Zuni facility may be contributing and causing
violations of the 1-hr NAAQS and 3-hr CAAQS if the facility is using fuel oil.
- There is no N02 cumulative impact analysis completed for Zuni that we are aware of. Based on the impacts
estimated for lowering the discharge height of Unit 2 boiler, it is reasonable to believe that the existing Zuni
facility could cause or contribute to a violation of the 1-hr N02 NAAQS.
- We need to consider that this facility site is situated in an area with high population exposure.
EPAOIG Appendix 53
https://mail.google. com/mail/?ui=2&ik=1144611 be0&view=pt&search=inbox&th=13c1705051 bfd43e
-------
Updated: Sun \&lley Steam Plant Permitting Effort Kick off Me...
Doris
EPAOIG Appendix 54
Forwarded message
III
Date: Thu, Jan 3, 2013 at 4:14 PM
Subject: Re: Updated: Sun Valley Steam Plant Permitting Effort Kick off Meeting
To: "Jung - CDPHE, Doris" |
OK, thanks.
On Thu, Jan 3, 2013 at 3:00 PM, Jung - CDPHE, Doris |
Chip,
wrote:
We are unable to attend since we already have a meeting scheduled for Jan 14 at 10 am.
Doris
~~wrote:
On Thu, Jan 3, 2013 at 2:46 PM, R K Hancock
Chuck, Doris,
Chad requested I forward this to you. Do not know if he has discussed with you or not.
Chip
Forwarded message
Date: Thu, Jan 3, 2013 at 2:38 PM
Subject: Updated: Sun Valley Steam Plant Permitting Effort Kick off Meeting
To: R K Hancock III ^^^^^^^^^¦/'Magno, Gary J"
"Christopher, "Reed,
When: Monday, January 14, 2013 10:00 AM-11:00 AM (GMT-07:00) Mountain Time (US & Canada).
Where: CDPHE Offices
Note: The GMT offset above does not reflect daylight saving time adjustments.
*
Chip,
This meeting will be to review the Sun Valley Steam Plant and the proposed permitting strategy for the
facility.
Thanks,
Chad
EPAOIG Appendix 54
https://mail.google. com/mail/?ui=2&ik=1144611 be0&view=pt&search=inbox&th=13c1705051 bfd43e
2/3
-------
STATE OF
COLORADO
State.co.us Executive Branch Mail - CC&V Modeling Analysis
EPAOIG Appendix 55
CC&V Modeling Analysis
1 message
Rosendo Fri, Oct 19,2018 at 2:59 PM
Garry Kaufman^^^^^^^^^^^^^^^^^
Garry,
On 09/27/18 I participated in a meeting with Robyn Wille and Jessica Lowrey, lawyers with the Colorado Attorney General's Office. While the meeting was to discuss the legal
challenge to the ColoWyo permit and how the revised modeling analysis for this facility should be conducted, the lawyers made some statements that are applicable to any air
permit application and that contradict what you have asked to accept as valid for the CC&V permit application (please see the email below).
There are also two technical arguments used to support the ColoWyo modeling analysis that are being accepted by consensus by everyone involved in this case, and that also
contradict what you have asked to accept in the CC&V application. I'm referring to the following topics.
1- The CO AG's Office lawyers indicated in very clear terms that the provisions in Appendix W to 40CFR51 are applicable to minor sources. They explained that while Colorado
regulations don't require air quality modeling for minor sources, if modeling is used to demonstrate compliance with the NAAQS, then it should be done according to the
procedures established by EPA and they explicitly mentioned that the provisions in Appendix W are applicable and should be followed regardless of the minor source status.
2- The CO AG's Office lawyers also indicated that the Air Division does not have the authority to regulate emissions from mobile sources because those are regulated at the
federal level, but that the Division does have the authority to regulate the usage of mobile sources within the facility when issuing a permit for such facility.
3- It was explained during the meeting that the Division is not trying to regulate the emissions from ColoWyo's mobile sources, but that in the context of conducting an air quality
modeling analysis to determine compliance with the NAAQS, those emissions need to be included and need to be characterized adequately to represent the worst-case
scenario. The CO AG's Office lawyers agreed and indicated that the modeling should be done in a way that withstands future legal challenge.
4- ColoWyo has requested to lower the N02/N0x in-stack ratio for all Tier 3 and below mobile sources from the previously approved value of 0.4 to a new value of 0.1. In doing
this request, ColoWyo has hired two different consulting companies which have presented solid arguments to support that value. Part of those arguments have been the
explanation that higher in-stack ratio values are valid only for engines with post-combustion diesel emissions control devices (i.e. Tier 4 diesel engines). While the Division has
not reached a final determination on this request, verbally everyone involved is in agreement that these arguments are sound and technically supported, and what is being
discussed is not whether those arguments are acceptable or not but rather how many units in the vehicle fleet have those control devices and would therefore have a higher in-
stack ratio.
5- ColoWyo has requested to use lower engine loads than the ones in EPA's MOVES for the non-road mobile sources at their site, and for that purpose their consultants have
provided actual data from their mining equipment. Some of the issues raised by Division staff, myself included, are: a)That the statistical treatment of the data to calculate the
average engine load was not adequate; b)The origin of the data and the procedures for estimating specific segment-specific engine loads used in the calculation of the average,
need to be explained in more detail to ensure that the data is representative of actual operations at the mine. Both ColoWyo's and Division's staff were in agreement and
ColoWyo has promised to correct the statistical procedures and to provide supporting explanations for the data.
A subsequent meeting took place on 10/11/18 involving not only the CO AG's Office lawyers but also ColoWyo's attorneys, as well as Division's and ColoWyo's technical and
management staff. The procedures to conduct the modeling analysis were discussed in detail, including items 3 through 5, and there was consensus that the modeling analysis
has to be conducted in a manner that is legally defensible in court. Particular emphasis was made by ColoWyo about the need to represent the worst-case scenario in the
model.
In stark contrast to what the Division is agreeing to in the ColoWyo case, for the CC&V air quality modeling analysis, Emmett, Gordon, and you, the three direct supervisors
above me in the Division's organizational chart, are asking me to accept as valid the following:
a) That the provisions in Appendix W are not applicable for minor sources and therefore the use of maximum allowable emissions is not pertinent for their mobile sources.
Appendix W states that maximum allowable emissions are conducive to the highest impacts on ambient air and consequently those are the emissions that should be used to
conduct a modeling analysis. I have already expressed my professional opinion that some of the emissions used by CC&V are not maximum allowable emissions.
b) That the Division does not have the authority to regulate mobile sources and therefore using maximum allowable emissions in the modeling analysis is at the discretion of the
Division and not regulatory.
c) That the in-stack ratios of five non-road mining vehicles with Tier 4 diesel engines at CC&V are very low, some as low as 0.01. CC&V has not provided any information to
clarify whether those engines have or not post-combustion control devices, yet the emission rates being used very low, corresponding to units with control devices that would
raise significantly the in-stack ratio. If the in-stack ratio is low, that means there are not control devices and the total emissions from the engine would be very high, and vice
versa, as accepted for the ColoWyo case.
d) That the engine loads for the non-road mining vehicles at CC&V are significantly lower than those in EPA's MOVES, that the statistical treatment of the raw data used by
CC&V is correct and that the data provided by CC&V are representative of the actual operations at the mine. This despite the fact that similar to the ColoWyo case, the Division's
subject matter experts also have objections to the statistical calculations and representativeness of the data in the CC&V case.
Considering that both ColoWyo and CC&V are both minor sources, and both are mining operations using similar type of non-road vehicles, how do I reconcile the contradictory
technical arguments that are being used to support the air quality modeling analysis in each one of those cases?
My main concern is that, in the context of a legal dispute between an external party and the Division, and as one of the Division's subject matter experts, I am providing
information and explanations to the CO AG's lawyers for them to use in the legal dispute, and then I would be turning around and contradict myself by approving as valid the
exact opposite arguments that I just provided to the CO AG's lawyers.
Wouldn't that place me at risk of perjuring myself?
What would happen if the CC&V permit were to be challenged in court?
Would I personally be in legal risk for knowingly approving technical arguments that I have previously stated to be different when acting as a subject matter expert on the
ColoWyo case?
These are questions that I would like to present to the CO AG's Office before continuing the review of the CC&V case.
The legal concerns are in addition to the possibility of compromising my ethical and professional integrity, and that of the Division, by approving technical arguments that several
Division subject matter experts, myself included, agree that are flawed.
In light of all the above I want to request to recuse myself from continuing to review the CC&V modeling analysis. There is a precedent for this back in 2011 when my
predecessor Doris Jung, and the previous Modeling Unit manager Chuck Machovec recused themselves from continuing to review a previous CC&V modeling analysis and it
ended up being reviewed and approved by a different person (Final Modeling Review report initialed by Gordon Pierce dated 06/20/2011).
EPAOIG Appendix 55
https://mail.google.com/mail/u/0?ik=1144611be0&view=pt&search=all&permthid=thread-a%3Ammiai-r1505214597860627835%7Cmsg-a%3As%3A-10... 1/4
-------
State.co.us Executive Branch Mail - CC&V Modeling Analysis
There is also the precedent of my supervisor refusing to conduct an additional modeling analysis for the previous ColoWyo appliElJoiAQJiQ actions
that would go against EPA guidance (email dated 02/03/2016 from Emmett Malone to Chip Hancock).
Note that in making this request I emphasize that I have already done my job and finalized the review of this application twice (Final Modeling Review Reports dated 09/16/2014
and 06/26/2018), in both cases concluding that compliance with the applicable NAAQS was not demonstrated. My technical analyses and conclusions in those reports have not
been challenged and no one has pointed out to any errors or concerns, other than the arguments in your email below that are now being contradicted through what is being
approved by the Division in the ColoWyo case.
At this point what I am concluding is that, without anyone pointing out to any errors in my work in the CC&V application, my professional expertise is just not being respected,
and consequently I find myself at a dead end with this application, being pressured to accept information that is not technically supported, and according to the aforementioned
conversations with the CO AG's Office lawyers, not legally supported. Thus, I am with the CC&V application being placed in an untenable position from an ethical perspective,
and possibly also from a legal point of view.
I think that given the circumstances the fresh perspective of a different person reviewing this modeling analysis will be beneficial.
Please let me know if you agree with my request.
Thank you.
Rosendo Majano
- Forwarded message -
From: Kaufman - CDPHE, Garrison|
Date: Thu, Jul 12, 2018 at 5:32 PM
Subject: Re: CC&V
To: "Pierce - CDPHE, Gordon"!
I, "Malone - CDPHE, Emmett"
Gordon and Emmett,
Thank you for the opportunity to review the draft modeling report and additional communications regarding the potential treatment of non-road engine emissions and other issues
with the CC&V modeling. The draft report and subsequent communications raise several issues regarding the latest round of modeling submitted by CC&V. In response to
these issues, I wanted to provide some perspective on the issues raised, as well as direction on how to proceed with finalizing this project.
The draft modeling report raises 4 basic concerns with the latest modeling that CC&V has submitted: 1) the N02/NOx in-stack ratios for non-road and stationary engines at the
facility; 2) the engine load values used in calculating emissions from non-road engines; 3) the use of source groups in combination with OLM for N02 modeling; and 4) the
exclusion of certain hiking trails within the facility boundaries from ambient air.
NQ2/NOX In-Stack Ratios
The modeling that CC&V submitted utilized a N02/NOx ratio of 0.05 for blasting and a weighted average ratio of 0.0927 for the non-road engines based on several ratios for
different non-road engines at the facility (0.09 for 55 "Her 1-3 non road engines; 0.5 for six Tier 4 non-road engines; 0.2 for four Tier 4 non-road engines; and 0.01 for one "Her 4
non-road engine). The weighted average of 0.0927 was also applied to the stationary engines at the facility, The draft report notes that the permit engineer looked at the ratios
and concluded that they were acceptable except the ratios of 0.2, and .01 used for certain of the Tier 4 engines, and application of the weighted average to the stationary
engines.
Regarding the five Tier 4 engines, CC&V submitted information from the representative for the engine manufacturer that for four of the engines an in-stack ration of between
0.15 and 0.20 would be reasonable and that an in-stack ratio of less than 0.01 would be reasonable for the fifth engine. CC&V used the high end of these values. I have passed
this information on to Roland Hea and Chip Hancock and they concluded that the ratios CC&V used for these five pieces of equipment was acceptable. With respect to the use
of the 0.0927 ratio for the stationary engines, CC&V relied on an e-mail from Chuck Pray in 2016 approving a ratio of .09 for both the non-road and stationary engines. Based on
this e-mail, Roland Hea concurred that the use of a more conservative ratio of 0.0927 was acceptable.
Given these determinations from permitting I believe that all of the in-stack ratios that CC&V used are acceptable and therefore the potential issues raised in the draft modeling
report are resolved.
Non-Road Engine Load
The draft modeling report questions the appropriateness of the engine load factors CC&V used for the various non-road engines at the facility. Engine load is important because
it has a large impact on the calculated emissions from the non-road engines. As acknowledged in the draft report and subsequent communications on this issue, determining
loads for these types of engines is challenging. Unlike stationary engines that can and do operate close to maximum load, numerous factors preclude non-road engines from
operating at or near full load for any extended period of time. In calculating engine loads, CC&V split the various non-road engines into broad categories (e.g. support trucks,
shovels, graders, haul trucks) and then calculated average engine loads for these categories utilizing 3 year averages of diesel fuel consumed for each of these categories. As
noted in the draft model report and Rebecca Simpson's underlying technical analysis dated May 23, 2018, these practices smooth out both temporal variability and variability
between different types and sizes of engines within the broad categories. This, along with the fact that CC&V did not remove outliers, means that CC&V's analysis does not
capture peak emissions from a given large category and may in some instances underestimate average load for individual pieces equipment and more narrow categories of
equipment of like type and size. Based on all of this, and the assertion that state and federal regulations require that emissions from the permitted facility correspond to potential
to emit, the draft modeling report concludes that the load factors CC&V used, and corresponding emissions included in the modeling based on these load factors was improper.
I appreciate the concerns raised and commend MMEIU staff on their excellent analysis of this issue. This analysis helps to advance our thinking on how to best model these
extremely challenging mining sources. At the same time, rejecting CC&V's modeling based on this analysis is problematic for a number of reasons. As an initial matter, I don't
agree with the conclusion that state and federal regulations require the use of PTE in characterizing emissions from the non-road engines for modeling purposes. Because this
is a minor source permit the federal regulations set forth in Appendix W do not directly apply. EPA has made this point on several occasions. Nor do our state regulations
mandate the use of PTE for the non-road engines at CC&V. Regulation No. 3 does provide that EPA approved modeling protocols (which presumably means Appendix W) be
used, but only where modeling is required under Regulation No. 3. Regulation No. 3 only requires modeling of major stationary sources under the PSD program, and does not
require modeling of minor sources. The regulation, does, however, give the Division broad discretion in assessing ambient impacts of any permitted source. As part of this
broad discretion, the Division has the discretion to determine if and how minor sources should be modeled including determining the emissions that should be used in that
modeling.
In addition to the technical complexities associated with determining emissions from non-road engines, the decision on how best to exercise our discretion is complicated by the
treatment on non-road engine mobile equipment under our permitting regulations. As currently written, Regulation No. 3 does not give the Division authority to require permits
for mobile equipment non-road engine emissions (this restriction is limited to the engine emissions themselves and does not extend to fugitive dust from roadways and other
locations). As you know, at least one operator has raised this issue and asserted that based on this we do not have the authority to require modeling of non-road engine
emissions at mining sites. While I would not agree with that conclusion, the treatment of these emission sources under our permitting regulations does raise the question of how
to properly characterize these emissions. Should they be treated as a part of the modeled source, where we typically have used our discretion to require modeling at or near
PTE levels, or should they treated more akin to background or non-source emissions which have typically been modeled at levels closer to actual or average emissions or
pollutant concentrations? If the latter approach is taken, the variability of fuel consumption over time, the lack of short term fuel consumption records, the non-uniformity of
usage of individual pieces or even categories of equipment from month to month, and the extreme variability of emissions over short time periods due to rapidly changing engine
loads resulting from different operating parameters, creates an extrengj^^r^j^iaj^nc^e m dejrmiigigactual emissions. In this case CC&v has used a fairly simplified
https://mail.google.com/mail/u/0?ik=1144611be0&view=pt&search=all&permthid=thread-a%3Ammiai-r1505214597860627835%7Cmsg-a%3As%3A-10... 2/4
-------
State.co.us Executive Branch Mail - CC&V Modeling Analysis
approach that could certainly be refined. At the same time, they have consistently been using this approach through multiple iteSin^MhUfeH^RRftQ^X^/^ral years,
and this approach is based on actual fuel use data. Given this, and the fact that even under the more sophisticated approach taken by MMEIU staff there are still significant
uncertainties about the emissions from this equipment on both a temporal and spatial basis, I believe allowing CC&V to use the engine load profiles, and resultant non-road
engine emissions reflected in their latest modeling submittal is a reasonable and appropriate exercise of the Division's discretion under the regulation.
Source Grouping and OLM
I am not familiar with this issue and defer to the two of you to determine the best approach for this particular case.
Treatment of Hiking Trails
As I understand it the latest CC&V submittal shows newly identified hiking trails within the ambient air boundary that are not included as ambient air in the modeling for at least
certain hours of the day. As indicated in the draft modeling report, and consistent with other areas, I agree that CC&V either needs to include these trails as part of ambient air
or they need to demonstrate to the Division's satisfaction how the general public will be excluded from these areas for any periods of time that they are not included as ambient
air.
Please let me know if you would like to discuss any of these issues
Garry Kaufman
Division Director
Air Pollution Control Division
Colorado Department of Public Health and Environment
Displaying2018_HL_Leader.jpg
Are you curious about ground-level ozone in Colorado? Visit our ozone webpage to learn more.
On Mon, Jul 2,2018 at 11:32 AM, Pierce - CDPHE, Gordon wrote:
Forwarded message
From: Rosendo Majanol
Date: Thu, Jun 28, 2018 al
Subject: Fwd: CC&V
To: Gordon Pierce
Cc: Emmett Malon
Gordon - Emmett just asked me to forward this to you.
Rosendo Majano
- Forwarded message
From: Rosendo Majanol
Date: Tue, Jun 26, 2018 a
Subject: CC&V
To: "Brickey, Jonathan]
Cc: Emmett Malone
Jonathan,
I have finished the review of the CC&V application and I will start preparing the final Modeling Review Report in our regular format. However, that is going to take several days
so in the meantime, and as we have done in the past, I'm sending you attached a memo with my findings and conclusions along with two supporting documents from one of
the Modeling Unit's mobile sources expert. Hopefully this information will allow you to move forward with your part of the work on this application while I finish my report.
In short, the Cripple Creek & Victor facility will cause modeled violations of the 1 -hr N02 NAAQS. In addition, there are several errors in the modeling analysis that require
correction before it can be determined whether the proposed project will or will not cause and/or contribute to modeled violations of the rest of applicable NAAQS.
Let me know if you have any questions.
Rosendo Majano
Gordon Pierce
Program Manager
Technical Services Program
COLORADO
Air Pollution Control Division
Department of Public Health & Environment
4300 Cherry Creek Drive South, Denver, CO 80246-1530
EPAOIG Appendix 57
https://mail.google.com/mail/u/0?ik=1144611be0&view=pt&search=all&permthid=thread-a%3Ammiai-r1505214597860627835%7Cmsg-a%3As%3A-10...
-------
APPENDIX A - Attachment 5
DRAFT Colorado Modeling Guideline for Air Quality Permits (May 2018)
-------
The Colorado Modeling Guideline for Air Quality Permits is currently undergoing a final
round of internal review by Division staff. However, a draft version is available to the
public during this review period to assist in the preparation of modeling analyses for
permit applications. Please note that some changes might still occur in the near future.
04/24/2018
-------
COLORADO
Department of Public
Health & Environment
Dedicated to protecting and improving the health and environment of the people of Colorado
COLORADO MODELING GUIDELINE FOR
AIR QUALITY PERMITS
MAY 2018
COLORADO DEPARTMENT OF PUBLIC HEALTH AND ENVIRONMENT
AIR POLLUTION CONTROL DIVISION, TECHNICAL SERVICES PROGRAM
MODELING AND EMISSIONS INVENTORY UNIT
-------
May 2018 Colorado Modeling Guideline for Air Quality Permits
Table of Contents
Preface ..3
What's New in this Document 4
Definitions 6
Section 1 - Introduction 10
Section 2 - Authority for Air Quality Impact Analyses 11
Attainment Areas 12
Nonattainment Areas 13
Requirements Unique to Colorado 14
Section 3 - Air Quality Impact Analysis 14
Air Dispersion Modeling 14
Air Quality Impact Analysis Process 15
Section 4 - Performing the Air Quality Impact Analysis 21
Modeling Thresholds 21
Modeling Protocol 22
Screening Modeling 23
Refined Modeling 23
Significant Impact Analysis 24
Significant Impact Level (SIL) 25
Cumulative Impact Analysis 27
NAAQS & CAAQS Analysis 29
PSD Increment Analysis 35
Increment Calculation 36
Unique Colorado PSD Increment Requirement 38
Additional Impacts Analysis 44
Impact Analysis for Water 44
Visibility Analysis 44
Soils and Vegetation Analysis 46
AQRV and Visibility Analysis in Federal Class I Areas 46
Air Quality Related Values Analysis for Major Stationary Sources 47
Pre- and Post-Construction Monitoring Analysis 50
Pre- Construction Monitoring Analysis ...50
Post-Construction Monitoring Analysis 50
i
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
Regulated, Non-Criteria Pollutant Analysis 52
Section 5 - Preferred Air Dispersion Models & Associated Inputs 52
Source Data .........52
Criteria Pollutant Recommendations 52
Carbon Monoxide 53
Lead 53
Nitrogen Dioxide 53
Ozone 54
Particulate Matter < 10|im (PMio) 54
Particulate Matter < 2.5|am (PM2.s) 54
Sulfur Dioxide 55
Mobile Sources Data 55
Nearby Sources Data 56
Background Concentrations 57
Elevation Data 58
Downwash Applicability 58
Receptor Network 59
Meteorological Data 60
Modeling Scenarios 61
Section 6- Reporting Requirements 62
Consistency in Geographic Coordinates 62
Exemptions from Submitting Modeling-Related Data 62
New Sources and Modifications Not Subject to PSD Rules 63
New Sources and Modifications Subject to PSD Rules 63
Recommended Additional Guidance 63
Appendix A - Modeling Threshold Supporting Documentation
ii
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
Preface
The Colorado Modeling Guideline for Air Quality Permits (Guideline) presents current Air Pollution
Control Division (Division) air quality modeling guidance for estimating impacts from stationary sources of
air pollution, it addresses modeling issues for source types ranging from small minor sources to major
sources such as those subject to Prevention of Significant Deterioration (PSD) review. Recommendations
in the Colorado Air Quality Modeling Guideline may not be applicable in all situations.
The Guideline is intended to help permit applicants, air quality specialists, and others understand the
Division's expectations for the ambient air impact analysis and to prevent unnecessary delays in the
permit process. It provides a starting point for modeling, but allows the use of professional judgement. To
avoid misunderstandings, obtain the most recent version of Colorado's guidance documents from
http://www.colorado.gov/airquality/permits.aspx. In addition, obtain current regulations and applicable
U.S. Environmental Protection Agency (U.S. EPA) guidance.
This guideline is not intended to describe the implications of modeling results. Such implications are
generally controlled by the permit rules or other relevant state and federal regulations, laws and
guidance. Nevertheless, the Guideline contains incidental discussion of the effects of certain modeling
results. Such discussion is for informational purposes only and shall not be construed to be authority
defining the regulatory impact of any modeling result. For that, the reader should refer to the applicable
rules and regulations.
This is a guide through modeling-related regulations and procedures. It is intended to promote technically
sound and consistent modeling techniques, while encouraging the use of improved and more accurate
techniques as they become available. The guideline helps permit applicants understand when modeling is
warranted. It clarifies what modeling-related information and data should be included with a permit
application. Supplemental guidance on specific technical issues and other modeling-related data and
information, including checklists and meteorological data, are available at
http://www.colorado.gov/airquality/permits.aspx. If modeling procedures other than those
recommended in Colorado and U.S. EPA guidance are used, there might be delays while the procedures
are reviewed. In some cases, U.S. EPA approval may be necessary.
This is only a guidance document. It has been published in accordance with §25-6.5-102, C.R.S. It is not
intended to supersede statutory/regulatory requirements or recommendations of the U.S. EPA.
U.S. EPA models and guidance are available on the Internet at http://www.epa.gov/scram.
Page 3 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
What's New in this Document
May 2018:
The following has been revised on May 17, 2018 from the April 2018 version of the Guideline.
¦ Ozone background criteria to use ARM2
¦ Figure 5 Title as the map only applies to PMio
¦ Table 8 footnote (a) corrected to increment
April 2018:
The following has been revised on April 20, 2018 from the March 2018 version of the Guideline.
¦ Spelling and syntax errors were corrected throughout
¦ PM2.5 SILs are now EPA approved based on recent guidance
March 2018:
The following has been revised on March 16, 2018 from the December 27, 2005, version of the Guideline.
¦ The overall document has been reformatted.
¦ The overall document has also been reorganized to improve the flow of information as well as
the understanding and retention of information presented.
¦ Hyperlinks have been updated to obtain the correct sites.
¦ Duplicate tables and figures have been removed. Only one version of each table and figure is
provided.
¦ Quotes from Regulation No. 3 and Appendix W have been updated to match the most recent
versions.
¦ Section 2 was added to address the applicable regulations that authorize Colorado to perform
modeling analyses.
¦ Section 3 was added to illustrate the full picture of the modeling analysis process.
¦ Section 4 was added to detail the different types of modeling analyses the applicant may be
required to perform.
¦ Section 5 was added to detail the information the applicant should use when performing a
modeling analysis.
¦ Section 6 was added to detail what information the Division is looking for when the applicant
submits a modeling analysis.
¦ Appendix A was added which includes the description of how the modeling thresholds in Table 1
were developed.
¦ Section 4 Additions:
D U.S. EPA's opinion on submitting a modeling protocol and language that a protocol is
intended to be legally binding.
Page 4 of 64
not
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
D Screening-level models were added with updated guidance on when screening models
cannot be used.
° Procedural steps were added to the Significant Impact Analysis
° Procedural steps were added to the NAAQS & CAAQ.S Analysis
D Procedural steps were added to the PSD Increment Analysis
¦ Section 5 Additions:
D The Ambient Ratio Method 2 (ARM2) is now the approved Tier 2 approach replacing the
Ambient Ratio Method (ARM). The approved ambient ratio for Colorado is discussed
more in detail. The Ozone Limiting Method is now a regulatory option.
D Annual PMi0 NAAQS compliance demonstration was removed as the NAAQS was
revoked.
D 24-hour and annual PM2.s NAAQS compliance demonstrations were added.
D 1-hour S02 compliance demonstrations as well as a discussion regarding the 24-hour and
annual S02 NAAQS demonstrations were added.
D PS Memo 10-01 discussion was added.
D Nearby source emission calculations have been updated in Appendix W from allowable
emissions to a subset of actual emissions. The threshold emission rates for nearby
sources to include was removed.
D Flagpole receptors guidance was added.
° The use of Digital Elevation Model (DEM) data for sources and receptors was removed
and language to use National Elevation Dataset (NED) files was added.
° Meteorological data will be provided by the Division in an AERMOD-ready format. The
applicant no longer needs to process meteorological data.
n Precursors to ozone was added.
n Secondary formations of PM2.5 was added.
D Mobile sources guidance was added.
D Modeling scenarios guidance was added.
¦ Language was updated throughout discussing when source and modifications are exempt from
modeling. The exemption now includes emissions below Table 1 thresholds AND not meeting the
scenarios (footnotes) described below Table 1.
¦ References to the Modeling Submittal Completeness Checklist to verify the necessary information
to submit with the modeling analysis was added.
¦ The sections relating to additional impacts analysis and AQRVs is currently still under review so
these sections have been greyed out.
Page 5 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
Definitions
The following explanation of terms are included solely for the reader's convenience; they do not take the
place of any definition in state or federal laws, rules, or regulations.
Air Quality Models. Computer codes for estimating ambient concentration levels (i.e., "impacts") from
new and existing sources of air pollution. They allow one to forecast future air quality levels from sources
that have not been constructed. They simulate in a simplified manner the complex behavior of emissions
injected into the atmosphere.
Air Quality Related Value (AQRV). A feature or property of a Class I area that may be affected by air
pollution. General categories of AQRV's include visibility, odor, flora, fauna, soil, water, geological
features, and cultural resources. https://www.nature.nps.gov/air/Pubs/pdf/flag/FLAG 2010.pdf
Ambient Air. Defined by 40 CFR 50.1(e) as "that portion of the atmosphere, external to the source, to
which the general public has access." NAAQS and PSD increments apply only in ambient air.
Appendix W, 40 CFR Part 51- Guideline on Air Quality Models. The U.S. EPA's Guideline on Air Quality
Models recommends air quality modeling techniques that should be applied to State Implementation Plan
(SIP) revisions for existing sources and to new source reviews, including prevention of significant
deterioration (PSD). It is intended for use by the U.S. EPA in judging the adequacy of modeling analyses
performed by U.S. EPA, state and local agencies, and industry. The Guideline identifies those techniques
and databases U.S. EPA considers acceptable. The guide is not intended to be a compendium of modeling
techniques. Rather, it serves as a basis by which air quality managers, supported by sound scientific
judgment, have a common measure of acceptable technical analysis. Appendix W was updated on
January 17, 2017. https://www3.epa.gov/ttn/scram/appendix w/2016/AppendixW 2017.pdf
Attainment Area. Any area that meets the national primary or secondary ambient air quality standard for
an applicable criteria pollutant.
Background. Air contaminant concentrations present in the ambient air that are not attributed to the
source or site being evaluated.
Building Downwash. Turbulence created by the wind flowing over buildings or structures that would
ordinarily not exist. This effect can alter ground-level concentration levels than would exist in the absence
of the building or structure.
Class I Area. An area defined by Congress that is afforded the greatest degree of air quality protection.
Class I areas are deemed to have special natural, scenic, or historic value. The Prevention of Significant
Deterioration (PSD) regulations provide special protection for Class I areas. Little deterioration of air
quality is allowed.
Class il Area. An area defined by Congress where a moderate degree of emissions growth is allowed.
Complex Terrain. Any terrain exceeding the height of the stack being modeled. This includes terrain
commonly referred to as intermediate terrain {receptorsbetween stack height and plume height).
Criteria Pollutant A pollutant for which a National Ambient Air Quality Standard (NAAQS) has been
defined.
Page 6 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
Cumulative Impact Analysis. A full modeling impact analysis that involves the facility under permit review,
nearby sources, and background concentrations to compare the facility's impact to the NAAQS.
Fugitive Emission. Any gaseous or particulate contaminant entering the atmosphere that could not
reasonably pass through a stack, chimney, vent, or other functionally equivalent opening designed to
direct or control its flow.
Good Engineering Practice (GEP) Stack Height. From Regulation No. 3, Part D, §VIII.D.3, "The greater of 65
meters or for stacks in existence on January 12, 1979 and for which the owner or operator had obtained
applicable pre-construction permits or approvals required, Hg = 2.5*H (provided the owner or operator
produces evidence that this equation was actually relied on in establishing an emission limitation and for
all other stacks Hg = H +1.5*L where,
Hg: good engineering practice stack height measured from the ground level elevation at the base of the
stack
H: height of nearby structure(s) measured from the ground level elevation at the base of the stack
L: lesser dimension (height or projected width) of nearby structure(s)."
Hazardous Air Pollutant (HAP). Any pollutant subject to a standard promulgated under the Federal Clean
Air Act (FCAA) section 112 (relating to hazardous air pollutants).
Major Stationary Source. The term major may refer to the total emissions at a stationary source or to a
specific facility.
¦ A named major stationary source is any source belonging to a list of 28 source categories in 40
CFR 52.21(b)(1) which emits or has the potential to emit 100 tons per year (tpy) or more of any
pollutant regulated by the Federal Clean Air Act (FCAA).
¦ A major stationary source is any source not belonging to the 28 named source categories which
emits or has the potential to emit such pollutants in amounts of 250 tpy or more.
¦ A major source is any source that emits 10 tpy or more of any single HAP or 25 tpy or more of any
combination of HAPs under FCAA section 112(b).
Major Modified Stationary Source. Used in the context of a PSD application, the phrase major modified
stationary source or facility refers to a change in operation that results in a significant net increase of
emissions for any pollutant for which a NAAQS has been defined. New sources at an existing major
stationary source are treated as modifications to the major stationary source.
Major New Source Review (NSR) Program. The major NSR program contained in parts C and D of title I of
the FCCA is a preconstruction review and permitting program applicable to new major sources and major
modifications at such sources. In areas meeting the NAAQS (attainment areas) or for which there is
insufficient information to determine whether they meet the NAAQS (unclassifiable areas), the NSR
requirements under part C of title I of the FCAA apply. The EPA calls this portion of the major NSR
program the Prevention of Significant Deterioration or PSD program. In areas not meeting the NAAQS, the
major NSR program is implemented under the requirements of part D of title I of the FCCA. The EPA calls
this program the "nonattainment" major NSR program. The EPA has promulgated rules in 40 CFR 52.21 to
implement PSD in portions of the country that do not have approved state or tribal PSD programs.
Page 7 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
Major Source Baseline Date. This is the date after which actual emissions associated with physical changes
or changes in the method of operation at a major stationary source affect the available increment.
Changes in actual emissions occurring at any stationary source after this date contribute to the baseline
concentration until the minor source baseline date is established.
Minor Source. Any stationary source that is not defined as a major stationary source in Regulation No. 3,
Part D §11.A.25. The term is sometimes used rather loosely. The definition may vary based on the context
in which it is used.
Minor Source Baseline Date. This is the earliest date after the PSD increment trigger date on which a PSD
application for a new major source or a major modification to an existing source is considered complete.
The minor source baseline date is pollutant- and geographically-specific.
Modeling and Emissions Inventory Unit (MEIU). This is the unit within the Technical Services Program (TSP)
of the Air Pollution Control Division that is responsible for review of air dispersion modeling.
Modified Stationary Source.
¦ When used in the context of modeling, the phrase modified stationary source or facility refers to
a change in the location or stack parameters of an emission point, including emission rate.
¦ When used in the context of a permit application, the phrase modified stationary source or
facility refers to a physical change in, or change in method of operation, that results in an
increase of emissions
National Ambient Air Quality Standards (NAAQS). Levels of air quality to protect the public health and
welfare (40 CFR 50.2). Primary standards are set to protect public health, including the health of
"sensitive" populations such as asthmatics, children, and the elderly from the effects of "criteria air
pollutants" and certain non-criteria pollutants. Secondary standards are set to protect public welfare,
including protection against decreased visibility, damage to animals, crops, vegetation, and buildings.
Nearby Sources. Any major source, major stationary source, or minor source that causes a significant
concentration gradient in the vicinity of a new or modified source.
Nonattainment Area (NAA). Any area that does not meet (or that contributes to ambient air quality in a
nearby area that does not meet) the national primary or secondary ambient air quality standard for a
criteria pollutant.
Other Background Sources. All sources of air pollution other than the source under review and those
identified as nearby sources. Examples include area and mobile sources, natural sources, most minor
sources, distant major sources, and major stationary sources. They usually are accounted for by using an
appropriate ambient background concentration as recommended in section 8 of Appendix W of 40 CFR
Part 51 or by application of a model using inventory recommendations in Table 8-2 of Appendix W.
Project. An operational and/or physical change that may affect air emission rates at a site.
PSD Increment. The maximum allowable increase of an air pollutant that is allowed to occur above the
applicable baseline concentration for that pollutant.
Qualitative Determination. Relies on descriptive generalized statements and made without regard to
quantity.
Page 8 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
Quantitative Determination. A numerical "estimate" of the air pollutant concentration in ambient air.
Reasonable Further Progress (RFP). From the Common Provisions Regulation, "The annual incremental
reductions in emissions of the applicable air pollutant (including substantial reductions in the early years
following approval or promulgation of plan provisions under the Federal Act, section 110(a)(2)(l) and
regular reductions thereafter) which are sufficient in the judgment of the commission and U.S. EPA, to
provide for attainment of the applicable National Ambient Air Quality Standards by the date required in
section 172(a) of the Federal Act."
Receptor. As used here, a receptor is a geographic location (point) at which the model calculates the
impact (i.e., air pollutant concentration) from a source of air pollution. In practice, a large number of
receptors (i.e., a grid of receptors) is used to estimate air quality impacts over the probable area of
impact from the source. Each receptor has a unique geographic coordinate and elevation.
Refined Model. An analytical technique that provides a detailed treatment of physical and chemical
atmospheric processes and requires detailed and precise input data. Specialized estimates are calculated
that are useful for evaluating source impact relative to air quality standards and allowable increments.
The estimates are more representative than those obtained from conservative screening techniques.
Requested Emission Rate. The emission rate calculated using the maximum rated (design) capacity of the
source or the emission rate specified as an enforceable permit condition.
Scenic and/or Important Views. An important or sensitive panorama or long-range view anywhere in
Colorado. This includes important views of landmarks or panoramas. The Division maintains a list of
scenic and/or important views in Colorado
(https://www.colorado.gov/airquality/permits/SCENICVW2005.pdf).
Screening Technique. A relatively simple analysis technique to determine whether a given source is likely
to pose a threat to air quality. Concentration estimates from screening techniques are conservative.
Significant Impact Analysis (SIA). Modeling analysis involving only the project sources to determine
whether a new and/or modified facility, or a combination of the two, could cause a significant ambient air
impact.
Significant Impact. A concentration in ambient air that exceeds a modeling significance level.
Significant Impact Levels (SILs). Values established by EPA to determine whether a proposed new or
modified source will cause or contribute to a violation of the NAAQS or PSD increments. When a facility
impact is above the applicable SILs, a refined cumulative impact analysis is required.
Simple Terrain. Any terrain with elevations lower than the top of the stack.
Stationary Sources Program (SSP). This is the program within the Air Pollution Control Division that is
responsible for air quality permitting and enforcement.
Unclassifiable Area. Any area that cannot be classified on the basis of available information as meeting or
not meeting the national primary or secondary ambient air quality standard for the pollutant.
Universal Transverse Mercator (UTM). A plane coordinate system that uses distances from a specified
reference point as the basis for all locations. It is based on a transverse Mercator projection that divides
Page 9 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
the Earth's surface into zones that are 6 degrees of longitude wide. Precise locations on the earth are
described in terms of north-south (northing) and east-west (easting) distances, measured in meters from
the origin of the appropriate UTM Zone.
Most of Colorado is zone 13, while the western seventh is in zone 12.
Air quality models are used to estimate impacts (air pollutant concentration levels) in ambient air to
determine if a proposed source or activity will comply with applicable ambient air standards and other
applicable regulatory requirements. Federal law requires that the Division have legally enforceable
procedures in place to prevent construction or modification of any source where the emissions from the
projected activity would violate control strategies or interfere with attainment and maintenance of the
National Ambient Air Quality Standards (NAAQS).1
All estimates of ambient concentrations required under Colorado Air Quality Control Commission (AQCC)
Regulation No. 3 must be based on U.S. EPA-approved air quality models, data bases, and other
requirements generally approved by the U.S. EPA and specifically approved by the Division. Case-by-case
approval from the Division and/or U.S. EPA is required if a non-EPA model is proposed.
Regulation No. 3, Part A, §VIII.A.l states that "all estimates of ambient concentrations required under this
Regulation No. 3 shall be based on the applicable air quality models, data bases, and other requirements
generally approved by U.S. EPA and specifically approved by the division. If a non-U.S. EPA approved
model, such as a wind tunnel study, Is proposed, the nature and requirements of such a model should be
outlined to the division at a pre-appllcatlon meeting. The application will be deemed incomplete until there
has been an opportunity for a public hearing on the proposed model and written approval of the U.S. EPA
has been received."
The primary U.S. EPA modeling guideline is 40 CFR Part 51, Appendix W - Guideline on Air Quality Models
(https://www3.epa.gov/ttn/scram/appendix w/2016/AppendixW 2017.pdf). There are many other U.S.
EPA guidance documents, memos, and U.S. EPA model clearinghouse decisions that explain modeling
procedures. This Guideline is intended to help permit applicants understand federal modeling procedures.
It also provides Colorado's interpretation of gray areas in federal guidance. As such, it presents
procedures that are "specifically approved" by the Division.
1 Pursuant to section 110(a)(2)(C) of the federal Clean Air Act, the State Implementation Plan (SIP)
needs to regulate the "modification and construction of any stationary source within the areas
covered by the plan as necessary to assure that national ambient air quality standards are
achieved." Similarly, 40 CFR section 51.160 requires the State to have the authority to prohibit
any construction or modification that would interfere with the attainment or maintenance of a
national standard. This includes PSD increments as well as NAAQS. See also 40 C.F.R 51.166.
There is no distinction in these provisions between major and minor sources.
Section 1 - Introduction
Page 10 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
The primary Colorado regulation for air quality permits is Regulation No. 3,2 Certain new/modified air
pollution sources are subject to the regulatory modeling requirements of Regulation No. 3 (authorized by
§25-7-114 to 25-7-114.7, Colorado Revised Statutes (C.R.S.)).
To avoid unnecessary delays in permit processing, pre-application meetings and communications (e.g.,
phone, e-mail, letter) are strongly recommended, particularly for new major sources and major
modifications. The Division does not routinely require or perform modeling to determine impacts from
hazardous air pollutants (HAPs).
Section 2 - Authority for Air Quality Impact Analyses
The Colorado AQCC developed regulations that require the Division's preliminary analysis for construction
permits to indicate the air quality impact from a proposed source or activity. In addition, the Division
must determine if the proposed source or activity will comply with applicable ambient air quality
standards. The recommended tools for determining impacts are air quality models. This section discusses
the regulatory requirements for air quality impact analyses.
While modeling is not required to obtain an operating permit, it may be performed or requested if the
operating permit is modified (Regulation No. 3, Part C, §X). Operating permits may also be subject to
modeling if the application is for a combined construction/operating permit (Regulation No. 3, Part C,
§III.C.12.d).
