Guidance on the Limited Maintenance
Plan Option for Moderate
PM Nonattainment Areas
and PM Maintenance Areas
£% United States
Environmental Protect
Agency
-------
Guidance on the Limited Maintenance
Plan Option for Moderate
PM Nonattainment Areas
and PM Maintenance Areas
Air Quality Policy Division
Air Quality Assessment Division
Office of Air Quality Planning and Standards
Transportation and Climate Division
Office of Transportation and Air Quality
Office of Air and Radiation
U.S. Environmental Protection Agency
United States
Environmental Protection
^1 Agency
EPA-420-B-22-044
October 2022
-------
Table of Contents
Section 1: Purpose and Applicability 4
Section 2: Critical Design Value for PM2 5 6
2.1 Overview 6
2.2 Event-Influenced Air Quality Data 8
2.3 Air Quality Review 9
Section 3: Transportation Conformity 10
Section 4: General Conformity 12
Appendix: PM10 LMP Guidance (2001) 13
-------
Section 1: Purpose and Applicability
This document clarifies the EPA's Limited Maintenance Plan (LMP) guidance for PM2.5
maintenance plan submissions by state, local, and tribal air agencies.1 Unless otherwise stated,
this guidance applies for any existing PM2.5 National Ambient Air Quality Standard (NAAQS)
and for any future PM2.5 NAAQS.
This PM2.5 LMP Guidance applies the attached 2001 Limited Maintenance Plan Option for
Moderate PMw Nonattainment Areas guidance2 (PM10 LMP Guidance) for PM2.5 LMP
submissions, except for the specific topics addressed below, where the 2001 guidance is
superseded. This document therefore focuses on distinctions specific for PM2.5 LMPs. For a
broader discussion on LMPs generally, see the PM10 LMP Guidance.
Moderate PM2.5 nonattainment areas or existing PM2.5 maintenance areas meeting the criteria in
this guidance may demonstrate maintenance for purposes of Clean Air Act (CAA) section 175 A
using the method described below. To show that an area is expected to continue to attain the
standard for the 10-year maintenance period, this method relies primarily on air quality analyses
indicating that there would be a low probability of violating the standard in the future, rather than
using air quality modeling or a projection of an area's emissions inventory for a future year. As
discussed in the PM10 LMP Guidance, an air agency submitting an LMP is not required to submit
a future year emissions inventory, but it is still required to submit the other elements of a
maintenance plan—an attainment year emissions inventory, provisions for continued operation
of the monitoring network, verification of continued attainment, and a contingency plan.3 Any
LMP for a PM2.5 area must also meet the applicable requirements of the exceptional events/data
modifications, transportation conformity, and general conformity programs, as set forth in
relevant implementing regulations for each program. Many of the requirements associated with
these programs are described further below.
As noted, the LMP is a tool that allows certain nonattainment and maintenance areas to provide
for maintenance under CAA section 175 A based on an analysis of current and historical air
quality data, rather than modeling or emissions projections. As such, using an LMP to provide
for maintenance is not appropriate where an area expects to experience significant emissions
growth, or even anticipates that such growth may be possible, during the relevant 10-year
maintenance time period. In those situations, in order to meet the statutory requirement to
provide for maintenance, the air agency should use the long-standing methods included in a "full
maintenance plan" to demonstrate that the area will maintain the NAAQS even considering those
projected emissions increases. There are a number of additional considerations that also may be
1 The remainder of this document will refer to "state, local, and tribal air agencies" as either "air agency" or "air
agencies."
2 The Limited Maintenance Plan Option for Moderate PMw Nonattainment Areas guidance (including attachments)
was issued on August 9, 2001 and can also be found at: www.epa.gov/state-and-toeal-transportation/2001-limited-
maintenance-plan-moderate-pmlO-and-attachment.
3 PM10 LMP Guidance at 6-7. See also Procedures for Processing Requests to Redesignate Areas to Attainment,
September 4, 1992 (Calcagni Memorandum), available at: www.epa.gov/sites/defaiilt/fites/2016-
03/documents/calcagni memo -
procedures for processing requests to redesignate areas to attainment 090492.pdf.
4
-------
relevant to whether an LMP is appropriate for a PM2.5 area. For example, because of the health
risks presented by exposure to PM2.5 and possibility of emissions growth,4 an LMP would likely
not be appropriate for the first maintenance plan for a Moderate PM2.5 area5 that includes a major
metropolitan area.6 However, an LMP may be appropriate for an area's first PM2.5 maintenance
plan in an isolated rural area, or in a smaller metropolitan area where the PM2.5 air quality
problem is due to a specific source or sources unrelated to on-road transportation emissions and
where emissions growth is not anticipated. Areas that have already been redesignated to
attainment and are submitting a second maintenance plan under CAA section 175A(b) may be
candidates particularly well-suited for an LMP, especially if air quality concentrations in the area
have been relatively stable during the first 10-year maintenance period, indicating that emissions
growth is unlikely. At a minimum, EPA intends to evaluate information provided by an air
agency against the criteria in this guidance and associated regulations to determine whether a
PM2.5 LMP is appropriate for a given area.
This document is intended solely as guidance. The statutory provisions and EPA regulations
discussed in this document contain legally binding requirements. However, this document is not
a regulation itself, nor does it change or substitute for statutory provisions and regulations. Thus,
it does not impose legally binding requirements on state, local, or tribal agencies or EPA. EPA
retains the discretion to consider and adopt approaches on a case-by-case basis that may differ
from this guidance, but still comply with the statute and regulations.
Questions about the application of this guidance for specific areas should be addressed to an
EPA Regional Office SIP program contact. See this site for a list of Regional Office contacts:
https://www.epa.eov/air-qiialitY-implementation-plans/fmd-reeional-contact-air-qiiality-
sipsfipstips.
A copy of this policy guidance can be found at the following websites:
• https://www.epa.eov/pm-pollution/implementation-national-ambient-air-qualitv-
standards-naaqs-fine-particulate-matter
• https://www.epa.eov/state-and-local-transportation/policy-and-technical-euidance-state-
and-local-transportation#state
4 For more information on the health and environmental effects of PM, see www. epa. gov/pm-po 1 Inti0 n/faealth-a nd~
enviromnental-effects-paiticulate-matter-pm.
5 Consistent with the PM10 LMP Guidance, air agencies in Serious PM25 nonattainment areas should submit
maintenance plans that meet EPA's guidance for submission of a full maintenance plan for their first maintenance
plan.
6 A major metropolitan area, for example, could be an area that has an urbanized area population greater than
200,000. (This population threshold is used in other transportation conformity provisions.)
