Industrial Waste
Management
-------
This Guide provides state-of-the-art tools and
practices to enable you to tailor hands-on
solutions to the industrial waste management
challenges you face.
WHAT'S AVAILABLE
• Quick reference to multimedia methods for handling and disposing of wastes
from all types of industries
• Answers to your technical questions about siting, design, monitoring, operation.
and closure of waste facilities
• Interactive, educational tools, including air and ground water risk assessment
models, fact sheets, and a facility siting tool.
• Best management practices, from risk assessment and public participation to
waste reduction, pollution prevention, and recycling
-------
;NOWLEDGEMENTS
The rdowing members of the Industrial Waste Focus Group and the Industrial Waste Steering Commiw are grateUy
acknowledged far al of their time and assistance in the development of this guidance document
Current Industrial Waste Focus
Group Members
Paul Bar*, The Dow Chemical
Company
Walter Carey. Nestle USA Inc and
New Miltord Farms
Rama Chaturvedi Bethlehem Steel
Corporation
H.C. Clark. Rice University
Barbara Dodds, League of Women
voters
Chuck Feerick. Exxon Mobil
Corporation
Stacey Ford. Exxon Mobil
Corporation
Robert Giraud OuPont Company
John Harney Citizens Round
Tabte/PURE
Kyle Isakower. American Petroleum
Institute
Richard Jarman, National Food
Processors Association
James Meiers, Cinergy Power
Generation Services
Scott Murto. General Motors and
American Foundry Society
James Roewer, Edison Electric
Institute
Edward Repa. Environmental
Industry Association
Tim Savior, International Paper
Amy Schaffer. Weyerhaeuser
Ed Skemofc, WMX Technologies. Inc
Michael Wach Western
Environmental Law Center
David Wens, University of South
Wabnms Medical Center
Pat Gwn Cherokee Nation of
Oklahoma
Past industrial Waste Focus
Group Members
Dora Cetofius. Sierra Club
Brian Forrestal. Laidlaw Waste
Systems
Jonathan Greenberg. Browning-
Ferris Industries
Michael Gregory, Arizona Toxics
Information and Sierra Club
Andrew Mites The Dexter
Corporation
Gary Robbins, Exxon Company
Kevin Sail. National Paint & Coatings
Association
Bruce SteJne. American Iron & Steel
Lisa Williams, Aluminum Association
Cuircnt Industrial Waste Steering
Committee Members
Keiiy Catalan Aaaocauon oi Slate
and Territorial Solid Waste
Management Officials
Marc Crooks, Washington State
Department ot Ecology
Cyndi Darling. Maine Department of
Environmental Protection
Jon DilDard Montana Department of
Environmental Qualty
Anne Dobbs. Texas Natural
Resources Conservation
Commission
Richard Hammond New York State
Department of Environmental
Conservation
Elizabeth Haven California State
Waste Resources Control Board
Jim Hul Missouri Department of
Natural Resources
Jim Knudson, Washington State
Department of Ecology
Chris McGuire, Florida Department
of Environmental Protection
Gene Mitchell Wisconsin
Department of Natural Resources
William Pounds, Pennsylvania
Department of Environmental
Protection
Bijan Sharafkhani Louisiana
Department of Environmental
Qualty
James Warner, Minnesota Pollution
Control Agency
ittustrial Waste Steering
Pamela um*. nianie
Environmental Protection
NormGumenik Arizona Department
of Environmental Qualty
Steve Jenkins, Alabama Department
of Environmental Management
Jim North Arizona Department of
Environmental Quality
-------
Industrial waste is generated by the production
of commercial goods, products, or services.
Examples include wastes from the production
of chemicals, iron and steel, and food goods.
-------
Friday
July 18, 1997
Part II
Environmental
Protection Agency
40 CFR Part 50
National Ambient Air Quality Standards
for Particulate Matter; Final Rule
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
ENVIRONMENTAL PROTECTION
AGENCY
40 CFR Part 50
[AD-FRL-5725-2]
RIN 2060-AE66
National Ambient Air Quality
Standards for Particulate Matter
AGENCY: Environmental Protection Agency
(EPA).
ACTION: Final rule.
SUMMARY: This document describes EPA's
decision to revise the national ambient air
quality standards (NAAQS) for particulate
matter (PM) based on its review of the
available scientific evidence linking
exposures to ambient PM to adverse health
and welfare effects at levels allowed by the
current PM standards. The current primary
PM standards are revised in several respects:
Two new PM2.5 standards are added, set at
15 |ig/m3 , based on the 3-year average of
annual arithmetic mean PM2.5 concentrations
from single or multiple community-oriented
monitors, and 65 |ig/m 3 , based on the 3-
year average of the 98th percentile of 24-hour
PM2.5 concentrations at each population-
oriented monitor within an area; and the
current 24-hour PMio standard is revised to
be based on the 99th percentile of 24-hour
PMio concentrations at each monitor within
an area. The new suite of primary standards
will provide increased protection against a
wide range of PM-related health effects,
including premature mortality and increased
hospital admissions and emergency room
visits, primarily in the elderly and individuals
with cardiopulmonary disease; increased
respiratory symptoms and disease, in children
and individuals with cardiopulmonary disease
such as asthma; decreased lung function,
particularly in children and individuals with
asthma; and alterations in lung tissue and
structure and in respiratory tract defense
mechanisms. The current secondary standards
are revised by making them identical in all
respects to the new suite of primary
standards. The new secondary standards, in
conjunction with a regional haze program,
will provide appropriate protection against
PM-related public welfare effects including
soiling, material damage, and visibility
impairment. In conjunction with the new
PM2.5 standards, a new reference method has
been specified for monitoring PM as PM2.5
EFFECTIVE DATD: This action is effective
September 16, 1997.
ADDRESSES: A docket containing
information relating to the EPA's review of
the PM primary and secondary standards
(Docket No. A-95-54) is available for public
inspection in the Central Docket Section of
the U.S. Environmental Protection Agency,
South Conference Center, Rm. 4, 401 M St.,
SW., Washington, DC. This docket
incorporates the docket established for the air
quality Criteria Document (Docket No.
ECAO-CD-92-0671). The docket may be
inspected between 8 a.m. and 3 p.m., Monday
through Friday, except legal holidays, and a
reasonable fee may be charged for copying.
The information in the docket constitutes the
complete basis for the decision announced in
this document. For the availability of related
information, see "SUPPLEMENTARY
INFORMATION."
FOR FURTHER INFORMATION CONTACT: John
H. Haines, MD-15, Air Quality Strategies
and Standards Division, Office of Air Quality
Planning and Standards, U.S. Environmental
Protection Agency, Research Triangle Park,
NC 27711; telephone: (919) 541-5533; e-
mail: haines.john@epamail.epa.gov.
SUPPLEMENTARY INFORMATION:
Related Final Rules on PM Monitoring
In a separate document published
elsewhere in this issue of the Federal
Register, EPA is amending its ambient air
quality surveillance requirements (40 CFR
part 58) and its ambient air monitoring
reference and equivalent methods (40 CFR
part 53) for PM.
Availability of Related Information
Certain documents are available from the
U.S. Department of Commerce, National
Technical Information Service, 5285 Port
Royal Road, Springfield, VA 22161.
Available documents include:
(1) Air Quality Criteria for Particulate
Matter (Criteria Document) (three volumes,
EPA/600/P-95-001aF thru EPA/600/P-95-
OOlcF, April 1996, NTJS #PB-96-168224,
$234.00 paper copy).
(2) Review of the National Ambient Air
Quality Standards for Particulate Matter:
Policy Assessment of Scientific and
Technical Information (Staff Paper) (EPA-
452/R-96-013, July 1996, NTIS #PB-97-
115406, $47.00 paper copy and $19.50
microfiche). (Add a $3.00 handling charge
per order.)
A limited number of copies of other
documents generated in connection with this
standard review, such as technical support
documents pertaining to air quality,
monitoring, and health risk assessment, can
be obtained from: Environmental Protection
Agency Library (MD-35), Research Triangle
Park, NC 27711, telephone (919) 541-2777.
These and other related documents are also
available for inspection and copying in the
EPA docket at the address under
"ADDRESSES," at the beginning of this
document.
Electronic Availability
The Staff Paper and human health risk
assessment support documents are available
on the Agency's Office of Air Quality
Planning and Standards' (OAQPS)
Technology Transfer Network (TTN) Bulletin
Board System (BBS) in the Clean Air Act
Amendments area, under Title I, Policy/
Guidance Documents. To access the bulletin
board, a modem and communications
software are necessary. To dial up, set your
communications software to 8 data bits, no
parity and one stop bit. Dial (919) 541-5742
and follow the on-screen instructions to
register for access. After registering, proceed
to choice ' ' Gateway to TTN Technical
Areas", then choose " CAAA BBS".
From the main menu, choose "<1> Title I:
Attain/Maint of NAAQS", then " Policy
Guidance Documents." To access these
documents through the World Wide Web,
click on "TTN BBSWeb", then proceed to
the Gateway to TTN Technical areas, as
above. If assistance is needed in accessing the
system, call the help desk at (919) 541-5384
in Research Triangle Park, NC.
Implementation Strategy for Revised Air
Quality Standards
On Wednesday, July 16, 1997, President
Clinton signed a memorandum to the
Administrator specifying his goals for the
implementation of the O3 and PM standards.
Attached to the President's memorandum is
a strategy prepared by an interagency
Administration group outlining the next steps
that would be necessary for implementing
these standards. The EPA will prepare
guidance and proposed rules consistent with
the President's memorandum. Copies of the
Presidential document are available in paper
copy by contacting the U.S. Environmental
Protection Agency Library at the address
under ' 'Availability of Related Information''
and in electronic form as discussed above in
"Electronic Availability."
The following topics are discussed in this
preamble:
I. Background
A. Legislative Requirements
B. Related Control Requirements
C. Review of Air Quality Criteria and Standards
forPM
D. Summary of Proposed Revisions to the PM
Standards
II. Rationale for the Primary PM Standards
A. Introduction
B. Need for Revision of the Current Primary PM
Standards
C. Indicators of PM
D. Averaging Time of PMi.s Standards
E. Form of PMi.s Standards
F. Levels for the Annual and 24-Hour PM2.5
Standards
G. Conclusions Regarding the Current PMio
Standards
H. Final Decisions on Primary PM Standards
III. Rationale for the Secondary Standards
A. Need for Revision of the Current
SecondaryStandards
B. Decision on the Secondary Standards
IV. Other Issues
A. Consideration of Costs
B. Margin of Safety
C. Data Availability
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
D. 1990 Amendments
V. Revisions to 40 CFR Part 50, Appendix K—
Interpretation of the PM NAAQS
A. PMi.5 Computations and Data Handling
Conventions
B. PMio Computations and Data Handling
Conventions
C. Changes that Apply to Both PM2.s and PMio
Computations
VI. Reference Methods for the Determination of
Particulate Matter as PMio and PMi.s in the
Atmosphere
A. Revisions to 40 CFR Part 50, Appendix J—
Reference Method for PMio
B. 40 CFR Part 50, Appendix L—New
Reference Method for PM2.5
VII. Effective Date of the Revised PM Standards
and Applicability of the Existing PMio
Standards
VIII. Regulatory and Environmental Impact
Analyses
A. Executive Order 12866
B. Regulatory Flexibility Analysis
C. Impact on Reporting Requirements
D. Unfunded Mandates Reform Act
E. Environmental Justice
F. Submission to Congress and the Comptroller
General
IX. Response to Petition for Administrator
Browner's Recusal
X. References
L Background
A. Legislative Requirements
Two sections of the Clean Air Act (Act)
govern the establishment, review, and
revision of NAAQS. Section 108 of the Act
(42 U.S.C. 7408) directs the Administrator to
identify certain pollutants which "may
reasonably be anticipated to endanger public
health and welfare'' and to issue air quality
criteria for them. These air quality criteria are
to "accurately reflect the latest scientific
knowledge useful in indicating the kind and
extent of all identifiable effects on public
health or welfare which may be expected
from the presence of [a] pollutant in the
ambient air * * *."