For both major sources and minor sources, Regulation No. 3, Part B, §111.B.5.d states, "the preliminary
analysis shall indicate what impact, if any, the new source will have (as of the projected date of
commencement of operation) on all areas (attainment, attainment/maintenance, nonattainment,
unclassifiable), within the probable area of influence of the proposed source...When the preliminary
analysis includes modeling, the model used shall be an appropriate one given the topography,
meteorology, and other characteristics of the region that the source will impact. Use of any non-guideline
model required U.S. EPA approval under Section VIII.A. of Part A of this regulation."
Regulation No. 3, Part B, §111. D.l states that the Division or the AQCC "shall grant the permit if it finds
that...the proposed source or activity will not cause an exceedance of any National Ambient Air Quality
Standards; and the source or activity will meet any applicable ambient air quality standards and all
applicable regulations..."
While Regulation No. 3 requires that the Division indicate the "impact, if any" in its preliminary analysis, it
does not explicitly require modeling; however, a demonstration of compliance with all NAAQS and CAAQS
is required. Thus, the impact analysis can be done using quantitative (modeling) or qualitative (non-
modeling) methods, as appropriate; however, U.S. EPA approved models and/or methods must be used if
3 numerical estimate (i.e., pollutant concentration in ambient air) of the impact is made.
2 Colorado air quality regulations are available on the CDPHE website
(https://www.colorado.gov/pacific/cdphe/aqcc-regs) or upon request. To obtain official copies,
please contact the Secretary of State's office.
Page 11 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
The modeling thresholds and scenarios outlined in Table 1 may be used to determine when modeling is
warranted. If it is unclear if modeling is warranted, please contact the Division. The thresholds are
applicable for sources located in nonattainment as well as attainment areas.
The impact analysis requirement in Regulation No. 3 applies to all areas: attainment,
attainment/maintenance, nonattainment, and unciassifiable.
Attainment Areas
New major sources and major modifications subject to PSD attainment area rules are required to submit
various types of modeling and/or analyses along with their permit application. The application must
include appropriate modeling and/or analyses to be ruled complete. Please refer to Regulation No. 3, Part
D, §VI.A.2 and §VI.A.6 for source impact analysis requirements.
With respect to ambient air standards, §VI.A.2 requires that "the owner or operator of the proposed
source or modification shall demonstrate to the Division that allowable emission increases from the
proposed source or modification in conjunction with all other applicable emissions increases or reductions
(including secondary emissions) will not cause or contribute to concentrations of air pollutants in the
ambient air in violation of either:
VI.A.2.a: any state or national ambient air quality standard in any baseline area or air quality
control region; or
Vi.A.2.b: any applicable maximum allowable increase over the baseline concentration in any area"
Regulation No. 3, Part D, §VI.D.l.b requires that "the proposed source or modification will achieve an
emissions rate that will ensure that the emissions of such pollutant from the source or modification will
not significantly affect ambient air quality in the nonattainment area." That is, the modeling that is
required should be used to determine if the source would have a significant impact in any nonattainment
area.
Major sources and major modifications are subject to additional requirements. See section 4 for more
details. The impact analysis requirement of Regulation No. 3 applies to all areas (attainment,
attainment/maintenance, nonattainment, unciassifiable).
Minor sources and minor modifications are not required by regulation to submit a modeling analysis that
demonstrates compliance along with their permit application; however, a demonstration of compliance
(quantitative or qualitative) with all NAAQ.S and CAAQ.S is required. Nevertheless, applicants may elect to
include modeling with the applications to prevent unnecessary delays.
If modeling is not submitted with the permit application, the Division will decide if modeling is warranted
to complete the impact analysis and compliance demonstration required by Regulation No. 3. If modeling
is warranted, the Division will perform a screening-level analysis if it is technically feasible to perform one.
If the screening-level analysis shows there could be modeled violations of applicable standards, the
Division will contact the applicant to discuss options. Since the Division does not usually perform refined-
level modeling as part of the permitting process, the Division will typically require that the applicant
perform any refined modeling that might be warranted.
If modeling is warranted, refer to sections 4, 5, and 6
Page 12 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
Nonattainment Areas
The impact analysis requirement of Regulation No. 3, Part B, §111. B,5.d applies in all areas ("attainment
attainment/maintenance, nonattainment unclassifiable"). Thus, modeling may sometimes be warranted
for sources in nonattainment areas. The goals of the impact analysis vary depending on the applicable
regulatory requirements. The regulations refer to the concept of reasonable further progress (RFP) for
sources located in nonattainment areas. If emissions from a new source or modification would prevent a
nonattainment area (NAA) from coming into compliance by the applicable date in the Federal Act or in
the SIP, then the source impairs RFP.
New major sources and major modifications subject to NSR nonattainment area rules are required to
submit various types of modeling and/or analyses along with their permit application. In nonattainment
areas, Regulation No. 3, Part D, §V contains a number of requirements for obtaining a permit. Refer to the
regulation for details. A few of the requirements follow:
¦ Offsets must represent reasonable further progress towards attainment of the National Ambient
Air Quality Standards when considered in connection with other new and existing sources of
emissions.
¦ In addition, offsets for PMio, sulfur oxides, and carbon monoxide must show, through
atmospheric modeling, a positive net air quality benefit in the area affected by the emissions.
¦ Provided, however, that offsets meeting the requirements of this section V.A.3 may also be
obtained from existing sources outside the nonattainment area if the applicant demonstrates:
D A greater air quality benefit may thus be achieved; or sufficient offsets are not available
from sources within the nonattainment area; and
° The other area has an equal or higher nonattainment classification than the area in which
the source is located; and
° Emissions from such other area contribute to a violation of the National Ambient Air
Quality Standard in the nonattainment area in which the source is located.
D With respect to offsets obtained from outside the non-attainment area, the division may
increase the ratio of the required offsets to new emissions the greater the distance such
offsets are from the new or modified source.
¦ Precursors to ozone must be analyzed and discussed. Please use EPA guidance regarding how to
perform an analysis for precursors to ozone.
If modeling is not submitted with the permit application, the Division will decide if modeling is warranted
to complete the impact analysis and compliance demonstration required by Regulation No. 3. If modeling
is warranted, the Division may perform a screening-level analysis if it is technically feasible to perform
one. If the screening-level analysis shows there could be modeled violations of applicable standards, the
Division wiil contact the applicant to discuss options. Since the Division does not usually perform refined-
level modeling as part of the permitting process, the Division will typically request that the applicant
perform any refined modeling that might be warranted. If modeling is warranted, refer to sections 4,
and 6.
Page 13 of 64
5,
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
in the event that compliance with standards cannot be demonstrated using typical attainment area
modeling procedures, a case-by-case approach should be developed in consultation with Division staff
familiar with the affected nonattainment area.
Requirements Unique to Colorado
The following are additional modeling-related regulatory requirements unique to Colorado:
¦ A major source by itself may not consume more than 75% of any applicable PSD increment
¦ Class I S02 increments apply to some pristine Class II areas
¦ For major sources subject to PSD review, water is included as one of the required elements in the
additional impact analysis; the requirement is intended to provide information on acid deposition
in high altitude lakes
¦ Sulfur dioxide (S02) 3-hour standard of 700 pg/m3
Section 3 - Air Quality Impact Analysis
An applicant must demonstrate that the proposed source or modification, as represented in the air
permit application, would not cause or contribute to a National Ambient Air Quality Standard (NAAQS) or
Prevention of Significant Deterioration (PSD) increment violation. An air quality impact analysis is the
means for the applicant to make the demonstration. It is an evaluation of the potential impact on the
environment associated with a new and/or modified facility. Additional analyses required by federal rule
would also be included in the air quality impact analysis.
The air quality impact analysis is a stand-alone report. Results from the report should be sufficient for
Division staff to evaluate the impact of the proposed operation without input from other reports. Division
staff should not refer to other documents or reports for data required to be in the report. In addition,
applicants should not exclude items normally required without coordination with the Division's Technical
Services Program (TSP) modeling staff unless the items are clearly not applicable to the project. Refer to
the Colorado Modeling Submittal Completeness Checklist
(https://www.colorado.Rov/airquality/permits/CompletenessChecklist-ModelinRSubmittall4Feb.pdf) to
determine what information needs to be submitted in the air quality impact analysis.
Air Dispersion Modeling
Air dispersion models are tools to approximate concentrations from one or more facilities or sources of
air contaminants. When an air contaminant is emitted into the atmosphere, it is transported and
dispersed by various atmospheric processes. Algorithms and equations have been developed to
approximate (model) these atmospheric processes and have been incorporated into various computer
codes (computer models). Division staff uses the results from these computer models in their review of
air permit applications. A modeled prediction is used to demonstrate if the new or modifying source will
show compliance with the NAAQS and CAAQS. If the model predicts an exceedance of the NAAQS and/or
CAAQS, the applicant is given the opportunity to adjust the facility allowable emissions, operating hours,
source parameters, and source configuration in order to demonstrate the predicted impact will be in
compliance with all state and federal standards.
Page 14 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
Procedures and models other than those recommended by U.S. EPA or in this guideline may be approved
on a case-by-case basis if there is sufficient technical justification. U.S. EPA approval may be required in
some cases. Refer to U.S. EPA guidance for use of alternative models.
If a non-EPA-approved model, such as a wind tunnel study, is proposed, the nature and requirements of
such a model should be outlined to the Division at a pre-application meeting. The permit application will
be deemed incomplete until there has been a public hearing on the proposed model and written approval
of the U.S. EPA has been received (Regulation No. 3 Part A §VII).
The most recent version of U.S. EPA-approved models should be used. Division approval should be
obtained if an older version is used.
For Class I area modeling, the Division generally supports the use of models and modeling techniques
recommended by the Interagency Workgroup on Air Quality Modeling (IWAQM)3. Written IWAQM
guidance does not always reflect their latest recommendations. In addition, recommendations for the
Class I analysis may vary from one area to another. Thus, work with Division staff and affected federal
land managers (FLMs) on a case-by-case basis to determine the appropriate methods to address impacts
at each affected Class I area.
Air Quality Impact Analysis Process
Division staff with the appropriate expertise reviews various aspects of the impact analysis. For example,
different specialists may provide comments on dispersion modeling, monitoring data, visibility modeling,
and air quality related values. Internal comments by reviewers are typically sent directly to the permit
engineer in the Stationary Sources Program who interprets the comments and, if necessary, brings staff
together to discuss or resolve issues.
Modeling submittals that accompany permit applications should generally be sent to the Division's
Stationary Sources Program where a permit engineer processes the permit application. The permit
engineer forwards modeling reports, date, modeling input/output files, the permit application, and other
relevant information to appropriate staff. As required by regulations, copies of the permit applications for
major stationary sources are sent to federal agencies such as U.S. EPA Region VIII and affected federal
land managers.
It is appropriate for applicants or their modelers to send modeling protocols directly to modeling staff in
the Division's Technical Services Program. A copy should also be sent to the Stationary Sources Program
permit review staff since they are responsible for the overall review of the permit.
The Division encourages phone conversations, e-mail, and other types of communication between staff
and the applicant's modeler(s) and other specialists to resolve issues once the actual review process
begins. It is assumed the applicant's modeler or other specialists will notify the applicant of important
3 IWAQM was formed to provide a focus for development of technically sound, regional air quality models
for regulatory assessments of pollutant source impacts on federal Class I areas. The guidance included
input from the U.S. Forest Service, National Park Service, U.S. Fish and Wildlife Service, the U.S. EPA, and
several states.
Page 15 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
modeling-related issues as necessary. It is recommended that significant issues and information transfers
be done in writing. Copies of any letters or e-mail messages shall be sent to the permit engineer.
When oversights, errors, or questionable assumptions and/or methods are found during the review
process, Division staff will use professional judgment to decide if deficiencies are sufficient to change the
outcome of the compliance demonstration. If the ramifications of a modeling-related issue are not
significant, the deficiencies are noted in the comments and appropriate language is included to justify
that a specific issue is minor. If it is difficult to assess the ramifications without redoing the analysis, the
Division may attempt to redo the analysis, while the deficiencies will be noted in the review comments
and the applicant will be asked to address the comments.
Any responses to comments may be sent directly to Technical Services Program modeling staff, but it is
recommended that a copy also be sent to the permit engineer. In cases where there are no modeling
issues, the Division's modeling comments are not usually forwarded to the applicant. Instead, the written
comments are added to the permit file.
Figure 1 graphically depicts the permit review process as it relates to air quality modeling. While the
flowchart is applicable to all permit applications for major stationary sources where modeling is required,
only certain portions of the flowchart are applicable for minor sources. For example, the loop involving
U.S. EPA Region VIII and the federal land manager (FLM) is not an integral part of the review process for
minor sources.
Figure 2 graphically shows the roles and responsibilities for the modeling review process within the
Division for air quality construction permits.
Figure 3 illustrates key aspects of the regulatory decision process for major stationary sources and major
modifications seeking construction permits. This figure is currently under review within the Division. Please
contact Division staff to confirm the review process for an AQRV and Visibility analysis in Class / areas.
Page 16 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
Figure 1, Permit Review Process
Page 17 of 64
-------
May 2018
Figure
Colorado Modeling Guideline for Air Quality Permits
2. Roles 81 Responsibilities within CDPHE
Page 18 of 64
-------
May 2018 Colorado Modeling Guideline for Air Quality Permits
Figure 3. Regulatory Decision Process for AQRVs
¦""is the proposed^
emission rate or
net emissions
increase signficant
for a given
s. pollutant? ^
and AQRV.
:«!~ n« appiy
Is pre-construction
AQRV monitoring
necessary?
or joint moniio:
Applicant must submit an An.»lysi» u(
Impairment with the permit t epilation
Is the application
complete in its
analysis and
monitoring of
AQRVs?
Consult with FlMs
t
Page 19 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
Page 20 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
Section 4 - Performing the Air Quality Impact Analysis
As discussed in Section 2, the Colorado AQCC developed regulations that require the Division's
preliminary analysis for construction permits to indicate the air quality impact from a proposed source or
activity. In addition, the Division must determine if the proposed source or activity will comply with
applicable ambient air quality standards. The recommended tools for determining impacts are air quality
models.
Figure 4 depicts the air quality impact analysis.
Modeling Thresholds
Modeling thresholds were developed to identify new sources and modifications that would have
relatively small impacts and do not warrant further analysis with respect to applicable air quality
standards. The development of these thresholds is intended to assist the Division staff, permit applicants,
air quality consultants, and others decide when modeling is warranted and to determine the impact from
a source. This section introduces de minimis emissions, which have low probability of causing or
contributing to an exceedance of an air quality standard. By using this approach, permitting costs
associated with the impact analysis required by Regulation No. 3 can be minimized.
Air quality modelers developed the modeling thresholds in Table 1 during a technical peer review of the
Division's modeling practices. The Division performed dispersion modeling to help demonstrate that the
thresholds in Table 1 are appropriate. This analysis can be seen in Appendix A at the end of this
document. Permit applicants and the Division should try to avoid situations where the decision to
perform modeling takes longer than actually performing a screening-level modeling analysis (screening-
level models can often be run quickly with minimal cost).
For a given pollutant, modeling is usually warranted if the long-term (tons per year) or short-term
(pounds per hour, etc.) requested emission rate for a new source or the facility-wide net emissions
increase for a modification is above the applicable emission threshold in Table 1. If the requested
emission rate and/or the facility-wide net emissions increase is below both of the thresholds, modeling is
usually not warranted unless one of the situations outlined in the footnotes of Table 1 applies. If there is
doubt regarding the need for modeling, the applicant should consult the Division.
The thresholds in Table 1 were not developed to address compliance with minor modifications to major
sources located within 10km of a Class I area. Thus, modeling decisions related to Regulation No. 3, Part
D, §11.A.44.c are made on a case-by-case basis. According to §II.A.44.c, any net emissions increase of a
regulated pollutant at a major stationary source located within 10 kilometers (6.2 miles) of a federal Class
I area should perform modeling to determine if the maximum 24-hour average impact in the Class I area
exceeds 1.0 |ig/m3 on a 24-hour basis. If it does, the emissions increase is significant and the modification
constitutes a major modification subject to PSD review.
The Class I significance level of 1.0 |ig/m3 on a 24-hour basis is only intended to determine if a
modification is major. It should not be used to determine if the impact in a Class I area is significant.
Page 21 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
Table 1. Modeling Thresholds
If emission rate is less than threshold, a qualitative description of impact
maybe adequate unless a situation warrants modeling.1
Pollutant
Requested Emission Rate from New Source or Facility-Wide
Net Emissions Increase from a Modification
Long Term (tons per year)
Short Term (pounds per hour)
Carbon Monoxide (CO)
23 pounds per hour
Nitrogen Oxides (NOx)
40
0.46
Sulfur Dioxide (S02)
40
0.46
Particulate Matter < 10 nm
(PMio)
82 pounds per day
Particulate Matter < 2.5 pirn
(PMzs)
5
11 pounds per day
Lead(Pb)
25 pounds per 3-months
Circumstances where source may cause or contribute to a violation of applicable ambient air
quality standards despite being below the thresholds:
(a) Sources of SO2, NO2, PM10, PM2.5, CO, or Pb where a substantial portion of the new or
modified emissions have poor dispersion characteristics (e.g., rain caps, horizontal stacks,
fugitive releases, or building downwash) in dose proximity to ambient air at the site
boundary
(b) Sources located in complex terrain (e.g., terrain above stack heights in dose proximity to
the source)
(c) Sources located in areas with poor existing air quality
(d) Modification at existing major sources, including grandfathered sources that have never
been modeled before
Modeling Protocol
The protocol is the primary mechanism by which all affected parties such as the applicant, the Division,
U.S. EPA, and federal land managers reach agreement on a modeling approach. The protocol
development process is intended to minimize the chances of misunderstandings and to avoid delays in
the permit process. It explains in detail how a modeling analysis will be performed, how the results will be
presented, and how compliance with applicable requirements will be demonstrated. The protocol is not
intended to be a binding, legal document as changes or deviations are often necessary as the data
collection and analysis progresses.
Submission of a modeling protocol is strongly recommended for all air quality impact analyses.
Page 22 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
Screening Modeling
The U.S. EPA developed screening-level modeling techniques to determine quickly whether a facility
should perform in-depth refined modeling analyses. Screening-level models produce estimates of worst-
case impacts from a single source without the need for hourly metrological data. Most applicants are
recommended to perform a screening-level analysis to show the facility is in compliance with the
applicable NAAQS and CAAQS. If there is doubt regarding the need for modeling, the applicant is
recommended to perform a screening-level analysis. If the screening-level analysis does not show
compliance with the NAAQS and CAAQS, then refined modeling is required.
The U.S. EPA has regulatory screening models that should be used for this analysis. These models can be
found on the U.S. EPA Support Center for Regulatory Atmospheric Modeling (SCRAM) website.
Screening models are designed to evaluate a single source. Most facilities, however, do not consist of a
single source, but screening models can still be used by summing the emissions from all sources at the
facility and model them as if they are being emitted from a single source. This method is only acceptable
when all sources are stacks and being emitted from the shortest stack, to represent worst-case.
When facilities consist mostly of fugitive emissions, screening models are not acceptable. When summing
these types of emissions together and modeling as a single source, the accuracy of the model is reduced
substantially and the results not credible. Therefore, the Division does not accept screening models from
the following source categories:
¦ Gravel Pits
¦ Quarries
¦ Landfills
* Mining Operations
¦ Any type of facility not mentioned above that involves multiple sources of fugitive emissions
Refined Modeling
Refined modeling requires detailed and precise input data along with more complex models in order to
provide refined impact estimates. If refined modeling is warranted, it should be performed in two distinct
phases.
The first phase is the significant impact analysis (SIA), which determines if the applicant can forego further
air quality analysis for a particular pollutant with respect to Colorado and National Ambient Air Quality
Standards and, for new major sources and major modifications, Prevention of Significant Deterioration
increments. The second phase is the cumulative impact analysis for the CAAQS, NAAQS, or applicable PSD
increments; it is sometimes referred to as the full impact analysis4.
4 U.S. EPA sometimes uses the phrase "full impact analysis" to refer to the National Ambient Air Quality
Standards (NAAQS) analysis and the Prevention of Significant Deterioration (PSD) increment analysis.
Page 23 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
Significant Impact Analysis
Individual facilities may be subject to different requirements depending on the proposed emission rates
of each facility. There are two general categories of permits: major NSR and minor NSR. The major NSR
permit is often referred to as a federal or PSD permit.
Technically, all Colorado APCD permits are federal in that the state must implement a minor NSR
permitting program to ensure the NAAQS and increments are attained. The air quality impact analyses for
major NSR and minor NSR permits begin with a significant impact analysis (SIA). The purpose of a SIA is to
determine whether a new and/or modified facility, or a combination of the two, could cause a significant
ambient air impact. Below are general steps for identifying emissions to include in the SIA.
SIA Step 1: Identify All Sources of Emissions. Include emissions from all new and/or modified sources at
the facility associated with the project.
SIA Step 2: Determine Whether There Is a Net Emissions Increase. Determination of the project
emissions may vary depending on the type of permit (minor NSR or major NSR). The determination of
the level of federal applicability is the first step in the technical review process and is performed by
the permit engineer. The federal applicability process determines whether a project is minor or
major. While the steps of the modeling process are consistent, requirements vary based on the type
of permit and pollutant.
SIA Step 3: Evaluate Modifications to Existing Sources at the Site. Carry out this step even if there is no
net increase in emissions. For both minor and major NSR modeling, include these sources in the SIA if
there is a change in operating hours or stack parameters, and previous modeling demonstrations
were limited to those operating hours or stack parameters. That is, the permit was based on those
limits.
SIA Step 4: Develop the Emission Inventory for the Site. In general, the requested allowable emission
rate, requested operating rate or maximum design rate should be modeled; however, the applicant
should consult with the permit engineer to verify that the appropriate emission rates were
developed. If the requested emission or operating rate used in the modeling is less than the
maximum design rate, it may become a permit condition. For modifications, the facility-wide net
emissions increase for the modification should be modeled in the SIA.
Major stationary sources do not need to include emissions from the commercial, residential, and
industrial growth analysis in the SIA. The growth analysis required by the PSD rules is only
recommended if a CAAQS and NAAQS analysis, a PSD increment analysis, or a similar air quality
impact analysis is triggered.
Carry out the SIA modeling.
For a given pollutant and averaging period, the highest estimated concentration at each receptor in
ambient air is compared to the modeling significance levels in Table 2 and Table 3, Impacts from nearby
and other background sources, including background concentrations, are not considered in the SIA. If the
estimated concentration levels are below the applicable modeling significance level, no further analysis is
recommended. The source is considered to have an insignificant impact. For example, if impacts are
below the significance levels in Table 3, a compliance demonstration for Colorado and National Ambient
Page 24 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
Air Quality Standards (CAAQS and NAAQS analysis) is not triggered. For major stationary sources subject
to PSD rules, a Class I or Class II PSD increment analysis is not triggered if the impacts are below the
significance levels in Table 2; however, other analysis requirements of the PSD rules must nevertheless be
addressed. If the impact exceeds the modeling significance levels, the source or modification has a
significant impact in ambient air and the next phase of analysis is triggered, as discussed below.
The SIA also provides a convenient way to define the "probable area of influence" of a source's emissions
(see Regulation No. 3, Part B, §111.B.5.d). In practice, it is sometimes useful to define the significant impact
radius or area for the source or activity of interest.
If modeling shows that no violation of a standard (or, for major stationary sources, an applicable PSD
increment) will occur within the significant impact area of a proposed source, as determined by a
comparison with the applicable modeling significance levels, no cumulative air quality impact analysis is
warranted.
Significant Impact Level (SIL)
PSD increment modeling significance levels (Table 2) are only used for major stationary sources subject to
PSD rules. The Class I PSD increment significance levels are based on U.S. EPA proposals from 1996.5 For
minor sources and minor modifications, the Division does not consider compliance with PSD increments
as a criterion in determining if a permit should be issued for a minor source or minor modification.
The modeling significance levels in Table 2 are only intended for the PSD increment analysis. Table 2 does
not include values for Class III areas as there are no Class III areas in Colorado. The modeling significance
levels were not developed to determine if there would be significant impacts to air quality related values
(AQRVs).
5 Federal Register: July 23, 1996 (Volume 61, Number 142), Proposed Rules, Page 38249-38344.
Page 25 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
Table 2. Significant Levels for PSD Increments (jjg/m3)
Pollutant
Class 1
Class II
Carbon Monoxide (CO)
8 hour
(a)
500
1 hour
(a)
2000
Nitrogen Dioxide (NO2)
Annual
0.1
1
1 hour
(a)
7.5
Sulfur Dioxide (S02)
Annual
0.1
1
24 hour
0.2
5
3 hour
1.0
25
1 hour
(a)
4
Particulate Matter < 10 p.m (PM10)
Annual
0.2
1
24 hour
0.3
5
Particulate Matter < 2.5 nm (PM2.5)
Annual
0.05
0.2
24 hour
0.27
1.2
(a) Modeling significant level has not been defined
For minor and major stationary sources, the modeling significance levels in Table 3 are used to determine
if a CAAQS and NAAQS analysis is triggered (see Figure 4). The significance levels in Table 3 are listed in
Regulation No. 3, Part D, §VI.D.2.
Page 26 of 64
-------
May 2018 Colorado Modeling Guideline for Air Quality Permits
Table 3. Significant Levels for NAAQS & CAAQS (fig/m3)
Pollutant
SIL
Carbon Monoxide (CO)
8 hour
500
1 hour
2000
Nitrogen Dioxide (NO2)
Annual
1
1 hour
7.5
Sulfur Dioxide (SO2)
Annual
1
24 hour
5
3 hour
25
1 hour
4a
Particulate Matter < 10 urn (PM10)
Annual
1
24 hour
5
Particulate Matter < 2.5 nm (PM2.5)
Annual
0.2
24 hour
1.2
interim modeling significance level developed by the Division:
https://www.colorado.gov/airquality/permits/lnteriml-
hrS02SIL.pdf
Cumulative Impact Analysis
The components of the cumulative impact analysis vary depending on the applicable regulatory
requirements. For minor sources and minor modifications, a compliance demonstration with Colorado
Ambient Air Quality Standards (CAAQS) and National Ambient Air Quality Standards (NAAQS) is usually the
only type of impact analysis that is requested. Refer to Table 1 to determine if modeling is warranted.
Table 4 summarizes the typical types of air quality analysis for new minor sources or minor modifications
that might be applicable. In attainment areas, all new sources and modifications with a significant impact
in ambient air should perform a cumulative CAAQS and NAAQS analysis. For nonattainment area
requirements, please refer to the Nonattainment Areas portion of Section 2.
Impact analysis requirements are stated in applicable regulations. Regulation No. 3, Part B, §111.D presents
the general requirements for all construction permit applications, including minor sources.
Page 27 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
For minor sources and minor modifications, a compliance demonstration with the Prevention of
Significant Deterioration (PSD) increments is not required to obtain a construction permit. A preliminary
opinion in June 1998 from the Colorado Attorney General's office suggests that rulemaking would be
necessary before compliance with PSD increments could be a permit issuance criterion for minor sources
and minor modifications. Therefore, increment consumption from minor source growth is assessed only
during the modeling process for new sources and modifications subject to PSD rules and during periodic
increment studies. Nevertheless, since all sources, including minor sources, can consume PSD increment
in areas where the PSD minor source baseline date has been triggered, new minor sources and minor
modifications are encouraged to voluntarily demonstrate compliance with applicable increments.
Table 4. Ambient Air Impact Analyses Applicable for New
Minor Source & Minor Modifications
Area Classification
Ambient Air Impact
Analysis
Attainment, Unclassifiable
NAAQS & CAAQS
NAAQS & CAAQS
Nonattainment
Reasonable Further
Progress (RFP)
The components of the major stationary source or major modification air quality impact analysis vary
depending on the applicable regulatory requirements. Permit applicants are encouraged to contact the
Division as early as possible to discuss permitting requirements. The Division and U.S. EPA encourages
applicants to submit modeling protocols.
All areas of Colorado are classified as Class II with the exception of the twelve federal Class I areas, which
are shown in Figures 6 and 7. Class I areas have the greatest protection from air quality deterioration;
Class III areas have the least protection; however, there are no Class III areas in Colorado. In addition to
demonstrating compliance with ambient air quality standards, major stationary source permit applicants
must demonstrate that they will not cause or contribute to violations of PSD increments. Major stationary
sources located within nonattainment areas are subject to additional requirements as discussed in the
Nonattainment Areas portion of Section 2.
Table 5 summarizes the typical types of air quality analysis for new major sources or major modifications
that might be applicable. The significant impact analysis must be performed if there is a possibility the
proposed source will impact a nonattainment area.
Table 5. Ambient Air Impact Analyses Applicable for New
Major Source & Major Modifications
Page 28 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
Area Classification
Ambient Air Impact Analysis
Attainment, Unclassifiable
NAAQS & CAAQS
PSD Increment
Additional Impacts Analysis in any area
(Visibility, Water, Soils, Vegetation, Growth)
AQRV Analysis in Class I Areas
Pre- and Post-Construction Monitoring
Nonattainment
NAAQS & CAAQS
Reasonable Further Progress (RFP)
Net Air Quality Benefit
AQRV Analysis in Class I Areas
Major stationary sources are required by regulation to submit an additional impacts analysis to address
potential impairment to soils, vegetation, water, visibility, and growth, if applicable; it applies in all areas,
including Class I and Class II areas. In addition, regulations require that applicants submit an analysis of
impairment to Air Quality Related Values (AQRVs) in affected Class I areas.
PSD applicants should also consult with the Division to determine if there will be any pre-construction
ambient monitoring requirements. Refer to Regulation No. 3, Part D, §VI to understand how the Division
decides if pre- or post-construction monitoring is required.
There are other regulatory requirements in addition to those required by PSD rules. For example,
Regulation No. 3, Part B, §111.D.l subparts a through g list general requirements for obtaining a permit.
While subpart e applies to major PSD sources, subparts c and d provide requirements that are more
general. Thus, the PSD modeling requirements of subpart e are only one of many requirements that may
be applicable.
Regulation No. 3, Part D, §VI.B states, "the [PSD] requirements of section VI.A do not apply to a major
stationary source or major modification with respect to a particular pollutant if the owner or operator
demonstrates that...the emissions from the source or modification would not be significant." Thus, the
impact analysis and monitoring requirements of the PSD rules are not applicable for a given pollutant if
the emission rate is not significant, as defined in Regulation No. 3, Part D, §11.A.44. In situations where the
requirements of §VI are waived, modeling for compliance with ambient air standards may nevertheless
be warranted under the requirements of Regulation No.,3, Part B, §111.
NAAQS & CAAQS Analysis
The federal Clean Air Act established two types of national air quality standards. Primary standards set
limits to protect public health, including the health of sensitive populations such as people with asthma,
Page 29 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
children, and the elderly. Secondary standards set limits to protect public welfare, including protection
against decreased visibility, damage to animals, crops, vegetation, and buildings. Colorado and National
Ambient Air Standards (CAAQS and NAAQS) are listed in Tables 6 and 7, respectively. Units of measure for
the standards are parts per million (ppm) by volume, parts per billion by volume (ppb), and micrograms
per cubic meter of air (|ig/m3).
The ambient air quality standards in Tables 6 and 7 are based on a reference temperature and pressure of
25 degrees Celsius and 760 millimeters of mercury (1,013.2 millibars or 1 atmosphere), respectively.
Correction of modeled concentration estimates to reference conditions (i.e., standard temperature and
pressure, STP) before comparison with ambient air quality standards is not required for air quality permit
modeling in Colorado. If it is necessary to perform unit conversions, the following formula may be used:
^ U CI tTL ~ ^ 9
Xnnm — ttxt; ; MW = Molecular weight of pollutant in —and. X is concentration6
vv (40.9 x MW) mole
If the impact is significant and a CAAQS and NAAQS modeling analysis is warranted, the modeling should
account for the source under review plus existing air pollution levels at the locations (receptors) where
the source has a significant impact. The purpose of the NAAQS/CAAQS analysis is to demonstrate that
proposed emissions of criteria pollutants from a new facility or from a modification of an existing facility
that does not trigger PSD increment review will not cause or contribute to an exceedance of the NAAQS
and CAAQS.
This can be done in several ways. In general, the compliance demonstration for standards should include:
NAAQS/CAAQS Step 1: Conduct a S/A. Perform a significant impact analysis to predict whether the
proposed source(s) could make a significant impact on existing air quality. That is, the model predicts
concentrations at one or more receptors in the modeling grid greater than or equal to a significant
impact level (SIL).
¦ Model all new and/or modified sources. Compare the predicted high concentration at or
beyond the property line for each criteria pollutant and each averaging time to the
appropriate SIL.
¦ If the sources do not make a significant impact for a pollutant of concern, the demonstration
is complete. If there is a significant impact, then the significant receptors define a significant
impact area and a full NAAQS analysis is required. Go to Step 2.
NAAQS/CAAQS Step 2: Determine Significant Impact Area. Each criteria pollutant and averaging period
subject to the NAAQS/CAAQS analysis may have a different significant impact area.
¦ The significant impact area is the set of receptors that have predicted concentrations at or
greater than the SIL for each applicable averaging time and criteria pollutant.
where = 0.0409 moles • L 1, where P =1 atm, T =298 K, R =0.08206 LatmK'mole"1, L = liters
61 ppmv =
1 ^pollutant _ 0.0409 moles • L 1 x 1000 L • m 3 x MW xlO6 jug- g 1
= (4Q.9xMW)/ug ¦ m~3
Page 30 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
¦ The full NAAQS analysis is carried out for each criteria pollutant and averaging time
separately and need only include the significant impact area for the associated criteria
pollutant and averaging time combination.
¦ Refinement of the significant impact area may be necessary as is discussed in Section 5,
Receptor Network.
NAAQS/CAAQS Step 3: Evaluate Nearby Sources. The applicant needs to request a nearby source
inventory from the Division. It is the responsibility of the applicant to obtain the data and ensure the
accuracy. Any changed made to the data must be documented and justified.
The nearby source inventory for major source and major modifications (e.g., sources subject to PSD
rules) should expand to 50km of the significant impact area of the new source or modification under
review. Identify nearby sources to explicitly model. Select additional background sources as
appropriate to account for impacts not reflected in the background concentration. Sources beyond
50 kilometers should be considered if long-range transport modeling is being performed for a federal
Class I area. Estimated impacts from growth in residential, commercial, and industrial sources
associated with, but not part of, the proposed source should be included in the analysis for major
sources and major modifications.
NAAQS/CAAQS Step 4: Conduct a CIA. Perform a cumulative impact analysis. Model all facility sources
with the nearby sources obtained from the Division. Model allowable emission rates for all sources
that emit the criteria pollutant.
NAAQS/CAAQS Step 5: Add Background concentration to CIA modeled result. The applicant needs to
request a representative background concentration from the Division. This background concentration
should be added to the modeled result from the CIA.
NAAQS/CAAQS Step 6: Compare to NAAQS/CAAQS. Compare the modeled CIA concentration plus
representative background concentration for each criteria pollutant and averaging time to the
appropriate NAAQS. Use the correct design value that follows the form of the applicable NAAQS or
the highest first high depending on the meteorological determination (discussed in Section 5, Criteria
Pollutants Recommendations).
If the maximum concentrations are at or below the NAAQS/CAAQS, the demonstration is complete. If
the concentration is above the NAAQS/CAAQS, perform a contribution analysis to demonstrate that
the proposed source will not exceed the applicable significant impact levels in Table 3 at the point
(receptor) and time of the modeled violation. If the proposed source will not exceed the applicable
SIL at the point and time of the modeled violation, the demonstration is complete. No further air
quality impact analysis is warranted for the new source or modification, even when a new violation
would result from its insignificant impact. If the proposed source has a significant impact at the point
and time of the modeled violations, review the demonstration and determine if any refinements can
be made or demonstrate that the project's impact will not be significant. The following options can
be considered to further refine the model to show compliance with the NAAQS/CAAQS:
¦ Emission Limits;
¦ Operating schedule restrictions;
Page 31 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
¦ Physical changes at the facility to improve dispersion characteristics;
¦ The use offences or physical barriers to preclude public access from contiguous land owned
or controlled by the operator (i.e., standards and increments only apply in ambient air);
¦ Additional pollution control equipment;
¦ The use of more refined modeling techniques, including nonguideline models (e.g., non-EPA
dispersion models, physical models, and monitoring-based methods)
The Common Provisions Regulation, §11 .A states that if emissions generated from sources in Colorado
cross the state line, such emissions shall not cause the air quality standards of the receiving state to be
exceeded, provided reciprocal action is taken by the receiving state. The Division is not aware of any
formal written agreements regarding reciprocal action. Nevertheless, if the impact from a new or
modified source will have a significant impact in another state as defined in section 5, or if it will likely
affect another state, the Division recommends contacting the appropriate agency in the affected state to
determine if there are any applicable state standards. If so, consult with the Division to determine what if
any analysis is recommended.
The Division may recommend that additional analysis be performed to show compliance with applicable
standards of that state. If modeling appears to be warranted, staff from the Division and the affected
state should discuss the situation to determine an acceptable modeling approach.
Table 6. Colorado Ambient Air Quality Standards (CAAQS)
Pollutant & Averaging
Period
Level
Form
Carbon Monoxide (CO)
1 hour
40,000 pg/m3
Not to be exceeded more than once per year
Nitrogen Dioxide (N02)
Annual
100 pg/m3
Annual Mean
Ozone (O3)
1 hour
235 pg/m3
Expected number of days per calendar year, with
maximum hourly average concentration greater
than 0.12 ppm, is equal to or less than 1
Sulfur Dioxide (SO2)
3 hour 700 |ig/m3 Not to be exceeded more than once per year
Page 32 of 64
-------
May 2018 Colorado Modeling Guideline for Air Quality Permits
Table 7. National Ambient Air Quality Standards (NAAQS)
Pollutant & _ . .
_ . , Primary/Secondary
Averaging Period
Level
Form
Carbon Monoxide (CO)
8 hour
Primary
9 ppm
Not to be exceeded more than once per
year
1 hour
Primary
35 ppm
Not to be exceeded more than once per
year
Lead (Pb)
Rolling 3 month
Primary &
Secondary
0.15 |ig/m3
Not to be exceeded
Nitrogen Dioxide (NO2)
Annual
Primary &
Secondary
53 ppb
Annual Mean
1 hour
Primary
100 ppb
98th percentile of 1-hour daily maximum
concentrations, average over 3 years
Ozone (03)
8 hour
Primary &
Secondary
0.070 ppm
Annual fourth-highest daily maximum 8-
hour concentration, averaged over 3
years
Sulfur Dioxide (S02)
Annual1
Primary
0.03 ppm
Annual mean
24 hour1
Primary
0.14 ppm
Not to be exceeded more than once per
year
3 hour
Secondary
0.5 ppm
Not to be exceeded more than once per
year
1 hour
Primary
75 ppb
99th percentile of 1-hour daily maximum
concentrations, averaged over 3 years
Particulate Matter < 10 |im (PM10)
Annual
Revoked in 2006
24 hour
Primary &
Secondary
150 pig/m3
Not to be exceeded more than once per
year on average over 3 years
Particulate Matter < 2.5 |im (PM2.5)
Annual
Primary
12.0 [ig/m3
Annual mean, averaged over 3 years
Annual
Secondary
15.0 |-ig/m3
Annual mean, averaged over 3 years
24 hour
Primary &
Secondary
35 |ig/m3
98th percentile, averaged over 3 years
^he 24-hour and Annual SO2 standards were revoked in 2010; however, they remain in effect in
Colorado until December 21, 2018. Please contact the Division if you have questions regarding SO2.
Page 33 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
Figure 4. Flowchart of the Air Quality Impact Analysis
Process for CAAQS and NAAQS
Is there a modeled
violation of the CAAQS
or NAAQS/
Does the new source
or modification have n
significant Impact at
the point (receptor)
and time of the
moddlngvlofctlion?
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
PSD Increment Analysis
The air quality analysis for new/modified sources subject to Prevention of Significant Deterioration (PSD)
rules must demonstrate compliance with PSD increments if the impact from the new source or
modification is significant. This section is not intended to provide a complete overview of PSD increment
consumption; for that, refer to EPA guidance documents.
The purpose of the PSD increment analysis is to demonstrate that emissions of applicable criteria
pollutants from a new major source or major modification of an existing source will not cause or
contribute to an exceedance of an increment. The PSD increment is the maximum allowable increase in
concentration that is allowed to occur above a baseline concentration for a pollutant. The following
discussion explains PSD increment analyses followed by the basic procedure for conducting the analyses.
Refer to Section 5, Criteria Pollutants Recommendations, for more information about the design value
that should be used to determine compliance with applicable PSD increments. Increment consumption is
a receptor-by-receptor concept. That is, the consumption of PSD increment by one particular source does
not necessarily preclude similar increment consumption by another nearby source if the consumption
occurs on a different day (i.e., under different meteorological conditions) and/or at a different location
(e.g., receptor).
All changes in emissions and related parameters7 after the minor source baseline date may affect PSD
increment consumption or expansion. This includes both stationary sources and mobile sources. In
addition, modifications at major stationary sources after the major source baseline date also may affect
increment consumption. Refer to U.S. EPA guidance and Division guidance8 for procedures.
Area and mobile sources may be important increment consuming sources. In most situations, the Division
can provide at least a county-level inventory of increment consuming area and mobile emissions;
however, because of the amount of time required by the Division to develop such inventories, the
Division will typically not develop increment inventories for an individual permit application until the
permit applicant and the Division agree that an area and mobile source inventory is actually warranted. If
the Division does not have the resources necessary to develop the inventory in the time frame needed by
the applicant, the burden of doing the area and mobile analysis may fall on the applicant.
All areas of Colorado are Class II areas except for the Class I areas shown in Figures 6 and 7. PSD baseline
areas for PMi0 are based on the Colorado Air Quality Control Regions (AQCRs) shown in Figure 5. It is
worth noting that there are both Colorado AQCRs (planning areas) and federal AQCRs. They are
comprised of different counties. While the Colorado AQCRs are used as PSD baseline areas for PMio, the
federal AQCRs are used in U.S. EPA's Air Quality System (AQS). The entire state serves as the baseline
7 "The creditable increase of an existing stack height or the application of any other creditable dispersion
technique may effect 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." (USEPA, 1990)
8 Refer to the Technical Guidance Series: PSD Increment Tracking System document for a detailed
discussion about the PSD increment tracking in Colorado.