5
-------
Section 2: Critical Design Value for PM2.5
2.1 Overview
It is important to note that this LMP guidance for PM2.5 areas does not include the concept of
broadly applicable LMP air quality concentration criteria for the annual and 24-hour PM2.5
NAAQS, as was included for the PM10 guidance.7 Rather, this PM2.5 LMP Guidance relies on the
critical design value (CDV) concept (explained in Appendix A of the PM10 guidance), which is
used to reflect the unique variability of air quality concentrations for each monitoring site. To be
eligible for a PM2.5 LMP, the air agency should calculate the site-specific CDV for the
monitoring site with the highest design value and all other active monitoring sites with complete
data in the relevant nonattainment or maintenance area. The air agency should demonstrate that
the average design value (ADV) for each site in the area, based on the most recent 5 consecutive
PM2.5 design values,8 does not exceed the associated CDV for each site. If each site in the
nonattainment area has an ADV that is less than the CDV, it would demonstrate that the area has
PM2.5 concentrations that will likely remain below the level of the standard in the future.
CDVs are described in the PM10 LMP Guidance as "an indicator of the likelihood of future
violations of the NAAQS given the current average design value and its variability." Consistent
with the approach described in the PM10 LMP Guidance, the CDV calculation for a particular
PM2.5 monitoring site involves parameters including: 1) the level of the relevant NAAQS; 2) the
co-efficient of variation of recent design values; and 3) a statistical parameter corresponding to a
10% probability of exceedance. CDVs are inversely related to the site's design value variability,
with higher variability resulting in a lower (or more stringent) CDV. The site's average design
value (ADV), calculated from the most recent 5 consecutive design values, is then compared to
the CDV. If the ADV is lower than the CDV, then the probability of a future exceedance is less
than 10%.
Although the PM10 LMP Guidance only included calculations for the PM10 CDV, the same
procedure has been applied to PM2.5 design values by Chu and Paisie in their 2006 evaluation of
current PM2.5 conditions across the United States.9 In addition to the conservative "10%
probability of exceedance" statistical parameter used in the CDV calculation, decreasing
7 The broadly applicable LMP air quality concentration criteria included in the 2001 PM10 Guidance were 98 |ig/m3
for the 24-hour PM10 standard and 40 |ig/m3 for the annual PM10 standard. In general, a PM10 LMP submission
would be approvable if the area average design value (ADV) did not exceed these levels. In Attachment B of the
2001 PM10 LMP guidance, these levels are referred to as "margin of safety" values. This PM2 5 guidance does not
include such national default air quality threshold qualification levels, but instead relies on area-specific critical
design values.
8 Attachment A of the 2001PM10 guidance refers to using "a minimum of five years of data" for calculating the
ADV and CDV. EPA recommends that the ADV be calculated using at least five years of design values, each
representing a three-year period, because this approach would rely on a more robust dataset. However, we
acknowledge that an alternative interpretation may be acceptable, where these variables could be calculated using
three years of design values, collectively representing five years of air quality data.
9 Chu, Shao-Hang and Joseph Paisie, 2006. An evaluation of current PM2 5 conditions in the U.S. Atmospheric
Environment, Volume 40, Supp. 2, Pages 206-211.
www.sciencedirect.com/science/art.icle/abs/pii/S1352231006005723.
6
-------
concentrations in recent years across much of the United States further reduces the probability of
future exceedances.10'11
Additionally, to the extent that the air agency is submitting a second 10-year maintenance plan
for PM2.5, a record showing that the area design value is lower than the CDV, coupled with air
quality data demonstrating the area has already been maintaining the NAAQS for at least 8 years,
provides EPA with further confidence that the area will continue to maintain the relevant PM2.5
standard.
Example Site Calculation: Comparing Average Design Value to the Critical Design Value
The following is an example calculation of the ADV for a single monitoring site in a
hypothetical 24-hour PM2.5 nonattainment area, and comparison to the site's CDV. In calculating
the ADV for a site, EPA recommends using the most recent 5 consecutive 3-year design values
to better account for variability of air quality data in a particular location. The air agency should
perform this calculation for the site in the area that commonly has the highest design value, and
for all other active monitoring sites. Notwithstanding consideration of other factors, the EPA
believes it would be appropriate to approve an LMP only when the ADV is less than the
associated CDV for each site in the area.
EQUATIONS
Critical Design Value: CDV = NAAQS / (l+(tcxCV))
Coefficient of Variation: CV = (standard deviation for sample / average design value) = o/ADV
VARIABLES
NAAQS (|ig/m3): Level of relevant annual or 24-hour PM2.5 NAAQS
tc (Critical t-value): 1.53312
YEARS DESIGN VALUES FOR SITE (in iig/1113)
2015-2017 17
2016-2018 14
2017-2019 13
2018-2020 15
2019-2021 18
Avg Design Value (ADV) =15.4
10 See https://www.epa.gov/air-trends.
11 Elizabeth A.W. Chan, Brett Gantt, Stephen McDow, 2018. The reduction of summer sulfate and switch from
summertime to wintertime PM25 concentration maxima in the United States, Atmospheric Environment, Volume
175, 2018, Pages 25-32. www.sciencedirect.com/science/article/pii/S1352231017308166.
12 The critical t-value ofl.533isbasedonan ADV calculation using five consecutive 3 -year design values and the
one-tail Student's t-distribution at a significance level of 0.10. If only three 3-year design values are used to
calculate the ADV, the critical t-value would be 1.886.
7
-------
CDV CALCULATION
24-hr NAAQS (|ig/m3) 35
ADV (|ig/m3) 15.4
g (std. deviation for sample) 2.07
CV = o/ADV = (2.07 / 15.4) = 0.13
CDV (|ig/m3) = 35 / (1+(1.533*0.13)) = 29.0
ADV < CDV? YES
2.2 Event-Influenced air quality data
The EPA's Exceptional Events Rule13 implements CAA section 319(b)(2), which requires the
Administrator to promulgate regulations "governing the review and handling of air quality
monitoring data influenced by an exceptional event." Pursuant to CAA section 319(b)(3)(B)(iv),
the Exceptional Events Rule provides "criteria and procedures for the Governor of a state to
petition the Administrator to exclude air quality monitoring data that is directly influenced by
exceptional events from use in determinations by the Administrator with respect to exceedances
or violations of the national ambient air quality standards [(NAAQS)]." The Rule specifies the
types of actions that qualify as "determinations by the Administrator" and therefore must follow
the process and requirements in the Exceptional Events Rule, but the Rule also identifies that it
may be appropriate to exclude atypical or unrepresentative data for other types of actions that do
not qualify as "determinations by the Administrator."
In April of 2019, EPA expanded on this concept by releasing the Additional Methods,
Determinations, and Analyses to Modify Air Quality Data Beyond Exceptional Events
(AdditionalMethods) guidance, which clarifies the types of regulatory determinations, actions
and analyses, including LMPs, for which EPA may consider certain modified air quality
monitoring data.14 The Additional Methods guidance supersedes any related prior approach for
data exclusion identified in the 2001 PMio LMP Guidance, and is the appropriate data exclusion
guidance to apply in the context of this PM2.5 LMP Guidance. Specifically, the Additional
Methods guidance indicates that atypical or unrepresentative monitoring data could qualify for
exclusion for use in calculating air quality design values in support of an LMP submission and any
subsequent yearly design value calculations for areas with approved LMPs. The Additional Methods
guidance identifies that air quality monitoring data above the NAAQS-specific LMP threshold will
be treated in a manner analogous to the treatment of exceedance data under the Exceptional Events
Rule provided the impacted data otherwise satisfy the general definition and criteria for exceptional
events. Because the PM2.5 LMP Guidance does not provide a NAAQS-specific LMP threshold, air
agencies are strongly encouraged to consult with their EPA Regional office counterparts where
exceptional/atypical events-related questions arise in the context of an LMP prior to investing
significant resources in developing exceptional events-like analyses.