Section 109 of the Act (42 U.S.C. 7409)
directs the Administrator to propose and
promulgate "primary" and "secondary"
NAAQS for pollutants identified under
section 108 of the Act. Section 109(b)(l) of
the Act defines a primary standard as one
"the attainment and maintenance of which in
the judgment of the Administrator, based on
[the] criteria and allowing an adequate margin
of safety, are requisite to protect the public
health." The margin of safety requirement
was intended to address uncertainties
associated with inconclusive scientific and
technical information available at the time of
standard setting, as well as to provide a
reasonable degree of protection against
hazards that research has not yet identified.
Both kinds of uncertainties are components of
the risk associated with pollution at levels
below those at which human health effects
can be said to occur with reasonable scientific
certainty. Thus, by selecting primary
standards that provide an adequate margin of
safety, the Administrator is seeking not only
to prevent pollution levels that have been
demonstrated to be harmful but also to
prevent lower pollutant levels that she finds
may pose an unacceptable risk of harm, even
if the risk is not precisely identified as to
nature or degree. The Act does not require the
Administrator to establish a primary NAAQS
at a zero-risk level, but rather at a level that
reduces risk sufficiently so as to protect
public health with an adequate margin of
safety. The selection of any particular
approach to providing an adequate margin of
safety is a policy choice left specifically to
the Administrator's judgment. Lead
Industries Ass 'n v. EPA, 647 F.2d 1130,
1161-1162 (B.C. Cir.1980).
A secondary standard, as defined in section
109 (b)(2) of the Act, must "specify a level
of air quality the attainment and maintenance
of which in the judgment of the
Administrator, based on [the] criteria, [are]
requisite to protect the public welfare from
any known or anticipated adverse effects
associated with the presence of [the] pollutant
in the ambient air." Welfare effects as
defined in section 302(h) of the Act (42
U.S.C. 7602(h)) include, but are not limited
to, "effects on soils, water, crops, vegetation,
manmade materials, animals, wildlife,
weather, visibility, and climate, damage to
and deterioration of property, and hazards to
transportation, as well as effects on economic
values and on personal comfort and well-
being."
Section 109(d)(l) of the Act requires
periodic review and, if appropriate, revision
of existing air quality criteria and NAAQS.
Section 109(d)(2) of the Act requires
appointment of an independent scientific
review committee to review criteria and
standards and recommend new standards or
revisions of existing criteria and standards, as
appropriate. The committee established under
section 109(d)(2) of the Act is known as the
Clean Air Scientific Advisory Committee
(CASAC), a standing committee of EPA's
Science Advisory Board.
B. Related Control Requirements
States are primarily responsible for
ensuring attainment and maintenance of
ambient air quality standards once EPA has
established them. Under section 110 of the
Act (42 U.S.C. 7410) and related provisions,
States are to submit, for EPA approval, State
implementation plans (SIP's) that provide for
the attainment and maintenance of such
standards through control programs directed
to sources of the pollutants involved. The
States, in conjunction with EPA, also
administer the prevention of significant
deterioration program (42 U.S.C. 7470-7479)
for these pollutants. In addition, Federal
programs provide for nationwide reductions
in emissions of these and other air pollutants
through the Federal Motor Vehicle Control
Program under Title II of the Act (42 U.S.C.
7521-7574), which involves controls for
automobile, truck, bus, motorcycle, nonroad
engine, and aircraft emissions; the new source
performance standards under section 111 of
the Act (42 U.S.C. 7411); and the national
emission standards for hazardous air
pollutants under section 112 of the Act (42
U.S.C. 7412).
C. Review of Air Quality Criteria and
Standards for PM
Particulate matter is the generic term for a
broad class of chemically and physically
diverse substances that exist as discrete
particles (liquid droplets or solids) over a
wide range of sizes. Particles originate from
a variety of anthropogenic stationary and
mobile sources as well as from natural
sources. Particles may be emitted directly or
formed in the atmosphere by transformations
of gaseous emissions such as sulfur oxides
(SOX), nitrogen oxides (NOX), and volatile
organic compounds (VOC). The chemical and
physical properties of PM vary greatly with
time, region, meteorology, and source
category, thus complicating the assessment of
health and welfare effects.
The last review of PM air quality criteria
and standards was completed in July 1987
with notice of a final decision to revise the
existing standards published in the Federal
Register (52 FR 24854, July 1, 1987). In that
decision, EPA changed the indicator for PM
from total suspended particles (TSP) to
PMio.1 Identical primary and secondary PMio
standards were set for two averaging times:
50 |j,g/m3, expected annual arithmetic mean,
averaged over 3 years, and 150 l-ig/m3, 24-
hour average, with no more than one expected
exceedance per year.2
The EPA initiated this current review of
the air quality criteria and standards for PM
in April 1994 by announcing its intention to
develop a revised Air Quality Criteria
Document for Particulate Matter (henceforth,
the "Criteria Document"). Thereafter, the
EPA presented its plans for review of the
criteria and standards for PM under a highly
accelerated, court-ordered schedule3 at a
public meeting of the CASAC in December
1994. Several workshops were held by EPA's
National Center for Environmental
Assessment (NCEA) to discuss important
new health effects information in November
1994 and January 1995. External review
drafts of the Criteria Document were made
available for public comment and were
1 PMio refers to particles with an aerodynamic diameter
less than or equal to a nominal 10 micrometers. Technical
details further specifying the measurement of PMio are
contained in 40 CFR part 50, Appendices J and M.
2 A more complete history of the PM NAAQS is
presented in section II.B of the OAQPS Staff Paper,
Review of National Ambient Air Quality Standards for
Particulate Matter: Assessment of Scientific and Technical
Information (U.S. EPA, 1996b).
3 A court order entered in American Lung Association
v. Browner, CIV-93-643-TUC-ACM (D. Ariz.,October 6,
1994), as subsequently modified, requires publication of
EPA's final decision on the review of the PM NAAQS
by July 19, 1997.
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
reviewed by CASAC at public meetings held
in August and December 1995 and February
1996. The CASAC came to closure in its
review of the Criteria Document, advising the
Administrator in a March 15, 1996 closure
letter (Wolff, 1996a) that "although our
understanding of the health effects of PM is
far from complete, a revised Criteria
Document which incorporates the Panel's
latest comments will provide an adequate
review of the available scientific data and
relevant studies of PM." CASAC and public
comments from these meetings, and from
subsequent written comments and the closure
letter, were incorporated as appropriate in the
final Criteria Document (U.S. EPA, 1996a).
External review drafts of a Staff Paper
prepared by the Office of Air Quality
Planning and Standards (OAQPS), Review of
the National Ambient Air Quality Standards
for Particulate Matter: Assessment of
Scientific and Technical Information
(henceforth, the "Staff Paper"), were made
available for public comment and were
reviewed by CASAC at public meetings in
December 1995 and May 1996.4 The CASAC
came to closure in its review of the Staff
Paper, advising the Administrator in a June
13, 1996 closure letter (Wolff, 1996b) that
"the Staff Paper, when revised, will provide
an adequate summary of our present
understanding of the scientific basis for
making regulatory decisions concerning PM
standards." CASAC and public comments
from these meetings, subsequent written
comments, and the CASAC closure letter
were incorporated as appropriate in the final
Staff Paper (U.S. EPA, 1996b).
On November 27, 1996, EPA announced
its proposed decision to revise the NAAQS
for PM (61 FR 65638, December 13, 1996)
(hereafter "proposal") as well as its proposed
decision to revise the NAAQS for ozone
(O3)(61 FR 65716, December 13, 1996). In
the proposal, EPA identified proposed
revisions, based on the air quality criteria for
PM, and solicited public comments on
alternative primary standards and on the
proposed forms of the standards.
To ensure the broadest possible public
input on the PM and O3 proposals, EPA took
extensive and unprecedented steps to
facilitate the public comment process beyond
the normal process of providing an
opportunity to request a hearing and receiving
written comments submitted to the
rulemaking docket. The EPA established a
national toll-free telephone hotline to
facilitate public comments on the proposed
revisions to the PM and O3 NAAQS, and on
related notices dealing with the
implementation of revised PM and O3
standards, as well as a system for the public
to submit comments on the proposals
electronically via the Internet. Over 14,000
calls and over 4,000 electronic mail messages
were received through these channels. The
public could also access key supporting
documents (including the Criteria Document,
Staff Paper, related technical documents and
fact sheets) via the Internet.
The EPA also held several public hearings
and meetings across the country to provide
direct opportunities for public comment on
the proposed revisions to the PM and O3
NAAQS and to disseminate information to
the public about the proposed standard
revisions. On January 14 and 15, 1997, EPA
held concurrent, 2-day public hearings in
Boston, MA, Chicago, IL, and Salt Lake City,
UT. A fourth public hearing, which focused
primarily on PM monitoring issues, was held
in Durham, NC on January 14, 1997. Over
400 citizens and organizations testified during
these public hearings. EPA also held two
national satellite telecasts to answer questions
on the standards and participated in meetings
sponsored by the Air and Waste Management
Association on the proposed revisions to the
standards at more than 10 locations across the
country. Beyond that, several EPA regional
offices held public meetings and workshops
and participated in hearings that States and
cities held around the country.
As a result of this intensive effort to solicit
public input, over 50,000 written and oral
comments were received on the proposed
revisions to the PM NAAQS by the close of
the public comment period on March 12,
1997. Major issues raised in the comments
are discussed throughout the preamble of this
final decision. A comprehensive summary of
all significant comments, along with EPA's
response to such comments (hereafter
"Response to Comments"), can be found in
the docket for this rulemaking (Docket No.
A-95-54).
The principal focus of this current review
of the air quality criteria and standards for
PM is on recent epidemiological evidence
reporting associations between ambient
concentrations of PM and a range of serious
health effects. Particular attention has been
given to several size-specific classes of
particles, including PMio and the principal
fractions of PMio, referred to as the fine
(PM2.5)5 and coarse (PMio-2.5)6 fractions. As
discussed in the Criteria Document, fine and
coarse fraction particles can be differentiated
by their sources and formation processes and
their chemical and physical properties,
including behavior in the atmosphere.
Detailed discussions of atmospheric
formation, ambient concentrations, and health
4 The Staff Paper evaluates policy implications of the
key studies and scientific information in the Criteria
Document, identifies critical elements that EPA staff
believes should be considered, and presents staff
conclusions and recommendations of suggested options for
the Administrator's consideration.
5 PM2.5 refers to particles with an aerodynamic diameter
less than or equal to a nominal 2.5 micrometers, as further
specified in 40 CFR part 50, Appendix L in this document.
6 PMio-2.5 refers to those particles with an aerodynamic
diameter less than or equal to a nominal 10 micrometers
but greater than 2.5 micrometers. In other words, it refers
to the inhalable particles that remain if fine (PM2.5)
particles are removed from a sample of PMio particles.
and welfare effects of PM, as well as
quantitative estimates of human health risks
associated with exposure to PM, can be found
in the Criteria Document and in the Staff
Paper.
D. Summary of Proposed Revisions to the PM
Standards
For reasons discussed in the proposal, the
Administrator proposed to revise the current
primary standards for PM (as indicated by
PMio), by adding two new primary PM2.5
standards set at 15 |ig/m3, annual mean, and
50 |ig/m3, 24-hour average. The proposed
annual PM2.5 standard would be based on the
3-year average of the annual arithmetic mean
PM2.5 concentrations, spatially averaged
across an area. The proposed 24-hour PM2.5
standard would be based on the 3-year
average of the 98th percentile of 24-hour
PM2.5 concentrations at each population-
oriented monitor within an area. The proposal
solicited comment on two alternative
approaches for selecting the levels of PM2.5
standards. The Administrator also proposed to
revise the current 24-hour primary PMio
standard of 150 |J.g/m3 by replacing the 1-
expected-exceedance form with a 98th
percentile form, averaged over 3 years at each
monitor within an area, solicited comment on
an alternative proposal to revoke the 24-hour
PMio standard, and proposed to retain the
current annual primary PMio standard of 50
|-ig/m3. The proposal also solicited comment
on proposed revisions to 40 CFR part 50,
Appendix K to establish new data handling
conventions for calculating 98th percentile
values and spatial averages, revisions to 40
CFR part 50, Appendix J to modify the
reference method for monitoring PM as
PMio, and a proposed new reference method
for monitoring PM as PM2.5 (40 CFR part 50,
Appendix L).