Page 35 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
area for SO2 and NO2. Figure 5 and Table 9 show the minor source baseline areas and trigger dates in
Colorado.
Increment Calculation
The baseline concentration does not need to be obtained to determine the amount of PSD increment
consumed or the amount of increment available. Instead, the amount of PSD increment that has been
consumed in an attainment or unclassified area is determined from the emissions increases and
decreases that have occurred from stationary sources in operation since the applicable minor source
baseline date. Modeled increment consumption calculations reflect the change in ambient pollutant
concentration attributable to increment-affecting emissions. Increment consumption (or expansion)
calculations are determined by evaluating the difference between the actual emissions at the applicable
minor source baseline date (Actuate) and actual emissions as of the date of the modeling demonstration
(Actually).
¦ Actuald. This is the representative 2-year average for long-term emission rates, or the maximum
short-term emission rate in the same 2-year period immediately before the applicable minor
source baseline date. For major sources permitted at or after the applicable major source
baseline date but not in operation as of the applicable minor source baseline date or for minor
sources not in operation as of the applicable minor source baseline date, ActualBo would be the
permit allowable emission rate.
¦ ActualMD. This is the most recent, representative 2-year average for long-term emissions rates, or
the maximum short-term emission rate in the same 2-year period immediately before the
modeling demonstration. If little or no operating data are available, as in the case of permitted
sources not yet in operation at the time of the increment analysis, ActualMD would be the permit
allowable emission rate.
A tiered approach is suggested for this analysis to limit the amount of research needed to determine
actual emission rates. The applicant should follow the basic procedure described in the following
paragraphs.
PSD Increment Step 1: Determine whether the modeled high concentration (excluding background
concentration) obtained in the PSD cumulative NAAQS analysis is equal to or less than the applicable
increment. If yes, the demonstration is complete because all sources were modeled at allowable
emission rates. This does not apply for criteria pollutants with NAAQS that are statistically-based (i.e.,
multi-year average).
PSD Increment Step 2: Determine the significant impact area for each criteria pollutant and averaging
period subject to the PSD increment analysis. The significant impact area will be the same one used in
the PSD NAAQS analysis, except for those criteria pollutants with NAAQS that are statistically-based.
For criteria pollutants with NAAQS that are statistically-based, determine the significant impact
analysis following the convention of exceedance-based NAAQS (i.e., maximum predicted
concentration).
PSD Increment Step 3: Obtain a listing of applicable increment-affecting sources and associated
parameters within 50km of the significant impact area from the Division to evaluate in the air quality
impact analysis. Sources beyond 50km should be considered if a long-range transport increment
analysis is being performed for a federal Class I area. It is the responsibility of the applicant to obtain
Page 36 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
these data and ensure their accuracy. Any changes made to the data must be documented and
justified,
PSD Increment Step 4: Adjust the emission inventory using professional judgment.
¦ Omit any source from the inventory that has a negative emission rate unless the source
existed and was in operation at the applicable minor source baseline date. A source must
have existed and been in operation on or before the applicable minor source baseline date to
be considered for increment expansion.
¦ Omit any source permitted after the applicable minor source baseline date that has shut
down or any source as part of the current project that will be shut down. A source that did
not exist or was not operating on or before the applicable minor source baseline date would
not have contributed to the air quality at that time, and there would be no need to model the
source with an emission rate of zero.
PSD Increment Step 5: C onduct the modeling demonstration using the same meteorological data set
used in the determination of the significant impact area using the following tiered approach, as
applicable.
Increment Modeling Tier I. Model all sources using their allowable emission rates. This approach
is conservative since the increment consumed is based on the entire allowable emission rate.
Compare the modeled high concentration to the appropriate increment. If the increment is not
exceeded, the demonstration is complete. Otherwise, go to Tier II.
Increment Modeling Tier II. Model selected sources with ActualMD emission rates and all other
sources at allowable emission rates. The selected sources are usually the applicant's sources. This
process assumes that the increment consumed for the selected sources is based on the entire
actual emission rate and the entire allowable emission rate for all other sources. If the increment
is not exceeded, the demonstration is complete. Otherwise, go to Tier III.
Increment Modeling Tier III. Model selected sources that existed and were in operation at the
applicable minor source baseline date with the difference between ActualMD and Actual-
¦ For major sources permitted at or after the applicable major source baseline date but not
in operation as of the applicable minor source baseline date or for minor sources not in
operation as of the applicable minor source baseline date, use the difference between
ActualMD and the allowable emission rate.
¦ For sources that existed at the applicable minor source baseline date, where a change in
actual emission rates involved a change in stack parameters, use the emission rates
associated with both the applicable minor source baseline date and the current and/or
proposed source configuration. That is, enter the Actual as negative numbers along
with the applicable minor source baseline source parameters, and enter Actual^ for the
same source as positive numbers along with the current and/or proposed source
parameters.
¦ Use emission rates found in Tiers I or II for other sources, as applicable.
Page 37 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
if the increment is not exceeded, the demonstration is complete. Otherwise, continue to refine increment
emission rates or demonstrate that the project's impact will not be significant.
Unique Colorado PSD Increment Requirement
As required by Regulation No. 3, Part D, §X,A,5,a, new sources and modifications subject to PSD rules
should demonstrate that the source by itself will not consume more than 75% of any applicable PSD
increment. The 75% rule does not apply to minor sources.
Along with the 75% increment consumption requirement, there are also Class II areas in Colorado that
have the same protections as Class I areas for S02. Refer to Regulation No. 3; Part D, §VI 11. B for more
information. Modeling is recommended for S02 sources that could impact these areas, based on
boundaries that existed on August 7, 1977:
a) Florissant Fossil Beds National Monument;
b) Colorado National Monument;
c) Dinosaur National Monument;
d) Black Canyon of the Gunnison National Park (areas that are not already Class I);
e) Great Sand Dunes National Park and Preserve (areas that are not already Class I);
f) Uncompahgre Mountain Primitive Area;
g) Wilson Mountain Primitive Area;
h) BLM lands in the Gunnison Gorge Recreation Area.
Figure 6 depicts these Class II areas.
Page 38 of 64
-------
May 2018 Colorado Modeling Guideline for Air Quality Permits
Table 8. PSD Increments (|ig/m3)
Pollutant Class 1
Class il
Nitrogen Dioxide (NO2)
Annual
2.5
25
1 hour
(a)
(a)
Sulfur Dioxide (SO2)
Annual
2
20
24 hour
5
91
3 hour
25
512
1 hour
(a)
(a)
Particulate Matter < 10 p.m (PM10)
Annual
4
17
24 hour
8
30
Particulate Matter < 2.5 |j.m (PM2.5)
Annual
1
4
24 hour
2
9
(a) PSD increment level has not been defined
Class III increment values have been removed as there are no
Class III areas in Colorado
Page 39 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
Table 9. PSD Baseline Dates in Colorado
As of March 20181
Pollutant
Major Source
Baseline Date
Trigger
Date
Minor Source
Baseline Date
Air Quality
Control Region
(AQCR)
Triggering Application
Sulfur Dioxide
(SO2)
01/06/1975
08/07/1977
10/12/1977
Entire State
Rio Blanco Oil Shale -
Tract C-a
Nitrogen Dioxide
(N02)
02/08/1988
02/08/1988
03/30/1989
Entire State
Amoco Production -
Wattenberg
11/01/1988
AQCR 1
Colorado Power Partners
- Brush
01/17/1980
AQCR 2
Piatte River Power
Authority - Rawhide
11/14/2000
AQCR 3
North American Power
Group - Kiowa Creek
11/22/1994
AQCR 4
Colorado Springs
Utilities - Nixon
11/09/2000
AQCR 5
Tri-State - Limon
Particulate
06/19/1989
AQCR 6
Cimarron Chemical -
Vilas
Matter < 10 pirn
(PM10)
01/06/1975
08/07/1977
04/04/1995
AQCR 7
Westplains Energy -
Pueblo
Not Triggered
AQCR 8
NA
Not Triggered
AQCR 9
NA
08/20/1984
AQCR 10
Colorado Ute - Nucla
10/12/1977
AQCR 11
Rio Blanco Oil Shale -
Tract C-a
07/01/1983
AQCR 12
Louisiana Pacific —
Kremmling
Not Triggered
AQCR 13
NA
Particulate
Black Hills - Pueblo
Matter < 2.5 nm
10/20/2010
10/20/2011
09/10/2013
Airport Generating
(PM2.5)
Station Unit 6
]Contact the Division for the latest information
PM2.5 increment is currently being reviewed within the Division. This table will be updated once the Division
makes a final decision regarding the AQCRs for PM2.5. Please contact the Division for more information.
-------
-------
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
Figure 7. Federal Class I Areas
110°0'0"W
105 WW
Gillette
JhermopolS-.
W>
Craters of the Moon NM
VWnd Cave Nl
Ilderness
South Dakota
I Ca$pe{(
Wyoming
Rock Springs
fgwllns
Icottsbh
lebraska
Vernal
Iderness
:ort Collins
.Greeley
SteTUng
Rocky i
werMetroArea
rand Junction
Goodidnd
Colorado
Colorado
\ Springs\
WeHs
VMIderness
Pueblo
La'Jurita-
ieive
Meafc fuur$ngo
Verde NP
Farmfngton
Alamosa
Trinidad
VWieeler Peak VMIdehiess
Oklahoma
San Pedro Parks Wildt
\ Gallup
35WN<
^j-Mazatzai
Wilderness
Siena Ancha/
mn&ith
ilderness I
Mourrt BbldyVU Iderness
New Mexico
Class I Federal Areas
> i * v f / r.
T Capitol Wk I ^ '
iReefNP Black Canyon of the Gunnison NP
(f Biyce \ Catiyciilafiis NP /
CanyoirNP—/ { La Ga
/ I
tf Grand
Jr Canyon NP
Flagstaff
Petrified forest NP J
0 50 100
Legend
_ US Forest Service Class I Federal Areas
| I Fish & Wildlife Service Class I Federal Areas
| National Park Service Class I Federal Areas
CD state
| cities and towns
J county
105 "WW
300
l Kilometers
Colorado Department of Public Health and Environment / Air Pollution Control Division/ Technical Services Program December 2005
Page 43 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
Additional Impacts Analysis
This section is currently under review within the Division. Please contact Division staff
to confirm the procedure for an additional impacts analysis.
Regulation No. 3, Part D, §VI.A.6 requires an additional impact analysis for major stationary sources and
major modifications. The additional impact analysis applies in all areas, including Clas^end Class II areas.
The regulation specifically requires an "analysis of the impairment to visibility, water, fBiis, and
vegetation." In some instances, a growth analysis is also required. The growth analysis is recommended
only if the new source or modification will have a significant impact, that isjjt is only required if an air
quality impact analysis (e.g., CAAQS and NAAQ.S analysis, PSD increment analysis) is triggered.
The additional impact analysis can be done using qualitative ox quantitativemelhods. The Division
generally views the analysis of impairment as a disclosure type or reauirement. The level on
depends on the situation and the likelihood that there couldrbe some type of impairment.
In general, if the additional impact analysis suggests there might be adverse impacts to soils, vegetation,
or visibility, the information may be used in the BACT review'proegsSi-lThiscloes not mean that the BACT
determination must fix the problem; it means that all the issues associated with BACT, including
economics and environmental impacts shouldibe balanced and considered.
Impact Analysis for Water
The inclusion of water in the additional impact analysis is a Colorado requirement. By regulation, the
water analysis in Class II areas does not affect permit approval or denial or control technology selection.
The water impact analysis isjntended to serve as a'SaW-gathering and analysis mechanism to allow the
Division and others to further investigate problems such as acid deposition in high altitude lakes. Refer to
the "Additional ImpactAnalysis" discussion in the "Statement of Basis and Purpose for the Prevention of
Significant Deterioration Program Regulations" (adopteWvlarch 10, 1983) of the Common Provisions
Regulation for more information|about the intent aftfYis requirement.
Visibility Analysis
I pa lysis c
sed iti t
In addition to the Class I visibility a'nalysis discussed in section 7.5, an analysis of impairment to visibility in
Class II areas should also be addressed in the permit application (see Regulation No. 3, Part D, §VI.A.6).
AccardingTo U.S. EPA guidance (USEPA 1990), "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 thaUihe visibility analysis required here is distinct from the Class I area visibility analysis
requirement. The suggested components of a good visibility impairment analysis are:
¦ 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."
Refer to U.S. EPA guidance for more specific recommendations. The focus of Class I visibility analysis is on
assessing visibility impacts within a Class I areas. The focus of the Class II visibility analysis is on sensitive
Page 44 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
views outside of Class I areas. The Division has developed a database of sensitive views to assess impacts
in Class II areas. These are called scenic and/or important views. They are not integral vistas.9The Class II
scenic and/or important views do not have the force and effect of the visibility rules in Class I areas. The
information regarding levels of change in visibility is used to track changes in visibility that might be
important to the public. A list of these views is available from the Division.
The Division does not appear to have the authority to deny a permit if adverse visibility-.impacts occur
outside a Class I area. Instead, the information may be used to consider the need for additional emission
controls. Therefore, it is important to keep the Class I visibility analysis distinct from tne Class II visibility
analysis in the modeling report.
In practice, when PSD applicants contact the Division, the Division will determine if there are any Class II
scenic views within the probable area of influence of the proposed soili c e. If there are, the analysis
approach should be determined on a case-by-case basis in consultatiOT with the Division. If modeling is
warranted, the modeling procedures for the scenic and/or important views are usually based on
techniques similar to those used for Class I visibility asses.,men'
The Division does not have specific thresholds or criteria fondeterrnining when there is impairment to a
Class II view. Impairment determinations are made on a case-by-case basis considering a number of
factors including the geographic extent, intensjly^duration, frequency, and time of modeled visibility
impairment. Other factors such as interference with a victor's visual experience, correlations between
time of impairment with natural conditions that reduce visibility, and other criteria might be considered.
Finally, limitations of the modeling system are considered, l-or example, results from a screening-level
model do not carry as much weight as results from a refined 'model. The ability of the modeling system to
properly account for relevant atmospheric chemistry and meteorology is also considered. If, after
considering all appropriatecifiteriajtis believed that Cl^ss II visibility may be impaired, the Division may
request that the "environmental impact analysis" portion,.of the "best available control technology"
(BACT) determination b# revisited.
A compliance demonstration1 ^iijfc^pfforados visibility standard, which is applicable in the AIR Program10
area, is not required to obtain a permit.
9 An integral vista adopted into regulation can be afforded the same level of protection from visibility
impairment as the Class I jrea itself or any lesser level or protection, as determined by a state on a case-
by-case basis. Because views in the Western U.S. commonly extend for great distances, integral vistas are
a controversial aspect of the Visibility SIP package. The Department of the Interior (DOI) preliminarily
identified integral vistas associated with Class I areas on January 15, 1981. However, both the DOI
(speaking for the National Park Service) and the Department of Agriculture (speaking for the U.S. Forest
Service) later declined to officially list any vistas. One reason given by the DOI was that states already had
sufficient opportunity through existing authority to protect integral vistas. Thus, the naming of integral
vistas and incorporation into SIPs was left to individual states (CDPHE, 1992).
10 The AIR program area is defined in 42-4-304, C.R.S. It generally includes all or part of the following
counties: Adams, Arapahoe, Boulder, Denver, Douglas, El Paso, Jefferson, Larimer, and Weld.
Page 45 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
Soils and Vegetation Analysis
Regulation No. 3 states that the owner or operator should provide an analysis of impairment to soils and
vegetation for each regulated pollutant emitted in a significant quantity. Only vegetation with commercial
or recreational value should be addressed. U.S. EPA's guidance states that, for most soils and vegetation,
ambient concentrations of criteria pollutants below the secondary National Ambient Air Quality
Standards (NAAQS) will not result in harmful effects. Nevertheless, the secondary NAA&S may not
adequately protect certain sensitive vegetation and soils, particularly for regulatecUnon-criteria pollutants
(USEPA 1990), see section 7.3. As recommended in U.S. EPA guidance, new sources or modifications
subject to PSD rules should:
a) provide an inventory of soils and vegetation w
of the facility (e.g., crops);
ith commercial of" recreational value in the vicinity
b) review peer-reviewed scientific literature to determine the concentration level (for appropriate
averaging times) of regulated pollutants that would-be Harmful to vegetation; if no information is
available in the literature, assume the secondary NAAQS is protective if one exists for the
regulated pollutant under review; if modeling has been'dfene< compare modeled impacts to the
secondary NAAQS and to other levels of concern identified through a literature search; if the
potential impact is determined to be bawriful, discuss the nature|Of the harm and its spatial
extent in the modeling report.
AQRV and Visibility Analysis in Federal Class I Areas
This section is currently under review within the Division. Please contact Division staff
to confirm the procedure for anAQRVand Visibility analysis in Class / areas.
as part of a PSD permit to estimate potential
.ss llJSreas. The goal of the Class I impact analysis is to
It of the installation of a new source or the
The Air Quality Related Values (AQRV), analysis is requ
changes in visibility,^deposition, soilf an
determine if the projected changes to AQR
modification of^^xisting source under the PSD regulations, are acceptable for a given Class I area (See
Regulation No. 3, Part CL§XIII and XIV for regulatory requirements). The decision to issue a permit is the
responsibility of the Division. A permit application can be denied if a proposed source would impair
visibility or other AQRVs 'Tn a Class I afea. It is important to note that the determination of impairment is
done on a case-by-case basis. In the case of visibility, this determination will be made based on the
magnitude, number of occurrences, time of year and if such changes would affect a visitor's experience in
the area>iFor more on the regulatory framework, refer to applicable regulations and section 10.1.
In general, the elements of the federal Class I AQRV, including visibility, analysis are determined on a
case-by-case basis.B
Regulation No. 3, Part D, §XIII.A states that "Within twenty days of receipt of a permit application for a
new major stationary source or major modification that may affect visibility or air quality related values in
any Federal Class I area, the division shall transmit a copy of the application to all affected Federal Land
Managers and consult with them as to its completeness in its analysis and monitoring (if required) of air
quality related values. If the division receives advance notification of a permit application of a source that
may affect visibility or air quality related values, it will notify all affected Federal Land Managers
Page 46 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
within thirty days of such notification. The division will consider any analysis performed by a Federal Land
Manager that indicates there will be an adverse impact on visibility or air quality related values if such
analysis is received within thirty days after the Federal Land Manager receives a copy of the complete
application. If the division disagrees with the Federal Land Manager, any notices for public comment or of
a public hearing on the application will explain the disagreement or state where the explanation can be
obtained."
If a protocol is submitted to the Division, as recommended in section 8.1, a copy should be provided for
each affected federal land manager.
As stated in Regulation No. 3, Part D, §XIII, the Division sends affected n Ms a cor>y of the permit
application for proposed new sources or modifications that may affect air quality related values (AQRVs)
in any federal Class I area. For relatively small and/or distant major stationary sources, the FLM may not
take an active role in the review or modeling process. In other cases where a significant impact may occur
or where there may be unacceptable levels of change to AQRVs, including visibility, the FLM usually takes
an active role.
While the Division's Stationary Sources Program is responsible for forwarding the permit application to
the appropriate FLMs, Technical Services Program staff typically contact affected FLMs to obtain Class I
significance levels and other recommendation^gr^the analysis reqiired^joy Regulation No. 3.
The initial contact with FLMs should occur early in the proces^fthere if a PSD pre-application meeting,
FLMs should be invited. Regulations require that tfae Divislwi consult with FLMs as to the
completeness of the permit application. If the applicant deciders directly contact affected FLMs for
recommendations, the Division skould.be kept in the ioop.
Air Quality Related Values Analysis for Major Stationary Sources
For proposed major^stationary sources and major modifications located in attainment areas, visibility
requirements for new sources and modifications subje^rto PSD rules are found in various sections of
Regulation No. 3, Part D including: WI.A.6, §XI^M^XIV.E
For proposed major stationar^sources and major modifications located in nonattainment areas, refer to
Regulation fa3|w D, §V
f igure 3 illustrates key aspects of the regulatory decision process for major stationary sources and major
modifications seeking construction permits:
The fkst step in the process is to determine those pollutants for which there will be a
sicmifictint emission rate increase for a new source or a significant net emissions
increase for a major modification.
If the proposed emission rate is not significant, the additional impact analysis
(Regulation No. 3, Part D, §VI.A.6) and the AQRV reguirements (§XI11 and §XIV.E) do
not apply. In practice, new sources are major for some pollutants and minor for
others. In some cases, the modification may not be major for all pollutants that
would affect AQRVs.
Page 47 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
But if th
as explaine'
If the Division concludes that an "analysis of impairment" (§VI,A,6) is necessary, there
are several key decisions that must be made. For example, the applicant should
discuss the project with the Division to decide if any AQRV monitoring is warranted
(§XIII.B). The Division will make this decision after consultation with the FLM. If
monitoring is required, a monitoring plan should be prepared and submitted for
Division approval. If monitoring is not warranted, which is usually the case, then the
applicant can move on to the next step in the flowchart.
The applicant should consult with the Division to determin^^Pixtent of the
"analysis of impairment." The regulations do notclearlvxlefine what constitutes an
"analysis of impairment." Thus, the extent of the analysis is decidedon a case-by-
case basis. The Division and U.S. EPA strongly rec^mWe'nd that trie applicant submit a
protocol.
When the permit application is submitted to tne Division, it must include '
"analysis of impairment" to be ruled comnrete.
Aplicants should be aware of Regulatiori^teiilR Part D, §XIII.A - Federal Class I Areas;
it states, "Within twenty days of receipt of a permit application for a new major
stationary source or major (acidification that may affect visibility or air quality related
values in any Federal Cldss I area, the Division sha!i transmit a copy of the application
to all affected Federal Land Managers and consult with them as to its completeness in
its analysis and monitoring (if required), of air qualityjelated values. If the Division
receives advance notification of a permit applicdtfori of a source that may affect
visibility or air quality related values, it will notify all affected Federal Land Managers
withinJhiriVdays of such notification."
The next stop is to determine if the source will cause or contribute to a
ation of applicabJj^Wass I PSD increments.
Pfhe source does not cause or contribute to a Class I increment violation
.A states, "The Division will consider any analysis performed by a
^Pgd®[£il Land Manager that indicates there will be an adverse impact on
visibility or air quality related values if such analysis is received within
thirtyaays after the Federal Land Manager receives a copy of the
complete application"
ne if there will be an adverse impact, the Division may perform the analysis,
o. 3, Part B §XIII.C.
If it is determined, through modeling provided by the applicant, that the
source will cause or contribute to a violation of applicable Class I PSD
increments, then the Division may still issue the permit if the
requirements of §XIII.D are met. Regulation No. 3, Part B, §XIII,D states,
"The owner or operator of a proposed major stationary source or major
modification may demonstrate to the satisfaction of the Federal Land
Manager that the emissions from such source or modification would not
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
have an adverse impact on the air quality related values (including
visibility) of Class I lands under the Federal Land Manager's jurisdiction,
notwithstanding that the change in air quality resulting from emissions
from such source or modification would cause or contribute to
concentrations that would exceed the maximum allowable increases for
Class I area. If the Federal Land Manager concurs with such
demonstration and so certifies to the Division, the Dlyision or the
Commission may, provided that applicable requirements are otherwise
met, issue the permit with such emission limitations as may be necessary
to assure that emissions of sulfur dioxide, arm PM10, PM2.5 and nitrogen
oxides would not exceed the following maximum allowable increases over
the minor source baseline concentraMw for such pollutant
PM2.5
~ Annual arithmetic^
~ Twenty-four hour maxi,
4- PM10
AnriMal arithmetigmean 17ju:
Twenty-j
Sulfur Dioxide
nnual arithmetic mean 20 pg/m
enty-four f\our maximum 91 pg/m3
ree-hour mpximum 325 pg/m3
"itrogen Dioxide
Annual arithmetic mean 25 pg/m3
Although the FLMs have an affirmative responsibility to protect AQRVs
they have no permitting authority under the federal Clean Air Act (CAA).
They also have no authority under the CAA to establish air quality-related
rules or standards. The FLM role consists of considering whether
emissions from a new source may have an adverse impact on AQRVs and
providing comments to permitting authorities. Thus, the final decision to
grant or deny a permit is made by the Division or AQCC. Regulation No.
3, Par B, §XI I LA states, "If the Division disagrees with the Federal Land
Manager, any notices for public comment or of a public hearing on the
application will explain the disagreement or state where the explanation
can be obtained."
If the FLMs disagree with the Division's decision to grant a permit, they
may request a hearing, see Regulation No. 3, Part D, §IV.A.6.
Page 49 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
Pre- and Post-Construction Monitoring Analysis
Division modeling and monitoring staff should be contacted as early as possible to discuss the need to
conduct pre-construction monitoring. If monitoring is proposed or required, a monitoring plan consistent
with applicable U.S. EPA and Division monitoring guidance (e.g., "Ambient Air Pollution and
Meteorological Monitoring Guidance") should be submitted for approval.
If the proposed emission rate from a new source or the net emissions increase from a modification is
significant for a given pollutant, as defined by Regulation No. 3, the estimated impact from the new
source or modification should be compared to the significant monitoring concentration (see Table 10 or
Regulation No. 3, Part D, §VI.D.2). In addition, if possible, existing air quality levels should be compared to
the significant monitoring concentration.
Pre-Construction Monitoring Analysis
Refer to Regulation No. 3, Part D, §VI.A.3 for details about how pre-construction monitoring requirements
are determined.
If existing air quality levels or the estimated impacts from the proposed source or modification are below
the applicable monitoring de minimis level, Regulation No. 3 states that the monitoring requirements may
not apply. If the levels are above the de minimis levels, pre-construction monitoring may be required if
the Division believes it is necessary.
Permit applicants should be aware that the time-line for submitting a PSD application could be affected
by the requirement to collect ambient data. For example, if the collection of site-specific meteorological
data is required, at least a full year of data must be collected. For air quality data, at least a full year of
data is typically required, although as little as four months of data may be allowed in some circumstances.
The Division must approve ambient data for use before the permit application can be ruled complete.
Post-Construction Monitoring Analysis
The modeling report submitted with the permit application should address the need for post-
construction monitoring (see Regulation No. 3, Part D, §VI.A.4).n As part of the permit review process,
the Division will, based on the language in the regulation, determine if post-construction monitoring is
necessary.
1140 CFR Part 51.166(v)(2) states that the source "shall...conduct ambient air monitoring as the reviewing
authority determines is necessary...."
Page 50 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
Table 10. PSD Significant Monitoring Concentration1
Pollutant & Averaging Period
Level
Carbon Monoxide (CO)
8 hour
575 |ig/m3
Nitrogen Dioxide (NO2)
Annual
14 pg/m3
Ozone (O3)
8 hour
100 tpy VOCs
1 hour
100 tpy VOCs
Sulfur Dioxide (S02)
24 hour
13 pg/m3
Particulate Matter < 10 (am (PM10)
24 hour
10 pg/m3
Particulate Matter < 2.5 pm (PM2.5)
24 hour
4 [ig/m3
Fluorides
24 hour
0.25 pg/m3
Total Reduced Sulfur
1 hour
10 pg/m3
Hydrogen Sulfide
1 hour
0.2 pg/m3
Reduced Sulfur Compounds
1 hour
10 pg/m3
^he significant monitoring concentrations (de minimis levels) apply only to
new sources and modifications subject to PSD review
(see Regulation No. 3,
Part D, §VI).
Page 51 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
Regulated, Non-Criteria Pollutant Analysis
For regulated, non-criteria pollutants (i.e., fluorides, total reduced sulfur, hydrogen sulfide, and reduced
sulfur compounds), a separate air quality analysis should be submitted if the applicant proposes to emit
the pollutant in a significant amount from a new source or proposes to cause a significant net emissions
increase from a modification. The PSD significant emission rates for these pollutants are as follows:
¦ Fluorides, 3 tons per year;
¦ Sulfuric Acid Mist, 7 tons per year;
¦ Flydrogen Sulfide, 10 tons per year;
¦ Total Reduced Sulfur (including hydrogen sulfide: 10 tons per year);
¦ Reduced Sulfur Compounds (including hydrogen sulfide: 10 tons per year);
Estimated impacts from regulated non-criteria pollutants should be presented and compared to the
significant monitoring concentrations (see Table 10 or Regulation No. 3, Part D, §VI.B.3). Existing
background concentration estimates should be determined in consultation with the Division. If ambient
measurements are available, they should be presented and compared to the significant monitoring
concentrations.
Section 5 - Preferred Air Dispersion Models & Associated Inputs
Source Data
Begin by clearly identifying and documenting all sources of emissions associated with the modeling
analysis. For each identified source, evaluate and discuss how emissions are generated and emitted. This
discussion will be the supporting basis for the source characterization used in the modeling analysis. Then
determine and document the appropriate source parameters associated with the source characterization.
Criteria Pollutant Recommendations
While this section is intended for sources located in attainment or unclassified areas of Colorado, it may,
in some cases, be used by sources located in nonattainment areas; however, sources in nonattainment
areas should read Section 2, Nonattainment Areas, first.
In a compliance demonstration, the applicable design concentration must be calculated. This is usually
done within the model or by using a post-processor. The design concentrations vary depending on the
available meteorological data. If there is not a meteorological dataset that is adequately representative of
the facility, then the design concentration needs to be the highest concentration for all pollutants and
averaging periods. This allows the worst-case impacts to be captured in the modeling analysis.
The design concentrations also vary depending on the impact analysis being performed. For a
NAAQS/CAAQS analysis Appendix W states, "the design concentration is the combination of the
appropriate background concentration with the estimated modeled impact of the proposed source,..The
specific form of the NAAQS for the pollutant(s) of concern will also influence how the background and
modeled data should be combined for appropriate comparison with respective NAAQS in such a modeling
demonstration. Given the potential for revision of the form of the NAAQS and the complexities of
Page 52 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
combining background and modeled data, specific details on this process can be found in the applicable
modeling guidance available on the EPA's SCRAM Web site." For a PSD increment analysis Appendix W
states, "the design concentration includes impacts occurring after the appropriate baseline data from all
increment-consuming and increment-expanding sources." For short-term increments, the maximum
allowable increases may be exceeded once per year at each site. For annual increments, the maximum
allowable increases may not be exceeded.
The facility should contact the Division's Stationary Sources Program (SSP) to determine what pollutants
need to be included in the air quality impact analysis.
Carbon Monoxide
Compliance demonstrations should address both the 1-hour and 8-hour NAAQS. The maximum highest
first high (H1H) modeled concentration from all receptors should be the design value from the SIA to
compare to the SILs. When using representative meteorological data, the maximum high second high
(H2H) modeled concentration from all receptors should be the design value for both 1-hour and 8-hour
averaging periods.
Compliance demonstrations should address the 3-month NAAQS. The NAAQS is significantly more
stringent than the CAAQS monthly value; therefore, the monthly CAAQS was revoked from Colorado
Regulation 8 Part C II.B in March 2010.
Nitrogen Dioxide
Compliance demonstrations should address both the 1-hour and annual NAAQS; however, the Division's
Stationary Sources Program (SSP) published a memo (PS Memo 10-01) that exempts facilities from
showing compliance with the 1-hour N02 NAAQS if the long-term modeling threshold is not exceeded.
Please be aware that the Division's TSP modeling staff will model the 1-hour averaging period if the
facility emissions are above the short-term threshold regardless of their long-term emissions rate. The
Division's TSP modeling staff will perform this analysis in support of Regulation No. 3 to demonstrate the
facility's impact will not cause or contribute to a violation of the NAAQS. The outcome of the analysis will
be included in the public final modeling report.
Both averaging periods are best performed with a tiered approach:
Tier 1:100 percent conversion of nitrogen oxides (NOx) to nitrogen dioxide (N02).
TierII:Ambient Ratio Method 2 (ARM2) uses an ambient ratio between 0.5 and 0.9 that must be
derived from U.S. EPA's Air Quality System. Colorado no longer accepts the EPA-recommended ARM
ratio of 0.8 as monitoring data has shown N02/NOx ambient ratios exceeding 0.8 conversions.
Justification for ambient ratio used is required. This method should also be used if the following are
true:
¦ Tier I results are within or below a range of 150- 200 ppb
¦ NO2 background concentrations are below EPA's recommended range of 20 - 30 ppb
¦ O3 background concentrations are below EPA's recommended range of 80 - 90 ppb
¦ In-stack NO^/NO* ratios at or below 0.2
Lead
Page 53 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
Tier III: Ozone Limiting Method (OLM) uses in-stack NCb/NO* ratios and background concentrations.
The EPA established a general acceptance of 0,5 as a default in-stack ratio of N02/NOxfor input to
OLM. If the applicant proposes to use an in-stack NOj/NOx ratio other than the EPA default, sufficient
justification and documentation will need to be provided to support the source-specific data. The
source-specific in-stack N02/N0X ratio needs approval from the permit engineer. Hourly by season
profiles of both NO2 and O3 should be requested from the Division for input to OLM.
The maximum highest first high (H1H) modeled concentration from all receptors should be the design
value from the SIA to compare to the SILs. When using representative meteorological data, the 1-hour
design value should be the maximum 5-year average of the 98th percentile of the annual distribution of
the maximum daily 1-hour modeled concentrations or the highest eighth high (H8) for each receptor.
When using representative meteorological data, the annual design value should be the maximum
modeled concentration for all receptors across all years of meteorological data.
Ozone
In general, accurate and cost effective methods for modeling ozone impacts from stationary point
sources are not available. Therefore, ozone modeling is not routinely requested for construction permits,
although it could be in unusual cases such as situations where the Division believes ozone standards couldl
realistically be violated by the proposed source or modification. If modeling is considered, the cost of
conducting such an analysis will be factored into the decision process.
Precursors to ozone need to be discussed with the Division's TSP modeling staff. The applicant should
review applicable EPA guidance regarding precursors to ozone.
Particulate Matter < 10|im (PM10)
Compliance demonstrations should address the 24-hour NAAQS. The annual PM10NAAQS was revoked in
2006 so compliance is no longer required for this averaging period. The maximum highest first high (H1H)
modeled concentration from all receptors should be the design value from the SIA to compare to the SILs.
When using representative meteorological data, the design value should be maximum highest sixth high
(H6H) modeled concentration for all years of meteorological data.
Particulate Matter < 2.5|im (PM2.5)
Compliance demonstrations should address both the 24-hour and annual NAAQS. The maximum highest
first high (H1H) modeled concentration from all receptors should be the design value from the SIA to
compare to the SILs. When using representative meteorological data, the 24-hour design value should be
the maximum 5-year average of the 98th percentile of the annual distribution of the maximum 24-hour
modeled concentration or the highest eighth high (H8H) for each receptor. When using representative
meteorological data, the annual design value should be the maximum 5-year average modeled
concentration from all receptors.
Secondary formation of PM2.5 needs to be discussed with the Division's TSP modeling staff. The applicant
should review applicable EPA guidance regarding when modeling secondary formation of PM2.5 is
necessary.
Page 54 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
Sulfur Dioxide
Compliance should be demonstrated with the 1-hour, 3-hour, 24-hour, and annual NAAQS as well as with
the Colorado 3-hour standard of 700 |ig/m3. The 24-hour and annual NAAQS remain in effect for "any
area for which it is not yet 1 year since the effective date of designation under the current (2010)
standards and any area for which an implementation plan providing for attainment of the current (2010)
standard has not been submitted and approved and which is designated nonattainment under the
previous S02 standards or is not meeting the requirements of a SIP call under the previous S02 standards."
The state of Colorado was designated as attainment/unclassifiable on December 21, 2017; therefore,
compliance demonstrations for the 24-hour and annual NAAQS are required until December 21, 2018.
Also, the Division's Stationary Sources Program (SSP) published a memo (PS Memo 10-01) that exempts
facilities from showing compliance with the 1-hour S02 NAAQS if the long-term modeling threshold is not
exceeded. This memo does not exempt a facility from showing compliance with the 3-hour and 24-hour
NAAQS and CAAQS. Please be aware that the Division's TSP modeling staff will model the 1-hour
averaging period if the facility emissions are above the short-term threshold regardless of their long-term
emissions rate. The Division's TSP modeling staff will perform this analysis in support of Regulation No. 3
to demonstrate the facility's impact will not cause or contribute to a violation of the NAAQS. The outcome
of the analysis will be included in the public final modeling report.
The maximum highest first high (H1H) modeled concentration from all receptors should be the design
value from the SIA to compare to the SILs. When using representative meteorological data, the 1-hour
design value should be the maximum 5-year average of the 99th percentile of the annual distribution of
the maximum daily 1-hour modeled concentration or the highest fourth high (H4H) for each receptor.
When using representative meteorological data, the 3-hour and 24-hour design values should be the
maximum highest second high (H2H) modeled concentration from all receptors. When using
representative meteorological data, the annual design value should be the maximum modeled
concentration from all receptors across all years of meteorological data.
Mobile Sources Data
Facilities that involve haul trucks need to include fugitive dust emissions in both the permit application
and the air quality impact analysis. Large mining equipment tailpipe emissions should also be included.
The Division is currently developing more guidelines to establish when to include tailpipe emissions from
haul road traffic and mining equipment. The Division has the current procedure if tailpipe emissions are
to be included in the air quality impact analysis. If the applicant is unsure whether tailpipe emissions
should be included, please contact the Division.
A facility is likely to have a fleet of trucks that is made up of a variety of different trucks. If the air quality
impact analysis involves N02 modeling using the Tier III approach, an in-stack N02/NOx ratio is necessary.
Different trucks will have different in-stack N02/NOx ratios. The Division recommends using a similar
tiered approach.
Tier A: Use the highest in-stack N02/NOx ratio of all the mobile engines in the fleet. This ratio should
be applied to all sources used to represent the truck traffic or non-road engines.
TierB:Calculate a weighted average in-stack N02/NOx ratio based on the total vehicle fleet and the
number of units with different in-stack N02/NOx ratios and use that value for the entire vehicle
Page 55 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
fleet. This accounts for the influence of the different types of engines according to the number of
units with higher or lower in-stack NO?yNOx ratios while at the same time keeping the modeling
analysis simple.
TierC: Represent vehicles with similar in-stack NOz/NOx ratios with separate sets of sources assigning
the corresponding in-stack NCb/NO* ratio to each set of sources. Each road segment could have
multiple sets of sources overlaid on top of each other. This is the most accurate representation of the
vehicle traffic.
All proposed in-stack N02/N0X ratios require sufficient justification and documentation to support the
source-specific data. The source-specific in-stack N02/N0X ratios need approval from the permit engineer.
Nearby Sources Data
U.S. EPA recommends that, at a minimum, all nearby sources should be explicitly modeled as part of the
NAAQS analysis. Other background sources usually are accounted for by using an appropriate ambient
background concentration (i.e., see §9.2.2 of the USEPA Guideline on Air Quality Models) or, if a suitable
ambient background concentration is not available, by application of a model. Nearby sources and other
background sources are terms used to reference all stationary sources except the new source or
modification under permit review.
The emissions estimates used in modeling nearby and other background sources should be consistent
with U.S. EPA recommendations in Table 8.2 of the USEPA Guideline on Air Quality Models and other
applicable U.S. EPA guidance. Table 8.2 recommends that actual operating levels averaged over 2 years
and federally enforceable permit limits should be used for all nearby sources. That is, emission rates
based on a combination of both allowable and actual data, if the actual data is available. A nearby source
is any major source, major stationary source, or minor source that causes a significant concentration
gradient in the vicinity of a new or modified source. All sources should be included if they are within 5
kilometers of the significant impact area of the source (significant impact area + 5km). Nevertheless, this
is not a bright line; in some cases, the 5-kilometer distance from the significant impact area should be
expanded. Professional judgment should be used when selecting sources to model.
The Division does not recommend a specific objective procedure for determining which sources should
be classified as nearby sources and which should be classified as other background sources. The
procedure used to select nearby sources should be based on professional judgment. In addition, it should
consider local conditions such as topography, meteorology, dispersion characteristics, availability of
ambient monitoring data, existing air quality, and other relevant factors. The procedure should include an
examination of the modeling results to ensure that all sources that should have been included were
included.
The nearby sources inventory provided by the Division may be missing key stack parameters as this
information is taken from submitted APENs. When the APENs are missing the stack parameters, this
information is left blank in the inventory. The Division has developed an initial approximation procedure
for applicants to use when the stack parameters are missing. Further refinement may be necessary in
order to demonstrate compliance.
¦ Determine the type of emission source: stack (point) or fugitive.
Page 56 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
¦ Point source: Find stack parameters for similar equipment in the inventory. Provide justification
for the stack parameters used.
¦ Fugitive source: Group all the fugitive (non-stack) emissions from a facility into one area source
with dimensions of 100m x 100m, release height of 2 m, and initial sigma-z of 3 m. The x and y
coordinates of the facility in the nearby source inventory can be used as the southwest corner of
the area source.
A nearby sources inventory will be provided to the applicant upon request from the Division's Inventory
and Support staff. The applicant must specify the following when requesting a nearby source inventory:
¦ Coordinates of the project site
¦ Pollutants to be modeled
¦ Extent of the area included in the inventory
Background Concentrations
In general, the background concentration is intended to account for sources not explicitly included in the
modeling.
For annual standards, the recommended background is typically based on the annual average value. For
shorter-term standards, selection of a background concentration can be more challenging. In general, the
background concentration should be one that can reasonably be assumed to occur with the modeled
concentration.
Determination of a background that can "reasonably be assumed to occur" is sometimes difficult. In
general, the niche being filled by the background concentration should be defined before a value is
selected. Since the background concentration field is usually assumed to be spatially uniform, the
background should account for elevated concentration levels that are expected to occur in the receptor
grid from non-modeled sources. Alternatively, a variable background field could be used if there is
sufficient data to generate one.
For purposes of addressing short-term standards, the total predicted concentration distribution should
represent combinations of impact and background that can reasonably be expected to occur
simultaneously in the particular application. The Division recognizes that the chance of two independently
caused short-term concentration maxima occurring simultaneously at any particular location may be low.