13 The Exceptional Events Rule was last revised by EPA in 2016. See 81 FR 68216 (Oct. 3, 2016).
14 See Additional Methods, Determinations, and Analyses to Modify Air Quality Data Beyond Exceptional Events
(Apr. 4, 2019), available at www.epa. gov/sites/defanlt/fi tes/20.1.9-
04/documents/clarifi cation_memo_on_data_modification_methods.pdf.
8
-------
2.3 Air Quality Review
As is the case for any maintenance plan, the LMP is expected to identify how the air agency
intends to track the progress of the maintenance plan. Consistent with the PMio LMP Guidance,
an air agency may do its periodic progress tracking by regularly recalculating the ADV (average
of 5 consecutive 3-year design values) for all the sites with complete data in the area, and
determining if the ADV is still less than the CDV for each site. Under this approach, if the air
agency determines that the ADV is not less than the CDV for all sites, the air agency should take
appropriate, early action to identify approaches to address the air quality trend and prevent a
violation of the NAAQS. Should a violation of the NAAQS occur, EPA may also use its
authority under the CAA to take actions necessary to ensure the area comes back into attainment.
9
-------
Section 3: Transportation Conformity
Transportation conformity is required under CAA section 176(c) (42 U.S.C. 7506(c)) to ensure
that federally funded or approved highway and transit activities are consistent with ("conform
to") the purpose of the SIP. Conformity to the purpose of the SIP means that transportation
activities will not cause or contribute to new air quality violations, worsen existing violations, or
delay timely attainment of the relevant NAAQS or any interim milestones.
The transportation conformity regulations (40 CFR Part 93, subpart A) establish criteria and
procedures for determining whether metropolitan transportation plans, transportation
improvement programs (TIPs), and federally supported highway and transit projects conform to
the SIP. These regulations provide for some flexibility when EPA has established an LMP policy
for a given NAAQS and pollutant, as explained in a previous EPA transportation conformity
rulemaking15 and the current transportation conformity regulations at 40 CFR 93.109(e). This
guidance establishes EPA's LMP policy for the PM2.5 NAAQS. The transportation conformity-
related portions of the attached PM10 LMP Guidance do not apply for PM2.5 transportation
conformity unless otherwise indicated.
The transportation conformity regulations require that:
A limited maintenance plan would have to demonstrate that it would be unreasonable to
expect that such an area would experience enough motor vehicle emissions growth for a
NAAQS violation to occur.16
As described above, a PM2.5 LMP may be submitted for a first and/or second 10-year
maintenance plan with documentation that supports the LMP demonstration described under the
transportation conformity regulations. The following are examples of how such an LMP
demonstration could be developed to address section 93.109(e) of the transportation conformity
regulations for a given area:
• As discussed above, an LMP for the first maintenance plan may be appropriate in isolated
rural areas or in smaller metropolitan areas where the PM2.5 air quality problem is due to
a specific source or sources unrelated to on-road transportation emissions (see footnote
6). Therefore, an LMP submission for an area's first maintenance plan should address, in
addition to air quality data trends, factors affecting the area's on-road mobile source
challenges, including its size, whether it includes a metropolitan planning organization,
its main sources of PM2.5 emissions, and its historical and projected vehicle miles
travelled (VMT).
• As noted in Section 1, an LMP may be particularly appropriate for a second maintenance
plan, as the area will have demonstrated attainment of the PM2.5 NAAQS for at least 8
years. To meet the requirement in the transportation conformity regulation, i.e.,
demonstrate that it would be unreasonable to expect that the area would experience
enough motor vehicle growth for a NAAQS violation to occur, an LMP submission for
15 See 69 FR 40063, July 1, 2004.
16 See 40 CFR 93.109(e).
10
-------
an area's second maintenance plan should again address the area's PM2.5 air quality
trends and its historical and projected VMT.
Finally, if emissions of re-entrained road dust have been found to be significant for PM2.5
transportation conformity purposes under 40 CFR 93.102(b)(3), e.g., those emissions have been
included in regional emissions analyses as part of transportation conformity determinations, then
the LMP submission from the air agency should also include an on-road PM2.5 emission analysis
consistent with the methodology in Attachment B of the PM10 LMP Guidance. EPA
acknowledges that this on-road emission analysis will not be needed for first or second LMP
submissions for most PM2.5 areas based on EPA's implementation of the PM2.5 NAAQS to date.
If the on-road emissions analysis is necessary, the LMP submission should only include on-road
emissions of direct PM2.5 (tailpipe, brake wear, tire wear and re-entrained road dust). As
discussed in Section 2.1 of this document, the concept of broadly applicable LMP air quality
concentration criteria for the annual and 24-hour PM2.5 NAAQS ("margins of safety") is not
included in this guidance. Therefore, when performing such an onroad emissions analysis, the air
agency should use the CDV for the area rather than the "margin of safety." If the onroad PM2.5
emissions analysis is required, the air agency must show that for each monitoring site in the area,
the ADV plus the on-road emissions growth estimate does not exceed the CDV.
The transportation conformity interagency consultation process must also be used to discuss the
development of any LMP submission.17 EPA Regional SIP and transportation conformity staff
will work together and provide technical assistance as needed for this component of the PM2.5
LMP.
Where an area has an adequate18 or approved PM2.5 LMP developed under this guidance, a
transportation plan or TIP conformity determination would not include a regional emissions
analysis for that PM2.5 NAAQS.19 However, transportation plan and TIP conformity
determinations that meet applicable requirements continue to be required in these areas (see
Table 1 in 40 CFR 93.109). The existing requirement for a regional emissions analysis also
continues to apply for any other pollutants or standards for which transportation conformity
applies in the area but which are not the subject of an LMP (40 CFR 93.109). In addition,
project-level conformity determinations must continue to be completed according to all
applicable requirements for federally supported highway and transit projects, including the hot-
spot requirements for projects in CO, PM10 and PM2.5 nonattainment and maintenance areas.20
17 See 40 CFR 93.105(b).
18 EPA's adequacy process is described in 40 CFR 93.118(e) and (f) with EPA's adequacy website at:
www.epa.gov/state-and-local-tEinsportation/adeanacY-review-state-implementation-plan-sip-snbmissions-
conformity.
19Per40 CFR 93.109(e): "Notwithstanding the other paragraphs of this section, an area is not required to satisfy the
regional emissions analysis for § 93.118 and/or 93.119 for a given pollutant and NAAQS, if the area has an adequate
or approved limited maintenance plan for such pollutant and NAAQS."