With regard to the secondary standards, the
Administrator proposed to revise the current
secondary standards by making them identical
to the suite of proposed primary standards, in
conjunction with the establishment of a
regional haze program under section 169A of
the Act.
II. Rationale for the Primary Standards
A. Introduction
1. Overview. This document presents the
Administrator's final decisions regarding the
need to revise the current primary ambient air
quality standards for PM, and, more
specifically, regarding the establishment of
new annual and 24-hour PM2.5 primary
standards and revisions to the form of the
current 24-hour PMio primary NAAQS.
These decisions are based on a thorough
review, in the Criteria Document, of the latest
scientific information on known and potential
human health effects associated with
exposure to PM at levels typically found in
the ambient air. These decisions also take into
account:
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
(1) Staff Paper assessments of the most
policy-relevant information in the Criteria
Document, upon which staff
recommendations for new and revised
primary standards are based.
(2) CASAC advice and recommendations,
as reflected in discussions of drafts of the
Criteria Document and Staff Paper at public
meetings, in separate written comments, and
in the CASAC's closure letters to the
Administrator.
(3) Public comments received during the
development of these documents, either in
connection with CASAC meetings or
separately.
(4) Extensive public comments received on
the proposed decisions regarding the primary
PM standards.
After taking this information and
comments into account, and for the reasons
discussed below in this unit, the
Administrator concludes that revisions to the
current primary standards to provide
increased public health protection against a
variety of health risks are appropriate. More
specifically, the Administrator has determined
that it is appropriate to establish new annual
and 24-hour PM2.5 standards, to revise the
current 24-hour PMio standard, and to retain
the current annual PMi0 standard. As
discussed more fully below in this unit, the
rationale for the final decisions regarding the
PM primary NAAQS includes consideration
of:
(1) Health effects information, and
alternative views on the appropriate
interpretation and use of the information, as
the basis for judgments about the risks to
public health presented by population
exposures to ambient PM.
(2) Insights gained from a quantitative risk
assessment conducted to provide a broader
perspective for judgments about protecting
public health from the risks associated with
PM exposures.
(3) Specific conclusions regarding the need
for revisions to the current standards and the
elements of PM standards (i.e., indicator,
averaging time, form, and level) that, taken
together, would be appropriate to protect
public health with an adequate margin of
safety.
As with virtually any policy-relevant
scientific research, there is uncertainty in the
characterization of health effects attributable
to exposure to ambient PM. As discussed in
the proposal, however, there is now a greatly
expanded body of health effects information
as compared with that available during the
last review of the PM standards. Moreover,
the recent evidence on PM-related health
effects has undergone an unusually high
degree of scrutiny and reanalysis over the
past several years, beginning with a series of
workshops held early in the review process
to discuss important new information. A
number of opportunities were provided for
public comment on successive drafts of the
Criteria Document and Staff Paper, as well as
for intensive peer review of these documents
by CASAC at several public meetings
attended by many knowledgeable individuals
and representatives of interested
organizations. In addition, there have been a
number of important scientific conferences,
symposia, and colloquia on PM issues,
sponsored by the EPA and others, in the U.S.
and abroad, during this period. While
significant uncertainties exist, the review of
the health effects information has been
thorough and deliberate. In the judgment of
the Administrator, this intensive evaluation of
the scientific evidence has provided an
adequate basis for regulatory decision making
at this time, as well as for the comprehensive
research needs document recently developed
by EPA, and reviewed by CASAC and others,
for improving our future understanding of the
relationships between ambient PM exposures
and health effects.
The health effects information and human
risk assessment were summarized in the
proposal and are only briefly outlined below
in this unit. Subsequent units provide a more
complete discussion of the Administrator's
rationale, in light of key issues raised in
public comments, for concluding that it is
appropriate to revise the current primary
standards (Unit II.B. of this preamble) and to
revise the specific elements of the standards
including indicator (Unit II.C. of this
preamble); averaging time, form, and level of
new PM2.5 standards (Units II.D., II.E., and
II.F. of this preamble); and averaging time,
form, and level of revised PMi0 standards
(Unit II.G. of this preamble).
2. Summary of the health effects evidence.
In brief, since the last review of the PM
criteria and standards, the most significant
new evidence on the health effects of PM is
the greatly expanded body of community
epidemiological studies. The Criteria
Document stated that these recent studies
provide "evidence that serious health effects
(mortality, exacerbation of chronic disease,
increased hospital admissions, etc.) are
associated with exposures to ambient levels
of PM found in contemporary U.S. urban
airsheds even at concentrations below current
U.S. PM standard" (U.S. EPA, 1996a; p. 13-
1). Although a variety of responses to
constituents of ambient PM have been
hypothesized to contribute to the reported
health effects, the relevant toxicological and
controlled human studies published to date
have not identified any accepted
mechanism(s) that would explain how such
relatively low concentrations of ambient PM
might cause the health effects reported in the
epidemiological literature.
Unit II. A. of the proposal further outlines
key information contained in the Criteria
Document, Chapters 10-13, and the Staff
Paper, Chapter V, on the known and potential
health effects associated with airborne PM,
alone and in combination with other
pollutants that are routinely present in the
ambient air. The information highlighted
there summarizes:
(1) The nature of the effects that have been
reported to be associated with ambient PM,
which include premature mortality,
aggravation of respiratory and cardiovascular
disease (as indicated by increased hospital
admissions and emergency room visits,
school absences, work loss days, and
restricted activity days), changes in lung
function and increased respiratory symptoms,
changes to lung tissues and structure, and
altered respiratory defense mechanisms.
(2) Sensitive subpopulations that appear to
be at greater risk to such effects, specifically
individuals with respiratory disease and
cardiovascular disease and the elderly
(premature mortality and hospitalization),
children (increased respiratory symptoms and
decreased lung function), and asthmatic
children and adults (aggravation of
symptoms).
(3) An integrated evaluation of the health
effects evidence, with an emphasis on the key
issues raised in assessing community
epidemiological studies, including alternative
interpretations of the evidence, both for
individual studies and for the evidence as a
whole.
(4) The PM fractions of greatest concern
to health.
The summary in the proposal will not be
repeated here. EPA emphasizes that the final
decisions on these standards take into account
the more comprehensive and detailed
discussions of the scientific information on
these issues contained in the Criteria
Document and Staff Paper, which were
reviewed by the CASAC and the public.
3. Key insights from the risk assessment.
The Staff Paper presents the results of a
quantitative assessment of health risks for two
example cities, including risk estimates for
several categories of health effects associated
with: existing PM air quality levels, projected
PM air quality levels that would occur upon
attainment of the current PMio standards, and
projected PM air quality levels that would
occur upon attainment of alternative PM2.5
standards. The risk assessment is intended as
an aid to the Administrator in judging which
alternative PM NAAQS would reduce risks
sufficiently to protect public health with an
adequate margin of safety, recognizing that
such standards will not be risk-free. The risk
assessment is described more fully in the
Staff Paper and summarized in the proposal.
Related technical reports and updates7 have
7 The risk assessment results that appear in the Staff
Paper and are summarized in the proposal have been
updated to include analyses of the particular forms of
standard alternatives contained in the proposal and to
correct estimates for one effects category (mortality from
long-term exposure) to reflect the actual statistics used in
the study upon which they were based (Pope et al., 1995).
The corrections, which cumulatively reduce estimates of
mortality associated with long-term exposures by 20 to
35%, have no effect on risk estimates for mortality
Continued
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
been placed in the docket (Abt Associates,
1996a,b; 1997a,b).
EPA emphasizes that it places greater
weight on the overall conclusions derived
from the studies—that PM air pollution is
likely causing or contributing to significant
adverse effects at levels below those
permitted by the current standards—than on
the specific concentration-response functions
and quantitative risk estimates derived from
them. These quantitative risk estimates
include significant uncertainty and, therefore,
should not be viewed as demonstrated health
impacts. EPA believes, however, that they do
represent reasonable estimates as to the
possible extent of risk for these effects given
the available information. Keeping in mind
the important uncertainties inherent in any
such analyses, the key insights from the risk
assessment that are most pertinent to the
current decision include:
(1) Fairly wide ranges of estimates of the
incidence of PM-related mortality and
morbidity effects and risk reductions
associated with attainment of alternative
standards were calculated for the two
locations analyzed when the effects of key
uncertainties and alternative assumptions
were considered. Significantly, the combined
analysis for these two cities alone found that
the risk remaining after attaining the current
PMio standards was on the order of hundreds
of premature deaths each year, hundreds to
thousands of respiratory-related hospital
admissions, and tens of thousands of
additional respiratory related symptoms in
children.
(2) Based on the results from the sensitivity
analyses of key uncertainties and the
integrated uncertainty analyses, the single
most important factor influencing the
uncertainty associated with the risk estimates
is whether or not a threshold concentration
exists below which PM-associated health
risks are not likely to occur.
(3) Over the course of a year, the few peak
24-hour PM2.5 concentrations appear to
contribute a relatively small amount to the
total health risk posed by the entire air quality
distribution as compared to the aggregated
risks associated with the low to mid-range
concentrations.
(4) There is greater uncertainty about both
the existence and the magnitude of estimated
excess mortality and other effects associated
with PM exposures as one considers
increasingly lower concentrations
approaching background levels.
associated with short-term exposures or the estimates for
any other effects. Because the key sensitivity analyses that
provide additional insights regarding thresholds,
copollutants, averaging time and related issues involved
the short-term exposure studies, none of these results are
affected by changes to the long-term exposure risk
estimates.
B. Need for Revision of the Current Primary
PM Standards
1. Introduction. The overarching issue in
the present review of the primary NAAQS is
whether, in view of the advances in scientific
knowledge reflected in the Criteria Document
and Staff Paper, the existing PM standards
should be revised and, if so, what revised or
new standards would be appropriate. The
concluding section of the integrative synthesis
of health effects information in the Criteria
Document, which CASAC characterized as
EPA's "best ever example of a true
integrative summary of the state of
knowledge about the health effects of
airborne PM," (Wolff, 1996b) provides the
following summary of the science with
respect to this issue:
The evidence for PM-related effects from
epidemiological studies is fairly strong, with most
studies showing increases in mortality, hospital
admissions, respiratory symptoms, and pulmonary
function decrements associated with several PM
indices. These epidemiological findings cannot be
wholly attributed to inappropriate or incorrect
statistical methods, misspecification of
concentration-effect models, biases in study design
or implementation, measurement errors in health
endpoint, pollution exposure, weather, or other
variables, nor confounding of PM effects with
effects of other factors. While the results of the
epidemiological studies should be interpreted
cautiously, they nonetheless provide ample reason
to be concerned that there are detectable health
effects attributable to PM at levels below the
current NAAQS. [U.S. EPA, 1996a, p. 13-92]
Given the nature of the health effects in
question, this finding, which is based on a
large number of studies that used PMio
measurements, as well as studies using other
indicators of PM, clearly indicates that
revision of the current PM NAAQS is
appropriate. Quite apart from the issue of
whether PMio should be the sole indicator for
the PM NAAQS, the extensive PM
epidemiological data base provides evidence
of serious health effects (e.g., mortality,
exacerbation of chronic disease, increased
hospital admissions) in sensitive populations
(e.g., the elderly, individuals with
cardiopulmonary disease), as well as
significant adverse health effects (e.g.,
increased respiratory symptoms, school
absences, and lung function decrements) in
children. Moreover, these effects associations
are observed in areas or at times when the
levels of the current PMio standards are met.