The Division can usually provide a background concentration upon request to account for other
background sources, including mobile sources and transport from distant sources. Determination of the
nearby sources accounted for by the background concentration can be rather subjective. Consequently,
the applicant should review the location and the collection date of the background data with respect to
nearby sources to determine how it should be incorporated into the overall modeling procedure.
The Division does not typically recommend the use of a background concentration to account for
increment consumption. Nevertheless, there may be situations where a statistical analysis or review of
trends in ambient air quality data would be useful to quantify local or regional changes in air quality since
the minor source baseline date.
Page 57 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
To streamline the background concentration requests, a form is available on the Division's website
(https://www.colorado.gov/airquality/permits.aspx). If the applicant would like seasonal background
data, please contact the Division's TSP modeling staff.
Elevation Data
Terrain elevations for sources and receptors should be used when appropriate. Discuss the source of
terrain data in the modeling report.
Terrain elevations for receptors as well as nearby and other background sources should be based on U.S.
Geological Survey (USGS) National Elevation Dataset (NED). A minimum resolution of 1/3 arc second (IO-
meter) files covering a minimum radius of 40 kilometers from the facility under review. NED files can be
downloaded using the CDPHE Elevation Data Quad Download Tool
(https://www.colorado.gov/airquality/quad selector map.aspx).
Some facility sites are graded (e.g., flat) so that actual site topography is or will be significantly different
from the topography that is found in a USGS NED or in other elevation data. Thus, it is appropriate to use
the site-specific graded elevations for the facility sources and buildings. A plot plan should be provided
that depicts the site-specific elevations. If NED files are used for facility sources and buildings, sufficient
justification and documentation will need to be provided to support the use of non-site-specific data.
Downwash Applicability
Downwash is a term used to represent the potential effects of a building on the dispersion of emissions
from a source. Downwash is considered for sources characterized as point sources. The stack height and
proximity of a point source to a structure can be used to determine the applicability of downwash.
Downwash does not apply to sources characterized as areas. Downwash is indirectly considered for
volume sources by adjusting the initial dispersion factors.
Point sources with stack heights less than good engineering practice (GEP) stack height should consider
dispersion impacts associated with building wake effects (downwash). GEP stack height is the greater of
(40 CFR § 51.100(ii)):
(1) 65 meters, measured from the ground-level elevation at the base of the stack:
(2)(i) For stacks in existence on January 12, 1979, and for which the owner or operator had obtained
all applicable permits or approvals required under 40 CFR parts 51 and 52.
Hg = 2.5H,
provided the owner or operator produces evidence that this equation was actually relied on in
establishing an emission limitation:
(ii) For all other stacks,
Hg = H + 1.5L
where
Hg is the GEP stack height;
Page 58 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
H is the structure height; and
L is the lesser of the structure height or maximum projected width (the width as seen from the source
looking towards either the wind direction or the direction of interest) of the structure.
These formulas define the stack height above which building wake effects on the stack gas exhaust may
be considered insignificant.
A structure is considered sufficiently close to a stack to cause downwash when the minimum distance
between the stack and the building is less than or equal to five times the lesser of the structure height or
maximum projected width of the structure (5L). This distance is commonly referred to as the structure's
region of influence. If the source is located near more than one structure, assess each structure and stack
configuration separately.
Once downwash applicability is determined, provide documentation to support that determination.
Receptor Network
The approach to creating a receptor network varies with the goals of the modeling study. Case-by-case
professional judgement should be used. Factors such as topography, density of nearby sources,
meteorology, and requirements of the selected model should be considered when selecting receptors. In
general, the network should be consistent with U.S. EPA's recommendations. It should extend far enough
to define the significant impact area for the source or modification under review. For elevated point
sources, it is sometimes useful to initially use a simple screening-level model to help determine how far
out to extend the receptor network.
If the concentration gradient is increasing at the edge of the network, the network should be extended. 1-
hour modeling analyses tend to result in large significant impact areas; therefore, professional judgement
should be used when extending and refining the receptor network. Refer to U.S. EPA Memos
(https://www.epa.gov/scram/air-quality-models-clarification-memos-dispersion-models) for guidance.
The Division generally considers a fine receptor grid to have receptor spacing of 100 meters or less. A
coarse receptor grid usually refers to receptor spacing greater than 100 meters.
While source-specific issues such as expected plume rise and topography should be considered when
deciding if the following recommendations are appropriate, the following recommendations often
provide a good starting point:
a. Up to 1 kilometer - grid with 100-meter receptor spacing (fine)
b. From 1 to 3 kilometers - grid with 250-meter spacing (coarse)
c. From 3 to 10 kilometers - grid with 500-meter spacing (coarse)
d. Beyond 10 kilometers - grid with 1-kilometer spacing (coarse)
e. Along fence line or ambient air boundary - 50 to 100 meter receptor spacing
f. If no fence or boundary - 50-meter receptor spacing within source facility
g. Discrete receptors for sensitive nearby sites (e.g., residences, schools) unless the grid is sufficient
Page 59 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
h. Flagpole receptors on balconies and rooftops of buildings not owned or operated by the facility
under review (e.g., balconies on apartment buildings, rooftop restaurants, rooftop pools)
i. If the modeled maximum values from the facility under review (or maximum values in an air
quality impact analysis such as a CAAQS and NMQS analysis) occur in a coarse receptor grid,
additional modeling should be performed with a fine grid to find the maximum concentrations
j. Additional fine receptor grids or discrete receptors may be necessary in complex terrain or
sensitive areas to clearly define the area of maximum impact
Receptors may be omitted from the property of the facility under review, provided that public access is
precluded by a fence or other physical barrier. Refer to the definition of ambient air in the definitions
section at the beginning of this document. If there is not a physical barrier (e.g., fence, wall), receptors
should be located on the property of the applicant. Division and/or U.S. EPA approval is necessary if the
applicant wants to use a physical barrier such as a canyon, river, tailings pile, intense terrain or other
physical features as the ambient air boundary. Intense terrain will be approved on a case-by-case basis to
preclude public access as a physical barrier. Intense terrain that acts as a physical barrier needs to have a
minimum slope of 5 to 1, per EPA guidance. If a physical barrier is approved by the Division to preclude
public access, frequent posting is usually necessary along with routine security patrols; in addition, points
of public access into the posted area (e.g., roads, trails) should be fenced or gated. Refer to U.S. EPA
memos on this subject.
Meteorological Data
Meteorological data should be collected, processed, and applied in ways that are consistent with the
most current federal regulations (https://www3.epa.gov/ttn/scram/guidance/guide/appw 17.pdf),
guidance and model user's guides. If representative meteorological data are not available, it may be
necessary to collect at least one (1) year of site-specific data. Any source intending to collect site-specific
data should contact the Division prior to setting up a monitoring program. The Division has monitoring
guidance available.
Meteorological data will be provided by the Division. The Technical Services Program modeling staff will
determine the most representative meteorological data appropriate to use for the facility under review.
The applicant should provide the following information to the Division to obtain AERMOD-ready
meteorological data:
¦ Coordinate (latitude/longitude or UTM) of source location, including datum
¦ Source location identified on l:24,000-scale topographic map(s)
¦ Brief description of the sources of emissions (i.e., stack vs fugitive, stack heights, source types)
The Division staff takes the above information and assesses the expected conditions at the source
location and for each source type. A dataset will be identified that best matches the conditions expected
at the source location from the available meteorological datasets known and that meet the completeness
requirement.
Per regulatory requirements, for PSD applications where the Division has required pre-construction
meteorological monitoring, the permit application will not be ruled complete until the data has been
submitted to the Division and approved.
Page 60 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
As stated in §8.4,2 of the USEPA Guideline on Air Quality Models, 5 years of adequately representative
NWS data, at least 1 year of site-specific data, or at least 3 years of prognostic meteorological data should
be used. If more than 1 year of site-specific data exist, multiple years (up to 5 years) should be used. For
long-range transport modeling and complex wind situations see §8.4.4.2 of the USEPA Guideline on Air
Quality Models.
The use of prognostic meteorological data is currently not accepted in Colorado due to complex terrain.
The Division is currently reviewing how prognostic meteorological data can be used to capture the effects
of complex terrain.
When deciding whether or not to recommend or require collection of site-specific meteorological data,
Division modeling staff considers:
a. Dispersion characteristics of the source under review
b. Meteorological and dispersion issues associated with complex terrain
c. Distance to the nearest Class I area (for new sources and modifications subject to PSD rules)
d. The likelihood that the source will have an adverse impact on ambient air quality
e. Whether or not the source is subject to PSD rules (monitoring is more likely to be requested for
new major stationary sources or major modifications subject to PSD rules than for minor sources)
f. Other relevant factors
To streamline the permit process and reduce the economic burden for minor sources and minor
modifications, collection of site-specific meteorological data is seldom requested for minor sources and
modifications. Nevertheless, it may be recommended if there is reason to believe the new source or
modification will cause or contribute to a violation of CAAQS or NAAQS.
If allowed under federal regulations and approved by the Division, conservative screening meteorological
data may be used in refined models instead of site-specific data for compliance demonstrations.
Modeling Scenarios
It is common for facilities to have sources that do not operate simultaneously with other sources at the
facility. This situation results in modeling different scenarios. For example, if a facility wants a permit that
allows operation of either flares or engines, but not both at one time, both the flare scenario and engine
scenario should be modeled.
If there are several sources that cannot operate simultaneously which would result in a significant
amount of scenarios, the applicant can simply include the worst-case source. Please be aware that using
this approach requires demonstration of the worst-case source. Comparing emission rates of these
sources does not equate to a worst-case analysis.
Permit conditions will be proposed based on the information used in the modeling. Restricted operating
schedules used to demonstrate compliance will be become permit conditions.
Page 61 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
Section 6 - Reporting Requirements
Include in the air quality impact analysis a written discussion covering the project, the modeling
performed, and the results.
The air quality impact analysis is a stand-alone report. Results from the report should be sufficient to
make a decision without input from other reports. Do not refer to other documents or reports for data
required to be in the report. In addition, do not exclude items without coordination with the Division's
TSP modeling staff unless the items are clearly not applicable to the project. Follow the reporting
requirements to expedite the technical review of the air quality impact analysis and to eliminate
unnecessary modeling.
Specific data are needed to review and perform modeling. The recommended list of data elements
presented here are often necessary to perform and/or review dispersion modeling. The applicant should
be prepared to provide these data with the application or upon request by the Division. If the data are
not provided with the application and cannot be provided upon request in a timely manner, the permit
process may be delayed. In addition, if data cannot be provided in a suitable format, additional staff time
may be necessary for data-processing tasks. Staff time is usually charged back to the applicant at the
permit processing hourly rate. While some of the data elements discussed here are already part of the
permit application and APEN forms, they are mentioned here for emphasis.
Send the air quality impact analysis to the permit engineer that requested the analysis. In addition, for
PSD applications send a copy of the air quality impact analysis to EPA Region 8.
Consistency in Geographic Coordinates
Geographic coordinates are used in modeling. Whenever possible, the datum upon which geographic
coordinates are based should be provided. For example, potentially significant discontinuities in source
and receptors coordinates may occur if some Universal Transverse Mercator (UTM) coordinates are
based on the North American Datum of 1927 (NAD27) while others are based on NAD83. Often, site
surveys are performed using GPS systems that are based upon WGS84 while UTMs might be based upon
a NAD27 topographic map. Therefore, a coordinate conversion should be performed when appropriate so
that receptors, source locations, and other coordinates reference a consistent system.
Exemptions from Submitting Modeling-Related Data
New sources and modifications with emissions less than the thresholds in Table 1 that do not meet any of
the situations described in the footnotes of Table 1 and sources not emitting any of the pollutants listed
in Table 1 do not need to provide any modeling-related data beyond what is requested in the permit
application and/or APEN forms.
Since ozone modeling and HAPs modeling are not routinely performed as part of the permit review
process, VOC sources do not need to provide any modeling-related data beyond what is requested in the
permit application and/or APEN forms.
Page 62 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
New Sources and Modifications Not Subject to PSD Rules
At a minimum, new sources and modifications not subject to PSD rules with emissions greater than the
thresholds in Table 1 should submit the data outlined in the Modeling Submittal Completeness Checklist
with the permit application or be prepared to provide the data upon request. The Modeling Submittal
Completeness Checklist can be found on the Division's Air Quality website
(https://www. Colorado, gov/airquality/permits. aspx).
New Sources and Modifications Subject to PSD Rules
New sources and modifications subject to PSD rules with emissions greater than the thresholds in Table 1
should submit the data outlined in the Modeling Submittal Completeness Checklist with the permit
application or be prepared to provide the data upon request. The Modeling Submittal Completeness
Checklist can be found on the Division's Air Quality website
(https://www.colorado.gov/airquality/permits.aspx).
The following additional items should be submitted as well:
¦ For each pollutant for which the new source or modification is subject to modeling under PSD
rules, provide a source history that clearly shows the start-up and shutdown dates of each unit
(e.g., emissions source) at the facility. Include current and historic stack parameters and
source/building configurations. Compare start-up and shutdown dates to applicable PSD baseline
dates to determine PSD increment consuming and expanding emissions (see the Division's "PSD
Increment Tracking System" for baseline dates and related information). Provide metadata (i.e.,
describe the methods used to generate the data). The applicant may choose to ignore this data
element if an air quality impact analysis is not requested or if PSD increment modeling is not
requested; however, the Division encourages applicants to provide these data so that PSD
increment consumption and expansion can be tracked.
¦ A table showing nearby increment consuming/expanding sources (only recommended if a PSD
increment analysis has been performed). Refer to the Division's "PSD Increment Tracking System"
guidance for details, in particular section 2.1.2.
¦ A table comparing maximum modeled impacts with appropriate thresholds such as modeling
significance levels, standards, PSD increments, significant monitoring concentrations, and levels
of acceptable change to AQRVs.
¦ UTM coordinates for maximum modeled concentration estimate(s) from the PSD increment
compliance demonstration modeling (if applicable). These data are used to help the Division track
increment consumption across the state.
Recommended Additional Guidance
Air Quality Models - Clarification Memos for Dispersion Models: https://www.epa.gov/scram/air-quality-
models-clarification-memos-dispersion-models
Federal Land Managers' Air Quality Related Values Work Group (FLAG):
https://www.nature.nps.gov/air/Pubs/pdf/flaR/FLAG 2010.pdf
Page 63 of 64
-------
May 2018
Colorado Modeling Guideline for Air Quality Permits
Model Clearinghouse Information Storage and Retrieval System:
https://cfpub.epa.gov/oarweb/MCHISRS/
Clean Air Act Permit Modeling Guidance: https://www.epa.gov/scram/clean-air-act-permit-modeling-
guidance
Air Quality Dispersion Modeling - Preferred and Recommended Models User Guides and Implementation
Guides: https://www.epa.gov/scram/air-quality-dispersion-modeling-preferred-and-recommended-
models
CDPHE APCD Air Quality Modeling Guidance for Permits:
https://www.colorado.gov/airquality/permits.aspx
Page 64 of 64
-------
Appendix A - Modeling Thresholds Supporting Documentation
Air Dispersion Modeling Analysis
to
Support the Modeling Thresholds
and Associated Language in Section 2
of the
Colorado Modeling Guideline for Air Quality Permits
(January 2002, April 2010)
Colorado Department of Public Health & Environment
Air Pollution Control Division
Technical Services Program
Modeling, Meteorology, and Emission Inventory Unit
4300 Cherry Creek Drive South
Denver, Colorado 80246
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
2
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
Table of Contents
1. PREFACE 5
2. INTRODUCTION 9
3. EFFECTS ON AMBIENT AIR IMPACT ESTIMATIONS 10
3.1. Buoyancy 11
3.2. Momentum 11
3.3. Stack-Tip Downwash 11
3.4. Building Downwash 11
4. METHODOLOGY (JANUARY 2002 ANALYSIS) 12
4.1. Receptor Spacing 12
4.2. Model Runs 12
4.2.1. Sensitivity Analysis (Runs 1 through 10) 12
4.2.2. Building Downwash (Runs 11 through 18) 12
4.2.3. Multiple changes in Stack Characteristics with Building Downwash (Runs 19 through 22)
13
4.2.4. Horizontal Stack (Runs 23 through 25) 13
4.2.5. Capped Stack (Runs 26 through 28) 13
4.2.6. Minimum and Maximum Range of Values (Runs 29 and 30) 13
4.3. Comparison with Modeling Significance Levels and AAQS 13
5. RESULTS (JANUARY 2002 ANALYSIS) 14
6. DISCUSSION (JANUARY 2002 ANALYSIS) 22
6.1. Sensitivity Analysis (Runs 2 through 10) 22
6.2. Building Downwash (Runs 11 through 18) 23
6.3. Multiple Changes in Stack Characteristics with Building Downwash (Runs 19 through 22)...23
6.4. Horizontal Stack (Runs 23 Through 25) and Capped Stack (Runs 26 through 28) 24
6.5. Minimum and Maximum Range of Values (Runs 29 and 30) 24
6.6. Other Modeling Variables Not Examined in this Modeling Analysis 24
7. METHODOLOGY, RESULTS, AND DISCUSSION OF THE APRIL 2010 ANALYSIS 25
7.1. Methodology 25
7.1.1. Meteorology 25
7.1.2. Receptor N etwork 25
7.1.3. Point Sources 25
7.2. Results and Discussion 27
7.2.1. 1-hr Concentrations 27
7.2.2. 24-hr Concentrations 29
7.2.3. Annual Concentrations 31
7.2.4. 3 -month C oncentrations 33
7.2.5. Other Modeling Variables Not Examined in this Modeling Analysis 37
8. CONCLUSION 37
APPENDIX 39
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
3
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
4
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
1. Preface
The Air Pollution Control Division (Division) participated in a review of the "Colorado
Modeling Guideline for Air Quality Permits" (Colorado Modeling Guideline). The review
process resulted in revisions to the modeling guideline based on comments from a technical peer
review conducted in 2000 and 2001, public comments, and comments from several stakeholder
meetings. A public hearing on the guideline was held on December 20, 2001.
As part of the review process, the Division performed air quality modeling to help in the
development of appropriate language and emission modeling thresholds for Table 1 of the
Colorado Modeling Guideline. This report provides the results of the Division's modeling study.
While the body of this report is focused on point source modeling, a series of graphical images
are provided in the appendix to illustrate the magnitude and spatial extent of strong concentration
gradients near fugitive sources. All of the fugitive source modeling is based on a continuous
emission rate of 15 tons per year, which is the PM-10 modeling threshold in Table 1 of the
Colorado Modeling Guideline.
Table 1 from the January 1, 2002 version of the Colorado Modeling Guideline and associated
language in Section 2.5 - Modeling Thresholds - is presented on the next two pages. The
Colorado Modeling Guideline was updated on December 27, 2005 to reflect revisions to
Colorado AQCC Regulation No. 3 and EPA's Appendix W to 40 CFR Part 51 - Guideline on Air
Quality Models and did not result in any material change to Table 1 or its associated language in
Section 2.5.
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
5
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
[Excerpts from the January 1, 2002 version of the Colorado Modeling Guideline.]
Section 2.5 Modeling Thresholds
The modeling thresholds in this section are applicable for sources located in
nonattainment and attainment areas (see sections2,l, 2,2, and 2,3), The thresholds were not
developed to address situations such as those described in section 2.4,
The modeling thresholds were developed to identify new sources and modifications that
would have relatively small impacts and do not warrant further analysis with respect to
applicable air quality standards. The development of these thresholds is intended to assist the
Division Staff, permit applicants, air quality consultants, and others decide when modeling is
warranted to determine the impact from a source. This section introduces de minimis emissions,
which have a low probability of causing or contributing to an exceedance of an air quality
standard. By using this approach, permitting costs associated with the impact analysis required
by Regulation No. 3 can be minimized.
Air quality modelers developed the modeling thresholds in Table 1 during a technical
peer review of the Division's modeling practices. The Division performed dispersion modeling to
help demonstrate that the thresholds in Table 1 are appropriate.' Permit applicants and the
Division should try to avoid situations where the decision to perform modeling takes longer than
actually performing a screening-level modeling analysis (screening-level models can often be run
quickly with minimal cost).
For a given pollutant, modeling is usually warranted if the long-term (tons per year) or
short-term (pounds per hour, etc.) requested emission rate for a new source or the facility-wide
net emissions increase for a modification is above the applicable emission threshold in Table 1. If
the requested emission rate and/or the facility-wide net emissions increase is below both of the
thresholds, modeling is usually not warranted unless one of the situations at the bottom of Table
1 applies. If there is doubt regarding the need for modeling, the applicant should consult with the
Division.
1 The Division's modeling study shows that the thresholds are appropriate in situations where a source has
reasonably good dispersion characteristics. In situations where a source has poor dispersion characteristics or in
areas with poor existing air quality, the thresholds might not be appropriate. In these situations, the Division will
work with the source to determine an appropriate threshold.
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
6
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
Table 1 [January 1, 2002]. Modeling Thresholds. Modeling is usually warranted to quantify the impact
if the emission rate is equal to or greater than these long-term (tons per year) and/or short-term (pound
per hour, etc.) emission thresholds. If the emission rate is less, a qualitative description of the impact is
tiHpniiai-A unlace ic a eitiiatinn ibat warrcmtc moHAlinn ^
Pollutant
Requested Emission Rate from a New Source
or
Facility-Wide Net Emissions Increase from a Modification
Carbon Monoxide (CO)
100 tons per year
or
23 pounds per hour
Nitrogen Oxides (NOx)
40 tons per year2
Sulfur Dioxide (SO2)
40 tons per year
or
27 pounds per 3-hours
Particulate Matter (PM-10)
15 tons per year
or
82 pounds per day
Lead(Pb)
0.6 tons per year
or
100 pounds per month
(1) Modeling is usually warranted, even though the source or modification does not exceed the modeling
thresholds in Table 1, if it is reasonable to believe the source will cause or contribute to a violation of
applicable ambient air quality standards in circumstances such as:
(a) Sources of S02, PM-10, CO, or Pb where a substantial portion of the new or modified emissions have
poor dispersion characteristics (e.g., rain caps, horizontal stacks, fugitive releases,3 or building
downwash4) in close proximity to ambient air at the site boundary;
(b) Sources of S02, PM-10, CO, or Pb located in complex terrain (e.g., terrain above stack height in close
proximity to the source);
(c) Sources located in areas with poor existing air quality;
(d) Modifications at existing major stationary sources, including grandfathered sources that have never
been modeled before.
2 For new sources or modifications, including those with poor dispersion characteristics, that emit less than 40 tons
per year (tpy) of NOx, modeling is usually warranted only in the situations described in caveats (1 )(c) and (l)(d),
provided that most (e.g., >85%) of the NOx is emitted as nitric oxide (NO). That is, because of near-field chemical
transformation assumptions, N02 impacts from a 40 tpy NOx source are usually expected to be below the N02
ambient air quality standard. Thus, modeling is only warranted in situations where existing N02 levels are high
enough that the significant impact from the new source or modification might "contribute" to a modeled violation of
the N02 air quality standard.
3 For sources without stacks (e.g., fugitive releases from area or volume sources), modeling may be warranted at
levels less than those in Table 1 if most of the emissions are from sources located less than 250-meters from the
limit to public access. The 250-meter recommendation is based on a modeling study performed by the Division,
4For sources with emission rates below those in Table 1 where the stack height is less than the U.S. EPA'sgood
engineering practice (GEP) stack height, modeling may be warranted; however, the presence of a non-GEP stack
height does not mean that modeling is automatically warranted. The degree (e.g., severity) of the downwash effects,
existing air quality levels, the distance to the boundary of ambient air, and any other relevant factors should be
considered.
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
7
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
Table 1 was updated in April 2010 to address NAAQS changes for lead, particulate matter less
than 2.5 microns, and nitrogen dioxide (the associated language in section 2.5 - Modeling
Thresholds remains unchanged since January 1, 2002).
[Excerpts from the April 2010 update of Table 1 in the Colorado Modeling Guideline.]
Table 1 [April 2010/. Modeling Thresholds. Modeling is usually warranted to quantify the impact if the emission
rate is equal to or greater than these emission thresholds. If the emission rate is less, a qualitative description of the
impact is adequate unless there is a situation that warrants modeling."1 [Note: The long-term (tons per year)
thresholds apply to modeling decisions regarding annual average ambient air quality standards. The short term
(pound per hour) thresholds apply to modeling decisions for short-term standards (i.e., < 24-hr average).]
Pollutant
Requested Emission Rate from a New Source
or
Facility-Wide Net Emissions Increase from a Modification
Carbon Monoxide (CO)
100 tons per year
or
23 pounds per hour
Nitrogen Oxides (NOx)
40 tons per year5
or
0.46 pound per hour
Sulfur Dioxide (S02)
40 tons per year
or
27 pounds per 3-hours
Particulate Matter < 10 |jm (PM10)
15 tons per year
or
82 pounds per day
Particulate Matter < 2.5 |rm (PM2.5)
5 tons per year of primary PM2 5
or
11 pounds per day of primary PM2 5
Lead (Pb)
25 pounds per 3-months
(1) Modeling is usually warranted, even though the source or modification does not exceed the modeling thresholds in Table 1, if it is
reasonable to believe the source will cause or contribute to a violation of applicable ambient air quality standards in circumstances
such as:
(a) Sources where a substantial portion of the new or modified emissions have poor dispersion characteristics (e.g., rain caps,
horizontal stacks, fugitive releases6, or building downwash7) in close proximity to ambient air at the site boundary;
(b) Sources located in complex terrain (e.g., terrain above stack height in close proximity to the source);
(c) Sources located in areas with poor existing air quality;
(d) Modifications at existing major stationary sources, including grandfathered sources that have never been modeled before.
5For new sources or modifications, including those with poor dispersion characteristics, that emit less than 40 tons per year (tpy) of NOx,
modeling for the annual N02 NAAQS is usually warranted only in the situations described in caveats (l)(c) and (l)(d), provided that most (e.g.,
>85%) of the NOx is emitted as nitric oxide (NO). That is, because of near-field chemical transformation assumptions, NOj impacts from a 40 tpy
NOx source are usually expected to be below the annual N02 ambient air quality standard. Thus, modeling is only warranted in situations where
existing annual N02 levels are high enough that the significant impact from the new source or modification might "contribute" to a modeled
violation of the annual N02 air quality standard.
SFor sources without stacks (e.g., fugitive releases from area or volume sources), modeling may be warranted at levels less than those in Table 1
if most of the emissions are from sources located less than 250-meters from the limit to public access. The 250-meter recommendation is based
on a modeling study performed by the Division.
7
For sources with emission rates below those in T able 1 where the stack height is less than the U.S. EPA's good engineering practice (GEP)
stack height, modeling may be warranted; however, the presence of a non-GEP stack height does not mean that modeling is automatically
warranted. The degree (e.g., severity) of the downwash effects, existing air quality levels, the distance to the boundary of ambient air, and any
other relevant factors should be considered.
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
8
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
2. Introduction
In determining compliance with Ambient Air Quality Standards (AAQS), impacts from
new/modified emission unit(s) are estimated with an air dispersion model. If estimated impacts
from the new/modified emission unit(s) are above modeling significance levels, they are added
to impacts from other emission units located at the facility, impacts from emission units located
nearby, if appropriate, and a background concentration to determine total ambient air
concentrations for compliance with the National Ambient Air Quality Standards (NAAQS) and
Colorado Ambient Air Quality Standards (CAAQS). If the estimated impacts from the
new/modified emission umt(s) are below modeling significance levels, the new/modified
o
emission unit(s) is not considered to have a significant impact in ambient air and no further
analysis is necessary. Table 2 lists the modeling significance levels and AAQS for nitrogen
dioxide (NO2), sulfur dioxide (SO2), and fine particulate matter (PMi0).
Table 2 [January 2002|. Modeling Significance Levels and AAQS for NO2, SO2, and PM10
Modeling Significance
Level (|Jg/m3)
NAAQS (Ug/m3)
CAAQS (Ug/m3)
Pollutant
3-hr
24-hr
Annual
3-hr
24-hr
Annual
3-hr
24-hr
Annual
NO,
1
100
100
S02
25
5
1
1300*
365
80
700
PMI0
5
1
150
50
150
50
'Secondary NAAQS
Table 2 lists the modeling significance levels and AAQS for nitrogen dioxide (NO2), sulfur
dioxide (SO2), particulate matter less than 10 microns (PM10), particulate matter less than 2.5
microns (PM2.5) effective in April 2010.
Table 2 [April 2010). Modeling Significance Levels and AAQS for NO2, SO2, PM10, PM2.5
Modeling Significance Level (|Jg/m3)
NAAQS (Mg/m3)
CAAQS
(Mg/m3)
Pollutant
1-hr
3-hr
24-hr
Annual
1-hr
3-hr
24-hr
Annual
3-hr
I no2
4a
1
-189
100
SO,
25
5
1
1300"
365
80
700
PM10
5
1
150
PM2.5
1.2°
0.3c
35
15
"Interim modeling significance level developed by the Division
bSecondary NAAQS
'Interim modeling significance level developed by the Division based on level proposed by EPA for NAAQS only
8 "Ambient air" is defined as "that portion of the atmosphere, external to the source, to which the general public has
access."
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
9
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
The "Colorado Modeling Guideline for Air Quality Permits" (Colorado Modeling Guideline)
does not require a quantitative impact analysis for every new source/modification. The Colorado
Modeling Guideline provides threshold emission levels that would trigger a quantitative impact
analysis. Some of the public comments argue that only new/modified emission units emitting
pollutants greater than Prevention of Significant Deterioration (PSD) Significant Emission Rates
(shown below in Table 3) should trigger a quantitative impact analysis. Others also support
raising the PMio emission threshold level from 15 tons per year (tpy) to 40 tpy. This implies that
new/modified emission units with emission rates equivalent to or greater than the PSD
Significant Emission Rates would not cause or contribute to an exceedence of the AAQS.Table
3. Current (1998) and proposed (2001) modeling emission rate thresholds in tons per year, tpy.
The proposed levels are the same as the PSD Significant Emission Rates of Criteria Pollutants in
Regulation No. 3 [Note: One exception is that the fugitive PMio threshold would remain at 5
tpy].
Current
Emission Rate (tpy)
Proposed
Emission Rate (tpy)
Thresholds
Thresholds
I Pollutant
(Table 1; 12/23/98 Guideline)
(Table 1; 2/14/01 Guideline)
CO
50 attainment,
100
25 nonattainment
NOx
20
40
S02
20
40
PM,o (Stack)
5
15
PMio (Fugitive)
5
5
Pb
0.1
0.6
The January 2002 modeling analysis was performed to determine if a point source emitting 40
tpy of NOx9, SO2, or PMio or 15 tpy of PMio would have a significant impact in ambient air
(refer to Sections 4, 5, and 6).
The April 2010 modeling analysis (refer to Section 7) was performed to determine if the
emission rate thresholds in Table 3 (above) are adequate to indicate when a quantitative impact
analysis is necessary to demonstrate if the proposed modification or source will or will not cause
or contribute to a violation of a recently promulgated NAAQS [24-hr and annual PM2.5
(December 18, 2006, includes retaining the 24-hr revoking of the annual PMi0 standard), 3-
month rolling Pb (January 12, 2009), and 1-hr NO2 (April 12, 2010)].
3. Effects on Ambient Air Impact Estimations
Ambient air impacts are a function of atmospheric dispersion. Various factors affect
atmospheric dispersion, including plume rise, building wake effects, and meteorological
9 The ambient air standards are for nitrogen dioxide (N02), not oxides of nitrogen (NO,). NO, includes both nitric
oxide (NO) and N02. While some N02 is directly emitted from the stacks of stationary sources, a significant portion
of the emissions usually occur as nitric oxide (NO). The NO is converted to N02 by chemical mechanisms in the
atmosphere. To account for possible chemical conversion in the atmosphere, the total NO, emission rate is used in
Table 2 instead of only the primary N02 emission rate.
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
10
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
conditions. Plume rise is due to the momentum or buoyancy of the exhaust gases. Factors that
hinder plume rise are stack-tip downwash and building wake effects.
3.1. Buoyancy
Stack gases exhausted into the atmosphere having a density less than that of ambient air will
experience plume rise due to buoyancy. Lower molecular weight or high stack gas exit
temperature will result in a stack gas density lower than that of ambient air. In most regulatory
air models, buoyancy is a function of the difference between stack gas exit temperature and
ambient temperature. Model inputs used to determine the magnitude of buoyant forces are stack
gas exit temperature, ambient temperature, stack diameter, and stack gas exit velocity. The
larger of buoyancy force and momentum force is used to determine the effective plume height.
3.2. Momentum
The force imparted on the stack gases provides the momentum necessary for successful
exhaustion into the atmosphere. Momentum is important if the temperature of the stack gases is
within a few degrees of ambient temperature or subject to building wake effects. Obstructions at
the top of a vertical stack, such as a rain cap, can reduce or eliminate vertical momentum and
affect plume rise. Horizontal discharges also have essentially no momentum plume rise. Model
inputs that affect momentum are stack gas exit velocity and stack diameter. Depending on
meteorological conditions, stack gas exit temperature and ambient temperature also affect
momentum calculations.
3.3. Stack-Tip Downwash
Stack-tip downwash occurs when the stack gas plume is drawn down to the low pressure or
slight vacuum region downwind of the stack. The area of low pressure/slight vacuum is cause by
wind flowing past the stack. Stack-tip downwash can be eliminated if exit velocities are greater
than or equal to 150% of the wind speed at the stack top. Model inputs that affect stack-tip
downwash are stack gas exit velocity and wind speed. Stack diameter is also used to determine
the effective plume height.
3.4. Building Downwash
Wind flow around a building creates turbulent eddies downwind of the building. Plumes
released near buildings can be caught in the turbulent wake of the building. For elevated
releases, plumes subject to building downwash usually result in increased ground-level
concentrations. To avoid the effect of building downwash, the general rule is to design a stack
that is 2.5 times the lesser of the height or projected width of nearby buildings.10 This is known
as the Good Engineering Practice (GEP) height. Building dimensions are input into modeling
systems to determine if the stack gas plume will be affected by downwash.
10 A building is considered to be nearby if it is within 5L (five times L, where L is the lesser of the building height or
the projected width of the building) of a building or structure [see 40 CFR 51.100 (jj)(l)].
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
11
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
4. Methodology (January 2002 Analysis)
Multiple model runs were performed using a range of values in stack parameters. The Industrial
Source Complex Model (ISCST3 version 00101) was used with 19B9 Denver Stapleton
Meteorological Data. The emission rate used for all runs is 1.15 g/s (40 tpy) to determine NQx,
S02, and PMio concentration levels. Since modeling was performed for only one emission unit
and concentration is directly proportional to emission rate, concentration levels determined with
a 40 tpy emission rate are scaled to obtain PMio concentrations at 15 tpy.
Table 5 summarizes the values of each parameter for each model run. Stack characteristics were
selected to illustrate the effects of each/combination of parameter(s) on impact estimates. The
range of values in Table 5 is not intended to represent all possible stack characteristics and
combinations. In practice, many emission units have stack parameters that are lower or higher
than the range of values used in this study.
4.1. Receptor Spacing
The receptor network is described in Table 4 below.
Table 4. Receptor spacing and location
Distance from Source Location
Receptor Spacing
Fenceline
8 receptors spaced 50 m, 30 m, or 15 m (see
Table 5) apart forming a square perimeter with
source location in the center; spacing varies per
run
50 m
8 receptors spaced 50 m apart forming a square
perimeter with source location in the center
0 to 5000m
100 m Cartesian grid
5000 m to 10,000 m
250 m Cartesian grid
4.2. Model Runs
4.2.1. Sensitivity Analysis (Runs 1 through 10)
A base case (Run 1) was selected to compare with Runs 2 through 10. The sensitivity analysis
consists of 9 runs where each run differed from the base case by only one modeling parameter.
The parameters are stack height, urban dispersion, stack diameter, stack gas exit velocity, and
stack gas exit temperature. These runs assume that the plume is not subject to building
downwash.
4.2.2. Building Downwash (Runs 11 through 18}
Runs 11 through 18 were performed to examine the effects of building downwash effects on the
impacts and their location from the source. The footprint of the building is 9.14 m x 9.14 m (30
ft x 30 ft). Building height of 50% and 75% of the stack height were used. Runs 13 and 14 use
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
12
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
urban instead of rural dispersion coefficients. Runs 15 through 18 with urban dispersion include
fenceline receptors closer to the source.
4.2.3. Multiple changes in Stack Characteristics with Building Down wash
(Runs 19 through 22)
Runs 19 through 22 represent vertical unobstructed stacks with stack and building configurations
that hinder plume rise.
4.2.4. Horizontal Stack (Runs 23 through 25)
The stack inputs were modified to follow EPA guidance for modeling horizontal stacks (July 9,
1993 memo from Joseph A Tikvart to Ken Eng). Stack diameter is set to 0.001 m. Actual stack
height is used.
4.2.5. Capped Stack (Runs 26 through 28)
The stack inputs were modified to follow EPA guidance for modeling capped stacks (July 9,
1993 memo from Joseph A. Tikvart to Ken Eng). Stack diameter is set to 0.001 m. Stack height
is reduced by 3 times the actual stack diameter.
4.2.6. Minimum and Maximum Range of Values (Runs 29 and 30)
Run 29 represents a vertical stack with no obstruction that is subject to building downwash with
the lowest stack parameters in Table 4. Run 30 represents a vertical stack with no obstruction
and no downwash effects with the highest stack parameters in Table 4.
4.3. Comparison with Modeling Significance Levels and AAQS
According to U.S. EPA guidance, the highest impact concentration of any averaging period
should be used to determine whether the emission unit will have a significant impact in ambient
air. That is, the modeling significance level is used to determine if a source "contributes" to a
modeled violation of AAQS. When impacts are significant for an averaging period at a specific
receptor, the impacts from the emission unit are added to the impacts from nearby sources, if
appropriate, and a reasonable background concentration to determine the total ambient air
concentration for the compliance demonstration with the AAQS. The maximum annual and
highest-2nd-highest (H2H) short-term SO2 and PM10 (the allowance of one exceedence of the 24-
hr PM10 when using one year of meteorological data) total ambient air concentrations are
compared to the AAQS. For simplicity in this modeling analysis, H2H short-term SO2 and
PM10, and maximum annual concentrations are compared to the modeling significance level for
significance determination and used to determine whether the impact itself would exceed the
AAQS.
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
13
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
5. Results (January 2002 Analysis)
The results are presented in tabular format for all runs by emission rate and averaging period in
Table 6. The 24-hr results of model Runs 1 through 10 are also presented in Figure 1 through
Figure 5 to examine the magnitude and location of impacts. Since no chemical transformations
or conversion factors were used, the impacts listed below apply to any pollutant.
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
14
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
Table 5. Summary of stack, building and fencelinc parameters for each model run
Model Run'
Dispersion2 |
Stack Height (m)
Stack Diameter (m)
Stack Gas Exit Temperature (K)
Stack Gas Exit Velocity (m/s)
Stack Orientation4 |
Building Height (m)3
01
u
s
ss
Q
at
s
"3
u
s
1 - Base Case
R
6.10
0.31
644
25.4
V
0
50
2 - Height Decrease
R
3.05
0.31
644
25.4
V
0
50
3 - Height Increase
R
9.14
0.31
644
25.4
V
0
50
4 - Urban
U
6.10
0.31
644
25.4
V
0
50
5 - Diameter Decrease
R
6.10
0.15
644
25.4
V
0
50
6 - Diameter Increase
R
6.10
0.46
644
25.4
V
0
50
7 - Velocity Decrease
R
6.10
0.31
644
9.14
V
0
50
8 - Velocity Increase
R
6.10
0.31
644
76.2
V
0
50
9 - Temperature Decrease
R
6.10
0.31
477
25.4
V
0
50
10 - Temperature Increase
R
6.10
0.31
811
25.4
V
0
50
11-BH 50% SH
R
6.10
0.31
644
25.4
V
3.05
50
12 - BH 75% SH
R
6.10
0.31
644
25.4
V
4.58
50
13 - BH 50% SH, urban
U
6.10
0.31
644
25.4
V
3.05
50
14 - BH 75% SH, urban
U
6.10
0.31
644
25.4
V
4.58
50
15 - BH 50% SH, urban, 30 m FL
u
6.10
0.31
644
25.4
V
3.05
30
16 - BH 75% SH, urban, 30 m FL
u
6.10
0.31
644
25.4
V
4.58
30
17 - BH 50% SH, urban, 15 m FL
u
6.10
0.31
644
25.4
V
3.05
15
18 - BH 75% SH, urban, 15 m FL
u
6.10
0.31
644
25.4
V
4.58
15
19 - T/D/V Decrease, BH 75% SH, 30 m FL
R
6.10
0.15
477
10
V
4.58
30
20 - H/T/D/V Decrease, BH 67% SH, 30 m FL
R
4.58
0.15
477
10
V
3.05
30
21 - T/D/V Decrease, BH 75% SH, urban, 30 m FL
u
6.10
0.15
477
15
V
4.58
30
22 - H Decrease, BH 100% SH, urban, 30 m FL
u
3.05
0.31
644
25.4
V
3.05
30
23 - Horizontal5
R
6.10
0.001
644
0.001
H
0
50
24 - Horizontal, BH 50% SH5
R
6.10
0.001
644
0.001
H
3.05
50
25 - Horizontal, BH 50% SH, 30 m FL5
R
6.10
0.001
644
0.001
H
3.05
30
26 - Capped6
R
5.17
0.001
644
0.001
C
0
50
27 - Capped, BH 50% SH6
R
5.17
0.001
644
0.001
C
3.05
50
28 - Capped, BH 50% SH, 30 m FL6
R
5.17
0.001
644
0.001
C
3.05
30
29 - Low range of values, Building 100% SH
R
3.05
0.15
477
9.14
V
3.05
30
30 - High range of values
R
9.14
0.46
811
76.2
V
0
30
'Model Run Codes: BH = Building Height, SH = Stack Height, D = Diameter, V = Exit Velocity, T = Exit Temperature, FL = Fenceline,
2Dispersion Codes: R = Rural, U = Urban.