20 See 40 CFR 93.109(e) (providing that, in areas with limited maintenance plans, a "conformity determination that
meets other applicable criteria in Table 1 of [40 CFR 93.109(b)] is still required, including the hot-spot requirements
for projects in CO, PM10, and PM2 5 areas"). See also EPA's guidance for transportation conformity hot-spot
analyses available on EPA's website at: https://www.epa.gov/state-and-locai-transportation/proiect-levei-
co nfo nn i tv -and-ho t-spot -analv ses.
11
-------
Section 4: General Conformity
EPA's general conformity regulations do not distinguish between maintenance areas with an
approved "full maintenance plan" and those with an approved LMP. Thus, maintenance areas
with an approved LMP are subject to the same general conformity requirements under 40 CFR
part 93, subpart B, as those covered by a "full maintenance plan." No statements included
elsewhere in this guidance or in the PMio LMP Guidance should be construed to require
anything less than full compliance with the general conformity program requirements.
12
-------
Appendix: PMio LMP Guidance (2001)
13
-------
UNHID STATES fcNVIRONME.NTAl PROTECTION AGENCY
RcStARCH THIANGlt PAR* NC 2771 1
klb
Of-fiCl OF
am uuauty p: anmnc-
ANDS'ANDA^DS
SUBJrCT: Limited Maintenance Plan Option for Moderate PM,,, Nonattainment Areas
FROM: "'Lydia Wegman. Director
AQSSD (MD-15)
TO Director, Office of Ecosystem Protection. Region 1
Director. Division of Environmental Planning & Protection. Region II
Director. Air Protection Division, Region Hi
Director, Air, Pesticides & Toxics Management Division, Region iV
Director. Air and Radiation Division. Region V
Director. Air Pesticides & Toxics, Region VI
Direclor, Air and Toxics Division, Regions VII, IX
Director. Air Program. Region VIII
Director. Office of Air Quality, Region X
i WMiMiimiMMiillSlSieslM
This memorandum sets forth new guidance1 on maintenance plan submissions for certain
moderate particulate matter ( PM i nonaftainmen? areas seeking redesignaiiun to attainment, (see
section IV for further details on qualifying for the policy ). If the area meets the criteria listed in
this policy the State may submit a maintenance plan at the time it is requesting redesignation that
is more streamlined than would ordinarily be permitted. This new option is being termed a
limited maintenance plan (LMPr.
IL Win is there a need for a limited maintenance plan policy ?
Before the U.S. Court of Appeals for the District of Columbia handed down its decision
vacating the 1997 PM;, national ambient air quality standards (NAAQSRsee American Trucking
Associations, et al. v. Environmental Proieciion Agency (EPA). 175 F.3d 1027 (D.C, Cir. 1999).
;,S32Z
' »*
MEMORANDUM
"This memorandum is intended to provide EPA's preliminary views on how certain moderate PM10 nonattainmcnt
areas ma> qualify lo submit a maintenance plan that meets certain limited requirements. Since it represents only the Agency's
preliminary thinking that is subject to modification, this guidance is not binding on Stales. Tribes, the public, or EPA. Issues
concerning the applicability o! the limited maintenance plan policv will be addressed in actions to redesignate moderate PM10
nonattainment areas under § 107 of the CAA. It is only when EPA promulgates redesignations applying this policy that those
determinations will become binding on States. Tribes, the public, and EPA as a matter of law,
"Moderate PM,. aieas that do not meet the applicability criteria of this poltcv. and all serious PM,„ nonattainmcnt
areas, should submit maintenance plans that meet our guidance for submission of a full maintenance plan as described in the
September 4. 1992 memorandum. ""Procedures for Processing Requests to Redesignate Areas to Attainment." from John
Cakagin. former Director of the Office of Air Qualn> Planning and Standards iQAQPS ) Air Qualm management D» !<>ion to the
Regioiwl Air Division Directors (hereafter known as the Calcagni Memo)
14
-------
we were prepared to make case-by-case determinations that would make the 1987 PMio NAAQS
no longer applicable in any area meeting the standards. In taking actions to remove the
applicability of the 1987 NAAQS, we would have removed, as well, the nonattainment
designation and Clean Air Act (CAA) part D requirements from qualifying areas. As a result of
the D.C. Circuit's decision, for areas subject to the 1987 NAAQS, the only route to recognized
attainment of the NAAQS and removal of nonattainment status and requirements is formal
redesignation to attainment, including submittal of a maintenance plan. Since many areas have
been meeting the PMio NAAQS for 5 years or more and have a low risk of future exceedances,
we believe a policy that would allow both the States and EPA to redesignate speedily areas that
are at little risk of PMio violations would be useful.
III. How did EPA develop the approach used in the LMP option?
The EPA has studied PMio air quality data information for the entire country over the
past eleven years (1989-1999) and has determined that some moderate PMio nonattainment areas
have had a history of low PMIO design values with very little inter-annual variation. When we
looked at all the monitoring sites reporting data for those years, the data indicate that most of the
average design values fall below 2 levels, 98 |ig/m3 for the 24-hr PMio NAAQS and 40 |ig/m3 for
the annual PMio NAAQS. For most monitoring sites these levels are also below their individual
site-specific critical design values (CDV). The CDV is an indicator of the likelihood of future
violations of the NAAQS given the current average design value and its variability. The CDV is
the highest average design value an area could have before it may experience a future
exceedance of the NAAQS with a certain probability. A detailed explanation of the CDV is
found in Attachment A 3 to this policy which, because of its length, is a separate document
accompanying this memorandum.
We believe that the very small amount of variation between the peaks and means in most
of the data indicates a very stable relationship that can be reasonably expected to continue in the
future absent any significant changes in emissions. The period we assessed provides a fairly
long historical record and the data could therefore be expected to have been affected by a full
range of meteorological conditions over the period. Therefore, the amount of emissions should
be the only variable that could affect the stability in the air quality data. We believe we can
reliably make estimates about the future variability of PMio concentrations across the country
based on our statistical analysis of this data record, especially in areas where the amount of
emissions is not expected to change.
IV. How do I qualify for the LMP option ?
To qualify for the limited maintenance plan option, an area should meet the following
applicability criteria. The area should be attaining the NAAQS and the average PMio design
3 Dr. Shao-Hang Chu's paper entitled "Critical Design Value and Its Applications" explains the CDV
approach and is included in its entirety in Attachment A. This paper has been accepted for publication and
presentation at the 94th Air and Waste Management Association (A&WMA) Annual Conference in June 2001 in
Orlando, Florida.
15
-------
value 4 for the area, based upon the most recent 5 years of air quality data at all monitors in the
area, should be at or below 40 ug/m3 for the annual and 98 |ig/m3 for the 24-hr PMio NAAQS
with no violations at any monitor in the nonattainment area 5. If an area cannot meet this test it
may still be able to qualify for the LMP option if the average design values of the site are less
than their respective site-specific CDV.