Although the increase in relative risk is small
for the most serious outcomes, EPA believes
it is significant from an overall public health
perspective, because of the large number of
individuals in sensitive populations that are
exposed to ambient PM, as well as the
significance of the health effects involved
(U.S. EPA, 1996a, p. 1-21). The results of the
two-city PM risk assessment reinforce these
conclusions regarding the significance of the
public health risk—even under a scenario in
which the current PMio standards are
attained.
While the lack of demonstrated
mechanisms that explain the extensive body
of epidemiological findings is an important
caution, which presents difficulties in
providing an integrated assessment of PM
health effects research, a number of potential
mechanisms have been hypothesized in the
recent literature (U.S. EPA, 1996b; p. V-5 to
V-8; appendix D). Moreover, qualitative
information from laboratory studies of the
effects of particle components at high
concentrations and dosimetry considerations
suggest that the kinds of effects observed in
community studies (e.g., respiratory- and
cardiovascular-related responses) are at least
plausibly related to inhalation of PM.8
Indeed, as discussed in the Criteria Document
and section V.E of the Staff Paper, the
consistency of the results of the
epidemiological studies from a large number
of different locations and the coherent nature
of the observed effects9 are suggestive of a
likely causal role of ambient PM in
contributing to the reported effects.
2. Comments on scientific basis for
revision. A majority of the public comments
received on the proposal agreed that, based
on the available scientific information, the
current PMio standards are not of themselves
sufficient to protect public health and it
would be appropriate to revise them. Included
in those calling for revisions to the current
standards are many public health
professionals, including numerous medical
doctors and academic researchers. For
example, a group of 27 members of the
scientific and medical community recognized
as having substantial expertise in conducting
research on the health effects of air pollution
stated:
Health studies conducted in the U.S. and around
the world have demonstrated that levels of
particulate and ozone air pollution below the
current U.S. National Air Quality Standards
exacerbate serious respiratory disease and
contribute to early death. A large body of scientific
and medical evidence clearly indicates that the
current NAAQS are not sufficiently protective of
public health. [Thurston, 1997]
Similar conclusions were reached in a letter
signed by more than 1,000 scientists,
clinicians, researchers, and other health care
professionals (Dickey, 1997). The cosigners
to this letter argued that tens of thousands of
hospital visits and premature deaths could be
8 As discussed more fully below in this unit,
epidemiological studies alone cannot be used to
demonstrate mechanisms of action, but they can provide
evidence useful in making inferences with regard to causal
relationships (U.S. EPA, 1996b, p. V-9).
9 As noted in the proposal, the kinds of effects observed
in the epidemiological studies are logically related. For
example, the association of PM with mortality is mainly
linked to respiratory and cardiovascular causes, which is
coherent with observed PM associations with respiratory
and cardiovascular hospital admissions and respiratory
symptoms. Further, similar categories of effects are seen
in long- and short-term exposure studies.
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
prevented with the proposed air quality
standard revisions. In fact, these commenters
argued that even stronger standards than those
proposed by EPA are needed to protect the
health of the most vulnerable residents of our
communities.
A number of State and local government
authorities also submitted comments in
support of adopting new air quality standards
for fine particulate matter. The commenters
concurred with conclusions reached through
the EPA's peer review process that the PM
standards should be revised to protect public
health. A number of these commenters
suggested that the standards proposed by EPA
should be even stronger, while several other
State agencies recommended that EPA adopt
PM2.5 standards, but at less stringent levels.
A number of the comments from states
supporting even stronger standards
acknowledged the lack of demonstrated
mechanism(s) and other uncertainties but
stressed the strength of the other evidence in
urging EPA to set protective standards.
Many comments were also received from
representatives of environmental or
community health organizations that
supported the adoption of air quality
standards for PM2.5. These commenters
agreed with EPA's finding that a large body
of compelling evidence demonstrates that
exposure to particulate matter pollution, in
general, is associated with premature death,
aggravation of heart and lung diseases,
increased respiratory illness and reduced lung
function. They agreed with EPA that these
studies present a consistent and coherent
relationship between exposure to PM and
both mortality and various measures of
morbidity. However, the majority of these
commenters argued that EPA's proposed
standards for PM2.5 were inadequate and
recommended adoption of more stringent
levels of the 24-hour and/or annual air quality
standards forPM2.5. Many of these
commenters also urged EPA to revise the
NAAQS for PMio to be more protective of
public health. These commenters based their
recommendations on the findings of the
studies that were reviewed in the preparation
of the Criteria Document and Staff Paper.
One commenter used results from five of
these studies as the basis for recommending
PM2.5 standards of 10 l-ig/m3 (annual) and 18
Hg/m3 (24-hour) (Dockery et al, 1993; Pope
et al., 1995; Schwartz et al., 1996; Schwartz
et al., 1994; Thurston et al., 1994). The
commenters agreed with EPA on the
significance of these studies' results and the
need to revise the PM standards, while
differing with EPA's interpretation of the
findings for purposes of developing the
proposed PM standards.
Several commenters made reference to the
conclusions of a number of international
scientific panels regarding the health effects
of exposure to airborne particulate matter—
the British Expert Panel on Air Quality
Standards, the British Committee on the
Medical Effects of Air Pollutants, the World
Health Organization, the Canadian Ministry
of Environment, Lands and Parks, and the
Health Council of the Netherlands — and
argued that all these panels found that PM
concentrations equivalent to the current U.S.
standards for PMio are not protective of
human health and made recommendations for
greater protection. One commenter noted that
the findings of the British Health Panel have
resulted in a British proposal to adopt a 24-
hour PMio standard of 50 |ig/m3, which is
one-third the level of the current U.S.
NAAQS.
In these comments, some toxicological
studies were cited as providing evidence for
toxicity of particulate pollution. These
commenters disagreed with arguments that
PM standards cannot be adopted due to a lack
of a sufficient understanding of the biological
mechanism of injury. The commenters argued
that there is sufficient evidence that
particulate pollution is associated with
adverse health effects to make it inappropriate
to delay the establishment of standards while
further studies are undertaken. This group of
commenters was also critical of arguments
against the establishment of additional PM
standards based on the possibility of
confounding by other pollutants, and urged
that more attention be paid instead to the
possible additive or synergistic effects of
multiple pollutant exposures.
In general, the EPA agrees with these
commenters' arguments regarding the need to
revise the PM standards. The scientific
studies cited by these commenters were the
same studies used in the development of the
Criteria Document and the Staff Paper, and
the EPA agrees that there is a sufficient body
of evidence that the current NAAQS for PM
are not adequately protective of the public
health. For reasons detailed in Unit IFF. of
this preamble and in the Response to
Comments, EPA disagrees with aspects of
these commenters' views on the level of
protection that is appropriate and supported
by the available scientific information.
Another body of commenters, including
almost all commenters representing
businesses and industry associations, many
local governmental groups and private
citizens, and some States opposed revising the
standards. Many of these commenters argued
that the available scientific evidence does not
provide an adequate basis for revising the
current standards. The central arguments
made by these commenters can be divided
into two categories: (1) General comments on
the appropriateness of relying on the
epidemiological evidence for making
regulatory decisions, and (2) more specific
comments challenging EPA's appraisal of the
consistency and coherence of the available
information, EPA's conclusions regarding
causality, and the use of these studies for risk
assessment and decisions on whether to revise
the standards. While EPA has included
comprehensive responses to these comments
in the Response to Comments, certain key
points are summarized below in this unit.
a. General comments on the use of
epidemiological studies. The first category of
comments was largely derived from ad hoc
panels of occupational and other
epidemiological experts, consulting groups,
and individual consultants. Most of these
individuals and groups commented on the use
of epidemiology in reaching scientific and
policy conclusions primarily from an
occupational or hazard assessment
perspective, in contrast to the perspective of
the review of ambient PM criteria and
standards, where the use of community air
pollution epidemiological studies are central.
Citing accepted criteria used in evaluating
epidemiological studies to assess the
likelihood of causality (most notably those of
Sir Austin Bradford Hill, 1965), these
commenters argued that in the absence of a
demonstrated biological mechanism, the
relative risks of effects in the PM
epidemiological studies are too low (less than
values variously cited as 1.5 to 2.0) to reach
any conclusions regarding causality or to
form the basis for regulations. In general, the
commenters applied these criteria to a subset
of studies evaluated in the Criteria Document,
including as few as two long-term exposure
studies (EOF Group) (API, 1997), a group of
9 selected studies (Greenland panel) (API,
1997), those studies cited in the proposal
(AIHC, 1997), or as many as 23 selected
short-term exposure studies examined in a
recently published review paper (Gamble and
Fewis, 1996).
Based on a careful review of these
comments, EPA notes a number of limitations
in these commenters' evaluations of the
epidemiological studies that they considered,
as discussed in detail in the Response to
Comments. In summary, EPA notes that these
commenters provided scientific advice and
conclusions that are in substantial
disagreement with the conclusions of the
review reflected in the Criteria Document and
Staff Paper. EPA stands behind the scientific
conclusions reached in these documents
regarding the appropriate use of the available
community epidemiological studies. These
documents were the product of an extended
public process that included conducting
public workshops involving the leading
researchers in the field, drafts of the Criteria
Document and Staff Paper providing
opportunities for public scrutiny and
comment on, and, not least, receiving the
advice of an independent panel of air
pollution experts, including epidemiologists.
EPA clearly specified the key criteria by
which it evaluated the available
epidemiological studies in section 12.1.2 of
the Criteria Document, with substantial
reliance on those specified by Hill (1965). In
rejecting results with relative risks less than
1.5 to 2 as meaningful absent demonstrated
biological mechanisms, the commenters fail
to note that Hill and other expert groups (U.S.
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
DHEW, 1964) have emphasized that no one
criterion is definitive by itself, nor is it
necessary that all be met in order to support
a determination of causality (U.S. EPA,
1996a, p. 12-3).
With respect to biological plausibility, Hill
noted that "this is a feature I am convinced
we cannot demand. What is biologically
plausible depends upon the biological
knowledge of the day" (Hill, 1965). This
statement is clearly pertinent to the
lexicological and mechanistic understanding
of the effects of PM and associated air
pollutants, especially at lower concentrations.
It is also important to stress that while the
mechanistic evidence published as of the time
the Criteria Document closed does not
provide quantitative support for the
epidemiological results, neither can such
limited evidence refute these findings. It is
also important to stress that our understanding
of biological mechanisms for PM pollution
effects is not sufficient to explain the effects
observed at much higher concentrations in air
pollution episodes, for which causality is
generally accepted. Moreover, the
lexicological literalure has only recenlly
begun lo examine animal models (or
controlled human sludies) lhal mighl reflecl
Ihe sensitive populations in question (Ihe
elderly, individuals wilh chronic respiratory
and cardiovascular disease) or lhal adequately
reproduce all of Ihe physico-chemical
properties of particles in Ihe ambienl
almosphere. In short, Ihe absence of evidence
of a particular mechanism is hardly proof lhal
Ihere are no mechanisms lhal could explain
Ihe effecls observed so consislenlly in Ihe
epidemiological sludies. The absence of
biological mechanisms did nol deter CASAC
from recommending revisions lo Ihe PM
standards in 1982, 1986, and again in 1996.
While Hill appropriately emphasized Ihe
slrenglh of Ihe association as importanl (e.g.,
size of Ihe relative risk), he also pointed oul
lhal' 'We musl nol be loo ready lo dismiss
a cause-and-effecl hypolhesis merely on Ihe
ground lhal Ihe observed association appears
lo be slight There are many occasions in
medicine when this in Irulh is so" (Hill,
1965). EPA believes lhal Ihe effecls of air
pollution containing PM is such a case.
Unlike Ihe "lexlbook" examples of unlikely
significanl associations provided by some
commenlers (e.g., ice cream consumption
correlated wilh heal stroke), Ihe abundanl
epidemiological literalure on combustion
particles documenls numerous occasions in
which single short-term episodes of high air
pollution produced unequivocally elevated
relative risks. For Ihe week of Ihe well
documented 1952 London air pollution
episode, for example, Ihe relative risk of
mortality for all causes was 2.6, while Ihe
relative risk for bronchitis mortality was as
high as 9.3 (Minislry of Heallh, 1954).