3Building Footprint Dimensions: 9.14 m x 9.14 m (30 ft x 30 ft).
"Stack Orientation Codes: V = Vertical, H = Horizontal, C = Capped, Vertical Obstructed.
5Stack parameters adjusted according to EPA Guidance (July 9, 1993 memo from Joseph A Tikvart to Ken Eng)
'Stack parameters adjusted according to EPA Guidance (July 9,1993 memo from Joseph A Tikvart to Ken Eng), assumes D = 0.31 m
Shaded Values - Values different than base case
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
15
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
Table 6. Summary of impacts for 40 tpy and 15 tpy emission rates. [Note: In a compliance
demonstration with ambient air quality standards (AAQS),11 impacts from nearby sources, if appropriate,
and background sources would be added to these results.]
Impact Concentration (pg/m3)
40 tpy
40 tpy
40 tpy
15 tpy
15 tpy
Model Run
H2H
H2H
Max
H2H
Max
3-hr
24-hr
Annual
24-hr
Annual
1 - Base Case
88.26
28.15
3.74
10.56
1.40
2 - Height Decrease
148.08
37.90
5.57
14.21
2.09
3 - Height Increase
61.71
17.88
2.68
6.71
1.01
4 - Urban
161.08
71.03
9.77
26.64
3.66
5 - Diameter Decrease
283.71
76.96
12.21
28.86
4.58
6 - Diameter Increase
46.37
12.35
1.55
4.63
0.58
7 - Velocity Decrease
203.94
59.79
8.75
22.42
3.28
8 - Velocity Increase
36.12
9.05
1.07
3.39
0.40
9 - Temperature Decrease
118.95
35.72
5.16
13.40
1.94
10 - Temperature Increase
74.16
24.62
3.14
9.23
1.18
11- BH 50% SH
128.26
28.35
3.77
10.63
1.41
12 - BH 75% SH
308.61
83.93
6.47
31.47
2.43
13 - BH 50% SH, urban
196.43
71.42
9.80
26.78
3.68
14 - BH 75% SH, urban
544.72
237.69
35.92
89.13
13.47
15 - BH 50% SH, urban, 30 m FL
208.77
71.42
9.80
26.78
3.68
16 - BH 75% SH, urban, 30 m FL
949.60
317.85
54.00
119.19
20.25
17 - BH 50% SH, urban, 15 m FL
196.43
71.42
9.80
26.78
3.68
18 - BH 75% SH, urban, 15 m FL
1045.30
237.69
35.92
89.13
13.47
19 - T/D/V Decrease, BH 75% SH, 30 m FL
1487.95
463.76
99.96
173.91
37.49
20 - H/T/D/V Decrease, BH 67% SH, 30 m FL
1444.40
582.41
109.67
218.40
41.13
21 - T/D/V Decrease, BH 75% SH, urban, 30 m FL
1683.62
626.98
114.99
235.12
43.12
22 - H Decrease, BH 100% SH, urban, 30 m FL
1606.12
654.28
101.58
245.36
38.09
23 - Horizontal5
1341.46
377.22
54.20
141.46
20.33
24 - Horizontal, BH 50% SH5
4308.35
1138.81
188.96
427.05
70.86
25 - Horizontal, BH 50% SH, 30 m FL5
5676.34
1546.20
239.56
579.83
89.84
26 - Capped6
1824.50
480.88
80.05
180.33
30.02
27 - Capped, BH 50% SH6
5990.40
1577.67
252.12
591.63
94.55
28 - Capped, BH 50% SH, 30 m FL6
8643.41
2357.32
364.89
884.00
136.83
29 - Low range of values, Building 100% SH
8693.97
2037.43
487.96
764.04
182.99
30 - High range of values
9.38
2.30
0.31
0.86
0.12
11 Modeling Significance Levels and AAQS for NQ2, S02, and PM10
Modeling Significance
Level (ug/m3)
NAAQS (Ug/m3)
CAAQS (gg/m3)
Pollutant
3-hr
24-hr
Annual
3-hr
24-hr
Annual
3-hr
24-hr
Annual
1 NO,.
1
100
100 |
so2
25
5
1
1300
365
80
700
PM10
5
1
150
50
150
50
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
16
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
Magnitude and Location of Impacts from Varying Stack Height
at 40 tpy Emission Rate
Highest-2nd-Highest 24-hr Average Concentration
(microgram per cubic meter)
-200
300-
200-
0) 100-
<1>
E
(1)
+-»
(S
O
o
o
-100-
-200-
28.15 ug/mC
4
f
37.90 ug/m;
J
Higher impact occurs closer
to the source location when
stack height is decreased.
Impact is lower when stack
height is increased.
17.88 ug/m3
•300
•200
¦100
-100
0 100
X coordinate (meter)
LEGEND
© Point Source Location
A Impact Location for Height = 3.05 m
9 Impact Location for Height = 6.1 m (Base Case)
¦ Impact Location for Height = 9.14 m
Modeling Significance Level: 5 ug/m3
S02 NAAQS: 365 ug/m3
PM10 NAAQS: 150ug/m3
These results are based on 1989 Denver
Stapleton Airport Meteorological Data.
Using different meteorological data will
result in different impact estimates.
Figure 1. Magnitude and Location of Impacts from Varying Stack Height
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
17
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
Magnitude and Location of Impacts from Varying Stack Diameter
at 40 tpy Emission Rate
Highest-2nd-Highest 24-hr Average Concentration
(microgram per cubic meter)
-200
-100
100
200
300
400
300
•C 200-
E
(1)
+-»
(S
O
o
o
1
12.35 ug/m3
I
1
28.15 ug/m3
¦
i
¥
Higher impact occurs closer
to the source location when
stack diameter is decreased.
Lower impact occurs further
from the source location when
stack diameter is increased.
/
JN
76.96 ug/m3
-300
-100
-100
-300
-200
-100 0 100
X coordinate (meter)
200
300
LEGEND
© Point Source Location
A. Impact Location for D = 0.15 m
% Impact Location for D = 0.31 m (Base Case)
¦ Impact Location for D = 0.46 m
Modeling Significance Level: 5 ug/m3
S02NAAQS: 365 ug/m3
PM10 NAAQS: 150ug/m3
These results are based on 1989 Denver
Stapleton Airport Meteorological Data.
Using different meteorological data will
result in different impact estimates.
Figure 2. Magnitude and Location of Impacts from Varying Stack Diameter
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
18
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
Magnitude and Location of Impacts from Varying Stack Exit Velocity
at 40 tpy Emission Rate
Highest-2nd-Highest 24-hr Average Concentration
(microgram per cubic meter)
-300
-200
-100
400
300-
200
0)
"5
E
-
100- ¦
-100
-200-'
100
200
300
9.05 ug/m3
9
¦
I
28.15 ug/m3
*
Higher impact occurs closer
to the source location when
stack exit velocity is decreased
Lower impact occurs further
from the source location when
stack exit velocity is increased.
59.79 ug/m3
V
¦400
•300
• 100
-0
-100
-200
-200
0 100
X coordinate (meter)
LEGEND
© Point Source Location
A Impact Location for V = 9.14 m/s
0 Impact Location for V = 25.4 m/s (Base Case)
¦ Impact Location for V = 76.2 m/s
Modeling Significance Level: 5 ug/m3
S02 NAAQS: 365 ug/m3
PM10NAAQS: 150ug/m3
These results are based on 1989 Denver
Stapleton Airport Meteorological Data.
Using different meteorological data will
result in different impact estimates.
Figure 3. Magnitude and Location of Impacts from Varying Stack Exit Velocity
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
19
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
Magnitude and Location of Impacts from Varying Stack Exit Temperature
at 40 tpy Emission Rate
Highest-2nd-Highest 24-hr Average Concentration
(microgram per cubic meter)
-100
100
E
0)
+-»
CD
c
T3 100
O
O
o
28.1
35.72 ug/m
n
ug/m 3
24.62 ug/m3
Impact is higher when stack
exit temperature is decreased.
Impact is lower when stack
exit temperature is increased.
200
100
-100
X coordinate (meter)
©
LEGEND
Point Source Location
Modeling Significance Level: 5 ug/m3
S02 NAAQS: 365 ug/m3
PM10NAAQS: 150ug/m3
A Impact Location for T = 477 K
Impact Location for T= 644 K (Base Case)
~ Impact Location for T = 811 K
These results are based on 1989 Denver
Stapleton Airport Meteorological Data.
Using different meteorological data will
result in different impact estimates.
Figure 4. Magnitude and Location of Impacts from Varying Stack Gas Exit Temperature
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
20
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
Magnitude and Location of Impacts for Urban and Rural Dispersion
at 40 tpy Emission Rate
Highest-2nd-Highest 24-hr Average Concentration
(microgram per cubic meter)
-200
-100
100
200
200
a>
a>
E
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
6. Discussion (January 2002 Analysis)
6.1. Sensitivity Analysis (Runs 2 through 10)
The results show that increases in stack height, stack diameter, stack gas exit velocity, and stack
gas exit temperature decrease ambient pollutant concentration levels and increase the distance of
impact from the source. Decreases in stack height, stack diameter, stack gas exit velocity, and
stack gas exit temperature increase ambient pollutant concentration levels and decrease the
distance of impact from the source. Tall and wide stacks with high velocity and temperature
promote plume rise. Short and narrow stacks with low velocity and temperature impede plume
rise. The modeling parameters used for these runs with an emission rate of 40 tpy resulted in
exceedances of the modeling significance levels for all averaging periods for SO2, NO2, and
PM 10. All impact concentrations for 15 tpy PM10, except for diameter and velocity increases, are
above the modeling significance levels for both averaging periods for PMio-
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
22
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
6.2. Building Down wash (Runs 11 through 18)
Increase in building height increases the magnitude of impact and decreases the distance of
impact from the source. Examining the concentrations for runs 13 through 18 in Table 7 reveals
the relationship between maximum impacts and fenceline receptors. The maximum impacts
obtained for a given emission unit can vary with the location of the fenceline. Thus, the
fenceline location is important because it usually determines the ambient air boundary.12 For
example, the maximum annual concentration for an emission unit subject to downwash from a
building height equal to 75% of the stack height with a fenceline at 50 m is 35.92 |Jg/m3. If the
same emission unit has a fenceline at 30 m, the maximum annual concentration is 54.00 |Jg/m3, a
50% increase. For the emission unit subject to downwash from a building height equal to 50%
of the stack height, the H2H 24-hr and maximum annual concentrations are the same for all
fenceline distances used. The modeling parameters used for these runs with an emission rate of
40 tpy resulted in exceedences of the modeling significance levels for all averaging periods for
SO2, NO2, and PM10. All impact concentrations for 15 tpy PM10 are above the modeling
significance levels for both averaging periods for PM10. The 3-hr SO2 CAAQS is exceeded by
the source impacts.
Table 7. Impacts from 40 tpy by fenceline distance from source
Impact Concentration from 40 tpy (|Jg/m3)
Fenceline Distance
Building Height =
Building Height =
from Source
50% Stack H
eight
75% Stack Height
H2H
H2H
Max
H2H
H2H
Max
3-hr
24-hr
Annual
3-hr
24-hr
Annual
50 m
196.43
71.42
9.8
544.72
237.69
35.92
30 m
208.77
71.42
9.8
949.60
317.85
54.00
15 m
196.43
71.42
9.8
1045.30
237.69
35.92
6.3. Multiple Changes in Stack Characteristics with Building Downwash
(Runs 19 through 22)
These runs were performed to determine impact concentrations resulting from vertical,
unobstructed stacks subject to building downwash with poor dispersion characteristics (low
temperature, velocity and stack diameter). Short stacks with fairly good dispersion can have
high impacts due to an overwhelming effect from building downwash. The modeling parameters
used for these runs with an emission rate of 40 tpy resulted in exceedances of the modeling
significance levels for all averaging periods for SO2, NO2, and PM10. All impact concentrations
for 15 tpy PM10 are above the modeling significance levels for both averaging periods for PM10.
The SO2 AAQS and 24-hr PM10 NAAQS (at 40 tpy and 15 tpy) have been exceeded by the
12 Ambient air quality standards apply only in "ambient air." That is, it is not necessary to place receptors (e.g., to
estimate impacts) within property owned or controlled by the facility if public access is precluded by a fence or
physical barrier.
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
23
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
source impacts. The NO2 impacts, using a 75% annual conversion to NO2 from NOx, range from
75 |Jg/nr to 86 |Jg/mJ, greater than 75% of the NO2 NAAQS.
6.4. Horizontal Stack (Runs 23 Through 25) and Capped Stack (Runs 26
through 28)
Horizontal and capped stacks do not promote plume rise. This is illustrated by the exceedances
of the modeling significance levels for all averaging periods as well as most of the AAQS for
SO2, NO2, and PM10 with a few exceptions (annual NAAQS for runs with no building
downwash).
6.5. Minimum and Maximum Range of Values (Runs 29 and 30)
These runs were performed to determine the range of impact concentrations for the range of
stack and building characteristics used in this modeling analysis. Run 29 is the poor dispersion
example with all impact concentrations exceeding the modeling significance levels and AAQS.
Run 30 is a good dispersion example with all impact concentrations below the modeling
significance levels.
6.6. Other Modeling Variables Not Examined in this Modeling Analysis
There are other parameters used in modeling that are not examined here, such as different
meteorological data sets, elevated terrain, and background concentrations. Typical yearly
variations of meteorological data at one location can result in modeled design concentration
differences of up to 25% or even higher in some locations.11 Higher impacts may result when
plume rise is insufficient to clear nearby terrain.
Contributors to ambient air concentration for determining compliance with AAQS are impacts
from the source of interest and nearby sources, and the background concentration. Even though
impacts are just above modeling significance levels or only a small fraction of the AAQS, a
complete compliance demonstrate must also take existing air pollutant concentration levels into
account. This may mean that, in addition to adding a background concentration, nearby sources
with strong concentration gradients should be included in the modeling. Since it's not reasonable
to model all sources, it is necessary to add a background concentration to account for the
emissions from all sources that have not been explicitly included in the modeling. Background
concentrations vary by geographic area. For areas with high background concentrations (and/or
strong concentration gradients from nearby sources) near the AAQS, a source impact that is
greater than the modeling significance levels, but still a relatively small percentage of the AAQS,
can result in a modeled violation of the AAQS.
n In a recent study conducted in Alaska, it was found that the modeled maximum annual average concentration varied by as
much as 200% over a five (5) year period at one particular site, depending on which year of meteorological data was used in the
model. At two other sites, the maxima varied by 139% and 122%, respectively. For short-term (24-hour) concentrations, the
maximum modeled concentration varied by 161%, 148%, and 121% at three different sites, depending on which one of the five
years of meteorological data were used. In addition to the variation in the maximum modeled impact, the location (geographic
location) of the modeled maxima varied significantly from one year to the next. [Reference: Presentation by Alan Schuler,
Alaska Department of Environmental Conservation, 2001 EPA/State/Local Modeler's Workshop, Chicago]
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
24
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
7. Methodology, Results, and Discussion of the
April 2010 Analysis
7.1. Methodology
Annual, 24-hr, and 1-hr impacts for 22 individual point source scenarios using 48 one-year
periods of hourly meteorology were estimated with AERMOD (09292) and SCREEN3 for a
range of emission rates. Since no chemical transformations or conversion factors were used, the
impacts in Figures #-# below are applicable for any pollutant. Urban effects were not modeled.
7.1.1. Meteorology
The following meteorological data (station/years) were used in this analysis.
DEN (Denver Stapleton) 1990-1994
Greely (Greeley) 2002-2006
Akron (Akron) 1990-1994
Pueblo (Pueblo Memorial Airport) 2002-2006
COSprings (Colorado Springs) 1987-1991
Sydney (Sydney) 2003-2007
Kodak 1993-1997
PRPA06 (Platte River Power-Rawhide)
Thermo/Ft Lupton
FtStVrain (Fort St Vrain Power)
PuebloDepot (Peublo Chemical Depot) 1998-2000
Portland
Asarcol993, 1994, 1998-2000
Naturita
7.1.2. Receptor Network
Receptors were placed every 10 degrees at the following distances (meters) from the point
source: 30, 50, 75, 100, 125, 150, 175, 200, 250, 300, 350, 400, 450, 500, 550, 600, 650, 700,
750, 800, 850, 900, 950, 1000, 1100, 1200, 1300, 1400, 1500, 1600, 1700, 1800, 1900, 2000,
2100, 2200, 2300, 2400, 2500, 2600, 2700, 2800, 2900, 3000, 3100, 3200, 3300, 3400, 3500,
3600, 3700, 3800, 3900, 4000, 4100, 4200, 4300, 4400, 4500, 4600, 4700, 4800, 4900, 5000,
5500, 6000, 6500, 7000, 7500, 8000, 8500, 9000, 9500, 10000. Flat terrain was assumed.
7.1.3. Point Sources
Table 8 summarizes the point source parameters (building dimensions, where applicable) for
each scenario/model run. The range of source types in this analysis (points, with and without
building downwash) is not intended to represent all possible stack characteristics and
combinations but is intended to illustrate the effects of each/combination of parameter(s) on
impact estimates.
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
25
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
Table 8. Summary of Point Source Inputs
Emissions
Rate (g/s)
Stack
Height
(m)
Temp
(K)
Exit
Velocity
(m/s)
Stack
Diameter
(m)
Bldg
Height
(m)
Bldg
Width
(m)
Bldg
Length
(m)
Location of Bldg
B1
100
10
293
1
2.4
B2
100
35
293
11.7
2.4
B3
100
35
432
11.7
2.4
B4
100
100
416
18.8
4.6
B5
100
150
425
26.5
5.6
B6
100
200
425
26.5
5.6
BC08
100
8
377
10
0.7
BC10
100
10
300
15
0.2
BCMD
100
40
325
10
1.5
BCLG
100
91.4
467.6
17.1
4.57
COl
100
6
400
15
0.5
ASOS1
100
55
432
11.7
2.4
D1
100
10
293
1
2.4
34
60
120
NE bldg corner =
stack location
D2
100
10
293
1
2.4
34
60
120
NE bldg corner SW of
stack (-96 m, -96 m)
D3
100
35
432
11.7
2.4
34
60
120
NE bldg corner =
stack location
D4
100
35
432
11.7
2.4
34
60
120
NE bldg corner SW of
stack (-96 m, -96 m)
D5
100
100
416
18.8
4.6
50
60
120
NE bldg corner =
stack location
D6
100
100
416
18.8
4.6
50
60
120
NE bldg corner SW of
stack (-140 m, -140 m)
BC08D
100
8
377
10
0.7
6
10
6
North side of building
centered on stack
BC10D
100
10
300
15
0.2
8
10
6
NE bldg corner
located 4 m south of
stack
COID
100
6
400
15
0.5
4
4.5
9
North side of building
centered on stack
ASOS1D
100
55
432
11.7
2.4
34
60
120
NE bldg corner =
stack location
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
26
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
7.2. Results and Discussion
Predicted concentrations from AERMOD and SCREEN3 for various emission rates are
compared to the NAAQS for Pb (3-month), PM2.5 (24-hr and annual), and NO2 (1-hr) in the
subsequent subsections. The SCREEN3 concentrations do not include estimates in the cavity
region, consistent with past and present Division practice.
~1-hr N02 NAAQS
SCREEN3
DEN91
¦DEN93
¦ Greely02
Greely04
GreelyOG
Akron91
AkronS 3
• Pueblo02
Puebto04
PuebloOO
• COSpnngs&B
¦ COSpnngsQO
¦SydneyOS
•Sydney05
¦SydneyO?
¦kodak94
¦Kodak99
•PRPA06
¦ FtStVrain
¦PuebloDepotSS
¦Portland
Asarco94
Asaroo99
• GreelyOS
GreelyOS
Akron90
-Akron92
¦ PuebtoOO
-PuebloC-3
• PuebloC5
¦ COSprings87
¦COSpr>ngs89
¦COSprmgsBI
¦SydneyO*
•SydneyOS
•Kodak93
- Kodak95
Kodak97
Thermo/Ft Lupton
- PaebloDepotBS
¦PuebloDepotOO
¦Asaroo93
¦Asaroo93
¦ AsarooCO
¦FRi'Natunta
7.2.1. 1-hr Concentrations
Figure 6 through Figure 9 present the 1-hr concentrations for emission rates of 9.13 pounds per
hr (annual NOx emission rate threshold equivalent - 40 tpy), 2.28 pounds per hr, 1.14 pounds per
hour, and 0.46 pound per hr. Based on these results, the 1-hr NO2 NAAQS could be threatened
by an individual emission unit with an emission rate around or greater than 2.28 pounds per hour.
At a point source emission rate of 1.14 pounds per hour (with or without building downwash), it
is reasonable to believe the source will cause or contribute to a violation of the 1-hr NO2
NAAQS. For a point source with an emission rate of 0.46 pound per hour with poor dispersion,
there will be situations (Table 1 footnotes and Section 7.2.5) when the modeling significance
level is exceeded and it is reasonable to believe the source will cause or contribute to a violation
of the 1-hr N02 NAAQS.
Maximum 1-hr Concentration vs. Point Source
AERMOD/Hourly Meteorology and SCREEN3
(Source only, background concentration not included)
7000
6000
Point Source
5000
£: 4000
3000
E 2000
Figure 6. Maximum 1-hr Concentrations - 9.13 pounds per hr (40 tpy equivalent)
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
27
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
Maximum 1-hr Concentration vs. Point Source
AERMOD/Hourly Meteorology and SCREEN3
{Source only, background concentration not included)
1200
1000
200
«N <9 <$" ^ rf? jP rx? rON 0 o° J§>
Point Source
Figure 7. Maximum 1-hr Concentrations -2.28 pounds per hr (10 tpy equivalent)
Maximum 1-hr Concentration vs. Point Source
AERMOD/Hourly Meteorology and SCREEN3
(Source only, background concentration not included)
900
700
^ 600
400
300
Point Source
Figure 8. Maximum 1-hr Concentrations - 1.14 pounds per hr (5 tpy equivalent)
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
28
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
Maximum 1-hr Concentration vs. Point Source
AERMOD/Hourly Meteorology and SCREEN3
(Source only, background concentration not included)
Point Source
Figure 9. Maximum 1-hr Concentrations - 0.46 pound per hr (2 tpy equivalent)
7.2.2. 24-hr Concentrations
Figure 10 through Figure 12 present the 24-hr concentrations for emission rates of 82 pounds per
day (24-hr PMio emission rate threshold), 27 pounds per day, and 11 pounds per day. Based on
these results, the 24-hr PM2 5 NAAQS could be threatened by an individual emission unit with
poor dispersion and an emission rate around or greater than 27 pounds per day. For a point
source with an emission rate of 11 pounds per day with poor dispersion, there will be situations
(Table 1 footnotes and Section 7.2.5) when the modeling significance level is exceeded and it is
reasonable to believe the source will cause or contribute to a violation of the 24-hr PMb 5
NAAQS.
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
29
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
Maximum 24-hr Concentration vs. Point Source
AERMOD/Hourly Meteorology and SCREEN3
(Source only, background concentration not included)
450
PuewoQI
—Pb»ooe
— :<33orff>at«e
— C02emjj=c
— SjareytH
—° <6° ° ^
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
Maximum 24-hr Concentration vs. Point Source
AERMOD/Hourly Meteorology and SCREEN3
(Source only, background concentration not included)
Point Source
Figure 12. Maximum 24-hr Concentration - 11 lb per day (2 tpy equivalent)
7.2.3. Annual Concentrations
Figure 13 through Figure 15 present the annual concentrations for emission rates of 15 tpy, 10
tpy, and 5 tpy. Based on these results, the annual PM2.5 NAAQS could be threatened by an
individual emission unit with poor dispersion and an emission rate around or greater than 10 tpy.
For a point source with an emission rate of 5 tpy with poor dispersion, there will be situations
(Table 1 footnotes and Section 7.2.5) when the modeling significance level is exceeded and it is
reasonable to believe the source will cause or contribute to a violation of the 24-hr PM2 5
NAAQS.
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
31
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
Figure 13. Maximum Annual Concentrations - 15 tpy
Maximum Annual Concentration vs. Point Source
AERMOD/Hourly Meteorology and SCREEN3
(Source only, background concentration not included)
Maximum Annual Concentration vs. Point Source
AERMOD/Hourly Meteorology and SCREEN3
(Source only, background concentration not included)
<& # jjP ^ S> oN <3* & & &
Point Source
Point Source
Figure 14. Maximum Annual Concentrations - 10 tpy
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
32
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
Maximum Annual Concentration vs. Point Source
AERMOD/Hourly Meteorology and SCREEN3
(Source only, background concentration not included)
& <&
Point Source
Figure 15. Maximum Annual Concentrations - 5 tpy
7.2.4. 3-month Concentrations
For the rolling 3-month Pb NAAQS, the annual and 24-hr concentrations (monthly average
concentrations were not obtained from the model) were reviewed for emission rates of 0.6
tpy/300 pounds per 3-months (Figure 16 and Figure 17), 0.1 tpy/50 pounds per 3-months (Figure
18 and Figure 19), and 0.05 tpy/25 pounds per 3-months (Figure 20 and Figure 21).
Concentrations for a 3-month average are greater than the annual average but less than the 24-hr
average. Based on these results, the 3-month Pb NAAQS could be threatened by an individual
emission unit with poor dispersion and an emission rate around or greater than 0.1 tpy/50 pounds
per 3-months. For a point source with an emission rate of 0.05 tpy/25 pounds per 3-months with
poor dispersion, there will be situations (Table 1 footnotes and Section 7.2.5) when it is
reasonable to believe the source will cause or contribute to a violation of the rolling 3-month Pb
NAAQS.
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
33
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
Maximum Annual Concentration vs. Point Source
AERMOD/Hourly Meteorology and SCREEN3
(Source only, background concentration not included)
2.5
0.5
"s-montfi Pfc NAAQ8
¦ 3CSEEN3
DEN91
—QEN93
emissions - D.6'
QSN33
—dsns:
—PueWcCS
—coapo'Pt Uj
—PueMaOepotM
—PueCloDepoCC
py cP
C? C> r° C&
V*
Point Source
Figure 16. Maximum Annual Concentrations - 0.6 tpy
Maximum 24-hr Concentration vs. Point Source
AERMOD/Hourly Meteorology and SCREEN3
(Source only, background concentration not included)
<£~ # <$* <& <& ^ ^ r*p _fv? foN JT # ou ^
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
Maximum Annual Concentration vs. Point Source
AERMOD/Hourly Meteorology and SCREEN3
{Source only, background concentration not included)
<$¦ <& <& <& rS?> oN0 J5 cs? c& o° <£ <&° ° G°
Point Source
Figure 18. Maximum Annual Concentrations - 0.1 tpy
Maximum 24-hr Concentration vs. Point Source
AERMOD/Hourly Meteorology and SCREEN3
(Source only, background concentration not included)
£
O)
3r
| 1.5
o
0.5
• 3Cfl£EN3
CENS1
— C€NS5
AtoonSS
— Puet»»D2
PueMoM
— C35or.nsi56
— COBBrnB»93
— BymerGi
— 3y*w>OS
—ayane>07
— KM8H9S
Tnerma'St Lw*or-
— PueMC»ot99
—PwtKiOeoaOa
f
9} ¦
m
<£' <& <$"
<6° & ° 4? <$> o° jr
Point Source
Figure 19. Maximum 24-hr Concentrations - 50 pounds per 3-months
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
35
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
0.3
0.25
0.2
Maximum Annual Concentration vs. Point Source
AERMOD/Hourly Meteorology and SCREEN3
(Source only, background concentration not included)
<
E
0.1
0.05
Pve»oG»
—Pveacos
-COSermjsSS
—cosofosiso
—3y«nej03
— ijare-yDS
—Ko33«SJ
—WOM56
— FSPAK !C«3 30: 0-2
<& & <& <& <& r& rf' jP ,s? ,0S n
Point Source
nv sy
r
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
7.2.5. Other Modeling Variables Not Examined in this Modeling Analysis
There are other parameters used in modeling that are not examined here, such as elevated terrain,
urban effects, and background concentrations. Higher impacts may result when plume rise is
insufficient to clear nearby terrain. As discussed in EPA's AERMOD Implementation Guide
(March 19, 2009), plumes emitted or entrained into an urban air mass would be affected by the
dispersive nature of the "convective-like" boundary layer that forms during nighttime conditions
due to the urban heat island effect. Contributors to ambient air concentration for determining
compliance with AAQS are impacts from the source of interest and nearby sources, and the
background concentration. Even though impacts are just above modeling significance levels or
only a small fraction of the AAQS, a complete compliance demonstrate must also take existing
air pollutant concentration levels into account. This may mean that, in addition to adding a
background concentration, nearby sources with strong concentration gradients should be
included in the modeling. Since it's not reasonable to model all sources, it is necessary to add a
background concentration to account for the emissions from all sources that have not been
explicitly included in the modeling. Background concentrations vary by geographic area. For
areas with high background concentrations (and/or strong concentration gradients from nearby
sources) near the AAQS, a source impact that is greater than the modeling significance levels,
but still a relatively small percentage of the AAQS, can result in a modeled violation of the
AAQS.
8. Conclusion
The results in the January 2002 study demonstrate that a point source emitting 40 tons per year
of nitrogen oxides (NOx), sulfur dioxide (SO2), or fine particulate matter (PM10) or 15 tons per
year of PM10 could have a significant impact in ambient air, and in certain stack and building
configurations, exceed ambient air quality standards by itself. Lead (Pb) modeling was not
investigated as part of this study. When compounding factors such as the presence of nearby
sources and existing air pollution levels are considered, it is reasonable to conclude that even
sources with relatively small emission rates (much lower than those in Table 1 of the Modeling
Guideline) could cause or contribute to modeled violations of ambient air quality standards.
The results in the April 2010 study demonstrate that a point source emitting 0.46 pounds per
hour of NOx, 5 tons per year of PM2.5, 11 pounds per day of PM2.5, or 25 pounds per 3-months of
Pb could have a significant impact in ambient air, and in certain stack and building
configurations, exceed ambient air quality standards by itself.
Clearly, these studies show that it is problematic to use only emission rates to determine when
modeling is warranted. Many factors (including dispersion characteristics of the proposed
source) should be considered in the decision to perform modeling. Consequently, the Division
opposes the adoption of bright line exemptions from modeling that are based solely on emission
rates. Furthermore, due to the complexity of pollution dispersion in the atmosphere, it is not
realistic to develop a simple look-up table that adequately accounts for all of the important
factors that affect air pollution dispersion.
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
37
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
The study shows that, in cases where a source has good dispersion characteristics and the
existing air quality is well below ambient air quality standards, there is a low probability that the
source will cause or contribute to a modeled violation of ambient air quality standards. Thus, it
is reasonable to conclude that modeling is not warranted for minor sources and minor
modifications with good dispersion at emission rates below the thresholds in Table 1 of the
Colorado Modeling Guideline.
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
38
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
Appendix
Air Dispersion Modeling Analysis
of Fugitive Sources
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
39
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
LEGEND
Hi ghest-Sth-Highest Concentration
¦5V 613 ug/m 3
eth-Highest Concentration
5 - 50 ug/m3
¦ 51 -1 00 ugAn3
¦ >100ugAr>3
Modeling Significance Le*/el
5 ug/m3
S02 NAAQS:
365 ug/m3
PM10NAAQS:
150 ug/)m3
These results are based on 1986-1990
Denver Stapleton Arport Meteorological
Data. Using different meteorological
data will result in different impact estimates.
Magnitude and Location of Impacts from an Area Source
1 Square Acre and Release Height of 0 m
at 15 tpy Emission Rate
24-hr Average Concentration
(micrograms per cubic meter)
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
40
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
Magnitude and Location of Impacts from an Area Source
1 Square Acre and Release Height of 10 m
at 15 tpy Emission Rate
24-hr Average Concentration
(micrograms per cubic meter)
LEGEND
Highest-6th-Highest Concentration
35 ug/m3
6th-Highest Concentration
5- 50 ug/m3
¦ 51-100ug/m3
¦ >100 ug/m3
Modeling Significance Level
5 ug/m3
S02 NAAQS:
365 ug/m3
PM10 NAAQS:
150 ug/m3
These results ate based on 1986-1990
Denver Stapleton Airport Meteorological
Data. tJsin g d ifferent met eoro lo gi ca I
data will result in different impact estimates.
N
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
41
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
LEGEND
Highest-6th-Highest Concentration
•fa 313 ug(m3
6th-Highest Concentration
5 - 50 ug/m3
¦ 51-100ug/m3
¦ > 100 ug/m3
Modeling Significance Level: 5 ug/m3
S02 NAAQS: 365 ug/m3
PM10 NAAQS: 150 ug/m3
These results ate based on 1986-1990
Denver Stapleton Airport Meteorological
Data. Usin g d ifferent met eoro lo gi ca I
data will result in different impact estimates.
Magnitude and Location of Impacts from an Area Source
1 Square Acre and Release Height of 2.5 m
at 15 tpy Emission Rate
24-hr Average Concentration
(micrograms per cubic meter)
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
42
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
LEGEND
Highest-6th-Highest Concentration
•fa 124 ug(m3
6th-Highest Concentration
5 - 50 ug/m3
¦ 51-100ug/m3
¦ > 100 ug/m3
Modeling Significance Level: 5 ug/m3
S02 NAAQS: 365 ug/m3
PM10 NAAQS: 150 ug/m3
These results ate based on 1986-1990
Denver Stapleton Airport Meteorological
Data. Usin g d ifferent met eoro lo gi ca I
data will result in different impact estimates.
Magnitude and Location of Impacts from an Area Source
1 Square Acre and Release Height of 5 m
at 15 tpy Emission Rate
24-hr Average Concentration
(micrograms per cubic meter)
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
43
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
LEGEND
Highest-6th-Highest Concentration
•fa 59 ug/m3
6th-Highest Concentration
5 - 50 ug/m3
¦ 51-100ug/m3
¦ > 100 ug/m3
Modeling Significance Level: 5 ug/m3
S02 NAAQS: 365 ug/m3
PM10 NAAQS: 150 ug/m3
These results ate based on 1986-1990
Denver Stapleton Airport Meteorological
Data. Usin g d ifferent met eoro lo gi ca I
data will result in different impact estimates.
Magnitude and Location of Impacts from an Area Source
1 Square Acre and Release Height of 7.5 m
at 15 tpy Emission Rate
24-hr Average Concentration
(micrograms per cubic meter)
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
44
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
Magnitude and Location of Impacts from an Area Source
1 Square Acre, Release Height of 2 m, and
Initial Vertical Dispersion of 3 m at 15 tpy Emission Rate
24-hr Average Concentration
(micrograms per cubic meter)
, '
, /
" x _
S
\
/
N
/'
"\
/
\
/
\
/
\
/"
W — —
\
/
\
i
~
/
\
\
\
i
i
/
\
V
/
\
\
N
i
i
i
i
V
i
> / \
¦: (#)
i
¦ !
i
I
\
f
i
\
/
i
\
\
N
N
/
/
\
/'
\
/
\
/
/
\
/
/
/
\
' N
" V.
¦
•
A
s'
Modeling Significance Level: 5 ug/m3
S02 NAAQS: 365 ug/m3
PM10 NAAQS: 150 ug/m3
These results ate based on 1986-1990
Denver Stapleton Airport Meteorological
Data. Usin g d ifferent met eoro lo gi ca I
data will result in different impact estimates.
LEGEND
Highest-6th-Highest Concentration
-jV 324 ug(m3
6th-Highest Concentration
5 - 50 ug/m3
¦ 51-100ug/m3
¦ > 100 ug/m3
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
45
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
Magnitude and Location of Impacts from an Area Source
1 Square Acre, Release Height of 2 m, and
Initial Vertical Dispersion of 3 m at 15 tpy Emission Rate
Annual Average Concentration
(micrograms per cubic meter)
LEGEND
Maximum Annual Concentration
98 ug/m3
Annual Concentration
1-15 ug/m3
¦ 16-30ug/m3
¦ >30 ug/m3
Modeling Significance Level: 1 ug/m3
N02NMQS: 100ug/m3
S02 NAAQS: 80 ug/m3
PM10 NAAQS: 50 ug/m3
These results are based on 1986-1990
Denver Stapleton Airport Meteorological
Data. Usin g d ifferent met eoro lo gi ca I
data will result in different impact estimates.
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
46
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
LEGEND
Highest-6th-Highest Concentration
-jV 468 ug(m3
6th-Highest Concentration
5 - 50 ug/m3
¦ 51-100ug/m3
¦ > 100 ug/m3
Modeling Significance Level: 5 ug/m3
S02 NAAQS: 365 ug/m3
PM10 NAAQS: 150 ug/m3
These results ate based on 1986-1990
Denver Stapleton Airport Meteorological
Data. Usin g d ifferent met eoro lo gi ca I
data will result in different impact estimates.
Magnitude and Location of Impacts from an Area Source
2 Square Acres and Release Height of 0 m
at 15 tpy Emission Rate
24-hr Average Concentration
(micrograms per cubic meter)
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
47
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
LEGEND
Highest-6th-Highest Concentration
•fa 31 ug/m3
6th-Highest Concentration
5 - 50 ug/m3
¦ 51-100ug/m3
¦ > 100 ug/m3
Modeling Significance Level: 5 ug/m3
S02 NAAQS: 365 ug/m3
PM10 NAAQS: 150 ug/m3
These results ate based on 1986-1990
Denver Stapleton Airport Meteorological
Data. Usin g d ifferent met eoro lo gi ca I
data will result in different impact estimates.
Magnitude and Location of Impacts from an Area Source
2 Square Acres and Release Height of 10 m
at 15 tpy Emission Rate
24-hr Average Concentration
(micrograms per cubic meter)
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
48
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
LEGEND
Highest-6th-Highest Concentration
-jV 245 ug(m3
6th-Highest Concentration
5 - 50 ug/m3
¦ 51-100ug/m3
¦ > 100 ug/m3
Modeling Significance Level: 5 ug/m3
S02 NAAQS: 365 ug/m3
PM10 NAAQS: 150 ug/m3
These results ate based on 1986-1990
Denver Stapleton Airport Meteorological
Data. Usin g d ifferent met eoro lo gi ca I
data will result in different impact estimates.
Magnitude and Location of Impacts from an Area Source
2 Square Acres and Release Height of 2.5 m
at 15 tpy Emission Rate
24-hr Average Concentration
(micrograms per cubic meter)
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
49
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
LEGEND
Highest-6th-Highest Concentration
-5V 111 ug 100 ug/m3
Modeling Significance Level: 5 ug/m3
S02 NAAQS: 365 ug/m3
PM10 NAAQS: 150 ug/m3
These results ate based on 1986-1990
Denver Stapleton Airport Meteorological
Data. Usin g d ifferent met eoro lo gi ca I
data will result in different impact estimates.
Magnitude and Location of Impacts from an Area Source
2 Square Acres and Release Height of 5 m
at 15 tpy Emission Rate
24-hr Average Concentration
(micrograms per cubic meter)
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
50
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
LEGEND
Highest-6th-Highest Concentration
•fa 49 ug/m3
6th-Highest Concentration
5 - 50 ug/m3
¦ 51-100ug/m3
¦ > 100 ug/m3
Modeling Significance Level: 5 ug/m3
S02 NAAQS: 365 ug/m3
PM10 NAAQS: 150 ug/m3
These results ate based on 1986-1990
Denver Stapleton Airport Meteorological
Data. Usin g d ifferent met eoro lo gi ca I
data will result in different impact estimates.
Magnitude and Location of Impacts from an Area Source
2 Square Acres and Release Height of 7.5 m
at 15 tpy Emission Rate
24-hr Average Concentration
(micrograms per cubic meter)
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
51
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
LEGEND
Highest-6th-Highest Concentration
-jV 248 ug(m3
6th-Highest Concentration
5 - 50 ug/m3
¦ 51-100ug/m3
¦ > 100 ug/m3
Modeling Significance Level
5 ug/m3
S02 NAAQS:
365 ug/m3
PM10 NAAQS:
150 ug/m3
These results ate based on 1986-1990
Denver Stapleton Airport Meteorological
Data. Usin g d ifferent met eoro lo gi ca I
data will result in different impact estimates.
Magnitude and Location of Impacts from an Area Source
2 Square Acres, Release Height of 2 m, and
Initial Vertical Dispersion of 3 m at 15 tpy Emission Rate
24-hr Average Concentration
(micrograms per cubic meter)
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
52
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
Magnitude and Location of Impacts from an Area Source
2 Square Acres, Release Height of 2 m, and
Initial Vertical Dispersion of 3 m at 15 tpy Emission Rate
Annual Average Concentration
(micrograms per cubic meter)
LEGEND
Maximum Annual Concentration
^ 78 ug/m3
Annual Concentration
1-15 ug/m3
¦ 16-30ug/m3
¦ >30 ug/m3
Modeling Significance Level: 1 ug/m3
N02 NAAAQS: 100ug/m3
S02 NAAQS: 80 ug/m3
PM10NAAQS: 50 ug/m3
These results ate based on 1986-1990
Denver Stapleton Airport Meteorological
Data. Using different meteorological
data will result in different impact estimates.
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
53
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
LEGEND
Highest-6th-Highest Concentration
¦5^ 263 ug100 ug/m3
Modeling Significance Level: 5 ug/m3
S02 NAAQS: 365 ug/m3
PM10 NAAQS: 150 ug/m3
These results ate based on 1986-1990
Denver Stapleton Airport Meteorological
Data. Usin g d ifferent met eoro lo gi ca I
data will result in different impact estimates.
Magnitude and Location of Impacts from an Area Source
5 Square Acres and Release Height of 0 m
at 15 tpy Emission Rate
24-hr Average Concentration
(micrograms per cubic meter)
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
54
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
LEGEND
Highest-6th-Highest Concentration
-5V 32 ug/m3
6th-Highest Concentration
5 - 50 ug/m3
¦ 51-100ug/m3
¦ > 100 ug/m3
Modeling Significance Level: 5 ug/m3
S02 NAAQS: 365 ug/m3
PM10 NAAQS: 150 ug/m3
These results ate based on 1986-1990
Denver Stapleton Airport Meteorological
Data. Usin g d ifferent met eoro lo gi ca I
data will result in different impact estimates.