We believe it is appropriate to offer this second method of qualifying for the LMP
because, based on the air quality data we have studied, we believe there are some monitoring
sites with average design values above 40 |ig/m3 or 98 |ig/m3, depending on the NAAQS in
question, that have experienced little variability in the data over the years. When the CDV
calculation was performed for these sites we discovered that their average design values are less
than their CDVs, indicating that the areas have a very low probability (1 in 10) of exceeding the
NAAQS in the future. We believe it is appropriate to provide these areas the opportunity to
qualify for the LMP in this circumstance since the 40 |ig/m3 or 98 |ig/m3 criteria are based on a
national analysis and don't take into account each local situation.
The final criterion is related to mobile source emissions. The area should expect only
limited growth in on-road motor vehicle PMio emissions (including fugitive dust) and should
have passed a motor vehicle regional emissions analysis test. It is important to consider the
impact of future transportation growth in the LMP, since the level of PM-10 emissions
(especially from fugitive dust) is related to the level of growth in vehicle miles traveled (VMT).
Attachment B (below) should be used for making the motor vehicle regional emissions analysis
demonstration.
If the State determines that the area in question meets the above criteria, it may select the
LMP option for the first 10 year maintenance period. Any area that does not meet these criteria
should plan to submit a full maintenance plan that is consistent with our guidance in the Calcagni
Memo in order to be redesignated to attainment. If the LMP option is selected, the State should
continue to meet the qualifying criteria until EPA has redesignated the area to attainment. If an
area no longer qualifies for the LMP option because a change in air quality affects the average
design values before the redesignation takes effect, the area will be expected to submit a full
maintenance plan.
Once an area selects the LMP option and it is in effect, the State will be expected to
recalculate the average design value for the area annually and determine if the criteria used to
qualify for the LMP will still be met. If, after performing the annual recalculation of the area's
average design value in a given year, the State determines that the area no longer qualifies for the
LMP, the State should take action to attempt to reduce PMio concentrations enough to requalify
for the LMP. One possible approach the State could take is to implement a contingency measure
4 The methods for calculating design values for PMio are presented in a document entitled the "PMio SIP Development
Guideline", EPA-450/2-86-001, June 1987. The State should determine the most appropriate method to use from this Guideline
in consultation with the appropriate EPA Regional office staff.
5 If the EPA determines that the meteorology was not representative during the most recent five-year period, we may
reject the State's request to use the LMP option and request, instead, submission of a full maintenance demonstration.
16
-------
or measures found in its SIP. If, in the next annual recalculation the State is able to re-qualify for
the LMP, then the LMP will go back into effect. If the attempt to reduce PMio concentrations
fails, or if it succeeds but in future years it becomes necessary again to address increasing PMio
concentrations in the area, that area no longer qualifies for the LMP. We believe that repeated
increases in PMio concentrations indicate that the initial conditions that govern air quality and
that were relied on to determine the area's qualification for the LMP have changed, and that
maintenance of the NAAQS can no longer be assumed. Therefore, the LMP cannot be reinstated
by further recalculations of the design values at this point. Once the LMP is determined to no
longer be in effect, a full maintenance plan should be developed and submitted within 18 months
of the determination.
Treatment of data used to calculate the design values.
Flagged Particulate Matter Data:
Three policies allow PM-10 data to be flagged for special consideration:
Exceptional Events Policy (1986) for data affected by infrequent
events such as industrial accidents or structural fires near a
monitoring site;
Natural Events Policy (1996) for data affected by wildfires, high
winds, and volcanic and seismic activities, and;
• Interim Air Quality Policy on Wildland and Prescribed Fires for
data affected by wildland fires that are managed to achieve
resource benefits.
We will treat data affected by these events consistently with these
previously-issued policies. We expect States to consider all data
(unflagged and flagged) when determining the design value. The EPA
Regional offices will work with the State to determine the validity of
flagged data. Flagged data may be excluded on a case-by-case basis
depending on State documentation of the circumstances justifying flags.
Data flagged as affected by exceptional or natural events will generally
not be used when determining the design value. However, in order for
data affected by a natural event to be excluded, an adequate Natural
Events Action Plan is required as described in the Natural Events policy.
Data flagged as affected by wildland and prescribed fires will be used in
determining the design value. If the State is addressing wildland and
prescribed fire use with the application of smoke management programs,
the State may submit an LMP if the design value is too high only as a
result of the fire-affected data.
We are in the process of developing a policy to address agricultural
burning. When it is finalized we will amend the LMP option to account
for the new policy.
17
-------
V.
What should an LMP consist of?
Under the LMP, we will continue to satisfy the requirements of Section 107(d)(3)(E) of
the Act which provides that a nonattainment area can be redesignated to attainment only if the
following criteria are met:
1. The EPA has determined that the NAAQS for the applicable pollutant has been
attained.
2. The EPA has fully approved the applicable implementation plan under section
110(k).
3. The EPA has determined that the improvement in air quality is due to permanent
and enforceable reductions in emissions.
4. The State has met all applicable requirements for the area under section 110 and
partD.
5. The EPA has fully approved a maintenance plan, including a contingency plan,
for the area under section 175 A.
However, there are some differences between what our previous guidance (the Calcagni
memo) recommends that States include in a maintenance plan submission and what we are
recommending under this policy for areas that qualify for the LMP. The most important
difference is that under the LMP the demonstration of maintenance is presumed to be satisfied.
The following is a list of core provisions which should be included in an LMP submission. Note
that any final EPA determination regarding the adequacy of an LMP will be made following
review of the plan submitted in light of the particular circumstances facing the area proposed for
redesignation and based upon all available information.
a. Attainment Plan
The State's approved attainment plan should include an emissions inventory (attainment
inventory) which can be used to demonstrate attainment of the NAAQS. The inventory should
represent emissions during the same five-year period associated with the air quality data used to
determine whether the area meets the applicability requirements of this policy (i.e., the most
recent five years of air quality data). If the attainment inventory year is not one of the most
recent five years, but the State can show that the attainment inventory did not change
significantly during that five-year period, it may still be used to satisfy the policy. If the
attainment inventory is determined to not be representative of the most recent 5 years, a new
inventory must be developed. The State should review its inventory every three years to ensure
emissions growth is incorporated in the attainment inventory if necessary.
b. Maintenance Demonstration
The maintenance demonstration requirement of the Act will be considered to be satisfied
for the moderate PMio nonattainment areas meeting the air quality criteria discussed above. If
the tests described in Section IV are met, we will treat that as a demonstration that the area will
maintain the NAAQS. Consequently, there is no need to project emissions over the maintenance
period.
18
-------
c. Important elements that should be contained within the redesignation request
1. Monitoring Network Verification of Continued Attainment
To verify the attainment status of the area over the maintenance period,
the maintenance plan should contain a provision to assure continued
operation of an appropriate, EPA-approved air quality monitoring
network, in accordance with 40 CFR part 58. This is particularly
important for areas using an LMP because there will be no cap on
emissions.