Hospital admissions also increased by more
lhan a factor of two. British epidemiologist
in Ihe 1950s concluded lhal increased
mortality was likely when PM (as mass
calibrated British Smoke <4.5 jam in
aerodynamic diameter) exceeded 500 l-ig/m3
(Martin and Bradley, 1960). This is only
aboul a factor of 3 higher lhan lhal allowed
by Ihe currenl PM standard. Unlike Ihe
"lexlbook" and olher unlikely statistical
associations noted by some commenlers,
where Ihe only evidence is for low relative
risk, clear and convincing links between high-
level PM concentrations and mortality and
morbidity buttress Ihe findings of similar
associations al much lower PM
concenlralions as suggested in Ihe more
recenl epidemiological literalure.
These commenlers also appear lo ignore
several epidemiological sludies conducted al
low PM concenlralions in U.S. and European
cities, including bolh short- and long-term
exposures lo PM air pollution, lhal find
statistically significanl relative risks of
respiratory symptom categories in children in
Ihe range of 1.5 lo 5 (Schwartz el al., 1994;
Pope and Dockery, 1992; Braun-Fahrlander el
al., 1992; Dockery el al., 1989; Dockery el
al., 1996). Concenlralions in Ihese sludies
extend from moderately above lo well below
Ihose permitted by Ihe currenl PMio
standards. While, as noted in Ihe proposal,
mosl of Ihe recenl epidemiological sludies of
mortality and hospital admissions report
comparatively small relative risks, the
findings of relative risks well in excess of Ihe
1.5 lo 2 criterion noted by commenlers for
earlier sludies of high PM episodes, as well
as Ihe relative risks of 1.5 lo 5 reported in
more recenl sludies of less serious, bul still
importanl effecls categories, lend credibility
lo EPA's interpretation of Ihe resulls.
In addition lo basing Iheir conclusions
primarily on their own assessmenl of a
limited sel of sludies, Ihis group of
commenlers reached differenl conclusions
aboul Ihe consistency of Ihe observed
associations because of Iheir assumptions lhal
all model building strategies by all authors
are equally valid. Even Ihe mosl thorough of
Ihese Irealmenls (Gamble and Lewis, 1996)
shared Ihis flaw, particularly in Ihe discussion
of the series of Philadelphia mortality sludies
and in Ihe discussion of modeling approaches.
The authors' Irealmenl of modeling and
confounding issues was further limited
because they did nol include the mosl recenl
Philadelphia resulls (Samel el al., 1996a,b)
sponsored by Ihe Heallh Effecls Institute
(HEI, 1997). One of the importanl functions
of the Criteria Documenl is lo evaluate the
strengths and limitations of various sludies.
As discussed more fully below in this unil,
the Criteria Documenl found lhal some of the
sludies cited by commenlers as suggesting a
lack of consistency had importanl limitations.
In general, these commenlers' analyses
suffered by ignoring Ihe much more Ihorough
critical review of Ihese sludies and issues
contained in the Criteria Documenl, notably
thai in section 12.6 on alternative modeling
approaches.
EPA also rejecls the notion advanced by
these commenlers lhal epidemiological
sludies musl use personal exposure
monitoring lo be considered for regulatory
purposes. In particular, commenlers ignore
the significanl strengths of Ihe lime-series
sludies and prospective cohort sludies relied
on by EPA as compared lo cross-sectional
epidemiological sludies. Time-series sludies,
such as the daily mortality sludies, look al
changes in response rale in relation lo
changes in wealher and air pollution over
lime intervals of a few days. This controls for
olher factors such as smoking and
socioeconomic slalus, which are little
changed during such short intervals.
Prospective cohort sludies (e.g., Pope el al.,
1995; Raizenne el al., 1996), on Ihe olher
hand, look al changes in heallh slalus in a
selected cohort of individuals, which allows
direcl adjuslmenl for smoking slalus,
socioeconomic slalus, and olher subjecl-
specific factors. The commenlers also ignore
the Criteria Documenl conclusions on how
properly conducted monitoring can provide
an adequate index of population exposure lo
ambienl air pollution in such sludies lhal, as
detailed below, is more relevanl lo
establishing ambienl air quality standards
(U.S. EPA 1996a, chapter 7). Allhough
personal monitoring may be practical for
some occupational and epidemiological
sludies, and has been employed in some pasl
sludies of air pollution, il is nol realistic lo
require personal monitors in air pollution
sludies of daily mortality, which require
urban scale population data over a period of
years. Furthermore, Ihe use of community
moniloring-based epidemiological sludies as a
basis for establishing standards and guidelines
has a long history in air pollution, including
the British aulhorilies' response lo the
London episodes and Ihe establishmenl of Ihe
original U.S. NAAQS in 1971. Rejecting the
use of the vasl array of such sludies on Ihis
basis alone would also go againsl the advice
of the independenl scientific experts on every
CASAC panel lhal has addressed Ihe subjecl
of PM pollution Ihrough Ihe years, each of
which has recommended general PM
standards based primarily on the resulls of
community epidemiological sludies
(Friedlander, 1982; Lippmann, 1986; Wolff,
1996b). As noted above in Ihis unil, EPA has
included a more detailed discussion of ils
responses lo these commenls in Ihe Response
lo Commenls.
b. Specific comments on epidemiologic
studies. The second group of commenlers
noted above made more specific challenges lo
EPA's assessmenl of Ihe epidemiological
sludies. These commenls, although
overlapping some of those made by the firsl
group, were generally made by commenlers
who have taken a more active role in the
review of the Criteria Documenl and Staff
Paper. These commenlers asserted lhal the
epidemiological evidence on PM is nol as
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
consistent and coherent as EPA has claimed,
and, in particular, charged that EPA ignored
or downplayed a number of studies that the
commenters argue contradict the evidence the
Agency cited as supporting the consistency
and coherence of PM effects. The studies, all
of which commenters contend do a better job
of addressing one or more key issues, such
as confounding pollutants, weather, exposure
misclassification, and model specification,
than earlier studies, include several that were
available during preparation of the Criteria
Document, and a number that appeared after
the Criteria Document and Staff Paper were
completed. Because the status of the later
studies differ from that of the earlier ones for
purposes of decisions under section 109 of
the Act, the two categories are discussed
separately below in this unit. Additional
responses to comments relating to both sets
of studies have been included in the Response
to Comments. In addition to the inclusion of
specific studies, commenters also raised other
issues regarding the limitations of the
epidemiological information and the use of
these studies in EPA's two-city risk
assessment. Both of these topics are also
discussed below in this unit.
(i) Studies available for inclusion in the
criteria review. With some exceptions, most
of the above commenters cited somewhat
similar lists of' 'negative'' studies that they
argue EPA ignored or downplayed in arriving
at conclusions on consistency and coherence.
Of the most commonly cited studies, the
following were available for inclusion in the
Criteria Document: daily mortality studies by
Styer et al. (1995), Lyon et al. (1995), Li and
Roth (1995), Moolgavkar (1995a,b), Wyzga
and Lipfert (1995), Lipfert and Wyzga
(1995), and Samet et al. (1995, 1996a,b); the
long-term exposure mortality study by Abbey
et al. (1991); and the re-examination of the
Six-City mortality results (Dockery et al.,
1993) by Lipfert (1995).
The written record of EPA's evaluations of
these studies effectively refutes the claim that
the Agency ignored any of these studies and
supports the treatment the Agency accorded
to each of them. All of the studies available
to EPA at the time of CASAC closure on the
PM Criteria Document (March 1996) were
examined for inclusion in the Criteria
Document and Staff Paper, which form the
basis for the PM proposal. "Negative"10
studies were evaluated in detail along with
"positive" studies when they were found to
have no critical methodological deficiencies,
or to point out strengths and limitations.
Studies that had more serious problems were
generally discussed in less detail, whether
positive or negative, than studies with fewer
or small deficiencies. The EPA assessments
were evaluated by peer reviewers, by
CASAC, and by the public.
Most of the short-term exposure studies
cited above in this unit are reanalyses and
extensions of PM/mortality studies that had
been published by other investigators. In
general, the Criteria Document concluded that
the most comprehensive and thorough
reanalyses were those in the series conducted
for the HEI, which reanalyzed data sets used
in studies from six urban areas in Phase LA
(Samet et al., 1995)11, with extended analyses
for Philadelphia in Phase I.B (Samet et al.,
1996a,b). The most important finding in the
HEI Phase LA reanalyses of the six areas is
' 'the confirmation of the numerical results of
the earlier analyses of all six data sets" (HEI,
1995)12. After replicating the original
investigators' analyses, Samet et al. (1995)
also found similar results analyzing the data
using an improved statistical model. The HEI
Oversight Committee found
[I]t is reasonable to conclude that, in these six
data sets, daily mortality from all causes combined,
and from cardiovascular and respiratory causes in
particular, increases as levels of particulate air
pollution indexes increase. [HEI, 1995]
It is important to note that these reanalyses
by respected independent scientists confirm
the reliability and reproducibility of the work
of the original investigators, particularly in
view of the concerns some commenters have
expressed about EPA's reliance on a number
of PM studies published by these authors.
The Phase LA HEI results for Philadelphia
also found that it was difficult to separate the
effects of PM from those of co-occurring
SO2, in agreement with the Moolgavkar et
al.(1995a) analysis. Subsequent HEI work,
and several of the other so-called "negative"
studies cited above in this unit, further
examined this issue in terms of confounding
or effects modification by one or more co-
occurring gaseous pollutants or weather.
Contrary to commenters' claims, this issue
and these studies received considerable
attention in the Criteria Document and Staff
Paper, and the overall implications and
conclusions from these assessments were
summarized in the proposal. In particular, the
so-called "negative" and other findings of
Moolgalvkar et al. (1995a,b) in their
Philadelphia and Steubenville studies were
discussed in great detail in section 12.6 of the
10 The term "negative" studies, as used in these
comments, should not be construed to mean those in which
there is a negative effects estimate (either significant or
non-significant) for the nominal cause. As used by these
commenters, the term also includes statistically non-
significant positive effect estimates. In other words, the
commenters define ' 'positive'' studies as including only
those in which the effect estimate is both positive and
statistically significant.
1! Data sets were those used in the original studies by
Dockery et al. (1992) for St. Louis and Eastern Tennessee;
Pope et al. (1992) for Utah Valley; Schwartz and Dockery
(1992a) for Philadelphia; Schwartz (1993) for
Birmingham; and a portion of the Santa Clara data from
Fairley (1990). The data set from the Moolgavkar et al.
(1995a) Philadelphia reanalysis was also included (Samet
etal., 1995).
12 The HEI Board of Directors appointed an eight
member Oversight Committee consisting of leading
scientists in several disciplines relevant to air pollution
epidemiology to oversee key aspects of the project and to
prepare HEI's assessment of the results.
PM Critera Document and compared to those
of the original investigators (Schwartz and
Dockery, 1992a,b) and other investigators (Li
and Roth, 1995; Wyzga and Lipfert, 1995).
Further analytical studies of the Philadelphia
data set were carried out by HEI (Samet et
al., 1996a,b) and have largely resolved many
of the uncertainties in the earlier analyses; in
EPA's opinion, these studies supersede the
results of the original investigators (Schwartz
and Dockery, 1992a) and the several earlier
reanalyses, including Moolgavkar (1995a),
Moolgavkar and Luebeck (1996), Li and Roth
(1995), Wyzga and Lipfert (1995), and Samet
et al. (1995). Even though TSP is not the best
PM indicator for health effects, since it
includes a substantial fraction of non-thoracic
particles, the extended Criteria Document
assessment (U.S. EPA, 1996a, pp. 12-291 to
-299; 12-327) of the Phase I.B HEI analyses
in Philadelphia (Samet et al., 1996a,b) serves
to support the following findings:
(1) The mortality effects estimates for TSP
do not depend heavily on statistical methods
when appropriate models are used.