Magnitude and Location of Impacts from an Area Source
5 Square Acres and Release Height of 10 m
at 15 tpy Emission Rate
24-hr Average Concentration
(micrograms per cubic meter)
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
55
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
Magnitude and Location of Impacts from an Area Source
5 Square Acres and Release Height of 2.5 m
at 15 tpy Emission Rate
24-hr Average Concentration
(micrograms per cubic meter)
Modeling Significance Level: 5 ug/m3
S02 NAAQS: 365 ug/m3
PM10 NAAQS: 150 ug/m3
These results ate based on 1986-1990
Denver Stapleton Airport Meteorological
Data. Usin g d ifferent met eoro lo gi ca I
data will result in different impact estimates.
LEGEND
Highest-6th-Highest Concentration
-jV 158ug(m3
6th-Highest Concentration
5 - 50 ug/m3
¦ 51-100ug/m3
¦ > 100 ug/m3
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
56
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
LEGEND
Highest-6th-Highest Concentration
•fa 76 ug/m3
6th-Highest Concentration
0 - 50 ug/m3
¦ 51-100ug/m3
¦ > 100 ug/m3
Modeling Significance Level: 5 ug/m3
S02 NAAQS: 365 ug/m3
PM10 NAAQS: 150 ug/m3
These results ate based on 1986-1990
Denver Stapleton Airport Meteorological
Data. Usin g d ifferent met eoro lo gi ca I
data will result in different impact estimates.
Magnitude and Location of Impacts from an Area Source
5 Square Acres and Release Height of 5 m
at 15 tpy Emission Rate
24-hr Average Concentration
(micrograms per cubic meter)
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
57
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
LEGEND
Highest-6th-Highest Concentration
•fa 49 ug/m3
6th-Highest Concentration
5 - 50 ug/m3
¦ 51-100ug/m3
¦ > 100 ug/m3
Modeling Significance Level: 5 ug/m3
S02 NAAQS: 365 ug/m3
PM10 NAAQS: 150 ug/m3
These results ate based on 1986-1990
Denver Stapleton Airport Meteorological
Data. Usin g d ifferent met eoro lo gi ca I
data will result in different impact estimates.
Magnitude and Location of Impacts from an Area Source
5 Square Acres and Release Height of 7.5 m
at 15 tpy Emission Rate
24-hr Average Concentration
(micrograms per cubic meter)
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
58
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
Magnitude and Location of Impacts from an Area Source
5 Square Acres, Release Height of 2 m, and
Initial Vertical Dispersion of 3 m at 15 tpy Emission Rate
Annual Average Concentration
(micrograms per cubic meter)
Modeling Significance Level: 5 ug/m3
S02 NAAQS: 365 ug/m3
PM10 NAAQS: 150 ug/m3
These results ate based on 1986-1990
Denver Stapleton Airport Meteorological
Data. Usin g d ifferent met eoro lo gi ca I
data will result in different impact estimates.
LEGEND
Highest-6th-Highest Concentration
•fa 153ug(m3
6th-Highest Concentration
5 - 50 ug/m3
¦ 51-100ug/m3
¦ > 100 ug/m3
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
59
-------
"Air Dispersion Modeling Analysis to Support the Modeling Thresholds and Associated Language in Section 2 of the Colorado
Modeling Guideline for Air Quality Permits"
Magnitude and Location of Impacts from an Area Source
5 Square Acres, Release Height of 2 m, and
Initial Vertical Dispersion of 3 m at 15 tpy Emission Rate
Annual Average Concentration
(micrograms per cubic meter)
LEGEND
Maximum Annual Concentration
•fa 52 ug/m3
Annual Concentration
1-15 ug/m3
¦ 16-30ug/m3
¦ >30 ug/m3
Modeling Significance Level: 1 ug/m3
N02NMQS: 100ug/m3
S02 NAAQS: 80 ug/m3
PM10NAAQS: 50 ug/m3
These results ate based on 1986-1990
Denver Stapleton Airport Meteorological
Data. Usin g d ifferent met eoro lo gi ca I
data will result in different impact estimates.
Colorado Department of Public Health & Environment / Air Pollution Control Division
April 2010
60
-------
APPENDIX B
Colorado Interim Modeling Guideline for Air Quality Permits (October 2021)
-------
COLORADO
Department of Public
Health & Environment
Dedicated to protecting and improving the health and environment of the people of Colorado
INTERIM COLORADO MODELING
GUIDELINE FOR AIR QUALITY PERMITS
OCTOBER 2021
COLORADO DEPARTMENT OF PUBLIC HEALTH AND ENVIRONMENT
AIR POLLUTION CONTROL DIVISION
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
Table of Contents
Preface 3
What's New in this Document 4
Definitions 7
Section 1 - Introduction 11
Section 2 - Authority for Air Quality Impact Analyses 12
Attainment Areas 13
Nonattainment Areas 14
Requirements Unique to Colorado 15
Section 3 - Air Quality Impact Analysis 17
Air Dispersion Modeling 17
Air Quality Impact Analysis Process 18
Section 4 - Performing the Air Quality Impact Analysis 23
Modeling Thresholds 23
Modeling Protocol 23
Screening Modeling 23
Refined Modeling 24
Significant Impact Analysis 24
Significant Impact Level (SIL) 25
Cumulative Impact Analysis 28
NAAQS & CAAQS Analysis 30
PSD Increment Analysis 36
Increment Calculation 37
Unique Colorado PSD Increment Requirement 39
Additional Impacts Analysis 45
Impact Analysis for Water 45
Visibility Analysis 45
Soils and Vegetation Analysis 47
AQRV and Visibility Analysis in Federal Class I Areas 47
Air Quality Related Values Analysis for Major Stationary Sources 48
Pre-and Post-Construction Monitoring Analysis 50
Pre- Construction Monitoring Analysis 51
Post-Construction Monitoring Analysis 51
-------
October 2021 Interim Colorado Modeling Guideline for Air Quality Permits
Regulated, Non-Criteria Pollutant Analysis 53
Section 5 - Preferred Air Dispersion Models & Associated Inputs 53
Source Data 53
Criteria Pollutant Recommendations 53
Carbon Monoxide 54
Lead 54
Nitrogen Dioxide 54
Ozone 55
Particulate Matter < 10|am (PMio) 55
Particulate Matter < 2.5|am (PM2.5) 55
Sulfur Dioxide 55
Mobile Sources Data 56
Nearby Sources Data 56
Background Concentrations 57
Elevation Data 58
Downwash Applicability 60
Receptor Network 61
Meteorological Data 61
Modeling Scenarios 62
Section 6 - Reporting Requirements 62
Consistency in Geographic Coordinates 63
Exemptions from Submitting Modeling-Related Data 63
New Sources and Modifications Not Subject to PSD Rules 63
New Sources and Modifications Subject to PSD Rules 63
Recommended Additional Guidance 64
Appendix A - Case by Case Modeling Threshold Supporting Documentation
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
Preface
The Interim Colorado Modeling Guideline for Air Quality Permits (Guideline) presents current Air Pollution
Control Division (Division) interim air quality modeling guidance for estimating impacts from stationary
sources of air pollution. It addresses modeling issues for source types ranging from small minor sources to
major sources such as those subject to Prevention of Significant Deterioration (PSD) review.
Recommendations in the Colorado Air Quality Modeling Guideline may not be applicable in all situations.
The Guideline is intended to help permit applicants, air quality specialists, and others understand the
Division's expectations for the ambient air impact analysis and to prevent unnecessary delays in the
permit process. It provides a starting point for modeling, but allows the use of professional judgment. To
avoid misunderstandings, obtain the most recent version of Colorado's guidance documents from
http://www.colorado.gov/airquality/permits.aspx. In addition, obtain current regulations and applicable
U.S. Environmental Protection Agency (U.S. EPA) guidance.
This guideline is not intended to describe the implications of modeling results. Such implications are
generally controlled by the permit rules or other relevant state and federal regulations, laws and
guidance. Nevertheless, the Guideline contains incidental discussion of the effects of certain modeling
results. Such discussion is for informational purposes only and shall not be construed to be authority
defining the regulatory impact of any modeling result. For that, the reader should refer to the applicable
rules and regulations.
This is a guide through modeling-related regulations and procedures. It is intended to promote technically
sound and consistent modeling techniques, while encouraging the use of improved and more accurate
techniques as they become available. The guideline helps permit applicants understand when modeling is
warranted. It clarifies what modeling-related information and data should be included with a permit
application. Supplemental guidance on specific technical issues and other modeling-related data and
information, including checklists and meteorological data, are available at
http://www.colorado.Rov/airquality/permits.aspx. If modeling procedures other than those
recommended in Colorado and U.S. EPA guidance are used, there might be delays while the procedures
are reviewed. In some cases, U.S. EPA approval may be necessary.
This is only a guidance document. It is not intended to supersede statutory/regulatory requirements or
recommendations of the U.S. EPA or Colorado law or regulations.
U.S. EPA models and guidance are available on the Internet at http://www.epa.gov/scram.
Page 3 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
What's New in this Document
October 2021:
The following has been revised on October 19, 2021 from the May 2018 version of the Guideline.
• Revisions were made to the Modeling Thresholds section and Table 1 to clarify the considerations
used on an interim basis to evaluate when minor sources of emissions undergo modeling.
• Revisions were made to remove references to outdated Colorado Ambient Air Quality Standards
The following has been revised on May 17, 2018 from the April 2018 version of the Guideline.
¦ Ozone background criteria to use ARM2
¦ Figure 5 Title as the map only applies to PMio
¦ Table 8 footnote (a) corrected to increment
April 2018:
The following has been revised on April 20, 2018 from the March 2018 version of the Guideline.
¦ Spelling and syntax errors were corrected throughout
¦ PM2.5 SILs are now EPA approved based on recent guidance
March 2018:
The following has been revised on March 16, 2018 from the December 27, 2005, version of the Guideline.
¦ The overall document has been reformatted.
¦ The overall document has also been reorganized to improve the flow of information as well as
the understanding and retention of information presented.
¦ Hyperlinks have been updated to obtain the correct sites.
¦ Duplicate tables and figures have been removed. Only one version of each table and figure is
provided.
¦ Quotes from Regulation No. 3 and Appendix W have been updated to match the most recent
versions.
¦ Section 2 was added to address the applicable regulations that authorize Colorado to perform
modeling analyses.
¦ Section 3 was added to illustrate the full picture of the modeling analysis process.
¦ Section 4 was added to detail the different types of modeling analyses the applicant may be
required to perform.
(Table 6)
• Other clarifying and grammatical changes were made.
May 2018:
Page 4 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
¦ Section 5 was added to detail the information the applicant should use when performing a
modeling analysis.
¦ Section 6 was added to detail what information the Division is looking for when the applicant
submits a modeling analysis.
¦ Appendix A was added which includes the description of how the modeling thresholds in Table 1
were developed.
¦ Section 4 Additions:
D U.S. EPA's opinion on submitting a modeling protocol and language that a protocol is not
intended to be legally binding.
D Screening-level models were added with updated guidance on when screening models
cannot be used.
D Procedural steps were added to the Significant Impact Analysis
D Procedural steps were added to the NAAQS & CAAQS Analysis
D Procedural steps were added to the PSD Increment Analysis
¦ Section 5 Additions:
D The Ambient Ratio Method 2 (ARM2) is now the approved Tier 2 approach replacing the
Ambient Ratio Method (ARM). The approved ambient ratio for Colorado is discussed
more in detail. The Ozone Limiting Method is now a regulatory option.
D Annual PMio NAAQS compliance demonstration was removed as the NAAQS was
revoked.
D 24-hour and annual PM2.5 NAAQS compliance demonstrations were added.
D 1-hour S02 compliance demonstrations as well as a discussion regarding the 24-hour and
annual S02 NAAQS demonstrations were added.
D PS Memo 10-01 discussion was added.
D Nearby source emission calculations have been updated in Appendix W from allowable
emissions to a subset of actual emissions. The threshold emission rates for nearby
sources to include was removed.
D Flagpole receptors guidance was added.
D The use of Digital Elevation Model (DEM) data for sources and receptors was removed
and language to use National Elevation Dataset (NED) files was added.
D Meteorological data will be provided by the Division in an AERMOD-ready format. The
applicant no longer needs to process meteorological data.
D Precursors to ozone was added.
D Secondary formations of PM2.5 was added.
D Mobile sources guidance was added.
D Modeling scenarios guidance was added.
Page 5 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
¦ Language was updated throughout discussing when source and modifications are exempt from
modeling. The exemption now includes emissions below Table 1 thresholds AND not meeting the
scenarios (footnotes) described below Table 1.
¦ References to the Modeling Submittal Completeness Checklist to verify the necessary information
to submit with the modeling analysis was added.
¦ The sections relating to additional impacts analysis and AQRVs is currently still under review so
these sections have been greyed out.
Page 6 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
Definitions
The following explanation of terms are included solely for the reader's convenience; they do not take the
place of any definition in state or federal laws, rules, or regulations.
Air Quality Models. Computer codes for estimating ambient concentration levels (i.e., "impacts") from
new and existing sources of air pollution. They allow one to forecast future air quality levels from sources
that have not been constructed. They simulate in a simplified manner the complex behavior of emissions
injected into the atmosphere.
Air Quality Related Value (AQRV). A feature or property of a Class I area that may be affected by air
pollution. General categories of AQRV's include visibility, odor, flora, fauna, soil, water, geological
features, and cultural resources. https://www.nature.nps.gov/air/Pubs/pdf/flag/FLAG 2010.pdf
Ambient Air. Defined by 40 CFR 50.1(e) as "that portion of the atmosphere, external to the source, to
which the general public has access." NAAQS and PSD increments apply only in ambient air.
Appendix W, 40 CFR Part 51- Guideline on Air Quality Models. The U.S. EPA's Guideline on Air Quality
Models recommends air quality modeling techniques that should be applied to State Implementation Plan
(SIP) revisions for existing sources and to new source reviews, including prevention of significant
deterioration (PSD). It is intended for use by the U.S. EPA in judging the adequacy of modeling analyses
performed by U.S. EPA, state and local agencies, and industry. The Guideline identifies those techniques
and databases U.S. EPA considers acceptable. The guide is not intended to be a compendium of modeling
techniques. Rather, it serves as a basis by which air quality managers, supported by sound scientific
judgment, have a common measure of acceptable technical analysis. Appendix W was updated on
January 17, 2017. https://www3.epa.gov/ttn/scram/appendix w/2016/AppendixW 2017.pdf
Attainment Area. Any area not designated as a Nonattainment Area, as defined below. Generally, an
Attainment Area meets the national primary or secondary ambient air quality standard for an applicable
criteria pollutant.
Background. Air contaminant concentrations present in the ambient air that are not attributed to the
source or site being evaluated.
Building Downwash. Turbulence created by the wind flowing over buildings or structures that would
ordinarily not exist. This effect can alter ground-level concentration levels than would exist in the absence
of the building or structure.
Class I Area. An area defined by Congress that is afforded the greatest degree of air quality protection.
Class I areas are deemed to have special natural, scenic, or historic value. The Prevention of Significant
Deterioration (PSD) regulations provide special protection for Class I areas. Little deterioration of air
quality is allowed.
Class II Area. An area defined by Congress where a moderate degree of emissions growth is allowed.
Complex Terrain. Any terrain exceeding the height of the stack being modeled. This includes terrain
commonly referred to as intermediate terrain (receptors between stack height and plume height).
Page 7 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
Criteria Pollutant. A pollutant for which a National Ambient Air Quality Standard (NAAQS) has been
defined.
Cumulative Impact Analysis. A full modeling impact analysis that involves the facility under permit review,
nearby sources, and background concentrations to compare the facility's impact to the NAAQS.
Fugitive Emission. Any gaseous or particulate contaminant entering the atmosphere that could not
reasonably pass through a stack, chimney, vent, or other functionally equivalent opening designed to
direct or control its flow.
Good Engineering Practice (GEP) Stack Height. From Regulation No. 3, Part D, §VIII.D.3, "The greater of 65
meters or for stacks in existence on January 12, 1979 and for which the owner or operator had obtained
applicable pre-construction permits or approvals required, Flg = 2.5*FH (provided the owner or operator
produces evidence that this equation was actually relied on in establishing an emission limitation and for
all other stacks FHg = FH +1.5*L where,
Hg: good engineering practice stack height measured from the ground level elevation at the base of the
stack
FH: height of nearby structure(s) measured from the ground level elevation at the base of the stack
L: lesser dimension (height or projected width) of nearby structure(s)."
Flazardous Air Pollutant (HAP). Any pollutant subject to a standard promulgated under the Federal Clean
Air Act (FCAA) section 112 (relating to hazardous air pollutants).
Major Stationary Source. The term major may refer to the total emissions at a stationary source or to a
specific facility.
1. A named major stationary source is any source belonging to a list of 28 source categories in 40 CFR
52.21(b)(1) which emits or has the potential to emit 100 tons per year (tpy) or more of any
pollutant regulated by the Federal Clean Air Act (FCAA).
2. A major stationary source is any source not belonging to the 28 named source categories which emits
or has the potential to emit such pollutants in amounts of 250 tpy or more.
3. A major source is any source that emits 10 tpy or more of any single HAP or 25 tpy or more of any
combination of HAPs under FCAA section 112(b).
Major Modified Stationary Source. Used in the context of a PSD application, the phrase major modified
stationary source or facility refers to a change in operation that results in a significant net increase of
emissions for any pollutant for which a NAAQS has been defined. New sources at an existing major
stationary source are treated as modifications to the major stationary source.
Major New Source Review (NSR) Program. The major NSR program contained in parts C and D of title I of
the FCCA is a preconstruction review and permitting program applicable to new major sources and major
modifications at such sources. In areas meeting the NAAQS (attainment areas) or for which there is
insufficient information to determine whether they meet the NAAQS (unclassifiable areas), the NSR
requirements under part C of title I of the FCAA apply. The EPA calls this portion of the major NSR
program the Prevention of Significant Deterioration or PSD program. In areas not meeting the NAAQS, the
major NSR program is implemented under the requirements of part D of title I of the FCCA. The
Page 8 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
EPA calls this program the "nonattainment" major NSR program. The EPA has promulgated rules in 40 CFR
52.21 to implement PSD in portions of the country that do not have approved state or tribal PSD
programs.
Major Source Baseline Date. This is the date after which actual emissions associated with physical changes
or changes in the method of operation at a major stationary source affect the available increment.
Changes in actual emissions occurring at any stationary source after this date contribute to the baseline
concentration until the minor source baseline date is established.
Minor Source. Any stationary source that is not defined as a major stationary source in Regulation No. 3,
Part D §11.A.25. The term is sometimes used rather loosely. The definition may vary based on the context
in which it is used.
Minor Source Baseline Date. This is the earliest date after the PSD increment trigger date on which a PSD
application for a new major source or a major modification to an existing source is considered complete.
The minor source baseline date is pollutant- and geographically-specific.
Modeling and Emissions Inventory Unit (MEIU). This is the unit within the Technical Services Program (TSP)
of the Air Pollution Control Division that is responsible for review of air dispersion modeling.
Modified Stationary Source.
¦ When used in the context of modeling, the phrase modified stationary source or facility refers to
a change in the location or stack parameters of an emission point, including emission rate.
¦ When used in the context of a permit application, the phrase modified stationary source or
facility refers to a physical change in, or change in method of operation, that results in an
increase of emissions
National Ambient Air Quality Standards (NAAQS). Levels of air quality to protect the public health and
welfare (40 CFR 50.2). Primary standards are set to protect public health, including the health of
"sensitive" populations such as asthmatics, children, and the elderly from the effects of "criteria air
pollutants" and certain non-criteria pollutants. Secondary standards are set to protect public welfare,
including protection against decreased visibility, damage to animals, crops, vegetation, and buildings.
Nearby Sources. Any major source, major stationary source, or minor source that causes a significant
concentration gradient in the vicinity of a new or modified source.
Nonattainment Area (NAA). Any area designated by EPA as nonattainment for a NAAQS.
Other Background Sources. All sources of air pollution other than the source under review and those
identified as nearby sources. Examples include area and mobile sources, natural sources, most minor
sources, distant major sources, and major stationary sources. They usually are accounted for by using an
appropriate ambient background concentration as recommended in section 8 of Appendix W of 40 CFR
Part 51 or by application of a model using inventory recommendations in Table 8-2 of Appendix W.
Project. An operational and/or physical change that may affect air emission rates at a site.
PSD Increment. The maximum allowable increase of an air pollutant that is allowed to occur above the
applicable baseline concentration for that pollutant.
Page 9 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
Qualitative Determination. Relies on descriptive generalized statements and made without regard to
quantity.
Quantitative Determination. A numerical "estimate" of the air pollutant concentration in ambient air.
Reasonable Further Progress (RFP). From the Common Provisions Regulation, "The annual incremental
reductions in emissions of the applicable air pollutant (including substantial reductions in the early years
following approval or promulgation of plan provisions under the Federal Act, section 110(a)(2)(l) and
regular reductions thereafter) which are sufficient in the judgment of the commission and U.S. EPA, to
provide for attainment of the applicable National Ambient Air Quality Standards by the date required in
section 172(a) of the Federal Act."
Receptor. As used here, a receptor is a geographic location (point) at which the model calculates the
impact (i.e., air pollutant concentration) from a source of air pollution. In practice, a large number of
receptors (i.e., a grid of receptors) is used to estimate air quality impacts over the probable area of
impact from the source. Each receptor has a unique geographic coordinate and elevation.
Refined Model. An analytical technique that provides a detailed treatment of physical and chemical
atmospheric processes and requires detailed and precise input data. Specialized estimates are calculated
that are useful for evaluating source impact relative to air quality standards and allowable increments.
The estimates are more representative than those obtained from conservative screening techniques.
Requested Emission Rate. The emission rate calculated using the maximum rated (design) capacity of the
source or the emission rate specified as an enforceable permit condition.
Scenic and/or Important Views. An important or sensitive panorama or long-range view anywhere in
Colorado. This includes important views of landmarks or panoramas. The Division maintains a list of
scenic and/or important views in Colorado
(https://www.colorado.Rov/airquality/permits/SCENICVW2005.pdf).
Screening Technique. A relatively simple analysis technique to determine whether a given source is likely
to pose a threat to air quality. Concentration estimates from screening techniques are conservative.
Significant Impact Analysis (SIA). Modeling analysis involving only the project sources to determine
whether a new and/or modified facility, or a combination of the two, could cause a significant ambient air
impact.
Significant Impact. A concentration in ambient air that exceeds a modeling significance level.
Significant Impact Levels (SILs). Values established by EPA to determine whether a proposed new or
modified source will cause or contribute to a violation of the NAAQS or PSD increments. When a facility
impact is above the applicable SILs, a refined cumulative impact analysis is required.
Simple Terrain. Any terrain with elevations lower than the top of the stack.
Stationary Sources Program (SSP). This is the program within the Air Pollution Control Division that is
responsible for air quality permitting and enforcement.
Unclassifiable Area. Any area that cannot be classified on the basis of available information as meeting or
not meeting the national primary or secondary ambient air quality standard for the pollutant.
Page 10 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
Universal Transverse Mercator (UTM). A plane coordinate system that uses distances from a specified
reference point as the basis for all locations. It is based on a transverse Mercator projection that divides
the Earth's surface into zones that are 6 degrees of longitude wide. Precise locations on the earth are
described in terms of north-south (northing) and east-west (easting) distances, measured in meters from
the origin of the appropriate UTM Zone.
Most of Colorado is zone 13, while the western seventh is in zone 12.
Air quality models are used to estimate impacts (air pollutant concentration levels) in ambient air to
evaluate whether a proposed source or activity will comply with applicable ambient air standards and
other applicable regulatory requirements. Federal law requires that the Division have legally enforceable
procedures in place to prevent construction or modification of any source where the emissions from the
projected activity would violate control strategies or interfere with attainment and maintenance of the
National Ambient Air Quality Standards (NAAQS).1
All estimates of ambient concentrations required under Colorado Air Quality Control Commission (AQCC)
Regulation No. 3 must be based on U.S. EPA-approved air quality models, data bases, and other
requirements generally approved by the U.S. EPA and specifically approved by the Division. Case-by-case
approval from the Division and/or U.S. EPA is required if a non-EPA model is proposed.
Regulation No. 3, Part A, §VI 11 .A. 1 states that "all estimates of ambient concentrations required under this
Regulation No. 3 shall be based on the applicable air quality models, data bases, and other requirements
generally approved by U.S. EPA and specifically approved by the division. If a non-U.S. EPA approved
model, such as a wind tunnel study, is proposed, the nature and requirements of such a model should be
outlined to the division at a pre-application meeting. The application will be deemed incomplete until there
has been an opportunity for a public hearing on the proposed model and written approval of the U.S. EPA
has been received."
The primary U.S. EPA modeling guideline is 40 CFR Part 51, Appendix W - Guideline on Air Quality Models
(https://www3.epa.Rov/ttn/scram/appendix w/2016/AppendixW 2017.pdf). There are many other U.S.
EPA guidance documents, memos, and U.S. EPA model clearinghouse decisions that explain modeling
procedures. This Guideline is intended to help permit applicants understand federal modeling procedures.
It also provides Colorado's interpretation of gray areas in federal guidance. As such, it presents
procedures that are "specifically approved" by the Division.
1 Pursuant to section 110(a)(2)(C) of the federal Clean Air Act, the State Implementation Plan (SIP)
needs to regulate the "modification and construction of any stationary source within the areas
covered by the plan as necessary to assure that national ambient air quality standards are
achieved." Similarly, 40 CFR section 51.160 requires the State to have the authority to prohibit
any construction or modification that would interfere with the attainment or maintenance of a
national standard. This includes PSD increments as well as NAAQS. See also 40 C.F.R 51.166.
There is no distinction in these provisions between major and minor sources.
Section 1 - Introduction
Page 11 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
The primary Colorado regulation for air quality permits is Regulation No. 3.2 Certain new/modified air
pollution sources are subject to the regulatory modeling requirements of Regulation No. 3 (authorized by
§25-7-114 to 25-7-114.7, Colorado Revised Statutes (C.R.S.)).
To avoid unnecessary delays in permit processing, pre-application meetings and communications (e.g.,
phone, e-mail, letter) are strongly recommended, particularly for new major sources and major
modifications. The Division does not routinely require or perform modeling to determine impacts from
hazardous air pollutants (HAPs).
Section 2 - Authority for Air Quality Impact Analyses
The Colorado AQCC developed regulations that require the Division's preliminary analysis for construction
permits to evaluate what impact, if any, the new source will have within the probable area of influence of
the proposed source. See Regulation No. 3, Part B, §111.B.5.d.. In addition, the Division must determine if
the proposed source or activity will comply with applicable ambient air quality standards. When a
source's requested emission rate exceeds certain levels, the recommended tools for determining impacts
are air quality models. This section discusses the regulatory requirements for air quality impact analyses.
While modeling is not required to obtain an operating permit, it may be performed or requested if the
operating permit is modified (Regulation No. 3, Part C, §X). Operating permits may also be subject to
modeling if the application is for a combined construction/operating permit (Regulation No. 3, Part C,
§ 111 .C. 12.d).
For both major sources and minor sources, Regulation No. 3, Part B, §111.B.5.d states, "the preliminary
analysis shall indicate what impact, if any, the new source will have (as of the projected date of
commencement of operation) on all areas (attainment, attainment/maintenance, nonattainment,
unclassifiable), within the probable area of influence of the proposed source... When the preliminary
analysis includes modeling, the model used shall be an appropriate one given the topography,
meteorology, and other characteristics of the region that the source will impact. Use of any non-guideline
model required U.S. EPA approval under Section VIII. A. of Part A of this regulation."
Regulation No. 3, Part B, §111.D.l states that the Division or the AQCC "shall grant the permit if it finds
that...the proposed source or activity will not cause an exceedance of any National Ambient Air Quality
Standards; and the source or activity will meet any applicable ambient air quality standards and all
applicable regulations..."
While Regulation No. 3 requires that the Division indicate the "impact, if any" in its preliminary analysis, it
does not explicitly require modeling; however, a demonstration of compliance with all NAAQS and CAAQS
is required. Thus, the impact analysis can be done using quantitative (modeling) or qualitative (non-
modeling) methods, as appropriate; however, U.S. EPA approved models and/or methods must be used if
a numerical estimate (i.e., pollutant concentration in ambient air) of the impact is made, unless specific
2 Colorado air quality regulations are available on the CDPHE website
(https://www.colorado.gov/pacific/cdphe/aqcc-regs) or upon request. To obtain official copies,
please contact the Secretary of State's office.
Page 12 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
approval is granted to use a non-EPA approved model. Regulation No. 3 is silent as to how a
demonstration can be made when modeling is not required - i.e. the qualitative method.
The modeling thresholds and scenarios outlined in Table 1 may be used to determine when modeling is
warranted. If it is unclear if modeling is warranted, please contact the Division. The thresholds are
applicable for sources located in nonattainment as well as attainment areas.
The impact analysis requirement in Regulation No. 3 applies to all areas: attainment,
attainment/maintenance, nonattainment, and unclassifiable.
Attainment Areas
New major stationary sources and major modifications subject to PSD attainment area rules are required
to submit various types of modeling and/or analyses along with their permit application. The application
must include appropriate modeling and/or analyses to be ruled complete. Please refer to Regulation No.
3, Part D, §VI.A.2 and §VI.A.6 for source impact analysis requirements.
With respect to ambient air standards, §VI.A.2 requires that "the owner or operator of the proposed
source or modification shall demonstrate to the Division that allowable emission increases from the
proposed source or modification in conjunction with all other applicable emissions increases or reductions
(including secondary emissions) will not cause or contribute to concentrations of air pollutants in the
ambient air in violation of either:
VI.A.2.a: any state or national ambient air quality standard in any baseline area or air quality
control region; or
VI.A.2.b: any applicable maximum allowable increase over the baseline concentration in any area"
Regulation No. 3, Part D, §VI.D.l.b requires that "the proposed source or modification will achieve an
emissions rate that will ensure that the emissions of such pollutant from the source or modification will
not significantly affect ambient air quality in the nonattainment area." That is, the modeling that is
required should be used to determine if the source would have a significant impact in any nonattainment
area.
Major sources and major modifications are subject to additional requirements. See section 4 for more
details. The impact analysis requirement of Regulation No. 3 applies to all areas (attainment,
attainment/maintenance, nonattainment, unclassifiable).
Minor sources and minor modifications are not required by regulation to submit a modeling analysis that
demonstrates compliance along with their permit application; however, a demonstration of compliance
(quantitative or qualitative) with all NAAQS and CAAQS is required. Nevertheless, applicants may elect to
include modeling with the applications to prevent unnecessary delays.
If modeling is not submitted with the permit application, the Division will decide if modeling is warranted
to complete the impact analysis and compliance demonstration required by Regulation No. 3. If modeling
is warranted, the Division will perform a screening-level analysis if it is technically feasible to perform one.
If the screening-level analysis shows there could be modeled violations of applicable standards, the
Division will contact the applicant to discuss options. Since the Division does not usually perform refined-
Page 13 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
level modeling as part of the permitting process, the Division will typically require that the applicant
perform any refined modeling that might be warranted.
If modeling is warranted, refer to sections 4, 5, and 6.
Nonattainment Areas
The impact analysis requirement of Regulation No. 3, Part B, §111.B.5.d applies in all areas ("attainment,
attainment/maintenance, nonattainment, unclassifiable"). Thus, modeling may sometimes be warranted
for sources in nonattainment areas. The goals of the impact analysis vary depending on the applicable
regulatory requirements. The regulations refer to the concept of reasonable further progress (RFP) for
sources located in nonattainment areas. If emissions from a new source or modification would prevent a
nonattainment area (NAA) from coming into compliance by the applicable date in the Federal Act or in
the SIP, then the source impairs RFP.
New major stationary sources and major modifications subject to NSR nonattainment area rules are
required to submit various types of modeling and/or analyses along with their permit application. In
nonattainment areas, Regulation No. 3, Part D, §V contains a number of requirements for obtaining a
permit. Refer to the regulation for details. A few of the requirements follow:
Offsets must represent reasonable further progress towards attainment of the National Ambient
Air Quality Standards when considered in connection with other new and existing sources of
emissions.
a) In addition, offsets for PMio, sulfur oxides, and carbon monoxide must show, through
atmospheric modeling, a positive net air quality benefit in the area affected by the emissions.
b) Provided, however, that offsets meeting the requirements of this section V.A.3 may also be
obtained from existing sources outside the nonattainment area if the applicant demonstrates:
¦ A greater air quality benefit may thus be achieved; or sufficient offsets are not available
from sources within the nonattainment area; and
¦ The other area has an equal or higher nonattainment classification than the area in which
the source is located; and
¦ Emissions from such other area contribute to a violation of the National Ambient Air
Quality Standard in the nonattainment area in which the source is located.
¦ With respect to offsets obtained from outside the non-attainment area, the division may
increase the ratio of the required offsets to new emissions the greater the distance such
offsets are from the new or modified source.
Precursors to ozone must be analyzed and discussed. Please use EPA guidance regarding how to perform
an analysis for precursors to ozone.
If modeling is not submitted with the permit application, the Division will decide if modeling is warranted
to complete the impact analysis and compliance demonstration required by Regulation No. 3. If modeling
is warranted, the Division may perform a screening-level analysis if it is technically feasible to perform
one. If the screening-level analysis shows there could be modeled violations of applicable
Page 14 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
standards, the Division will contact the applicant to discuss options. Since the Division does not usually
perform refined-level modeling as part of the permitting process, the Division will typically request that
the applicant perform any refined modeling that might be warranted. If modeling is warranted, refer to
sections 4, 5, and 6.
In the event that compliance with standards cannot be demonstrated using typical attainment area
modeling procedures, a case-by-case approach should be developed in consultation with Division staff
familiar with the affected nonattainment area.
Requirements Unique to Colorado
The following are additional modeling-related regulatory requirements unique to Colorado:
¦ A major source by itself may not consume more than 75% of any applicable PSD increment
¦ Class I S02 increments apply to some pristine Class II areas
¦ For major sources subject to PSD review, water is included as one of the required elements in
the additional impact analysis; the requirement is intended to provide information on acid
deposition in high altitude lakes
¦ Sulfur dioxide (S02) 3-hour standard of 700 |ag/m3
Interim Modeling Thresholds
Table 1 sets forth interim modeling thresholds, both short-term (e.g. Ib/hr) and long-term (e.g. tpy) (for
NOx, S02, or PM2.5), that are intended to assist the Division staff, permit applicants, air quality
consultants, and others in deciding when modeling is warranted and to determine the impact from a
source. This section introduces de minimis emissions, which have low probability of causing or
contributing to an exceedance of an air quality standard. If a source or project has emissions below these
thresholds, the Division considers that the sources would have relatively small impacts and do not
warrant further analysis with respect to applicable air quality standards. By using this approach,
permitting costs associated with the impact analysis required by Regulation No. 3 can be minimized.
For a given pollutant, modeling must be performed if the requested emission rate is above the listed
thresholds (for NOx, S02, and PM2.5, if above the long-term thresholds). If the requested emission rate
and/or the facility-wide net emissions increase is below the long-term threshold for NOx, S02, and PM2.5,
or below the listed threshold for the other pollutants, modeling may not be warranted; however, the
Division may determine that modeling is required based on the factors set forth in the "Notes" section of
Table 1. In all cases, if there is doubt regardingthe need for modeling, the applicant should consult the
Division before submitting the application.
For sources with a requested emission rate between the short-term thresholds and the long-term
thresholds for NOx, S02, and PM2.5, or below the listed threshold for the other pollutants, the source must
select one of three available options to demonstrate that the proposed source will not interfere with the
NAAQS:
Option 1:
The source may conduct modeling in accordance with this Guideline.
Page 15 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
Option 2:
The source may consult with permitting staff and the MEIU and obtain a written determination from the
MEIU that modeling is not required based on the factors set forth in the "Notes" section of Table 1.
Option 3:
The source may propose a NAAQS monitoring plan to the Division's monitoring unit that ensures non-
interference with the NAAQS. Sources should consult with Division staff on the contents of a NAAQS
monitoring plan prior to submittal. As of the date of release of this Interim Guideline, Nancy Chick is the
appropriate point of contact, nancy.chick(5)state.co.us. The Division may update this Guideline with more
specific information.
The thresholds in Table 1 do not address compliance with minor modifications to major sources located
within 10km of a Class I area. Thus, modeling decisions related to Regulation No. 3, Part D, §II.A.44.c are
made on a case-by-case basis. According to §II.A.44.c, any net emissions increase of a regulated pollutant
at a major stationary source located within 10 kilometers (6.2 miles) of a federal Class I area should
perform modeling to determine if the maximum 24-hour average impact in the Class I area exceeds 1.0
|ag/m3 on a 24-hour basis. If it does, the emissions increase is significant and the modification constitutes
a major modification subject to PSD review.
The Class I significance level of 1.0 |ag/m3 on a 24-hour basis is only intended to determine if a
modification is major. It should not be used to determine if the impact in a Class I area is significant.
Table 1. Interim Modeling Thresholds
Pollutant
Requested Emission Rate from New Source or Facility-Wide
Net Emissions Increase from a Modification
Long Term (tons per year)
Short Term (pounds per hour)
Carbon Monoxide (CO)
23 pounds per hour
Nitrogen Oxides (NOx)
40
0.46
Sulfur Dioxide (S02)
40
0.46
Particulate Matter < 10 |am
(PMio)
82 pounds per day
Particulate Matter < 2.5 |am
(PM2.5)
5
11 pounds per day
Lead(Pb)
25 pounds per 3-months
Page 16 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
1Circumstances where modeling may be required despite being below the long-term thresholds:
(a) Sources of SO 2, NO2, PM10, PM2.5, CO, orPb where a substantial portion of the new or
modified emissions have poor dispersion characteristics (e.g., rain caps, horizontal stacks,
fugitive releases, or building downwash) in close proximity to ambient air at the site
boundary
(b) Sources located in complex terrain (e.g., terrain above stack heights in close proximity to
the source)
(c) Sources located in areas with poor existing air quality (i.e. sources located in areas with
monitored values ofS02, N02, PM10, PM2.5, CO, or Pb close to the NAAQS or CAAQS)
(d) Modification at existing major sources, including grandfathered sources that have never
been modeled before
The Division is working to further develop these factors to offer more clarity to the regulated
community about when these factors might result in modeling of an application. Until that work
is completed, applicants with questions should consult the Division prior to submitting an
An applicant must demonstrate that the proposed source or modification, as represented in the air
permit application, would not cause or contribute to a National Ambient Air Quality Standard (NAAQS) or
Prevention of Significant Deterioration (PSD) increment violation. When modeling is required, an air
quality impact analysis is the means for the applicant to make the demonstration. It is an evaluation of
the potential impact on the environment associated with a new and/or modified facility. Additional
analyses required by federal rule would also be included in the air quality impact analysis.
The air quality impact analysis is a stand-alone report. Results from the report should be sufficient for
Division staff to evaluate the impact of the proposed operation without input from other reports. Division
staff should not refer to other documents or reports for data required to be in the report. In addition,
applicants should not exclude items normally required without coordination with the Division's Technical
Services Program (TSP) modeling staff unless the items are clearly not applicable to the project. Refer to
the Colorado Modeling Submittal Completeness Checklist
(https://www.colorado.gov/airquality/permits/CompletenessChecklist-ModelingSubmittall4Feb.pdf) to
determine what information needs to be submitted in the air quality impact analysis.
Air Dispersion Modeling
Air dispersion models are tools to approximate concentrations from one or more facilities or sources of
air contaminants. When an air contaminant is emitted into the atmosphere, it is transported and
dispersed by various atmospheric processes. Algorithms and equations have been developed to
approximate (model) these atmospheric processes and have been incorporated into various computer
codes (computer models). Division staff uses the results from these computer models in their review of
air permit applications. A modeled prediction is used to evaluate whether the new or modifying source
will show compliance with the NAAQS and CAAQS. If the model predicts an exceedance of the NAAQS
and/or CAAQS, that does not automatically mean the application "fails" the air quality impact analysis. For
application.
Section 3 - Air Quality Impact Analysis
Page 17 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
example, the applicant is given the opportunity to adjust the facility allowable emissions, operating hours,
source parameters, and source configuration in order to demonstrate the predicted impact will be in
compliance with all state and federal standards.
Procedures and models other than those recommended by U.S. EPA or in this guideline may be approved
on a case-by-case basis if there is sufficient technical justification. U.S. EPA approval may be required in
some cases. Refer to U.S. EPA guidance for use of alternative models.
If a non-EPA-approved model, such as a wind tunnel study, is proposed, the nature and requirements of
such a model should be outlined to the Division at a pre-application meeting. The permit application will
be deemed incomplete until there has been a public hearing on the proposed model and written approval
of the U.S. EPA has been received (Regulation No. 3 Part A §VII).
The most recent version of U.S. EPA-approved models should be used. Division approval should be
obtained if an older version is used.
For Class I area modeling, the Division generally supports the use of models and modeling techniques
recommended by the Interagency Workgroup on Air Quality Modeling (IWAQM)3. Written IWAQM
guidance does not always reflect their latest recommendations. In addition, recommendations for the
Class I analysis may vary from one area to another. Thus, work with Division staff and affected federal
land managers (FLMs) on a case-by-case basis to determine the appropriate methods to address impacts
at each affected Class I area.
Air Quality Impact Analysis Process
Division staff with the appropriate expertise reviews various aspects of the impact analysis. For example,
different specialists may provide comments on dispersion modeling, monitoring data, visibility modeling,
and air quality related values. Internal comments by reviewers are typically sent directly to the permit
engineer in the Stationary Sources Program who interprets the comments and, if necessary, brings staff
together to discuss or resolve issues.
Modeling submittals that accompany permit applications should generally be sent to the Division's
Stationary Sources Program where a permit engineer processes the permit application. The permit
engineer forwards modeling reports, date, modeling input/output files, the permit application, and other
relevant information to appropriate staff. As required by regulations, copies of the permit applications for
major stationary sources are sent to federal agencies such as U.S. EPA Region VIII and affected federal
land managers.