2. Contingency Plan
Section 175 A of the Act states that a maintenance plan must include
contingency provisions, as necessary, to promptly correct any violation of
the NAAQS which may occur after redesignation of the area to
attainment. These contingency measures do not have to be fully adopted
at the time of redesignation. However, the contingency plan is considered
to be an enforceable part of the SIP and the State should ensure that the
contingency measures are adopted as soon as possible once they are
triggered by a specific event. The contingency plan should identify the
measures to be adopted, and provide a schedule and procedure for
adoption and implementation of the measures if they are required.
Normally, the implementation of contingency measures is triggered by a
violation of the NAAQS but the State may wish to establish other triggers
to prevent a violation of the NAAQS, such as an exceedance of the
NAAQS.
19
-------
3. Approved attainment plan and section 110 and part D CAA
requirements:
In accordance with the CAA, areas seeking to be redesignated to
attainment under the LMP policy must have an attainment plan that has
been approved by EPA, pursuant to section 107(d)(3)(E). The plan must
include all control measures that were relied on by the State to
demonstrate attainment of the NAAQS. The State must also ensure that
the CAA requirements for PMio pursuant to section 110 and part D of the
Act have been satisfied. To comply with the statute, the LMP should
clearly indicate that all controls that were relied on to demonstrate
attainment will remain in place. If a State wishes to roll back or
eliminate controls, the area can no longer qualify for the LMP and the
area will become subject to full maintenance plan requirements within 18
months of the determination that the LMP is no longer in effect.
VI. How is Conformity treated under the LMP option?
The transportation conformity rule (40 CFR parts 51 and 93) and the general conformity
rule (58 FR 63214; November 30, 1993) apply to nonattainment areas and maintenance areas
operating under maintenance plans. Under either conformity rule one means of demonstrating
conformity of Federal actions is to indicate that expected emissions from planned actions are
consistent with the emissions budget for the area. Emissions budgets in LMP areas may be
treated as essentially not constraining for the length of the maintenance period because it is
unreasonable to expect that an area satisfying the LMP criteria will experience so much growth
during that period of time such that a violation of the PMio NAAQS would result. While this
policy does not exempt an area from the need to affirm conformity, it does allow the area to
demonstrate conformity without undertaking certain requirements of these rules. For
transportation conformity purposes, EPA would be concluding that emissions in these areas need
not be capped for the maintenance period, and, therefore, a regional emissions analysis would not
be required. Similarly, Federal actions subject to the general conformity rule could be
considered to satisfy the "budget test" specified in section 93.158 (a)(5)(i)(A) of the rule, for the
same reasons that the budgets are essentially considered to be unlimited.
EPA approval of an LMP will provide that if the LMP criteria are no longer satisfied and
a full maintenance plan must be developed to meet CAA requirements (see Calcagni Memo
referenced in footnote #2 for full maintenance plan guidance), the approval of the LMP would
remain applicable for conformity purposes only until the full maintenance plan is submitted and
EPA has found its motor vehicle emissions budgets adequate for conformity purposes under 40
CFR parts 51 and 93. EPA will condition its approval of all LMPs in this fashion because in the
case where the LMP criteria are not met and a full maintenance plan is required EPA believes
that LMPs would no longer be an appropriate mechanism for assuring maintenance of the
standards.
20
-------
ATTACHMENT A
Critical Design Value Estimation and Its Applications
Shao-Hang Chu
US Environmental Protection Agency
Office of Air Quality Planning and Standards
Air Quality Strategies and Standards Division (MD-15)
Research Triangle Park, NC 27711
ABSTRACT
The air quality design value is the mathematically determined pollutant concentration at a
particular site that must be reduced to, or maintained at or below the National Ambient Air
Quality Standards (NAAQS) in order to assure attainment. The design value may be calculated
based on ambient measurements observed at a local monitor in a 3-year period or on model
estimates. The design value, however, varies from year to year due to both the pollutant
emissions and natural variability such as meteorological conditions, wildfires, dust storms,
volcanic activities etc. In order to investigate certain polity options related to pollution controls
it would be desirable to estimate a critical design value above which the NAAQS is likely to be
violated with a certain probability.
In this paper, a statistical technique has been developed to estimate a critical design value that is
based on the average design value and its variability in the past. The critical design value could
be used as a planning tool for regulatory agencies because it is an indicator of the likelihood of
future violations of the NAAQS given the current average design value and its variability. The
approach is general and could be applied to estimate the critical design value for any pollutant.
As an example, elevenyears (1989-1999) of PM10 data nationwide were extracted from the US
EPA AIRS database to estimate the PM10 critical design values. The analyses indicate that
PM10 design values in the West have much larger inter-annual variability than those in the East
as reflected in their much lower critical design values. This, in turn, suggests that the interannual
variability in meteorology, wildfires, and dust storms may have played a more significant role in
the West, and also this larger variability could be partly explained by the once every six days
sampling schedule at most PM10 monitoring sites.
INTRODUCTION
The air quality design value is the mathematically determined pollutant concentration at a
particular site that must be reduced to, or maintained at or below the National Ambient Air
Quality Standards (NAAQS) in order to assure attainment1. The design value may be calculated
based on ambient measurements observed at a local monitor in a 3-year period or on model
estimates. The detailed calculation of the design values for various criteria pollutants is described
in the Appendices of the Code of Federal Regulations2. In certain cases, the design value has
been used for regulatory purposes to determine whether the local pollutant concentration has
violated the National Ambient Air Quality Standard (NAAQS). Most often, however, the design
21
-------
value is used to determine the level of control needed to reduce the pollutant concentration to the
NAAQS3,4,5.
The design value, however, varies from year to year due to both the pollutant emissions and
natural variability such as meteorological conditions, wildfires, dust storms, volcanic activities
etc. In order to investigate certain policy options related to pollution controls it would be
desirable to define a critical design value above which future violations of the air quality
standard are likely to occur with a certain probability.
In this paper, an effort has been made to statistically estimate a critical design value based on the
average of these yearly design values and their variability in the past. This critical design value
is defined in such a way as it is the highest average design value any monitoring site could have
before it runs a risk of violating the NAAQS in the future at a certain probability. The technical
basis of this estimation approach and its applications will be discussed in the following
paragraphs.
CRITICAL DESIGN VALUE ESTIMATION
Our intention is to find a critical design value (CDV) that is the highest possible average design
value (ADV) any site could have before it risks a future violation of the standard at a certain
probability. First, we try to formulate a relationship among a set of variables involved: such as
the CDV, NAAQS, the ADV, the standard deviation of the design values in the past, and a
desirable risk factor. We find that if we assume that the design values are normally distributed
and the coefficient of variation (CV), which is the ratio of the standard deviation versus the mean
of the design values, does not change in the near future, then we can write the relationship as:
CDV = NAAQS/(l+tc*CV) (1)
Where CDV is the critical design value, CV is the coefficient of variation of the annual design
values (the ratio of standard deviation divided by the mean design value in the past), and tc is the
critical t-value corresponding to a probability, c %, of exceeding the NAAQS in the future and
the degree of freedom in the estimate to the CV. Equation (1) says that based on the variability
of the design values in the past, the probability of any monitoring site with an ADV less than or
equal to the CDV to exceed the NAAQS in the future would be no more than c % given the same
CV. In other words, the CDV is the highest ADV any monitoring site could have before it may
record a future violation of the NAAQS with a certain probability. The percent probability, c, is
the chosen risk factor. One can choose either a more, or less, conservative c value depending on
how much risk one is willing to take.