(2) Estimated PM effects are not highly
sensitive to appropriate methods for adjusting
for time trends and for weather.
(3) Air pollution has significant health
effects above and beyond those of weather.
(4) Copollutants such as ozone, CO, and
NO2 may be important predictors of
mortality, but their effects can be
substantially separated from those of TSP and
SO2
(5) The health effects of TSP in
Philadelphia cannot be completely separated
from SO2, which is itself a precursor of fine
particles, based solely on the epidemiological
analyses in this single city.
The most recent HEI Oversight Committee
comments on these studies (HEI, 1997),
which were submitted to the docket by HEI,
state that:
Although individual air pollutants (TSP, SCh,
and ozone) are associated with increased daily
mortality in these data, the limitations of the
Philadelphia data make it impossible to establish
that particulate air pollution alone is responsible for
the widely observed associations between increased
mortality and air pollution in that city. All we can
conclude is that it appears to play a role. [HEI,
1997; p.38.]
While recognizing the limitations in the
conclusions that can be made based on
studies in a single city, the Oversight
Committee endorses the approach taken by
EPA in evaluating a broader set of
epidemiological studies:
Consistent and repeated observations in locales
with different air pollution profiles can provide the
most convincing epidemiological evidence to
support generalizing the findings from these
models. This has been the approach reported by the
EPA in its recent Criteria Document and Staff
Paper. [HEI, 1997; p. 38.]
As noted in the proposal, based on this
approach, EPA's assessment of numerous
mortality studies concludes that when studies
-------
10
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
are evaluated on an individual basis, the PM-
effects associations are valid and, in a number
of studies, not seriously confounded by co-
pollutants (U.S. EPA, 1996a; p. 13-57); and
when a collection of studies from multiple
areas with differing concentrations of PM and
co-pollutants are examined together, the
association with PMio remains reasonably
consistent across a wide range of
concentrations of these potentially influential
pollutants (U.S. EPA, 1996a; p. 12-33; U.S.
EPA, 1996b;p. V-55).
In addition to relying on the most
comprehensive and best analyses in
evaluating the reanalysis in Philadelphia and
other areas, the Criteria Document gave less
weight to both so-called "negative" and
"positive" studies with methodogical
limitations. In particular, EPA agreed with the
epidemiological experts on CASAC
(Lippmann et al, 1996; Samet, 1995) that the
Li and Roth (1995) study approach of using
a "panoply" of different modeling strategies
to produce seemingly conflicting findings
provides little useful insight and is superseded
by the HEI report. The attempt by Lipfert and
Wyzga (1995) to address relative effects of
different pollutants was considered
inconclusive (Lippmann et al., 1996) and
flawed by the use of a metric (elasticity) that
ignores the absolute concentrations of the
pollutants being compared (see Response to
Comments).
Further, the Steubenville studies and
reanalyses (Schwartz and Dockery, 1992b;
Moolgavkar, 1995b) were discussed in detail
to examine methodologies, and the
differences in relative risks between the two
were regarded as small (U.S. EPA, 1996a, p.
12-280 to 283). Both studies used TSP as the
PM indicator variable, and they are
augmented by the more recent findings of
Schwartz et al. (1996) that examine PMio and
its components. The mixed results by Lyon
et al. (1995) in Utah Valley are compromised
by loss of information related to the
methodology (U.S. EPA, 1996a, p. 12-58). As
noted above, subsequent reanalyses of the
Utah Valley study by HEI (Samet et al.,
1995) as well as by Pope and Kalkstein
(1996) confirmed the original findings of
Pope et al. (1992) using different model
specifications. The Salt Lake City study by
Styer et al. (1995) was mentioned in the PM
Criteria Document, but received little
discussion because aspects of the
methodological approach limited its statistical
power to detect effects. The analysis of
Chicago mortality data in the same paper
shared these problems, particularly for
seasonal analyses; in this larger city, they
nonetheless found significant associations on
an annual basis between PMio and mortality
that are consistent with other studies. In short,
the record shows that EPA did not ignore
these short-term exposure studies cited by
commenters; moreover, EPA's assessment of
these studies is consistent with the views of
four researchers on the CASAC panel who
have extensive involvement in conducting
population studies of air pollution (Lippmann
etal., 1996).13
Similarly, EPA believes that appropriate
treatment and weight were given to studies of
long-term exposure and mortality. EPA
concluded that the lack of associations in the
Abbey et al. (1991) prospective cohort study
were not inconsistent with two other such
studies because the use of days of peak TSP
levels as the PM indicator (instead of PMio
or PM2.s) is inappropriate for California
cohorts exposed to both urban smog and
fugitive dust episodes, and the overall sample
size may have been too small to detect
significant effects (U.S. EPA, 1996b; pp. V-
17 to -18). The inadequacy of Lipfert's
(1995) application of state-wide average
sedentary lifestyle data to adjust mortality for
the six cities studied by Dockery et al. (1993),
in which superior subject-specific body mass
index data had already been considered, was
also noted and addressed in the Staff Paper
(U.S. EPA, 1996b; p. V-16). Again, EPA did
not ignore these studies; the rationale for
giving them less weight was clearly
articulated in the documents reviewed by
CASAC and judged appropriate for use in
standard setting.
While the proposal presents only a
summary discussion of key Criteria
Document and Staff Paper findings, EPA
believes that discussion is fully consistent
with the state of the science. Furthermore, the
proposal highlights the nature of alternative
viewpoints on the epidemiology in a
quotation from the Criteria Document (61 FR
65644, December 13, 1996) and cites
explicitly the views of most of the authors
noted above in this unit (Moolgavkar et al.,
1995b; Moolgavkar and Luebeck, 1996; Li
and Roth, 1995; Samet et al., 1996; Wyzga
and Lipfert, 1995). The proposal also
summarizes EPA conclusions based on all of
the literature as assessed in the Criteria
Document and Staff Paper with respect to
issues raised in these and other studies,
including potential confounding by
independent risk factors such as weather and
other pollutants, choice of statistical models,
use of outdoor monitors, and exposure
misclassification.
More specifically, in the proposal EPA has
not ignored the view advanced by some that
the results of individual studies of multiple
pollutants, such as the HEI Philadelphia
studies, are more suggestive of an "air
pollution" effect than an effect of PM alone.
Indeed, the proposal notes that it is reasonable
to expect that other pollutants may play a role
in modifying the magnitude of the estimated
effects of PM on mortality, either through
pollutant interactions or independent effects
(61 FR 65645, December 13, 1996). Based on
the large body of evidence at hand, however,
EPA cannot accept the suggestion that such
multi-pollutant studies are in any way
"negative" with respect to EPA's
conclusions that PM, alone or in combination
with other pollutants, is associated with
adverse effects at levels below those allowed
by the current standards. This conclusion is
based not only on the consistency of PM
effects across areas with widely varying
concentrations of potentially confounding
copollutants, but also on the extended
analyses of the Philadelphia studies in the
Criteria Document and Staff Paper.
Because commenters have tended to ignore
the latter analyses, it is appropriate to
summarize them here briefly. As noted above
in this unit, the Criteria Document assessment
of the Philadelphia studies finds that PM can
reasonably be distinguished from potential
effects of all pollutants except SO2. The Staff
Paper builds on this analysis through an
integrated assessment that draws on
information from atmospheric chemistry,
human exposure studies, and respiratory tract
penetration results to provide insight as to
which of these two pollutants is more likely
to be responsible for mortality in the elderly
and individuals with cardiopulmonary disease
(U.S. EPA 1996b; pp. V-46 to -50). That
assessment notes that the inhalable (PMio),
including the fine (PM2.5), components of
TSP are more likely than SC>2 to penetrate
and remain indoors where the sensitive
population resides most of the time.14 In
addition, these PM components, especially
PM2.5, penetrate far more effectively to the
airways and gas exchange regions of the lung
than does SC>2. Furthermore, in Philadelphia,
it is possible that SO2 is a surrogate for fine
particulate acid sulfates. For these reasons,
even though statistical analyses of the
Philadelphia data set cannot fully distinguish
between these two highly correlated
pollutants, EPA believes that the weight of
the available evidence from an integrated
assessment more strongly supports the notion
that PM is playing an important direct role
in the observed mortality effects associations
in Philadelphia. Moreover, as noted above in
this unit, in some other locations with
significant PM-mortality associations,
ambient SC>2 levels are too low to confound
PM.
(ii) Recent studies available after
completion of criteria review. As noted above
in this unit, other studies cited by some
commenters as so-called ' 'negative''
evidence ignored by EPA were published or
otherwise made available only after
completion of the PM Criteria Document.
EPA agrees that it did not rely on these
13 Their March 20, 1996 letter to the Administrator
concludes that the HEI analysis of Philadelphia supersedes
earlier analyses, specifically Moolgavkar et al. (1995a),
Lipfert and Wyzga (1995), and Li and Roth (1995), and
points out the limitations of Styer et al. (1995).
14 In response to comments on this rulemaking, some
papers submitted by industry commenters make statements
that are in substantial agreement with these staff
conclusions with respect to the likelihood of 862
penetrating to indoor environments and the lesser
likelihood of affecting sensitive populations indoors
(Lipfert and Wyzga, 1997; Lipfert and Urch, 1997).
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
11
studies, based on its long-standing practice of
basing NAAQS decisions on studies and
related information included in the pertinent
air quality criteria and available for CASAC
review.15 Although EPA has not relied on
such studies in this review and decision
process, the Agency nevertheless has
conducted a provisional examination of these
and other recent studies to assess their general
consistency with the much larger body of
literature evaluated in the Criteria
Document.16 EPA has placed its examination
of recent studies in the rulemaking docket.
Among the most frequently cited new
studies relied on by commenters were Davis
et al. (1996), Moolgavkar et al. (1997), and
Roth and Li (1997). Davis et al. (1996)
conducted a reanalysis of the Birmingham
mortality data set originally investigated in
Schwartz (1993). At the time of the close of
the public comment period, the paper based
on this manuscript had not been accepted for
publication in a peer reviewed journal (Sacks,
1997). Commenters nevertheless highlight the
authors' claim that "when humidity is
included among the meteorological variables
(it is excluded in the analysis by Schwartz
[1993]), we find that the PMio effect is not
statistically significant." EPA's review found
important factual errors in this study.
Contrary to Davis et al., Schwartz did include
humidity in his 1993 study, and his finding
of a hot-and-humid-day effect was reported
there. In addition, the PM-related variables
used by Davis et al. in their manuscript were
not, as the authors claimed, the same as those
in Schwartz (1993). Davis et al. also used a
different humidity indicator, specific
humidity. Reanalysis by one of the co-authors
(R. Smith, personal communication, February
15 Since the 1970 amendments, the EPA has taken the
view that NAAQS decisions are to be based on scientific
studies that have been assessed in air quality criteria [see
e.g., 36 FR 8186 (April 30, 1971) (EPA based original
NAAQS for six pollutants on scientific studies discussed
in the air quality criteria and limited consideration of
comments to those concerning validity of scientific basis);
38 FR 25678, 25679-25680 (September 14, 1973) (EPA
revised air quality criteria for sulfur oxides to provide basis
for reevaluation of secondary NAAQS)]. This longstanding
interpretation was strengthened by new legislative
requirements enacted in 1977 (section 109(d)(2) of the
Act; section 8(c) of the Environmental Research,
Development, and Demonstration Authorization Act of
1978) for CASAC review of air quality criteria and
reaffirmed in EPA's decision not to revise the ozone
standards in 1993. 58 FR 13008, 13013-13014 (March 9,
1993). Some of the commenters now criticizing EPA for
not considering the most recent PM studies strongly
supported the Agency's interpretation in the 1993 decision
(UARG, 1992).