It is appropriate for applicants or their modelers to send modeling protocols directly to modeling staff in
the Division's Technical Services Program. A copy should also be sent to the Stationary Sources Program
permit review staff since they are responsible for the overall review of the permit.
3 IWAQM was formed to provide a focus for development of technically sound, regional air quality models
for regulatory assessments of pollutant source impacts on federal Class I areas. The guidance included
input from the U.S. Forest Service, National Park Service, U.S. Fish and Wildlife Service, the U.S. EPA, and
several states.
Page 18 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
The Division encourages phone conversations, e-mail, and other types of communication between staff
and the applicant's modeler(s) and other specialists to resolve issues once the actual review process
begins. It is assumed the applicant's modeler or other specialists will notify the applicant of important
modeling-related issues as necessary. It is recommended that significant issues and information transfers
be done in writing. Copies of any letters or e-mail messages shall be sent to the permit engineer.
When oversights, errors, or questionable assumptions and/or methods are found during the review
process, Division staff will use professional judgment to decide if deficiencies are sufficient to change the
outcome of the compliance demonstration. If the ramifications of a modeling-related issue are not
significant, the deficiencies are noted in the comments and appropriate language is included to justify
that a specific issue is minor. If it is difficult to assess the ramifications without redoing the analysis, the
Division may attempt to redo the analysis, while the deficiencies will be noted in the review comments
and the applicant will be asked to address the comments.
Any responses to comments may be sent directly to Technical Services Program modeling staff, but it is
recommended that a copy also be sent to the permit engineer. In cases where there are no modeling
issues, the Division's modeling comments are not usually forwarded to the applicant. Instead, the written
comments are added to the permit file.
Figure 1 graphically depicts the permit review process as it relates to air quality modeling. While the
flowchart is applicable to all permit applications for major stationary sources where modeling is required,
only certain portions of the flowchart are applicable for minor sources. For example, the loop involving
U.S. EPA Region VIII and the federal land manager (FLM) is not an integral part of the review process for
minor sources.
Figure 2. Under revision but will graphically shows the roles and responsibilities for the modeling review
process within the Division for air quality construction permits.
Figure 3. Under revision but will illustrate key aspects of the regulatory decision process for major
stationary sources and major modifications seeking construction permits. This figure is currently under
review within the Division. Please contact Division staff to confirm the review process for an AQRV and
Visibility analysis in Class I areas.
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
Figure 1. Permit Review Process
Page 20 of 64
CDPHE
APCD
Duties
Send copy of
application to
EPA and/or
FLMs, if
necessary
Determine if
application is
complete
Complete
Per mili nary
analysis
performed by
SSP&MEIU
Request
additional
information
from applicant
Determine if
Public Notice
is required
Public
Notice
Required
APCD or
AQCC
grants or
denies
permit
Public Notice
Process
-------
October 2021 Interim Colorado Modeling Guideline for Air Quality Permits
Figure 2. Roles & Responsibilities within CDPHE
Under Review
Page 21 of 64
-------
October 2021 Interim Colorado Modeling Guideline for Air Quality Permits
Figure 3. Regulatory Decision Process for AQRVs
Under Review
Page 22 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
Section 4 - Performing the Air Quality Impact Analysis
Figure 4 on page 35, below, depicts the air quality impact analysis.
Modeling Protocol
The protocol is the primary mechanism by which all affected parties such as the applicant, the Division,
U.S. EPA, and federal land managers reach agreement on a modeling approach. The protocol
development process is intended to minimize the chances of misunderstandings and to avoid delays in
the permit process. It explains in detail how a modeling analysis will be performed, how the results will be
presented, and how compliance with applicable requirements will be demonstrated. The protocol is not
intended to be a binding, legal document as changes or deviations are often necessary as the data
collection and analysis progresses.
Submission of a modeling protocol is strongly recommended for all air quality impact analyses.
Screening Modeling
The U.S. EPA developed screening-level modeling techniques to determine quickly whether a facility
should perform in-depth refined modeling analyses. Screening-level models produce estimates of worst-
case impacts from a single source without the need for hourly metrological data. Most applicants are
recommended to perform a screening-level analysis to show the facility is in compliance with the
applicable NAAQS and CAAQS. If there is doubt regarding the need for modeling, the applicant is
recommended to perform a screening-level analysis. If the screening-level analysis does not show
compliance with the NAAQS and CAAQS, then refined modeling is required.
The U.S. EPA has regulatory screening models that should be used for this analysis. These models can be
found on the U.S. EPA Support Center for Regulatory Atmospheric Modeling (SCRAM) website.
Screening models are designed to evaluate a single source. Most facilities, however, do not consist of a
single source, but screening models can still be used by summing the emissions from all sources at the
facility and model them as if they are being emitted from a single source. This method is only acceptable
when all sources are stacks and being emitted from the shortest stack, to represent worst-case.
When facilities consist mostly of fugitive emissions, screening models are not acceptable. When summing
these types of emissions together and modeling as a single source, the accuracy of the model is reduced
substantially and the results not credible. Therefore, the Division does not accept screening models from
the following source categories:
c) Gravel Pits
Quarries
Landfills
Page 23 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
Mining Operations
Any type of facility not mentioned above that involves multiple sources of fugitive emissions
Refined Modeling
Refined modeling requires detailed and precise input data along with more complex models in order to
provide refined impact estimates. If refined modeling is warranted, it should be performed in two distinct
phases.
The first phase is the significant impact analysis (SIA), which determines if the applicant can forego further
air quality analysis for a particular pollutant with respect to Colorado and National Ambient Air Quality
Standards and, for new major sources and major modifications, Prevention of Significant Deterioration
increments. The second phase is the cumulative impact analysis for the CAAQS, NAAQS, or applicable PSD
increments; it is sometimes referred to as the full impact analysis4.
Significant Impact Analysis
Individual facilities may be subject to different requirements depending on the proposed emission rates
of each facility. There are two general categories of permits: major NSR and minor NSR. The major NSR
permit for sources in an Attainment Area is often referred to as a PSD permit.
The air quality impact analyses for major NSR and minor NSR permits begin with a significant impact
analysis (SIA). The purpose of a SIA is to determine whether a new and/or modified facility, or a
combination of the two, could cause a significant ambient air impact. Below are general steps for
identifying emissions to include in the SIA.
SIA Step 1: Identify All Sources of Emissions. Include emissions from all new and/or modified sources at
the facility associated with the project.
SIA Step 2: Determine Whether There Is a Net Emissions Increase. Determination of the project
emissions may vary depending on the type of permit (minor NSR or major NSR). The determination of
the level of federal applicability is the first step in the technical review process and is performed by
the permit engineer. The federal applicability process determines whether a project is minor or
major. While the steps of the modeling process are consistent, requirements vary based on the type
of permit and pollutant.
SIA Step 3: Evaluate Modifications to Existing Sources at the Site. Carry out this step even ifthere is no
net increase in emissions. For both minor and major NSR modeling, include these sources in the SIA if
there is a change in operating hours or stack parameters, and previous modeling demonstrations
were limited to those operating hours or stack parameters. That is, the permit was based on those
limits.
SIA Step 4: Develop the Emission Inventory for the Site. In general, the requested allowable emission
rate, requested operating rate or maximum design rate should be modeled; however, the applicant
should consult with the permit engineer to verify that the appropriate emission rates were
4 U.S. EPA sometimes uses the phrase "full impact analysis" to refer to the National Ambient Air Quality
Standards (NAAQS) analysis and the Prevention of Significant Deterioration (PSD) increment analysis.
Page 24 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
developed. If the requested emission or operating rate used in the modeling is less than the
maximum design rate, it may become a permit condition. For modifications, the facility-wide net
emissions increase for the modification should be modeled in the SIA.
Major stationary sources do not need to include emissions from the commercial, residential, and
industrial growth analysis in the SIA. The growth analysis required by the PSD rules is only
recommended if a CAAQS and NAAQS analysis, a PSD increment analysis, or a similar air quality
impact analysis is triggered.
Carry out the SIA modeling.
For a given pollutant and averaging period, the highest estimated concentration at each receptor in
ambient air is compared to the modeling significance levels in Table 2 and Table 3. Impacts from nearby
and other background sources, including background concentrations, are not considered in the SIA. If the
estimated concentration levels are below the applicable modeling significance level, no further analysis is
recommended. The source is considered to have an insignificant impact. For example, if impacts are
below the significance levels in Table 3, a compliance demonstration for Colorado and National Ambient
Air Quality Standards (CAAQS and NAAQS analysis) is not triggered. For major stationary sources subject
to PSD rules, a Class I or Class II PSD increment analysis is not triggered if the impacts are below the
significance levels in Table 2; however, other analysis requirements of the PSD rules must nevertheless be
addressed. If the impact exceeds the modeling significance levels, the source or modification has a
significant impact in ambient air and the next phase of analysis is triggered, as discussed below.
The SIA also provides a convenient way to define the "probable area of influence" of a source's emissions
(see Regulation No. 3, Part B, §111.B.5.d). In practice, it is sometimes useful to define the significant impact
radius or area for the source or activity of interest.
If modeling shows that no violation of a standard (or, for major stationary sources, an applicable PSD
increment) will occur within the significant impact area of a proposed source, as determined by a
comparison with the applicable modeling significance levels, no cumulative air quality impact analysis is
warranted.
Significant Impact Level (SIL)
PSD increment modeling significance levels (Table 2) are only used for major stationary sources subject to
PSD rules. The Class I PSD increment significance levels are based on U.S. EPA proposals from 1996.5 For
minor sources and minor modifications, the Division does not consider compliance with PSD increments
as a criterion in determining if a permit should be issued for a minor source or minor modification.
The modeling significance levels in Table 2 are only intended for the PSD increment analysis. Table 2 does
not include values for Class III areas as there are no Class III areas in Colorado. The modeling significance
levels were not developed to determine if there would be significant impacts to air quality related values
(AQRVs).
5 Federal Register: July 23, 1996 (Volume 61, Number 142), Proposed Rules, Page 38249-38344.
Page 25 of 64
-------
October 2021 Interim Colorado Modeling Guideline for Air Quality Permits
Page 26 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
Table 2. Significant Levels for PSD Increments (ng/m3)
Pollutant
Class 1
Class II
Carbon Monoxide (CO)
8 hour
(a)
500
1 hour
(a)
2000
Nitrogen Dioxide (NO2)
Annual
0.1
1
1 hour
(a)
7.5
Sulfur Dioxide (SO2)
Annual
0.1
1
24 hour
0.2
5
3 hour
1.0
25
1 hour
(a)
4
Particulate Matter < 10 |im (PM10)
Annual
0.2
1
24 hour
0.3
5
Particulate Matter < 2.5 |im (PM
2.5)
Annual
0.05
0.2
24 hour
0.27
1.2
(a) Modeling significant level has not been defined
For minor and major stationary sources, the modeling significance levels in Table 3 are used to determine
if a CAAQS and NAAQS analysis is triggered (see Figure 4). The significance levels in Table 3 are listed in
Regulation No. 3, Part D, §VI.D.2.
Page 27 of 64
-------
October 2021 Interim Colorado Modeling Guideline for Air Quality Permits
Table 3. Significant Levels for NAAQS & CAAQS (|ig/m3)
Pollutant
SIL
Carbon Monoxide (CO)
8 hour
500
1 hour
2000
Nitrogen Dioxide (NO2)
Annual
1
1 hour
7.5
Sulfur Dioxide (SO2)
Annual
1
24 hour
5
3 hour
25
1 hour
4a
Particulate Matter < 10 |im (PM10)
Annual
1
24 hour
5
Particulate Matter < 2.5 |im (PM2.5)
Annual
0.2
24 hour
1.2
interim modeling significance level developed by the Division:
https://www.colorado.Rov/airqualitv/permits/lnteriml-
hrS02SIL.pdf
Cumulative Impact Analysis
The components of the cumulative impact analysis vary depending on the applicable regulatory
requirements. For minor sources and minor modifications, a compliance demonstration with Colorado
Ambient Air Quality Standards (CAAQS) and National Ambient Air Quality Standards (NAAQS) is usually the
only type of impact analysis that is requested. Refer to Table 1 to determine if modeling is warranted.
Table 4 summarizes the typical types of air quality analysis for new minor sources or minor modifications
that might be applicable. In attainment areas, all new sources and modifications with a significant impact
in ambient air should perform a cumulative CAAQS and NAAQS analysis. For nonattainment area
requirements, please refer to the Nonattainment Areas portion of Section 2.
Impact analysis requirements are stated in applicable regulations. Regulation No. 3, Part B, §111.D presents
the general requirements for all construction permit applications, including minor sources.
Page 28 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
For minor sources and minor modifications, a compliance demonstration with the Prevention of
Significant Deterioration (PSD) increments is not required to obtain a construction permit. A preliminary
opinion in June 1998 from the Colorado Attorney General's office suggests that rulemaking would be
necessary before compliance with PSD increments could be a permit issuance criterion for minor sources
and minor modifications. Therefore, increment consumption from minor source growth is assessed only
during the modeling process for new sources and modifications subject to PSD rules and during periodic
increment studies. Nevertheless, since all sources, including minor sources, can consume PSD increment
in areas where the PSD minor source baseline date has been triggered, new minor sources and minor
modifications are encouraged to voluntarily demonstrate compliance with applicable increments.
Table 4. Ambient Air Impact Analyses Applicable for New
Minor Source & Minor Modifications
Area Classification
Ambient Air Impact
Analysis
Attainment, Unclassifiable
NAAQS & CAAQS
NAAQS & CAAQS
Nonattainment
Reasonable Further
Progress (RFP)
The components of the major stationary source or major modification air quality impact analysis vary
depending on the applicable regulatory requirements. Permit applicants are encouraged to contact the
Division as early as possible to discuss permitting requirements. The Division and U.S. EPA encourages
applicants to submit modeling protocols.
All areas of Colorado are classified as Class II with the exception of the twelve federal Class I areas, which
are shown in Figures 6 and 7. Class I areas have the greatest protection from air quality deterioration;
Class III areas have the least protection; however, there are no Class III areas in Colorado. In addition to
demonstrating compliance with ambient air quality standards, major stationary source permit applicants
must demonstrate that they will not cause or contribute to violations of PSD increments. Major stationary
sources located within nonattainment areas are subject to additional requirements as discussed in the
Nonattainment Areas portion of Section 2.
Table 5 summarizes the typical types of air quality analysis for new major sources or major modifications
that might be applicable. The significant impact analysis must be performed if there is a possibility the
proposed source will impact a nonattainment area.
Table 5. Ambient Air Impact Analyses Applicable for New
Major Source & Major Modifications
Page 29 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
Area Classification
Ambient Air Impact Analysis
Attainment, Unclassifiable
NAAQS & CAAQS
PSD Increment
Additional Impacts Analysis in any area
(Visibility, Water, Soils, Vegetation, Growth)
AQRV Analysis in Class I Areas
Pre-and Post-Construction Monitoring
Nonattainment
NAAQS & CAAQS
Reasonable Further Progress (RFP)
Net Air Quality Benefit
AQRV Analysis in Class I Areas
Major stationary sources are required by regulation to submit an additional impacts analysis to address
potential impairment to soils, vegetation, water, visibility, and growth, if applicable; it applies in all areas,
including Class I and Class II areas. In addition, regulations require that applicants submit an analysis of
impairment to Air Quality Related Values (AQRVs) in affected Class I areas.
PSD applicants should also consult with the Division to determine if there will be any pre-construction
ambient monitoring requirements. Refer to Regulation No. 3, Part D, §VI to understand how the Division
decides if pre- or post-construction monitoring is required.
There are other regulatory requirements in addition to those required by PSD rules. For example,
Regulation No. 3, Part B, §111.D.l subparts a through g list general requirements for obtaining a permit.
While subpart e applies to major PSD sources, subparts c and d provide requirements that are more
general. Thus, the PSD modeling requirements of subpart e are only one of many requirements that may
be applicable.
Regulation No. 3, Part D, §VI.B states, "the [PSD] requirements of section VI.A do not apply to a major
stationary source or major modification with respect to a particular pollutant if the owner or operator
demonstrates that...the emissions from the source or modification would not be significant." Thus, the
impact analysis and monitoring requirements of the PSD rules are not applicable for a given pollutant if
the emission rate is not significant, as defined in Regulation No. 3, Part D, § 11 .A.44. In situations where the
requirements of §VI are waived, modeling for compliance with ambient air standards may nevertheless
be warranted under the requirements of Regulation No.,3, Part B, §111.
NAAQS & CAAQS Analysis
The federal Clean Air Act established two types of national air quality standards. Primary standards set
limits to protect public health, including the health of sensitive populations such as people with asthma,
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
children, and the elderly. Secondary standards set limits to protect public welfare, including protection
against decreased visibility, damage to animals, crops, vegetation, and buildings. Colorado and National
Ambient Air Standards (CAAQS and NAAQS) are listed in Tables 6 and 7, respectively. Units of measure for
the standards are parts per million (ppm) by volume, parts per billion by volume (ppb), and micrograms
per cubic meter of air (|ag/m3).
The ambient air quality standards in Tables 6 and 7 are based on a reference temperature and pressure of
25 degrees Celsius and 760 millimeters of mercury (1,013.2 millibars or 1 atmosphere), respectively.
Correction of modeled concentration estimates to reference conditions (i.e., standard temperature and
pressure, STP) before comparison with ambient air quality standards is not required for air quality permit
modeling in Colorado. If it is necessary to perform unit conversions, the following formula may be used:
; MW = Molecular weight of pollutant in—and X is concentration6
ppm (40.9 x MW) mole
If the impact is significant and a CAAQS and NAAQS modeling analysis is warranted, the modeling should
account for the source under review plus existing air pollution levels at the locations (receptors) where
the source has a significant impact. The purpose of the NAAQS/CAAQS analysis is to demonstrate that
proposed emissions of criteria pollutants from a new facility or from a modification of an existing facility
that does not trigger PSD increment review will not cause or contribute to an exceedance of the NAAQS
and CAAQS.
This can be done in several ways. In general, the compliance demonstration for standards should include:
NAAQS/CAAQS Step 1: Conduct a SIA. Perform a significant impact analysis to predict whether the
proposed source(s) could make a significant impact on existing air quality. That is, the model predicts
concentrations at one or more receptors in the modeling grid greater than or equal to a significant
impact level (SIL).
¦ Model all new and/or modified sources. Compare the predicted high concentration at or
beyond the property line for each criteria pollutant and each averaging time to the
appropriate SIL.
¦ If the sources do not make a significant impact for a pollutant of concern, the demonstration
is complete. If there is a significant impact, then the significant receptors define a significant
impact area and a full NAAQS analysis is required. Go to Step 2.
NAAQS/CAAQS Step 2: Determine Significant Impact Area. Each criteria pollutant and averaging period
subject to the NAAQS/CAAQS analysis may have a different significant impact area.
¦ The significant impact area is the set of receptors that have predicted concentrations at or
greater than the SIL for each applicable averaging time and criteria pollutant.
\L_
-i
fi1 -pollutant 0.0409 moles ¦ L xlOOOL • m - xMJTxlO /Jg • g~ ,
6 1 ppmv = , g T— = 777-7 — = {40.9xMW)jug -m '
10 "Z,
10" Z,
where = %-p = 0.0409 moles • L 1, where P =1 atm. T =298 K, R =0.08206 L-atm-K^mole1, L = liters
Page 31 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
¦ The full NAAQS analysis is carried out for each criteria pollutant and averaging time
separately and need only include the significant impact area for the associated criteria
pollutant and averaging time combination.
¦ Refinement of the significant impact area may be necessary as is discussed in Section 5,
Receptor Network.
NAAQS/CAAQS Step 3: Evaluate Nearby Sources. The applicant needs to request a nearby source
inventory from the Division. It is the responsibility of the applicant to obtain the data and ensure the
accuracy. Any changed made to the data must be documented and justified.
The nearby source inventory for major source and major modifications (e.g., sources subject to PSD
rules) should expand to 50km of the significant impact area of the new source or modification under
review. Identify nearby sources to explicitly model. Select additional background sources as
appropriate to account for impacts not reflected in the background concentration. Sources beyond
50 kilometers should be considered if long-range transport modeling is being performed for a federal
Class I area. Estimated impacts from growth in residential, commercial, and industrial sources
associated with, but not part of, the proposed source should be included in the analysis for major
sources and major modifications.
NAAQS/CAAQS Step 4: Conduct a CIA. Perform a cumulative impact analysis. Model all facility sources
with the nearby sources obtained from the Division. Model allowable emission rates for all sources
that emit the criteria pollutant.
NAAQS/CAAQS Step 5: Add Background concentration to CIA modeled result. The applicant needs to
request a representative background concentration from the Division. This background concentration
should be added to the modeled result from the CIA.
NAAQS/CAAQS Step 6: Compare to NAAQS/CAAQS. Compare the modeled CIA concentration plus
representative background concentration for each criteria pollutant and averaging time to the
appropriate NAAQS. Use the correct design value that follows the form of the applicable NAAQS or
the highest first high depending on the meteorological determination (discussed in Section 5, Criteria
Pollutants Recommendations).
If the maximum concentrations are at or below the NAAQS/CAAQS, the demonstration is complete. If
the concentration is above the NAAQS/CAAQS, perform a contribution analysis to demonstrate that
the proposed source will not exceed the applicable significant impact levels in Table 3 at the point
(receptor) and time of the modeled violation. If the proposed source will not exceed the applicable
SIL at the point and time of the modeled violation, the demonstration is complete. No further air
quality impact analysis is warranted for the new source or modification, even when a new violation
would result from its insignificant impact. If the proposed source has a significant impact at the point
and time of the modeled violations, review the demonstration and determine if any refinements can
be made or demonstrate that the project's impact will not be significant. The following options can
be considered to further refine the model to show compliance with the NAAQS/CAAQS:
¦ Emission Limits;
¦ Operating schedule restrictions;
Page 32 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
¦ Physical changes at the facility to improve dispersion characteristics;
¦ The use offences or physical barriers to preclude public access from contiguous land owned
or controlled by the operator (i.e., standards and increments only apply in ambient air);
¦ Additional pollution control equipment;
¦ The use of more refined modeling techniques, including nonguideline models (e.g., non-EPA
dispersion models, physical models, and monitoring-based methods)
The Common Provisions Regulation, § 11 .A states that if emissions generated from sources in Colorado
cross the state line, such emissions shall not cause the air quality standards of the receiving state to be
exceeded, provided reciprocal action is taken by the receiving state. The Division is not aware of any
formal written agreements regarding reciprocal action. Nevertheless, if the impact from a new or
modified source will have a significant impact in another state as defined in section 5, or if it will likely
affect another state, the Division recommends contacting the appropriate agency in the affected state to
determine if there are any applicable state standards. If so, consult with the Division to determine what if
any analysis is recommended.
The Division may recommend that additional analysis be performed to show compliance with applicable
standards of that state. If modeling appears to be warranted, staff from the Division and the affected
state should discuss the situation to determine an acceptable modeling approach.
Table 6. Colorado Ambient Air Quality Standards (CAAQS)
Pollutant & Averaging
Period
Level
Form
Sulfur Dioxide (SO2)
3 hour
700 |ag/m3 Not to be exceeded more than once per year
Page 33 of 64
-------
October 2021 Interim Colorado Modeling Guideline for Air Quality Permits
Table 7. National Ambient Air Quality Standards (NAAQS)
Pollutant & „ . .
„ . . Primary/Secondary
Averaging Period
Level
Form
Carbon Monoxide (CO)
8 hour
Primary
9 ppm
Not to be exceeded more than once per
year
1 hour
Primary
35 ppm
Not to be exceeded more than once per
year
Lead (Pb)
Rolling 3 month
Primary &
Secondary
0.15 |ag/m3
Not to be exceeded
Nitrogen Dioxide (NO2)
Annual
Primary &
Secondary
53 ppb
Annual Mean
1 hour
Primary
100 ppb
98th percentile of 1-hour daily maximum
concentrations, average over 3 years
Ozone (O3)
8 hour
Primary &
Secondary
0.070 ppm
Annual fourth-highest daily maximum 8-
hour concentration, averaged over 3
years
Sulfur Dioxide (SO2)
Annual1
Primary
0.03 ppm
Annual mean
24 hour1
Primary
0.14 ppm
Not to be exceeded more than once per
year
3 hour
Secondary
0.5 ppm
Not to be exceeded more than once per
year
1 hour
Primary
75 ppb
99th percentile of 1-hour daily maximum
concentrations, averaged over 3 years
Particulate Matter < 10 pim (PM10)
Annual
Revoked in 2006
24 hour
Primary &
Secondary
150 |ag/m3
Not to be exceeded more than once per
year on average over 3 years
Particulate Matter < 2.5 pim (PM2.5)
Annual
Primary
12.0 |ag/m3
Annual mean, averaged over 3 years
Annual
Secondary
15.0 |ag/m3
Annual mean, averaged over 3 years
24 hour
Primary &
Secondary
35 |ag/m3
98th percentile, averaged over 3 years
1_rhe 24-hour and Annual S02 standards were revoked in 2010; however, they remain in effect in
Colorado until December 21, 2018. Please contact the Division if you have questions regarding S02.
Page 34 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
Figure 4. Flowchart of the Air Quality Impact Analysis
Process for CAAQS arid NAAQS
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
PSD Increment Analysis
The air quality analysis for new/modified sources subject to Prevention of Significant Deterioration (PSD)
rules must demonstrate compliance with PSD increments if the impact from the new source or
modification is significant. This section is not intended to provide a complete overview of PSD increment
consumption; for that, refer to EPA guidance documents.
The purpose of the PSD increment analysis is to demonstrate that emissions of applicable criteria
pollutants from a new major source or major modification of an existing source will not cause or
contribute to an exceedance of an increment. The PSD increment is the maximum allowable increase in
concentration that is allowed to occur above a baseline concentration for a pollutant. The following
discussion explains PSD increment analyses followed by the basic procedure for conducting the analyses.
Refer to Section 5, Criteria Pollutants Recommendations, for more information about the design value
that should be used to determine compliance with applicable PSD increments. Increment consumption is
a receptor-by-receptor concept. That is, the consumption of PSD increment by one particular source does
not necessarily preclude similar increment consumption by another nearby source if the consumption
occurs on a different day (i.e., under different meteorological conditions) and/or at a different location
(e.g., receptor).
All changes in emissions and related parameters7 after the minor source baseline date may affect PSD
increment consumption or expansion. This includes both stationary sources and mobile sources. In
addition, modifications at major stationary sources after the major source baseline date also may affect
increment consumption. Refer to U.S. EPA guidance and Division guidance8 for procedures.
Area and mobile sources may be important increment consuming sources. In most situations, the Division
can provide at least a county-level inventory of increment consuming area and mobile emissions;
however, because of the amount of time required by the Division to develop such inventories, the
Division will typically not develop increment inventories for an individual permit application until the
permit applicant and the Division agree that an area and mobile source inventory is actually warranted. If
the Division does not have the resources necessary to develop the inventory in the time frame needed by
the applicant, the burden of doing the area and mobile analysis may fall on the applicant.
All areas of Colorado are Class II areas except for the Class I areas shown in Figures 6 and 7. PSD baseline
areas for PMio are based on the Colorado Air Quality Control Regions (AQCRs) shown in Figure 5. It is
worth noting that there are both Colorado AQCRs (planning areas) and federal AQCRs. They are
comprised of different counties. While the Colorado AQCRs are used as PSD baseline areas for PMio, the
federal AQCRs are used in U.S. EPA's Air Quality System (AQS). The entire state serves as the baseline
7 "The creditable increase of an existing stack height or the application of any other creditable dispersion
technique may effect 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." (USEPA, 1990)
8 Refer to the Technical Guidance Series: PSD Increment Tracking System document for a detailed
discussion about the PSD increment tracking in Colorado.
Page 36 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
area for S02 and N02. Figure 5 and Table 9 show the minor source baseline areas and trigger dates in
Colorado.
Increment Calculation
The baseline concentration does not need to be obtained to determine the amount of PSD increment
consumed or the amount of increment available. Instead, the amount of PSD increment that has been
consumed in an attainment or unclassified area is determined from the emissions increases and
decreases that have occurred from stationary sources in operation since the applicable minor source
baseline date. Modeled increment consumption calculations reflect the change in ambient pollutant
concentration attributable to increment-affecting emissions. Increment consumption (or expansion)
calculations are determined by evaluating the difference between the actual emissions at the applicable
minor source baseline date (ActualBD) and actual emissions as of the date of the modeling demonstration
(ActualMD).
a. Actuakd- This is the representative 2-year average for long-term emission rates, or the maximum
short-term emission rate in the same 2-year period immediately before the applicable minor
source baseline date. For major sources permitted at or after the applicable major source
baseline date but not in operation as of the applicable minor source baseline date or for minor
sources not in operation as of the applicable minor source baseline date, ActualBD would be the
permit allowable emission rate.
b. ActualMD• This is the most recent, representative 2-year average for long-term emissions rates, or
the maximum short-term emission rate in the same 2-year period immediately before the
modeling demonstration. If little or no operating data are available, as in the case of permitted
sources not yet in operation at the time of the increment analysis, ActualMD would be the permit
allowable emission rate.
A tiered approach is suggested for this analysis to limit the amount of research needed to determine
actual emission rates. The applicant should follow the basic procedure described in the following
paragraphs.
PSD Increment Step 1: Determine whether the modeled high concentration (excluding background
concentration) obtained in the PSD cumulative NAAQS analysis is equal to or less than the applicable
increment. If yes, the demonstration is complete because all sources were modeled at allowable
emission rates. This does not apply for criteria pollutants with NAAQS that are statistically-based (i.e.,
multi-year average).
PSD Increment Step 2: Determine the significant impact area for each criteria pollutant and averaging
period subject to the PSD increment analysis. The significant impact area will be the same one used in
the PSD NAAQS analysis, except for those criteria pollutants with NAAQS that are statistically-based.
For criteria pollutants with NAAQS that are statistically-based, determine the significant impact
analysis following the convention of exceedance-based NAAQS (i.e., maximum predicted
concentration).
PSD Increment Step 3: Obtain a listing of applicable increment-affecting sources and associated
parameters within 50km of the significant impact area from the Division to evaluate in the air quality
impact analysis. Sources beyond 50km should be considered if a long-range transport increment
analysis is being performed for a federal Class I area. It is the responsibility of the applicant to obtain
Page 37 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
these data and ensure their accuracy. Any changes made to the data must be documented and
justified.
PSD Increment Step 4: Adjust the emission inventory using professional judgment.
¦ Omit any source from the inventory that has a negative emission rate unless the source
existed and was in operation at the applicable minor source baseline date. A source must
have existed and been in operation on or before the applicable minor source baseline date to
be considered for increment expansion.
¦ Omit any source permitted after the applicable minor source baseline date that has shut
down or any source as part of the current project that will be shut down. A source that did
not exist or was not operating on or before the applicable minor source baseline date would
not have contributed to the air quality at that time, and there would be no need to model the
source with an emission rate of zero.
PSD Increment Step 5: Conduct the modeling demonstration using the same meteorological data set
used in the determination of the significant impact area using the following tiered approach, as
applicable.
Increment Modeling Tier I. Model all sources using their allowable emission rates. This approach
is conservative since the increment consumed is based on the entire allowable emission rate.
Compare the modeled high concentration to the appropriate increment. If the increment is not
exceeded, the demonstration is complete. Otherwise, go to Tier II.
Increment Modeling Tier II. Model selected sources with ActualMD emission rates and all other
sources at allowable emission rates. The selected sources are usually the applicant's sources. This
process assumes that the increment consumed for the selected sources is based on the entire
actual emission rate and the entire allowable emission rate for all other sources. If the increment
is not exceeded, the demonstration is complete. Otherwise, go to Tier III.
Increment Modeling Tier III. Model selected sources that existed and were in operation at the
applicable minor source baseline date with the difference between ActualMD and ActualBD-
• For major sources permitted at or after the applicable major source baseline date but not
in operation as of the applicable minor source baseline date or for minor sources not in
operation as of the applicable minor source baseline date, use the difference between
ActualMD and the allowable emission rate.
• For sources that existed at the applicable minor source baseline date, where a change in
actual emission rates involved a change in stack parameters, use the emission rates
associated with both the applicable minor source baseline date and the current and/or
proposed source configuration. That is, enter the ActualBD as negative numbers along
with the applicable minor source baseline source parameters, and enter ActualMD for the
same source as positive numbers along with the current and/or proposed source
parameters.
• Use emission rates found in Tiers I or II for other sources, as applicable.
Page 38 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
If the increment is not exceeded, the demonstration is complete. Otherwise, continue to refine increment
emission rates or demonstrate that the project's impact will not be significant.
Unique Colorado PSD Increment Requirement
As required by Regulation No. 3, Part D, §X.A.5.a, new sources and modifications subject to PSD rules
should demonstrate that the source by itself will not consume more than 75% of any applicable PSD
increment. The 75% rule does not apply to minor sources.
Along with the 75% increment consumption requirement, there are also Class II areas in Colorado that
have the same protections as Class I areas for S02. Refer to Regulation No. 3, Part D, §VI11.B for more
information. Modeling is recommended for S02 sources that could impact these areas, based on
boundaries that existed on August 7, 1977:
• Florissant Fossil Beds National Monument;
• Colorado National Monument;
• Dinosaur National Monument;
• Black Canyon of the Gunnison National Park (areas that are not already Class I);
• Great Sand Dunes National Park and Preserve (areas that are not already Class I);
• Uncompahgre Mountain Primitive Area;
• Wilson Mountain Primitive Area;
• BLM lands in the Gunnison Gorge Recreation Area.
Figure 6 depicts these Class II areas.
Page 39 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
Table 8. PSD Increments (pig/m3)
Pollutant
Class 1
Class II
Nitrogen Dioxide (NO2)
Annual
2.5
25
1 hour
(a)
(a)
Sulfur Dioxide (SO2)
Annual
2
20
24 hour
5
91
3 hour
25
512
1 hour
(a)
(a)
Particulate Matter < 10 |im (PM10)
Annual
4
17
24 hour
8
30
Particulate Matter < 2.5 |im (PM
2.5)
Annual
1
4
24 hour
2
9
(a) PSD increment level has not been defined
Class III increment values have been removed as there are no
Class III areas in Colorado
Page 40 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
Table 9. PSD Baseline Dates in Colorado
As of March 20181
Pollutant
Major Source
Baseline Date
Trigger
Date
Minor Source
Baseline Date
Air Quality
Control Region
(AQCR)
Triggering Application
Sulfur Dioxide
(S02)
01/06/1975
08/07/1977
10/12/1977
Entire State
Rio Blanco Oil Shale -
Tract C-a
Nitrogen Dioxide
(NOz)
02/08/1988
02/08/1988
03/30/1989
Entire State
Amoco Production -
Wattenberg
11/01/1988
AQCR 1
Colorado Power Partners
- Brush
01/17/1980
AQCR 2
Platte River Power
Authority - Rawhide
11/14/2000
AQCR 3
North American Power
Group - Kiowa Creek
11/22/1994
AQCR 4
Colorado Springs
Utilities - Nixon
11/09/2000
AQCR 5
Tri-State - Limon
Particulate
06/19/1989
AQCR 6
Cimarron Chemical -
Vilas
Matter < 10 |im
(PMio)
01/06/1975
08/07/1977
04/04/1995
AQCR 7
Westplains Energy -
Pueblo
Not Triggered
AQCR 8
NA
Not Triggered
AQCR 9
NA
08/20/1984
AQCR 10
Colorado Ute - Nucla
10/12/1977
AQCR 11
Rio Blanco Oil Shale -
Tract C-a
07/01/1983
AQCR 12
Louisiana Pacific -
Kremmling
Not Triggered
AQCR 13
NA
Particulate
Black Hills - Pueblo
Matter < 2.5 nm
10/20/2010
10/20/2011
09/10/2013
Airport Generating
(PM2.5)
Station Unit 6
1Contact the Division for the latest information
PM2.5 increment is currently being reviewed within the Division. This table will be updated once the Division
makes a final decision regarding the AQCRs for PM2.5- Please contact the Division for more information.
-------
October 2021
In terim Colorado Modeling Guideline for Air Quality Permits
Figure 5. Colorado PMio PSD Baseline Areas
-------
October 2021 Interim Colorado Modeling Guideline for Air Quality Permits
Figure 6. Class II Areas with "Class I Protection" for S02 Increment
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
Figure 7. Federal Class I Areas
105 WW
QiHette
rand Teton NP
W } [
Craters of the Moon NM
South Dakota
Iderness
Idaho
Lander
Wyoming
Rock Springs
Mount /irkeT
VMIderness
Craig \
Vernal
4D10'0"N«
Port Collins
.Greeley
Stetling
Rockyi
Metro Arei
rand Junction
inowniass Wilderness
Elk Wilderness
Maoon Boils
Capitol
Reef NP
Color adq
Springs
Cbeyem
Weils
hack Canyon of the Gunnison NP
' Bryce
Canyon NP
La Gsita VWIderness
Sand bunes
id Piupeivo
WbriinuchtWiOwn.
/burpngo f I
Trinidad
Farmtngton
Wheeler Peak Vtf Idetness
Oklahoma
ildemess I
Mount Bnldy VMIderness
New Mexico
/
X Grand
J Canyon NP
San Pedro Parks Wilderness T'\
Taos
Flagstaff / / / J0a,up i cJ- 1 (tfftj, vwitfyrness
dandeller NM * .T*
Pi*i if teller est NP / pswa F*e
Class I Federal Areas
110WW 105TOW
i „no,0 50 100 200 300
Legend — —
M US Forest Service Class I Federal Areas
~ Fish & Wildlife Service Class I Federal Areas
| National Park Service Class I Federal Areas
i I state
| cities and towns
| county
Colorado Department of Public Health and Environment / A Ir Pollution Control Division/ Technical Services Program
December 2005
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
Additional Impacts Analysis
This section is currently under review within the Division. Please contact Division staff
to confirm the procedure for an additional impacts analysis.
Regulation No. 3, Part D, §VI.A.6 requires an additional impact analysis for major stationary sources and
major modifications. The additional impact analysis applies in all areas, including Class I and Class II areas.
The regulation specifically requires an "analysis of the impairment to visibility, water, soils, and
vegetation." In some instances, a growth analysis is also required. The growth analysis is recommended
only if the new source or modification will have a significant impact, that is, it is only required if an air
quality impact analysis (e.g., CAAQS and NAAQS analysis, PSD increment analysis) is triggered.
The additional impact analysis can be done using qualitative or quantitative methods. The Division
generally views the analysis of impairment as a disclosure type of requirement. The level of analysis
depends on the situation and the likelihood that there could be some type of impairment.
In general, if the additional impact analysis suggests there might be adverse impacts to soils, vegetation,
or visibility, the information may be used in the BACT review process. This does not mean that the BACT
determination must fix the problem; it means that all the issues associated with BACT, including
economics and environmental impacts should be balanced and considered.
Impact Analysis for Water
The inclusion of water in the additional impact analysis is a Colorado requirement. By regulation, the
water analysis in Class II areas does not affect permit approval or denial or control technology selection.
The water impact analysis is intended to serve as a data-gathering and analysis mechanism to allow the
Division and others to further investigate problems such as acid deposition in high altitude lakes. Refer to
the "Additional Impact Analysis" discussion in the "Statement of Basis and Purpose for the Prevention of
Significant Deterioration Program Regulations" (adopted March 10, 1983) of the Common Provisions
Regulation for more information about the intent of this requirement.
Visibility Analysis
In addition to the Class I visibility analysis discussed in section 7.5, an analysis of impairment to visibility in
Class II areas should also be addressed in the permit application (see Regulation No. 3, Part D, §VI.A.6).
According to U.S. EPA guidance (USEPA 1990), "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:
¦ 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."
Refer to U.S. EPA guidance for more specific recommendations. The focus of Class I visibility analysis is on
assessing visibility impacts within a Class I areas. The focus of the Class II visibility analysis is on sensitive
Page 45 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
views outside of Class I areas. The Division has developed a database of sensitive views to assess impacts
in Class II areas. These are called scenic and/or important views. They are not integral vistas.9The Class II
scenic and/or important views do not have the force and effect of the visibility rules in Class I areas. The
information regarding levels of change in visibility is used to track changes in visibility that might be
important to the public. A list of these views is available from the Division.
The Division does not appear to have the authority to deny a permit if adverse visibility impacts occur
outside a Class I area. Instead, the information may be used to consider the need for additional emission
controls. Therefore, it is important to keep the Class I visibility analysis distinct from the Class II visibility
analysis in the modeling report.
In practice, when PSD applicants contact the Division, the Division will determine if there are any Class II
scenic views within the probable area of influence of the proposed source. If there are, the analysis
approach should be determined on a case-by-case basis in consultation with the Division. If modeling is
warranted, the modeling procedures for the scenic and/or important views are usually based on
techniques similar to those used for Class I visibility assessments.
The Division does not have specific thresholds or criteria for determining when there is impairment to a
Class II view. Impairment determinations are made on a case-by-case basis considering a number of
factors including the geographic extent, intensity, duration, frequency, and time of modeled visibility
impairment. Other factors such as interference with a visitor's visual experience, correlations between
time of impairment with natural conditions that reduce visibility, and other criteria might be considered.
Finally, limitations of the modeling system are considered. For example, results from a screening-level
model do not carry as much weight as results from a refined model. The ability of the modeling system to
properly account for relevant atmospheric chemistry and meteorology is also considered. If, after
considering all appropriate criteria, it is believed that Class II visibility may be impaired, the Division may
request that the "environmental impact analysis" portion of the "best available control technology"
(BACT) determination be revisited.
A compliance demonstration with Colorado's visibility standard, which is applicable in the AIR Program10
area, is not required to obtain a permit.
9 An integral vista adopted into regulation can be afforded the same level of protection from visibility
impairment as the Class I area itself or any lesser level or protection, as determined by a state on a case-
by-case basis. Because views in the Western U.S. commonly extend for great distances, integral vistas are
a controversial aspect of the Visibility SIP package. The Department of the Interior (DOI) preliminarily
identified integral vistas associated with Class I areas on January 15, 1981. However, both the DOI
(speaking for the National Park Service) and the Department of Agriculture (speaking for the U.S. Forest
Service) later declined to officially list any vistas. One reason given by the DOI was that states already had
sufficient opportunity through existing authority to protect integral vistas. Thus, the naming of integral
vistas and incorporation into SIPs was left to individual states (CDPHE, 1992).