The inter-annual variability of the air quality design values at a monitoring site can be estimated
from historical data at that station. Using the air quality data in the past, one can calculate the
design values for each year. With these design values one can calculate the ADV and its
22
-------
variability in terms of the coefficient of variation (CV). Thus, one can calculate the CDV for any
site with a minimum of five years of data.
CHARACTERISTICS OF THE CRITICAL DESIGN VALUE
From equation (1) we see that the CDV is a nonlinear function of the NAAQS of the pollutant,
the critical t-value, to, and the coefficient of variation, CV, of the design values. The normalized
CRITICAL DESIGN VALUE FUNCTION
CWB HMQS/lltm). DniD
relationship of the CDV to the product of to and CV is shown in figure 1.
Figure 1.
The dependency of CDV on the other two variables can be summarized as:
1. The larger the variability (CV) of the design values in the past, the smaller the CDV will
be;
2. The lower the probability of risk for future violations (PX), the lower the CDV will be;
3. If CV=0, i.e., no variability in the design values in the past, then from Figure 1 and
Equation (1) we find the highest CDV equal to the NAAQS;
4. As CV increases, the CDV approaches zero;
5. If CV is not zero but to = 0, then we will also have a CDV equal to the NAAQS, but it
will have a 50% chance of violating the standard in the future because to = 0 corresponds
to a probability of 50%.
In Figure 2 we have chosen a risk factor of 10% probability of future violation and plotted two
examples using generated data with significantly different variability in the annual PM10 design
values. It is intended to illustrate the relationship among design values, ADV, CDV, and the
PM10 annual NAAQS of 50 ug/m3. In this example we see that the CDV depends strongly on
the inter-annual variability of the design values rather than on their means. Also, from the upper
23
-------
panel of Figure 2 we see that once the ADV is higher than the CDV, the probability of violating
the standard will be higher than the risk we have chosen (in this case, it is one out of ten).
Figure 2.
Contrasting the two panels of Figure 2, we see that whether a site will have a higher or lower risk
of violating the NAAQS in the future depends on how much higher or lower the ADV is to the
CDV. Thus, unless some drastic change in emissions occurred in the past or should occur in the
future, the CDV can be used to assess the likelihood of violating the NAAQS in the future in that
area based on normal probability predictions. For this reason, this technique and the estimated
24
-------
CDV could be used as a planning tool for regulatory agencies to decide whether more or fewer
pollutant controls are needed in a specific area.
PM10 CRITICAL DESIGN VALUES AND DISCUSSIONS
To demonstrate this approach, eleven years (1989-1999) of PM10 data nationwide were
extracted from the United States Environmental Protection Agency AIRS database. The annual
and 24-hr PM10 design values were calculated following the US EPA Guidance1. Then the
methodology described in the previous section was applied using a tolerable risk factor of 10%
probability of future violation of the NAAQS to calculate the CDVs for all monitor sites with
more than five years of valid data The analyses are discussed and presented in the following
figures.
Figure 3 is a frequency distribution of these calculated annual and 24-hr CDVs. We see that the
distributions of both the annual and the 24-hr CDVs are skewed to the left with a median annual
CDV of 45.3 ug/m3 and a median 24-hr CDV of 123.2 ug/m3. The long tails to the left (low
values) suggest that there are places where the inter-annual variability of the design values are
quite large. It also suggests that these areas are likely to have a higher probability of violating
the standards if they are already in a major PM10 source region with relatively high PM10
concentrations.
In Figure 4 a longitudinal scatter plot of both the ADVs and the CDVs at all sites spanning from
Maine to California, was produced to see whether there is a difference from the East to the West.
Comparing the differences between these overlaid ADVs and CDVs we see clearly that most of
the higher risk areas (i.e., the areas where the ADVs are greater than the CDVs) are in the West
and Midwest. The geographical distribution of the CDVs and the actual ADVs are shown in
Figures 5 and 6 respectively. For comparison purposes, the ADVs in Figure 6 are color coded to
show their probability of future violation of the NAAQS. The probability of future violation of
the NAAQS at each site is calculated by inverting the t-values using equation (1).
The East-West difference in CDVs can be explained largely by the fact that the West, in general,
has a much larger inter-annual variability of the design values than the East. However, since the
anthropogenic emissions in a region usually do not change very much from year to year, the
large variability in the inter-annual PM10 design values in the West may be largely attributable
to the inter-annual variation in natural conditions such as meteorology, wildfires, dust storms,
and volcanic emissions, etc. The higher occurrences of wildfires and dust storms in the West are
known to be associated with its much drier climate, meteorological conditions, and topography.
Another influencing factor on the inter-annual variability could be related to the sampling
frequency of the PM10 data, which for many sites is only once every six days. However, this is
more likely in the East because fewer sites are in non-attainment status and thus not required to
sample more frequently than once in six days.
25
-------
3 or SITES
300 :
200-
100 :
DISTRIBUTION OF ESTIMATED CRITICAL ANNUAL PM10 DESIGN VALUES
BASED ON ARS DATA NATIONWIDE
t J
11
-KZL
10.0 20J3 29.D *5 32.0 3« 35.0 3S5 3J) 315 41.0 «5 44.0 455 47.0 485 5DjO
CtynCALDES W VUJJF (UG/M3)
few
DISTRIBUTION Of ESTIMATED CRITICAL 24-HOUR MO DESIGN VALUES
BASED ON ARS IKTA NATBIWtt
NO. OF SITES
XD
ZW
100 -
' 1 1
30.0 575 4SuO Mi BOA 47.5 75J0 B2J 9D.D 575 T05jO 1115 12M 127.5 155J0 142.5 150.0
CRITICAL DTO VALUE (Ufi/»)
Figure 3.
26
-------
I I I
LONGITUDINAL DISTRIBUTION OF AVERAGE PM ANNUAL DV AND THEIR CDV
AIRS DATA [IBS-IMS)
SO
70
!¦
3 so
h
I-
0
o 20
o
10
0
»•» AVfiw
critov
i 1 i 1 i 1 i 1 i 1 i 1 r i 1 i 1 i 1 i 1 i ¦ i
IS 120 115 110 105 IW * 9D 85 AO 75 70 55
MCNUHR L0CAIXM (DEC K)
LONGITUDINAL DISTRIBUTION OF AVERAGE PM 24-HR DV AND THEIR CDV
AIRS DATA (I9W-1999)
550
£00
550
3 500
\ 450
0
3 400
\ 550
< 300
E
Z 250
y
o 200
§ 150
100
SO
»•» AVSW
CRUDV
I I I I I i I I I I I I I
125 120 115 110 105 100 95 90 » 50 75 70 65
MONITOR IjOCttCK (MS 10
Figure 4.