16 As discussed in EPA's 1993 decision not to revise
the NAAQS for ozone, new studies may sometimes be of
such significance that it is appropriate to delay a decision
on revision of NAAQS and to supplement the pertinent
air quality criteria so the new studies can be taken into
account. 58 FR at 13014, March 9, 1993. In the present
case, EPA's provisional examination of recent studies
suggests that reopening the air quality criteria review
would not be warranted even if there were time to do so
under the court order governing the schedule for this
rulemaking. Accordingly, EPA believes that the
appropriate course of action is to consider the newly
published studies during the next periodic review cycle.
8, 1997) showed that when Schwartz's PM
metric was used, the estimated PMi0 effect
was of about the same magnitude, and
statistically significant at the 0.05 level, even
using the characterization of humidity effect
proposed by Davis et al. It therefore appears
that the Davis et al. PMio result was, in fact,
consistent with that of Schwartz, and robust
against a very different weather model
specification.
Based on its examination of both the
content and the publication status of this
study, EPA believes the heavy reliance and
attention given to it are misguided. In contrast
to commenters' assertions, this study does not
contradict EPA's conclusions with respect to
consistency of the epidemiological evidence
and confounding by weather variables;
indeed, the consideration of the corrected
results would actually support EPA's
conclusions. EPA believes this example
reinforces the importance of relying on peer
reviewed studies and also conducting the kind
of critical examination of such studies that
takes place in the criteria and standards
review process.
Several commenters note that Roth and Li
(1997) also reexamined the Birmingham
mortality data, as well as hospital admissions
data from Schwartz (1994), and produced a
number of negative and inconsistent results
that depend on temperature effects and choice
of statistical model. Preliminary findings from
this study were presented by Roth at the May
1996 CASAC meeting. CASAC
epidemiologists and statisticians at the
meeting pointed out a number of
shortcomings, both in the analytical strategy
and in details of the models being
evaluated.17 As discussed in more detail in
the Response to Comments, the materials
from Roth and Li (1997) recently provided to
EPA as attachments to public comments show
that the deficiencies pointed out at the May
1996 CASAC meeting have not been
adequately addressed. EPA concludes that
this study does not support commenters'
claims.
The paper recently accepted for publication
by Moolgavkar et al. (1997) examines
hospital admissions and air pollution in
Minneapolis and Birmingham and comes to
different conclusions than earlier investigators
with respect to the role of PMio. While the
paper is a useful addition to the literature, the
authors clearly do not attempt to replicate the
original studies, making the kind of direct
comparisons suggested by commenters
17 For example, commenting on the Roth examination
of alternative model specifications, Dr. Stolwijk noted "If
you select out of his [Roth's] matrix the things that other
people have done, he comes to a different conclusion than
when he takes his whole matrix * * *. [Y]ou are going
to get a random effect that shows that there is no effect.
He [Roth] did this, I think, on purpose in this case. Most
epidemiologists, I think, have been trained to limit their
observations to something that they can state or would
have stated before they started and observe that and base
their conclusions on it" [U.S. EPA 1996(c); May 17, 1996
Transcript, pages 45-46].
difficult. The paper finds an air pollution
effect in one city that implicates ozone but
is unable to separate effects of PM from a
group of other pollutants. EPA's provisional
examination of this study raises some
questions about the methodology, which
might usefully be supplemented to further
separate pollutants as was done by Samet et
al. (1996a,b) in Philadelphia, and about the
authors' interpretation of the results in both
cities. In any event, EPA does not believe this
study negates the PM associations with
hospital admissions reported in a number of
other studies cited in the Criteria Document.
Another recent paper by Lipfert and Wyzga
(1997) provides analyses suggesting that
differential measurement error might account
for some or all of the observation by
Schwartz et al. (1996) that daily mortality is
more strongly associated with fine (PM2.5)
than with coarse (PMi 0-2.5) PM. EPA staff
and CASAC accounted for this possibility,
however, and it was factored into both the
Staff Paper and CASAC recommendations.18
Some commenters have highlighted
selected individual papers or summaries from
the APITEA19 project conducted in Europe,
and from Roth (1996), calling attention
particularly to negative results found in
heavily polluted regions of Eastern Europe.
EPA notes that a number of the recent
APITEA and other studies in Western Europe
have shown significant associations between
mortality and air pollution including PM, and
that a meta-analysis of 12 Western and
Central-eastern European studies "is
supportive of a causal association between
PM and SO2 exposure and all-cause
mortality" (Katsouyanni et al., 1997). The
Eastern and Western European studies used
differing measurement methods for PM,
including PMio, gravimetric "suspended
18 CASAC panelists recommended a discussion of this
issue in the Staff Paper. The Staff Paper notes: "While
greater measurement error for the coarse fraction could
depress a potential coarse particle effect, this would not
explain the results in Topeka relative to other cities. Even
considering relative measurement error, these results
provide no clear evidence implicating coarse particles in
the reported effects." (U.S. EPA, 1996b p. V-64). EPA's
provisional examination of the Lipfert and Wyzga (1997)
paper in the Response to Comments, finds that it is
implausible that most of the effect attributed to PM2.5
could in fact be due to PMio-2.5, since differential
measurement error cannot make a weaker effect appear
stronger than a stronger one, except under extremely
unusual circumstances.
19 The APHEA (Air Pollution and Health: a European
Approach) project was supported by the European Union
Environment 1991-1994 Programme to investigate the
possible short-term health effects of exposure to low or
moderate levels of ambient air pollutants. Eleven European
research groups carried out studies in 15 cities
(Amsterdam, Athens, Barcelona, Bratislava, Cracow,
Helsinki, Koln, Lodz, London, Lyon, Milan, Paris, Poznan,
Rotterdam and Wroclaw) in which air pollutant
concentration data had been collected for at least 5 years.
Initial findings of studies on mortality and hospital
admissions were published in a series of papers in
Supplement 1 to the Journal of Epidemiology and
Community Health in 1996 and a meta-analysis of the
mortality data from 12 cities is currently in press
(Katsouyanni et al., 1997).
-------
12
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
particles," and the British Smoke method.20
The differences in aerometry and the
substantial differences in location and
strength of primary PM emissions sources in
central and eastern Europe as compared to
western Europe or the U.S. might well
explain the different results in these unique
areas. Consequently, integration of these
results would involve comprehensive
examination of the various PM instruments
used, monitor siting in relation to sources,
mass calibration procedures and other aspects
of these studies.21 EPA notes that a number
of European authorities, who are familiar with
this recent literature, have proceeded with
recommendations to strengthen their health
guidelines, risk assessments, or regulations
for PM.22
Aside from the recent literature cited by
these commenters, there are a number of
other recent epidemiological studies that, if
considered in today's decision, would tend to
support EPA's conclusions about the effects
of PM at lower concentrations, assuming their
results were accepted following a full review
in the criteria and CASAC process. For
example, in addition to the APEEA studies,
several other recent epidemiologic studies
have reported significant positive associations
between PM and health effects (Lipsett et al.,
1997; Peters et al., 1997; Borja-Aburto et al.,
1997; Delfino et al., 1997; Scarlett et al.,
1996; Woodruff et al., 1997; Wordley et al.,
1977). In addition, a number of recent
toxicologic papers have been accepted or
appear in proceedings (Costa and Dreher,
1997; Killingsworth et al., 1997; Godleski et
al., 1997) that involve exposure to
concentrated ambient fine particles or PM
constituents and appear to provide supportive
evidence as to the plausibility of the effects
that have been reported epidemic logically. If
considered in this decision, these studies
would also provide biological support for the
epidemiological observation that certain
susceptible groups (notably those with
cardiopulmonary disease) are most likely to
be affected by PM, again assuming the results
were sustained in the full criteria and CASAC
review process.
20 The Roth et al. (1997) study in Prague used a
measurement termed "suspended particles" that appears
to be close to TSP. The relation of this indicator to PMio
or PM2.5 in this city is not reported. Moreover, this study
uses a variant of the problematic methodology in the Roth
analyses cited above.
21 These concerns are consistent with EPA's treatment
of a number of European and South American studies that
are included in the Criteria Document and contributed to
the evaluation of the epidemiology in Chapter 12. Because
of differences in aerometry methods and characteristic
source classes between North America and other regions
of the world, however, the integrative assessment chapter
reported results only from studies conducted in the U.S.
and Canada (cf Tables 13-3 to 13-5) in reaching
quantitative conclusions for effects estimates.
22 See, for example, the United Kingdom Air Quality
Strategy, 1997; Swiss Federal Commission of Air Hygiene,
1996; World Health Organization Revised Air Quality
Guidelines for Europe, In Press).
In summary, EPA has conducted a
provisional assessment of the more recent
scientific literature. Based on this provisional
assessment, EPA disagrees with commenters'
assertion that full consideration of selected
new studies in this decision would materially
change the Criteria Document and Staff Paper
conclusions on the consistency and coherence
of the PM data, or on the need to revise the
current standards.
(iii) Other specific comments on the
epidemiological studies. Aside from their
assertion that EPA ignored or downplayed
particular studies, this second group of
commenters raise additional objections, based
on the statistical modeling strategies used and
the potential importance of personal exposure
misclassification, to EPA's conclusions
regarding the consistency of the
epidemiological evidence. EPA conclusions
on these topics were summarized in the
proposal and supported by extensive
treatments in the Criteria Document and Staff
Paper. With respect to the first issue,
commenters argued that sufficient flexibility
exists in the analyses of large data sets that
it may be possible to obtain almost any result
desired through choice of statistical method.
Analytical choices include the specific
statistical model; methods used to adjust for
seasonal variation and the trends in the data;
treatment of other variables (e.g., other
pollutants, weather, and day of week); "lag"
structure; and study population.
A more detailed discussion of this issue,
which expands on the assessment summarized
in the Criteria Document, is included in the
Response to Comments. In summary, EPA
must reject commenters' contention that
legitimate alternative analyses can obtain
"almost any result." As outlined above in
this unit, EPA's detailed reviews of individual
studies have shown that not all methods are
equally valid or legitimate. Moreover, strong
arguments can be made that the methods and
analytical strategies in the studies EPA relied
upon are more appropriate approaches than
those cited by commenters (e.g., Li and Roth,
1995; Lipfert and Wyzga, 1995; Davis et al.,
1996; Roth and Li, 1997). While not all
studies have addressed each of the above
issues in this unit equally well, the most
comprehensive analyses of these issues (e.g.,
Samet et al., 1995, 1996a,b; Pope and
Kalkstein, 1996), as well as the EPA analyses
comparing study results for each issue (U.S.
EPA, 1996a, pp. 12-261 to 12-305) found that
the authors of studies on which EPA chiefly
relied made appropriate modeling choices.
The Criteria Document concludes that:
"[T]he largely consistent specific results,
indicative of significant positive associations
of ambient PM exposures and human
mortality/morbidity effects, are not model
specific, nor are they artifactualy derived due
to misspecification of any specific model. The
robustness of the results of different modeling
strategies and approaches increases our
confidence in their validity [U.S. EPA, 1996a,
p. 13-54]." While it is true, as evidenced in
Li and Roth (1995), that PM-effects data can
be randomly manipulated to produce
apparently conflicting results, commenters
have provided no evidence that different
plausible model specifications could lead to
markedly different conclusions.
Some commenters have expressed concerns
about the reliability of the epidemiological
results because some studies showed a lack
of correlation in cross-sectional comparisons
between outdoor PM measured at central
locations and indoor or personal exposures to
PM (which includes PM from the outdoor,
indoor and personal environments).23 EPA
acknowledged and responded to this issue in
chapter 7 of the Criteria Document and the
proposal (61 FR 65645, December 13, 1996).
The major premise underlying commenters'
arguments on this issue is incorrect.24 The
question is not whether central monitoring
site measurements contain a signal reflecting
actual exposures to total PM from both
outdoor and indoor sources at the individual
level; the relevant question is whether central
monitoring site measurements contain a
signal reflecting actual exposures to ambient
PM for the subject population, including both
ambient PM, while individuals are outdoors,
and ambient PM that has infiltrated indoors,
while individuals are indoors. The PM
standards are intended to protect the public
from exposure to ambient PM, not PM
generated by indoor or personal sources.