10 The AIR program area is defined in 42-4-304, C.R.S. It generally includes all or part of the following
counties: Adams, Arapahoe, Boulder, Denver, Douglas, El Paso, Jefferson, Larimer, and Weld.
Page 46 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
Soils and Vegetation Analysis
Regulation No. 3 states that the owner or operator should provide an analysis of impairment to soils and
vegetation for each regulated pollutant emitted in a significant quantity. Only vegetation with commercial
or recreational value should be addressed. U.S. EPA's guidance states that, for most soils and vegetation,
ambient concentrations of criteria pollutants below the secondary National Ambient Air Quality
Standards (NAAQS) will not result in harmful effects. Nevertheless, the secondary NAAQS may not
adequately protect certain sensitive vegetation and soils, particularly for regulated non-criteria pollutants
(USEPA 1990), see section 7.3. As recommended in U.S. EPA guidance, new sources or modifications
subject to PSD rules should:
^ provide an inventory of soils and vegetation with commercial or recreational value in the vicinity
of the facility (e.g., crops);
^ review peer-reviewed scientific literature to determine the concentration level (for appropriate
averaging times) of regulated pollutants that would be harmful to vegetation; if no information is
available in the literature, assume the secondary NAAQS is protective if one exists for the
regulated pollutant under review; if modeling has been done, compare modeled impacts to the
secondary NAAQS and to other levels of concern identified through a literature search; if the
potential impact is determined to be harmful, discuss the nature of the harm and its spatial
extent in the modeling report.
AQRV and Visibility Analysis in Federal Class I Areas
This section is currently under review within the Division. Please contact Division staff
to confirm the procedure for an AQRV and Visibility analysis in Class I areas.
The Air Quality Related Values (AQRV), analysis is required as part of a PSD permit to estimate potential
changes in visibility, deposition, soils and water in Class I areas. The goal of the Class I impact analysis is to
determine if the projected changes to AQRVs, as a result of the installation of a new source or the
modification of an existing source under the PSD regulations, are acceptable for a given Class I area (See
Regulation No. 3, Part D, §XI 11 and XIV for regulatory requirements). The decision to issue a permit is the
responsibility of the Division. A permit application can be denied if a proposed source would impair
visibility or other AQRVs in a Class I area. It is important to note that the determination of impairment is
done on a case-by-case basis. In the case of visibility, this determination will be made based on the
magnitude, number of occurrences, time of year and if such changes would affect a visitor's experience in
the area. For more on the regulatory framework, refer to applicable regulations and section 10.1.
In general, the elements of the federal Class I AQRV, including visibility, analysis are determined on a
case-by-case basis.
Regulation No. 3, Part D, §XI 11 .A states that "Within twenty days of receipt of a permit application for a
new major stationary source or major modification that may affect visibility or air quality related values in
any Federal Class I area, the division shall transmit a copy of the application to all affected Federal Land
Managers and consult with them as to its completeness in its analysis and monitoring (if required) of air
quality related values. If the division receives advance notification of a permit application of a source that
may affect visibility or air quality related values, it will notify all affected Federal Land Managers
Page 47 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
within thirty days of such notification. The division will consider any analysis performed by a Federal Land
Manager that indicates there will be an adverse impact on visibility or air quality related values if such
analysis is received within thirty days after the Federal Land Manager receives a copy of the complete
application. If the division disagrees with the Federal Land Manager, any notices for public comment or of
a public hearing on the application will explain the disagreement or state where the explanation can be
obtained."
If a protocol is submitted to the Division, as recommended in section 8.1, a copy should be provided for
each affected federal land manager.
As stated in Regulation No. 3, Part D, §XI 11, the Division sends affected FLMs a copy of the permit
application for proposed new sources or modifications that may affect air quality related values (AQRVs)
in any federal Class I area. For relatively small and/or distant major stationary sources, the FLM may not
take an active role in the review or modeling process. In other cases where a significant impact may occur
or where there may be unacceptable levels of change to AQRVs, including visibility, the FLM usually takes
an active role.
While the Division's Stationary Sources Program is responsible for forwarding the permit application to
the appropriate FLMs, Technical Services Program staff typically contact affected FLMs to obtain Class I
significance levels and other recommendations for the analysis required by Regulation No. 3.
The initial contact with FLMs should occur early in the process. If there is a PSD pre-application meeting,
FLMs should be invited. Regulations require that the Division consult with the FLMs as to the
completeness of the permit application. If the applicant decides to directly contact affected FLMs for
recommendations, the Division should be kept in the loop.
Air Quality Related Values Analysis for Major Stationary Sources
For proposed major stationary sources and major modifications located in attainment areas, visibility
requirements for new sources and modifications subject to PSD rules are found in various sections of
Regulation No. 3, Part D including: §VI.A.6, §XI 11, and §XIV.E
For proposed major stationary sources and major modifications located in nonattainment areas, refer to
Regulation No. 3, Part D, §V.
Figure 3 illustrates key aspects of the regulatory decision process for major stationary sources and major
modifications seeking construction permits:
1. The first step in the process is to determine those pollutants for which there will be a significant
emission rate increase for a new source or a significant net emissions increase for a major
modification.
2. If the proposed emission rate is not significant, the additional impact analysis (Regulation No. 3,
Part D, §VI.A.6) and the AQRV reguirements (§XIII and §XIV.E) do not apply. In practice, new
sources are major for some pollutants and minor for others. In some cases, the modification may
not be major for all pollutants that would affect AQRVs.
3. If the Division concludes that an "analysis of impairment" (§VI.A.6) is necessary,
key decisions that must be made. For example, the applicant should discuss the
Page 48 of 64
there are several
project with the
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
Division to decide if any AQRV monitoring is warranted (§XIII.B). The Division will make this
decision after consultation with the FLM. If monitoring is required, a monitoring plan should be
prepared and submitted for Division approval. If monitoring is not warranted, which is usually the
case, then the applicant can move on to the next step in the flowchart.
4. The applicant should consult with the Division to determine the extent of the "analysis of
impairment." The regulations do not clearly define what constitutes an "analysis of impairment."
Thus, the extent of the analysis is decided on a case-by-case basis. The Division and U.S. EPA
strongly recommend that the applicant submit a protocol.
5. When the permit application is submitted to the Division, it must include the "analysis of
impairment" to be ruled complete.
6. Aplicants should be aware of Regulation No. 3, Part D, §XI 11 .A - Federal Class I Areas; it states,
"Within twenty days of receipt of a permit application for a new major stationary source or major
modification that may affect visibility or air quality related values in any Federal Class I area, the
Division shall transmit a copy of the application to all affected Federal Land Managers and consult
with them as to its completeness in its analysis and monitoring (if required) of air quality related
values. If the Division receives advance notification of a permit application of a source that may
affect visibility or air quality related values, it will notify all affected Federal Land Managers within
thirty days of such notification."
1. The next step is to determine if the source will cause or contribute to a violation of applicable
Class I PSD increments.
2. If the source does not cause or contribute to a Class I increment violation §XI 11 .A states, "The
Division will consider any analysis performed by a Federal Land Manager that indicates there will
be an adverse impact on visibility or air quality related values if such analysis is received within
thirty days after the Federal Land Manager receives a copy of the complete application"
But if the FLM fails to determine if there will be an adverse impact, the Division may perform the analysis,
as explained in Regulation No. 3, Part B §XIII.C.
3. If it is determined, through modeling provided by the applicant, that the source will cause or
contribute to a violation of applicable Class I PSD increments, then the Division may still issue the
permit if the requirements of §XI 11. D are met. Regulation No. 3, Part B, §XI 11. D states, "The owner
or operator of a proposed major stationary source or major modification may demonstrate to the
satisfaction of the Federal Land Manager that the emissions from such source or modification
would not have an adverse impact on the air quality related values (including visibility) of Class I
lands under the Federal Land Manager's jurisdiction, notwithstanding that the change in air
quality resulting from emissions from such source or modification would cause or contribute to
concentrations that would exceed the maximum allowable increases for Class I area. If the Federal
Land Manager concurs with such demonstration and so certifies to the Division, the Division or the
Commission may, provided that applicable requirements are otherwise met, issue the permit with
such emission limitations as may be necessary to assure that emissions of sulfur dioxide, and
PM10, PM2.5 and nitrogen oxides would not exceed the following maximum allowable increases
over the minor source baseline concentration for such pollutants..."
Page 49 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
1. PM2.5
1. Annual arithmetic mean 4 ng/m3
2. Twenty-four hour maximum 9 ng/m3
2. PM10
1. Annual arithmetic mean 17 ng/m3
2. Twenty-four hour maximum 30 ng/m3
3. Sulfur Dioxide
1. Annual arithmetic mean 20 ng/m3
2. Twenty-four hour maximum 91 ng/m3
3. Three-hour maximum 325 ng/m3
4. Nitrogen Dioxide
1. Annual arithmetic mean 25 ng/m3
5. Although the FLMs have an affirmative responsibility to protect AQRVs they have no permitting
authority under the federal Clean Air Act (CAA). They also have no authority under the CAA to
establish air quality-related rules or standards. The FLM role consists of considering whether
emissions from a new source may have an adverse impact on AQRVs and providing comments to
permitting authorities. Thus, the final decision to grant or deny a permit is made by the Division
or AQCC. Regulation No. 3, Par B, §XI 11 .A states, "If the Division disagrees with the Federal Land
Manager, any notices for public comment or of a public hearing on the application will explain the
disagreement or state where the explanation can be obtained."
6. If the FLMs disagree with the Division's decision to grant a permit, they may request a hearing,
see Regulation No. 3, Part D, §IV.A.6.
Pre-and Post-Construction Monitoring Analysis
Division modeling and monitoring staff should be contacted as early as possible to discuss the need to
conduct pre-construction monitoring. If monitoring is proposed or required, a monitoring plan consistent
with applicable U.S. EPA and Division monitoring guidance (e.g., "Ambient Air Pollution and
Meteorological Monitoring Guidance") should be submitted for approval.
If the proposed emission rate from a new source or the net emissions increase from a modification is
significant for a given pollutant, as defined by Regulation No. 3, the estimated impact from the new
source or modification should be compared to the significant monitoring concentration (see Table 10 or
Regulation No. 3, Part D, §VI.D.2). In addition, if possible, existing air quality levels should be compared to
the significant monitoring concentration.
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
Pre-Construction Monitoring Analysis
Refer to Regulation No. 3, Part D, §VI.A.3 for details about how pre-construction monitoring requirements
are determined.
If existing air quality levels or the estimated impacts from the proposed source or modification are below
the applicable monitoring de minimis level, Regulation No. 3 states that the monitoring requirements may
not apply. If the levels are above the de minimis levels, pre-construction monitoring may be required if
the Division believes it is necessary.
Permit applicants should be aware that the time-line for submitting a PSD application could be affected
by the requirement to collect ambient data. For example, if the collection of site-specific meteorological
data is required, at least a full year of data must be collected. For air quality data, at least a full year of
data is typically required, although as little as four months of data may be allowed in some circumstances.
The Division must approve ambient data for use before the permit application can be ruled complete.
Post-Construction Monitoring Analysis
The modeling report submitted with the permit application should address the need for post-
construction monitoring (see Regulation No. 3, Part D, §VI.A.4).U As part of the permit review process,
the Division will, based on the language in the regulation, determine if post-construction monitoring is
necessary.
11 40 CFR Part 51.166(v)(2) states that the source "shall...conduct ambient air monitoring as the reviewing
authority determines is necessary...."
Page 51 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
Table 10. PSD Significant Monitoring Concentration1
Pollutant & Averaging Period
Level
Carbon Monoxide (CO)
8 hour
575 |ag/m3
Nitrogen Dioxide (NO2)
Annual
14 |ag/m3
Ozone (O3)
8 hour
100 tpy VOCs
1 hour
100 tpy VOCs
Sulfur Dioxide (SO2)
24 hour
13 |ag/m3
Particulate Matter < 10 pim (PM10)
24 hour
10 |ag/m3
Particulate Matter < 2.5 pim (PM2.5)
24 hour
4 |ag/m3
Fluorides
24 hour
0.25 |ag/m3
Total Reduced Sulfur
1 hour
10 |ag/m3
Hydrogen Sulfide
1 hour
0.2 |ag/m3
Reduced Sulfur Compounds
1 hour
10 |ag/m3
1_rhe significant monitoring concentrations (de
new sources and modifications subject to PSD
Part D, §VI).
minimis levels) apply only to
review (see Regulation No. 3,
Page 52 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
Regulated, Non-Criteria Pollutant Analysis
For regulated, non-criteria pollutants (i.e., fluorides, total reduced sulfur, hydrogen sulfide, and reduced
sulfur compounds), a separate air quality analysis should be submitted if the applicant proposes to emit
the pollutant in a significant amount from a new source or proposes to cause a significant net emissions
increase from a modification. The PSD significant emission rates for these pollutants are as follows:
d) Fluorides, 3 tons per year;
e) Sulfuric Acid Mist, 7 tons per year;
f) Hydrogen Sulfide, 10 tons per year;
g) Total Reduced Sulfur (including hydrogen sulfide: 10 tons per year);
h) Reduced Sulfur Compounds (including hydrogen sulfide: 10 tons per year);
Estimated impacts from regulated non-criteria pollutants should be presented and compared to the
significant monitoring concentrations (see Table 10 or Regulation No. 3, Part D, §VI.B.3). Existing
background concentration estimates should be determined in consultation with the Division. If ambient
measurements are available, they should be presented and compared to the significant monitoring
concentrations.
Section 5 - Preferred Air Dispersion Models & Associated Inputs
Source Data
Begin by clearly identifying and documenting all sources of emissions associated with the modeling
analysis. For each identified source, evaluate and discuss how emissions are generated and emitted. This
discussion will be the supporting basis for the source characterization used in the modeling analysis. Then
determine and document the appropriate source parameters associated with the source characterization.
Criteria Pollutant Recommendations
While this section is intended for sources located in attainment or unclassified areas of Colorado, it may,
in some cases, be used by sources located in nonattainment areas; however, sources in nonattainment
areas should read Section 2, Nonattainment Areas, first.
In a compliance demonstration, the applicable design concentration must be calculated. This is usually
done within the model or by using a post-processor. The design concentrations vary depending on the
available meteorological data. If there is not a meteorological dataset that is adequately representative of
the facility, then the design concentration needs to be the highest concentration for all pollutants and
averaging periods. This allows the worst-case impacts to be captured in the modeling analysis.
The design concentrations also vary depending on the impact analysis being performed. For a
NAAQS/CAAQS analysis Appendix W states, "the design concentration is the combination of the
appropriate background concentration with the estimated modeled impact of the proposed source...The
specific form of the NAAQSfor the pollutant(s) of concern will also influence how the background and
modeled data should be combined for appropriate comparison with respective NAAQS in such a modeling
demonstration. Given the potential for revision of the form of the NAAQS and the complexities of
Page 53 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
combining background and modeled data, specific details on this process can be found in the applicable
modeling guidance available on the EPA's SCRAM Web site." For a PSD increment analysis Appendix W
states, "the design concentration includes impacts occurring after the appropriate baseline data from all
increment-consuming and increment-expanding sources." For short-term increments, the maximum
allowable increases may be exceeded once per year at each site. For annual increments, the maximum
allowable increases may not be exceeded.
The facility should contact the Division's Stationary Sources Program (SSP) to determine what pollutants
need to be included in the air quality impact analysis.
Carbon Monoxide
Compliance demonstrations should address both the 1-hour and 8-hour NAAQS. The maximum highest
first high (H1H) modeled concentration from all receptors should be the design value from the SIA to
compare to the SILs. When using representative meteorological data, the maximum high second high
(H2H) modeled concentration from all receptors should be the design value for both 1-hour and 8-hour
averaging periods.
Compliance demonstrations should address the 3-month NAAQS. The NAAQS is significantly more
stringent than the CAAQS monthly value; therefore, the monthly CAAQS was revoked from Colorado
Regulation 8 Part C II.B in March 2010.
Nitrogen Dioxide
Compliance demonstrations should address both the 1-hour and annual NAAQS. The outcome of the
analysis will be included in the public final modeling report.
Both averaging periods are best performed with a tiered approach:
Tier I: 100 percent conversion of nitrogen oxides (NOx) to nitrogen dioxide (N02).
Tier II: Ambient Ratio Method 2 (ARM2) uses an ambient ratio between 0.5 and 0.9 that must be
derived from U.S. EPA's Air Quality System. Colorado no longer accepts the EPA-recommended ARM
ratio of 0.8 as monitoring data has shown N02/NOx ambient ratios exceeding 0.8 conversions.
Justification for ambient ratio used is required. This method should also be used if the following are
true:
1. Tier I results are within or below a range of 150 - 200 ppb
2. N02 background concentrations are below EPA's recommended range of 20 - 30 ppb
3. 03 background concentrations are below EPA's recommended range of 80 - 90 ppb
4. In-stack N02/NOx ratios at or below 0.2
Tier III: Ozone Limiting Method (OLM) uses in-stack N02/NOx ratios and background concentrations.
The EPA established a general acceptance of 0.5 as a default in-stack ratio of N02/NOxfor input to
OLM. If the applicant proposes to use an in-stack N02/NOx ratio other than the EPA default, sufficient
justification and documentation will need to be provided to support the source-specific data. The
Lead
Page 54 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
source-specific in-stack N02/N0X ratio needs approval from the permit engineer. Hourly by season
profiles of both N02 and 03 should be requested from the Division for input to OLM.
The maximum highest first high (H1H) modeled concentration from all receptors should be the design
value from the SIA to compare to the SILs. When using representative meteorological data, the 1-hour
design value should be the maximum 5-year average of the 98th percentile of the annual distribution of
the maximum daily 1-hour modeled concentrations or the highest eighth high (H8) for each receptor.
When using representative meteorological data, the annual design value should be the maximum
modeled concentration for all receptors across all years of meteorological data.
Ozone
In general, accurate and cost effective methods for modeling ozone impacts from stationary point
sources are not available. Therefore, ozone modeling is not routinely requested for construction permits,
although it could be in unusual cases such as situations where the Division believes ozone standards could
realistically be violated by the proposed source or modification. If modeling is considered, the cost of
conducting such an analysis will be factored into the decision process.
Precursors to ozone need to be discussed with the Division's TSP modeling staff. The applicant should
review applicable EPA guidance regarding precursors to ozone.
Particulate Matter < lOpim (PMio)
Compliance demonstrations should address the 24-hour NAAQS. The annual PMio NAAQS was revoked in
2006 so compliance is no longer required for this averaging period. The maximum highest first high (H1H)
modeled concentration from all receptors should be the design value from the SIA to compare to the SILs.
When using representative meteorological data, the design value should be maximum highest sixth high
(H6H) modeled concentration for all years of meteorological data.
Particulate Matter < 2.5\im (PM2.5)
Compliance demonstrations should address both the 24-hour and annual NAAQS. The maximum highest
first high (H1H) modeled concentration from all receptors should be the design value from the SIA to
compare to the SILs. When using representative meteorological data, the 24-hour design value should be
the maximum 5-year average of the 98th percentile of the annual distribution of the maximum 24-hour
modeled concentration or the highest eighth high (H8H) for each receptor. When using representative
meteorological data, the annual design value should be the maximum 5-year average modeled
concentration from all receptors.
Secondary formation of PM2.5 needs to be discussed with the Division's TSP modeling staff. The applicant
should review applicable EPA guidance regarding when modeling secondary formation of PM2.5 is
necessary.
Sulfur Dioxide
Compliance should be demonstrated with the 1-hour, 3-hour, 24-hour, and annual NAAQS as well as with
the Colorado 3-hour standard of 700 |ag/m3. The 24-hour and annual NAAQS remain in effect for "any
area for which it is not yet 1 year since the effective date of designation under the current (2010)
standards and any area for which an implementation plan providing for attainment of the current
Page 55 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
(2010) standard has not been submitted and approved and which is designated nonattainment under the
previous S02 standards or is not meeting the requirements of a SIP call under the previous S02 standards."
The state of Colorado was designated as attainment/unclassifiable on December 21, 2017; therefore,
compliance demonstrations for the 24-hour and annual NAAQS are required until December 21, 2018.
The maximum highest first high (H1H) modeled concentration from all receptors should be the design
value from the SIA to compare to the SILs. When using representative meteorological data, the 1-hour
design value should be the maximum 5-year average of the 99th percentile of the annual distribution of
the maximum daily 1-hour modeled concentration or the highest fourth high (H4H) for each receptor.
When using representative meteorological data, the 3-hour and 24-hour design values should be the
maximum highest second high (H2H) modeled concentration from all receptors. When using
representative meteorological data, the annual design value should be the maximum modeled
concentration from all receptors across all years of meteorological data.
Mobile Sources Data
Facilities that involve haul trucks need to include fugitive dust emissions in both the permit application
and the air quality impact analysis. Large mining equipment tailpipe emissions should also be included.
The Division is currently developing more guidelines to establish when to include tailpipe emissions from
haul road traffic and mining equipment. The Division has the current procedure if tailpipe emissions are
to be included in the air quality impact analysis. If the applicant is unsure whether tailpipe emissions
should be included, please contact the Division.
A facility is likely to have a fleet of trucks that is made up of a variety of different trucks. If the air quality
impact analysis involves N02 modeling using the Tier III approach, an in-stack N02/N0X ratio is necessary.
Different trucks will have different in-stack N02/N0X ratios. The Division recommends using a similar
tiered approach.
Tier A: Use the highest in-stack N02/N0X ratio of all the mobile engines in the fleet. This ratio should
be applied to all sources used to represent the truck traffic or non-road engines.
TierB: Calculate a weighted average in-stack N02/N0X ratio based on the total vehicle fleet and the
number of units with different in-stack N02/N0X ratios and use that value for the entire vehicle fleet.
This accounts for the influence of the different types of engines according to the number of units with
higher or lower in-stack N02/N0X ratios while at the same time keeping the modeling analysis simple.
TierC: Represent vehicles with similar in-stack N02/N0X ratios with separate sets of sources assigning
the corresponding in-stack N02/N0X ratio to each set of sources. Each road segment could have
multiple sets of sources overlaid on top of each other. This is the most accurate representation of the
vehicle traffic.
All proposed in-stack N02/N0X ratios require sufficient justification and documentation to support the
source-specific data. The source-specific in-stack N02/N0X ratios need approval from the permit engineer.
Nearby Sources Data
U.S. EPA recommends that, at a minimum, all nearby sources should be explicitly modeled as part of the
NAAQS analysis. Other background sources usually are accounted for by using an appropriate ambient
background concentration (i.e., see §9.2.2 of the USEPA Guideline on Air Quality Models) or, if a
Page 56 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
suitable ambient background concentration is not available, by application of a model. Nearby sources
and other background sources are terms used to reference all stationary sources except the new source
or modification under permit review.
The emissions estimates used in modeling nearby and other background sources should be consistent
with U.S. EPA recommendations in Table 8.2 of the USEPA Guideline on Air Quality Models and other
applicable U.S. EPA guidance. Table 8.2 recommends that actual operating levels averaged over 2 years
and federally enforceable permit limits should be used for all nearby sources. That is, emission rates
based on a combination of both allowable and actual data, if the actual data is available. A nearby source
is any major source, major stationary source, or minor source that causes a significant concentration
gradient in the vicinity of a new or modified source. All sources should be included if they are within 5
kilometers of the significant impact area of the source (significant impact area + 5km). Nevertheless, this
is not a bright line; in some cases, the 5-kilometer distance from the significant impact area should be
expanded. Professional judgment should be used when selecting sources to model.
The Division does not recommend a specific objective procedure for determining which sources should
be classified as nearby sources and which should be classified as other background sources. The
procedure used to select nearby sources should be based on professional judgment. In addition, it should
consider local conditions such as topography, meteorology, dispersion characteristics, availability of
ambient monitoring data, existing air quality, and other relevant factors. The procedure should include an
examination of the modeling results to ensure that all sources that should have been included were
included.
The nearby sources inventory provided by the Division may be missing key stack parameters as this
information is taken from submitted APENs. When the APENs are missing the stack parameters, this
information is left blank in the inventory. The Division has developed an initial approximation procedure
for applicants to use when the stack parameters are missing. Further refinement may be necessary in
order to demonstrate compliance.
¦ Determine the type of emission source: stack (point) or fugitive.
¦ Point source: Find stack parameters for similar equipment in the inventory. Provide justification
for the stack parameters used.
¦ Fugitive source: Group all the fugitive (non-stack) emissions from a facility into one area source
with dimensions of 100m x 100m, release height of 2 m, and initial sigma-z of 3 m. The x and y
coordinates of the facility in the nearby source inventory can be used as the southwest corner of
the area source.
A nearby sources inventory will be provided to the applicant upon request from the Division's Inventory
and Support staff. The applicant must specify the following when requesting a nearby source inventory:
¦ Coordinates of the project site
¦ Pollutants to be modeled
¦ Extent of the area included in the inventory
Background Concentrations
In general, the background concentration is intended to account for sources not explicitly included in the
modeling.
Page 57 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
For annual standards, the recommended background is typically based on the annual average value. For
shorter-term standards, selection of a background concentration can be more challenging. In general, the
background concentration should be one that can reasonably be assumed to occur with the modeled
concentration.
Determination of a background that can "reasonably be assumed to occur" is sometimes difficult. In
general, the niche being filled by the background concentration should be defined before a value is
selected. Since the background concentration field is usually assumed to be spatially uniform, the
background should account for elevated concentration levels that are expected to occur in the receptor
grid from non-modeled sources. Alternatively, a variable background field could be used if there is
sufficient data to generate one.
For purposes of addressing short-term standards, the total predicted concentration distribution should
represent combinations of impact and background that can reasonably be expected to occur
simultaneously in the particular application. The Division recognizes that the chance of two independently
caused short-term concentration maxima occurring simultaneously at any particular location may be low.
The Division can usually provide a background concentration upon request to account for other
background sources, including mobile sources and transport from distant sources. Determination of the
nearby sources accounted for by the background concentration can be rather subjective. Consequently,
the applicant should review the location and the collection date of the background data with respect to
nearby sources to determine how it should be incorporated into the overall modeling procedure.
The Division does not typically recommend the use of a background concentration to account for
increment consumption. Nevertheless, there may be situations where a statistical analysis or review of
trends in ambient air quality data would be useful to quantify local or regional changes in air quality since
the minor source baseline date.
To streamline the background concentration requests, a form is available on the Division's website
(https://www.colorado.gov/airquality/permits.aspx). If the applicant would like seasonal background
data, please contact the Division's TSP modeling staff.
Elevation Data
Terrain elevations for sources and receptors should be used when appropriate. Discuss the source of
terrain data in the modeling report.
Terrain elevations for receptors as well as nearby and other background sources should be based on U.S.
Geological Survey (USGS) National Elevation Dataset (NED). A minimum resolution of 1/3 arc second (IO-
meter) files covering a minimum radius of 40 kilometers from the facility under review. NED files can be
downloaded using the CDPHE Elevation Data Quad Download Tool
(https://www.colorado.Rov/airquality/quad selector map.aspx).
Some facility sites are graded (e.g., flat) so that actual site topography is or will be significantly different
from the topography that is found in a USGS NED or in other elevation data. Thus, it is appropriate to use
the site-specific graded elevations for the facility sources and buildings. A plot plan should be provided
that depicts the site-specific elevations. If NED files are used for facility sources and buildings, sufficient
justification and documentation will need to be provided to support the use of non-site-specific data.
Page 58 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
Downwash Applicability
Downwash is a term used to represent the potential effects of a building on the dispersion of emissions
from a source. Downwash is considered for sources characterized as point sources. The stack height and
proximity of a point source to a structure can be used to determine the applicability of downwash.
Downwash does not apply to sources characterized as areas. Downwash is indirectly considered for
volume sources by adjusting the initial dispersion factors.
Point sources with stack heights less than good engineering practice (GEP) stack height should consider
dispersion impacts associated with building wake effects (downwash). GEP stack height is the greater of
(40 CFR § 51.100( ii)):
(1) 65 meters, measured from the ground-level elevation at the base of the stack:
(2)(i) For stacks in existence on January 12, 1979, and for which the owner or operator had obtained
all applicable permits or approvals required under 40 CFR parts 51 and 52.
Hg = 2.5H,
provided the owner or operator produces evidence that this equation was actually relied on in
establishing an emission limitation:
(ii) For all other stacks,
Hg = H + 1.5L
where
Hg is the GEP stack height;
H is the structure height; and
L is the lesser of the structure height or maximum projected width (the width as seen from the source
looking towards either the wind direction or the direction of interest) of the structure.
These formulas define the stack height above which building wake effects on the stack gas exhaust may
be considered insignificant.
A structure is considered sufficiently close to a stack to cause downwash when the minimum distance
between the stack and the building is less than or equal to five times the lesser of the structure height or
maximum projected width of the structure (5L). This distance is commonly referred to as the structure's
region of influence. If the source is located near more than one structure, assess each structure and stack
configuration separately.
Once downwash applicability is determined, provide documentation to support that determination.
Receptor Network
The approach to creating a receptor network varies with the goals of the modeling study. Case-by-case
professional judgement should be used. Factors such as topography, density of nearby sources,
meteorology, and requirements of the selected model should be considered when selecting receptors. In
general, the network should be consistent with U.S. EPA's recommendations. It should extend far
Page 59 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
enough to define the significant impact area for the source or modification under review. For elevated
point sources, it is sometimes useful to initially use a simple screening-level model to help determine how
far out to extend the receptor network.
If the concentration gradient is increasing at the edge of the network, the network should be extended. 1-
hour modeling analyses tend to result in large significant impact areas; therefore, professional judgement
should be used when extending and refining the receptor network. Refer to U.S. EPA Memos
(https://www.epa.gov/scram/air-quality-models-clarification-memos-dispersion-models) for guidance.
The Division generally considers a fine receptor grid to have receptor spacing of 100 meters or less. A
coarse receptor grid usually refers to receptor spacing greater than 100 meters.
While source-specific issues such as expected plume rise and topography should be considered when
deciding if the following recommendations are appropriate, the following recommendations often
provide a good starting point:
1) Up to 1 kilometer - grid with 100-meter receptor spacing (fine)
2) From 1 to 3 kilometers - grid with 250-meter spacing (coarse)
3) From 3 to 10 kilometers - grid with 500-meter spacing (coarse)
4) Beyond 10 kilometers - grid with 1-kilometer spacing (coarse)
5) Along fence line or ambient air boundary - 50 to 100 meter receptor spacing
6) If no fence or boundary - 50-meter receptor spacing within source facility
7) Discrete receptors for sensitive nearby sites (e.g., residences, schools) unless the grid is sufficient
8) Flagpole receptors on balconies and rooftops of buildings not owned or operated by the facility
under review (e.g., balconies on apartment buildings, rooftop restaurants, rooftop pools)
9) If the modeled maximum values from the facility under review (or maximum values in an air
quality impact analysis such as a CAAQS and NAAQS analysis) occur in a coarse receptor grid,
additional modeling should be performed with a fine grid to find the maximum concentrations
10) Additional fine receptor grids or discrete receptors may be necessary in complex terrain or
sensitive areas to clearly define the area of maximum impact
Receptors may be omitted from the property of the facility under review, provided that public access is
precluded by a fence or other physical barrier. Refer to the definition of ambient air in the definitions
section at the beginning of this document. If there is not a physical barrier (e.g., fence, wall), receptors
should be located on the property of the applicant. Division and/or U.S. EPA approval is necessary if the
applicant wants to use a physical barrier such as a canyon, river, tailings pile, intense terrain or other
physical features as the ambient air boundary. Intense terrain will be approved on a case-by-case basis to
preclude public access as a physical barrier. Intense terrain that acts as a physical barrier needs to have a
minimum slope of 5 to 1, per EPA guidance. If a physical barrier is approved by the Division to preclude
public access, frequent posting is usually necessary along with routine security patrols; in addition, points
of public access into the posted area (e.g., roads, trails) should be fenced or gated. Refer to U.S. EPA
memos on this subject.
Page 60 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
Meteorological Data
Meteorological data should be collected, processed, and applied in ways that are consistent with the
most current federal regulations (https://www3.epa.Rov/ttn/scram/Ruidance/Ruide/appw 17.pdf),
guidance and model user's guides. If representative meteorological data are not available, it may be
necessary to collect at least one (1) year of site-specific data. Any source intending to collect site-specific
data should contact the Division prior to setting up a monitoring program. The Division has monitoring
guidance available.
Meteorological data will be provided by the Division. The Technical Services Program modeling staff will
determine the most representative meteorological data appropriate to use for the facility under review.
The applicant should provide the following information to the Division to obtain AERMOD-ready
meteorological data:
¦ Coordinate (latitude/longitude or UTM) of source location, including datum
¦ Source location identified on l:24,000-scale topographic map(s)
¦ Brief description of the sources of emissions (i.e., stack vs fugitive, stack heights, source types)
The Division staff takes the above information and assesses the expected conditions at the source
location and for each source type. A dataset will be identified that best matches the conditions expected
at the source location from the available meteorological datasets known and that meet the completeness
requirement.
Per regulatory requirements, for PSD applications where the Division has required pre-construction
meteorological monitoring, the permit application will not be ruled complete until the data has been
submitted to the Division and approved.
As stated in §8.4.2 of the USEPA Guideline on Air Quality Models, 5 years of adequately representative
NWS data, at least 1 year of site-specific data, or at least 3 years of prognostic meteorological data should
be used. If more than 1 year of site-specific data exist, multiple years (up to 5 years) should be used. For
long-range transport modeling and complex wind situations see §8.4.4.2 of the USEPA Guideline on Air
Quality Models.
The use of prognostic meteorological data is currently not accepted in Colorado due to complex terrain.
The Division is currently reviewing how prognostic meteorological data can be used to capture the effects
of complex terrain.
When deciding whether or not to recommend or require collection of site-specific meteorological data,
Division modeling staff considers:
c. Dispersion characteristics of the source under review
d. Meteorological and dispersion issues associated with complex terrain
e. Distance to the nearest Class I area (for new sources and modifications subject to PSD rules)
f. The likelihood that the source will have an adverse impact on ambient air quality
Page 61 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
g. Whether or not the source is subject to PSD rules (monitoring is more likely to be requested for
new major stationary sources or major modifications subject to PSD rules than for minor sources)
h. Other relevant factors
To streamline the permit process and reduce the economic burden for minor sources and minor
modifications, collection of site-specific meteorological data is seldom requested for minor sources and
modifications. Nevertheless, it may be recommended if there is reason to believe the new source or
modification will cause or contribute to a violation of CAAQS or NAAQS.
If allowed under federal regulations and approved by the Division, conservative screening meteorological
data may be used in refined models instead of site-specific data for compliance demonstrations.
Modeling Scenarios
It is common for facilities to have sources that do not operate simultaneously with other sources at the
facility. This situation results in modeling different scenarios. For example, if a facility wants a permit that
allows operation of either flares or engines, but not both at one time, both the flare scenario and engine
scenario should be modeled.
If there are several sources that cannot operate simultaneously which would result in a significant
amount of scenarios, the applicant can simply include the worst-case source. Please be aware that using
this approach requires demonstration of the worst-case source. Comparing emission rates of these
sources does not equate to a worst-case analysis.
Permit conditions will be proposed based on the information used in the modeling. Restricted operating
schedules used to demonstrate compliance will be become permit conditions.
Include in the air quality impact analysis a written discussion covering the project, the modeling
performed, and the results.
The air quality impact analysis is a stand-alone report. Results from the report should be sufficient to
make a decision without input from other reports. Do not refer to other documents or reports for data
required to be in the report. In addition, do not exclude items without coordination with the Division's
TSP modeling staff unless the items are clearly not applicable to the project. Follow the reporting
requirements to expedite the technical review of the air quality impact analysis and to eliminate
unnecessary modeling.
Specific data are needed to review and perform modeling. The recommended list of data elements
presented here are often necessary to perform and/or review dispersion modeling. The applicant should
be prepared to provide these data with the application or upon request by the Division. If the data are
not provided with the application and cannot be provided upon request in a timely manner, the permit
process may be delayed. In addition, if data cannot be provided in a suitable format, additional staff time
may be necessary for data-processing tasks. Staff time is usually charged back to the applicant at the
permit processing hourly rate. While some of the data elements discussed here are already part of the
permit application and APEN forms, they are mentioned here for emphasis.
Section 6 - Reporting Requirements
Page 62 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
Send the air quality impact analysis to the permit engineer that requested the analysis. In addition, for
PSD applications send a copy of the air quality impact analysis to EPA Region 8.
Consistency in Geographic Coordinates
Geographic coordinates are used in modeling. Whenever possible, the datum upon which geographic
coordinates are based should be provided. For example, potentially significant discontinuities in source
and receptors coordinates may occur if some Universal Transverse Mercator (UTM) coordinates are
based on the North American Datum of 1927 (NAD27) while others are based on NAD83. Often, site
surveys are performed using GPS systems that are based upon WGS84 while UTMs might be based upon
a NAD27 topographic map. Therefore, a coordinate conversion should be performed when appropriate so
that receptors, source locations, and other coordinates reference a consistent system.
Exemptions from Submitting Modeling-Related Data
New sources and modifications with emissions less than the thresholds in Table 1, for which modeling is
not required, and sources not emitting any of the pollutants listed in Table 1 do not need to provide any
modeling-related data beyond what is requested in the permit application and/or APEN forms.
Since ozone modeling and HAPs modeling are not routinely performed as part of the permit review
process, VOC sources do not need to provide any modeling-related data beyond what is requested in the
permit application and/or APEN forms.
New Sources and Modifications Not Subject to PSD Rules
At a minimum, new sources and modifications not subject to PSD rules with emissions greater than the
thresholds in Table 1 should submit the data outlined in the Modeling Submittal Completeness Checklist
with the permit application or be prepared to provide the data upon request. The Modeling Submittal
Completeness Checklist can be found on the Division's Air Quality website
(https://www.colorado.gov/airquality/permits.aspx).
New Sources and Modifications Subject to PSD Rules
New sources and modifications subject to PSD rules with emissions greater than the thresholds in Table 1
should submit the data outlined in the Modeling Submittal Completeness Checklist with the permit
application or be prepared to provide the data upon request. The Modeling Submittal Completeness
Checklist can be found on the Division's Air Quality website
(https://www.colorado.gov/airquality/permits.aspx).
The following additional items should be submitted as well:
¦ For each pollutant for which the new source or modification is subject to modeling under PSD
rules, provide a source history that clearly shows the start-up and shutdown dates of each unit
(e.g., emissions source) at the facility. Include current and historic stack parameters and
source/building configurations. Compare start-up and shutdown dates to applicable PSD baseline
dates to determine PSD increment consuming and expanding emissions (see the Division's "PSD
Increment Tracking System" for baseline dates and related information). Provide metadata (i.e.,
describe the methods used to generate the data). The applicant may choose to ignore this data
element if an air quality impact analysis is not requested or if PSD increment modeling is
Page 63 of 64
-------
October 2021
Interim Colorado Modeling Guideline for Air Quality Permits
not requested; however, the Division encourages applicants to provide these data so that PSD
increment consumption and expansion can be tracked.
¦ A table showing nearby increment consuming/expanding sources (only recommended if a PSD
increment analysis has been performed). Refer to the Division's "PSD Increment Tracking System"
guidance for details, in particular section 2.1.2.
¦ A table comparing maximum modeled impacts with appropriate thresholds such as modeling
significance levels, standards, PSD increments, significant monitoring concentrations, and levels
of acceptable change to AQRVs.
¦ UTM coordinates for maximum modeled concentration estimate(s) from the PSD increment
compliance demonstration modeling (if applicable). These data are used to help the Division track
increment consumption across the state.
Air Quality Models - Clarification Memos for Dispersion Models: https://www.epa.gov/scram/air-quality-
models-clarification-memos-dispersion-models
Federal Land Managers' Air Quality Related Values Work Group (FLAG):
https://www.nature.nps.Rov/air/Pubs/pdf/flaR/FLAG 2010.pdf
Model Clearinghouse Information Storage and Retrieval System:
https://cfpub.epa.Rov/oarweb/MCHISRS/
Clean Air Act Permit Modeling Guidance: https://www.epa.Rov/scram/clean-air-act-permit-modelinR-
Ruidance
Air Quality Dispersion Modeling - Preferred and Recommended Models User Guides and Implementation
Guides: https://www.epa.Rov/scram/air-qualitv-dispersion-modelinR-preferred-and-recommended-
models
CDPHE APCD Air Quality Modeling Guidance for Permits:
https://www.colorado.Rov/airquality/permits.aspx
Recommended Additional Guidance
Page 64 of 64
-------
APPENDIX C
History of Denver Metro North Front Range 2008 Ozone Standard with Corresponding Ozone
Nonattainment New Source Review Requirements for Major Sources
-------
History of Denver Metro North Front Range 2008 Ozone Standard with Corresponding Ozone
Nonattainment New Source Review Requirements for Major Sources (40 CFR 51.165)
Major Source
Marginal
Moderate
Serious
Severe
Thresholds//Requirements
(tons per year)
(tons per
(tons per
(tons per
year)
year)
year)
DMNFR Designation or
2012
2016
2020
Projected
Bump-Up Year for 2008
2022
Ozone Standard
(75 ppb)
Volatile Organic Compounds
100
100
50
25
Nitrogen Oxides
100
100
50
25
Offset Ratios
1.1:1
1.15:1
1.2:1
1.3:1
Control Equipment for
Major New or Modified
Existing
Existing
Existing
VOC/NOx
Sources:
Sources:
Sources:
Sources:
Lowest Achievable
Reasonably
Reasonably
Reasonably
Emission Rate
Available
Available
Available
(LAER)
Control
Control
Control
Technology
Technology
Technology
(RACT)"
(RACT)"
(RACT)"
Major New
Major New
Major New
or Modified
or Modified
or Modified
Sources:
Sources:
Sources:
Lowest
Lowest
Lowest
Achievable
Achievable
Achievable
Emission
Emission
Emission
Rate
Rate
Rate
(LAER)
(LAER)
(LAER)
------- |