27
-------
I I I
PWta AVEIUSE ANNUAL CRniCAlDESICN VALUE COMCENIKATHN
HLCUUID) FHI Ml DATA (IKI -1 lit)
REE6<«.H]ITU:4l>-«,Ul.«-<5.W«K >-<5
PH1DAVERAEC 24-HR CpmCALDESBN VALUE COHCEWRfflCN
OLWUlttJ (KM AID Wk (19® - 19*)
Kts=!»
Figure 5.
28
-------
I I I
COIOH CODEC PMIO IWWkSE ANNUAL DESIGN VALUE COWMTRMIOH
ouiiis nownmr w rcsjujm hwk due fume
Kit W7tx HIIPLE: 10-7®!, BLUE: 10-KB, (UCft <1K
COLOR CODED FM10 AVERAGE 24-HR DESIGN VALVE CONCENMTON
conn® n»M»aiiHOfDraicinEiiuQS hihefwure
IBs^W.PUmi O-7H7.BUIL10-4IR, BLWfcOR
Figure 6.
29
-------
CONCLUSIONS
In this paper a statistical technique has been developed to determine the CDV which is the highest
possible average design value any monitoring site could have before it may record a future violation
of the NAAQS with a certain probability. The critical design value is calculated based on the average
design value and its variability in the past, and it also involves a risk factor of our choice in the
estimation. The difference between the ADV and CDV is a good indicator of whether the site is
running a higher or lower risk of violating the NAAQS in the future than one is willing to take. Using
this approach, one can even predict the probability of violating the NAAQS in the near future at any
given site with adequate data length. Thus, this technique could be used as a planning tool for
regulatory agencies to assess the risk of future violation of the NAAQS at any monitoring site and to
make decisions about emissions controls. Further, since this technique is very general, it can be
applied to any pollutant with a minimum of five years of valid data.
As an example, 11 years (1989-1999) of PM10 data were analyzed using this technique. The results
suggest that the inter-annual variability of the design values in the West is, on the average, much
larger than that in the East, which is reflected in the calculated CDVs. Since anthropogenic
emissions in a region usually do not change very much from year to year, the large variability in the
inter-annual PM10 design values in the West may be largely attributable to the inter-annual variation
in natural conditions such as meteorology, wildfires, dust storms, and volcanic activities, etc. The
higher occurrences of wildfires and dust storms in the West are known to be associated with its much
drier climate, meteorological conditions, and topography. The once every six days sampling practice
of PM10 monitoring may also have some influence on the inter-annual variability of PM10 design
values.
FUTURE WORK
Some further studies have been planned which include applying the same technique to other pollutants,
and searching for a better estimate of CV in case when significant trend exists in the yearly design
values. Since the variance estimate could be affected by an underlying trend and that a better estimate
could be made of the CV if the trend and/or serial correlation could be removed from the estimate.
ACKNOWLEDGMENTS
The author would like to give thanks to Dr. Terence Fitz-Simons and Mr. Bill Cox for reading the
draft manuscript and their helpful discussions.
REFERENCES
1. US EPA, PM10 SIP Development Guideline. EPA-450/2-86-001, US Environmental Protection
Agency, Research Triangle Park, NC 27711, 1986.
2. Code of Federal Regulations, Protection of Environment, 40 CFR part 50, 1998.
3. US EPA, Guideline on Air Quality Models (Revised). EPA-450/2-78-27 R. U.S. Environmental
Protection Agency, Research Triangle Park, NC 27711,1986.
4. US EPA, Guideline for the Interpretation of Ozone Air Quality Standards. EPA450/4-79-003, US
Environmental Protection Agency, Research Triangle Park, NC 27711, 1979.
5. Curran, T. C. and W.M. Cox, Data Analysis Procedures for the Ozone NAAQS Statistical Format. J.
Air Pollution Control Association, 1980.
30
-------
KEYWORDS
Critical design value, design value, inter-annual variability, PM10, probability
31
-------
ATTACHMENT B:
MOTOR VEHICLE REGIONAL ANALYSIS METHODOLOGY
The following methodology is used to determine whether increased emissions from on-road mobile
sources could, in the next 10 years, increase concentrations in the area and threaten the assumption of
maintenance that underlies the LMP policy. This analysis must be submitted and approved in order to
be eligible for the LM P option.
The following equation should be used:
DV + (VMTpi x DVmv) < MOS
Where:
DV = the area's design value based on the most recent 5 years of quality
assured data in (ig/m3
VMT = *he projected % increase in vehicle miles traveled (VMT) over the next
10 years motor vehicle design value based on on-road mobile portion
of the attainment year inventory in (ig/m3 margin of safety for the
DV™ ~ relevant PM-10 standard for a given area:
40 (ig/m3 for the annual standard or 98 (ig/m3 for the 24-hour standard
MOS =
Please note that DVmv is derived by multiplying DV by the percentage of the attainment year inventory
represented by on-road mobile sources. This variable should be based on both primary and secondary
PMio emissions of the on-road mobile portion of the attainment year inventory, including re-entrained
road dust.
States should consult with EPA regarding the three inputs used in the above calculation, and all EPA
comments and concerns regarding inputs and results should be addressed prior to submitting a limited
maintenance plan and redesignation request.
The VMT growth rate (VMTpi) should be calculated through the following methods:
1) an extrapolation of the most recent 10 years of Highway Performance Monitoring System
(HPMS) data over the 10-year period to be addressed by the limited maintenance plan; and
2) a projection of VMT over the 10-year period that would be covered by the limited maintenance plan,
using whatever method is in practice in the area (if different than #1).
Areas where method #1 is the current practice for calculating VMT do not also have to do calculation #2,
although this is encouraged. All other areas should use methods #1 and #2, and VMTpi is whichever
growth rate produced by methods #1 and #2 is highest. Areas will be expected to use transportation
models for method #2, if transportation models are available.
32
-------
Areas without transportation models should use reasonable professional practice.
Examples
1. DV 80 (ig/m3
wk.t 36%
VMTpi= , ,
_ 30 ng/m
L/ V mv — -3
98 (ig/m for 24-hour PM-10 standard
MOS
80+ (.36 *30) = 91
Less than 98 - Area passes regional analysis criterion.
2. DV = 35 (ig/m3
VMT^mvPi= =25406 (ig/m% (ig/m33 for annual PM-10 standard
MOS
35 + (.25 *6) = 37
Less than 40 - Area passes regional analysis criterion.
3. DV = 115 g/m3
VMT^mvPi= =256098% (ig(ig/m/m33 for 24-hour PM-10 standard
MOS
115 + (.25 * 60) = 130
More than 98 - Area does not pass criterion. Full section 175A maintenance plan required.
33
------- |