There is ample evidence, as discussed in
chapter 7 of the Criteria Document, that
personal exposure to ambient PM, while
23 Paradoxically, some commenters have argued (e.g.,
Valdberg, 1997) that the PM results are confounded
because the weather and other factors that cause daily
variations in outdoor pollution will cause similar daily
variations in indoor generated air pollution. For this to be
true, outdoor ambient pollution concentrations would have
to be correlated with personal exposure to indoor generated
air pollution such as that from smoking, cleaning, and
cooking. This argument is logically inconsistent with the
other comments on the lack of any such correlation with
personal exposure, and these commenters have offered no
scientific evidence to support their claim. In response, EPA
has performed and included in the Response to Comments
a numerical analysis of the relevant information from the
PTEAM exposure study that finds no evidence for such
a correspondence in the actual data.
24 As documented in Chapter 7 of the Criteria
Document, time-series community studies observe the
effects of varying levels of ambient air pollution; therefore
the effects of indoor-generated air pollution would be
independent of and in addition to the effects found in these
epidemiological studies. Commenters apparently believe
EPA is claiming such studies are detecting the effects of
daily variations in total PM personal exposure from indoor
and outdoor sources. This misunderstanding is evidenced,
for example, by Wyzga and Lipfert's (1995) treatment of
the difference between ambient monitors and actual
personal exposures as "exposure errors" and Brown's
comment for API that "if (ambient) PM is causally related
to mortality/morbidity, then it is personal PM exposure that
must be reduced to have an effect." On the contrary, it
is personal exposure to ambient PM that must be reduced
to address the risk identified in community air pollution
studies. Any lack of significant correlation between
outdoor PM concentrations and personal exposure to total
PM from all sources is irrelevant, except to the extent it
may decrease the power of time-series studies to detect the
effects of ambient pollution.
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
13
outdoors and while in indoor micro-
environments, does correlate on a day-to-day
basis with concentrations measured at
properly sited central monitors (U.S. EPA,
1996a, p. 1-10). EPA has, therefore,
concluded that it is reasonable to presume
that a reduction in ambient PM concentrations
will reduce personal exposure to ambient PM,
and that this will protect the public from
adverse health outcomes associated with
personal exposure to ambient PM.
Commenters have also restated
theoretically based concerns on a related
issue, namely errors in the measurement of
the concentrations of air pollutants, that was
summarized in the proposal. In multiple
pollutant analyses, measurement error or,
more generally, exposure misclassification,
could theoretically bias effects estimates of
PM or co-pollutants in either direction,
introducing further uncertainties in the
estimated concentration-response
relationships for all pollutants (U.S. EPA,
1996b, pp. V-39 to V-43). Relevant insights
on this issue in material appended to public
comments (Ozkaynak and Spengler, 1996)
have prompted an expanded statistical
analysis of the conditions under which such
errors could inflate the magnitude of the
effects estimates or the significance of PM
relative to gaseous pollutants, as has been
suggested by Lipfert and Wyzga (1995). This
analysis, which is summarized in the
Response to Comments, finds that the
conditions under which measurement error
could inflate the effects estimates or
significance of PM relative to other pollutants
are restricted to a limited set of statistical
relationships. Commenters have not provided
evidence that suggest such conditions are
likely to occur with respect to the
measurement of ambient PM in relation to
those for gaseous co-pollutants commonly
used in epidemiological studies.25 Therefore,
it appears unlikely that measurement and
exposure errors for PM and other pollutants
have inflated the estimated effects of PM,
even in multivariate analyses. More
importantly, the available evidence on the
consistency of the PM-effects relationships in
multiple urban locations, with widely varying
indoor/outdoor conditions and a variety of
monitoring approaches, makes it less likely
that the observed associations of PM with
serious health effects at levels allowed under
the current NAAQS are an artifact of errors
25 The EPA analysis finds that in order for measurement
errors in one pollutant variable to significantly bias the
estimated effect of another pollutant, three conditions are
necessary: (1) The measurement error in the poorly
measured pollutant must be very large, roughly at least the
same size as the population variability in that pollutant;
(2) the poorly measured pollutant must be highly
correlated with the other pollutant, either positively or
negatively; and (3) the measurement errors for the two
pollutants must be highly negatively correlated (Response
to Comments, Appendix D). This important factor was not
considered in Lipfert and Wyzga (1995) or by Commenters.
in measurement of pollution or of exposure
(U.S. EPA 1996b, pp. V-39 to V-43).
(iv) Comments on the PM risk assessment.
As noted in the proposal, uncertainties about
measurement errors, exposure
misclassification, and the relative effects of
copollutants are more important to the
quantitative estimates of risk associated with
PM than to the existence of valid PM-effects
associations at levels found in recent studies.
A number of commenters argued that EPA's
risk assessment is flawed and incomplete.
Chief among the reasons they advanced is
that the assessment is based on the same
epidemiological studies these commenters
argued are inadequate for the reasons
summarized and responded to above. Specific
comments also addressed the extent to which
the risk assessment might overstate risk
estimates because it assumes a linear no-
threshold relationship and the use of studies
that might inflate PM risk due to inadequate
consideration of co-pollutants and other
potential confounders. The full risk
assessment acknowledges these issues and
uncertainties, however, and it illustrates the
potential influence of such uncertainties in
sensitivity analyses (U.S. EPA 1996b; chapter
6, appendix F; Abt Associates, 1996a,b;
1997a,b). For example, Figure 2c in the
proposal (61 FR 65653, December 13, 1996)
illustrates the potential influence of what
appears to be the most significant uncertainty
in current information, whether a population
threshold exists below which the effects of
PM no longer occur (61 FR 65653, December
13, 1996). EPA notes that a full consideration
of the uncertainties, including the analysis
summarized above on measurement error,
suggests that the epidemiological studies
might well have understated the total effects
of air pollution; thus, both the direction and
the extent of any bias in the risk estimates
are less clear than commenters suggest.
EPA believes that, even recognizing the
large uncertainties, the key qualitative
insights derived from the risk assessment and
summarized in Unit II.A.3. of this preamble
remain appropriate. While not placing great
weight on the specific numerical estimates,
EPA believes that the risk analysis confirms
the general conclusions drawn primarily from
the epidemiological results themselves, that
there is ample reason to be concerned that
exposure to ambient PM at levels allowed
under the current air quality standards
presents a serious public health problem.
3. Key considerations informing the
decision. Having carefully considered the
public comments on the above matters, EPA
believes the fundamental scientific
conclusions on the effects of PM reached in
the Criteria Document and Staff Paper, and
restated in the introduction to this unit,
remain valid. That is, the epidemiological
evidence for ambient PM, alone or in
combination with other pollutants, shows
associations with premature mortality,
hospital admissions, respiratory symptoms,
and lung function decrements. Despite
extensive critical examination in the criteria
and standards review, these findings cannot
be otherwise explained by analytical, data, or
other problems inherent in the conduct of
such studies. Although the evidence from
lexicological studies available during the
criteria review has not revealed demonstrated
mechanisms that explain the range of effects
reported in epidemiological studies, it does
not and cannot refute the observation of such
effects in exposed populations. Moreover, the
effects observed in the recent epidemiological
studies at lower PM concentrations are both
coherent with each other and plausible based
on the categories of effects observed at much
higher concentrations in historic air pollution
episodes, laboratory studies of PM effects at
high doses, and particle dosimetry studies.
The consistency of the results from a large
number of locations and the coherent nature
of the observed results suggest a likely causal
role of ambient PM in contributing to the
reported effects (U.S. EPA, 1996a; p. 13-1).
Many of the studies showing PM effects were
conducted in areas where the current PMi0
standards are largely met, and both the studies
and EPA's risk assessment suggest that the
collective magnitude of the effects reflects a
significant public health problem.
For these reasons, and having considered
public comments on this issue, the
Administrator concludes that the review of
the criteria and standards provides strong
evidence that the current PMio standards do
not adequately protect public health, and that
revision of the standards is not only
appropriate, but necessary.
Aside from that conclusion, the
appropriateness of continuing to rely on the
use of PMio as the sole indicator for revised
PM standards is also relevant here. While the
basis for decisions on specific indicators is
discussed more fully in Unit II.C. of this
preamble, this issue is related to the
Administrator's decision on the need to revise
the standards. Based on both the staff review
(U.S. EPA, 1996b, p. VII-3) and the
recommendations of some commenters (e.g.,
California EPA), there are two alternative
approaches for providing additional health
protection in revising the standards: Adopt
tighter PMio standards and/or recognize the
fundamental differences between fine and
coarse particles and develop separate
standards for the major components of PMio,
including fine particles. Conceptually, the
first approach would give weight to
comments that standards should be based on
pollutant indicators for which the most data
have been collected, with less consideration
of the evidence that suggests that the current
standards provide adequate protection against
the effects of coarse particles, and that
tightening the current PMio standards in an
attempt to control fine particles would place
unnecessary requirements on coarse particles.
Because the PMio network is in place, a more
stringent PMio standard would also respond
-------
14
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
to commenters who have expressed a desire
for more immediate implementation of
revised standards. The second approach is
based on the view that, in the long run, more
effective and efficient protection can be
provided by separately targeting appropriate
levels of controls to fine and coarse PM.
The Staff Paper examined this issue in
detail (U.S. EPA 1996b, pp. VII-3 to VII-11),
and concluded that the available information
was sufficient to develop separate indicators
for fine and coarse fractions of PMi0, based
on the recent health evidence, the
fundamental differences between fine- and
coarse-fraction particles, and implementation
experience with PMio. Further, the staff
concluded that:
[Consideration of comparisons between fine and
coarse fractions suggests that fine fraction particles
are a better surrogate for those particle components
linked to mortality and morbidity effects at levels
below the current standards. In contrast, coarse
fraction particles are more likely linked with certain
effects at levels above those allowed by the current
PMio standards. In examining alternative
approaches to increasing the protection afforded by
PMio standards, the staff concludes that reducing
the levels of the current PMio standards would not
provide the most effective and efficient protection
from these health effects. [U.S. EPA 1996b; p. 7-
45]
As discussed in Unit II.C. of this preamble,
the Administrator believes that it is more
appropriate to provide additional protection
against the risk posed by PM by adding new
standards for the fine fraction of PMio, as
opposed to tightening the current PMio
standards. Although fewer epidemiological
studies have used PM2.5 and other fine
particle indicators (e.g., sulfates, acids), there
are nonetheless significant indications from
the scientific evidence - drawn from the
physicochemical studies of PM, air quality
and exposure information, toxicological
studies, and respiratory tract deposition data
- that this approach will provide the most
effective and efficient protection of public
health.
Several commenters have argued that the
decision on whether to revise the PM
standards should be deferred, particularly
with regard to fine particle standards, pending
establishment and operation of a national
monitoring network to characterize fine PM
and a research program to reduce
uncertainties in the effects information. These
commenters expressed concerns that
establishing fine PM standards now might
result in needless regulation of PM
components that may be unrelated to
observed health effects. As discussed more
fully in Unit IFF. of this preamble, such
commenters recommended, at most, that if
fine PM standards were established, they
should be set at a level "equivalent" to the
current PM standards.
EPA strongly disagrees that the decision on
revising the standards should be delayed to
await the results of new PM monitoring and
research programs. Under section 109(d) of
the Act, EPA's obligation after reviewing the
existing criteria and standards for PM is to
make such revisions in the standards and to
promulgate such new standards as are
appropriate under section 109(b) of the Act.
Based on her review of the criteria and
standards for PM, the Administrator has
concluded that the current standards are not
adequate to protect public health and that
revisions are appropriate. In the face of the
available evidence, a delay in revising the
standards would not only be inconsistent with
the statute but — even under the optimistic
assumption that the same extensive
monitoring and strategy assessment as now
contemplated would occur in the absence of
a revised standard — would add
approximately 2 years to the time when
significant health benefits can be realized,
resulting in potentially significant numbers of
additional premature deaths and even larger
numbers of children and individuals with air
pollution-related illness and symptoms. On
the other hand, establishing standards now
will set into motion the development of
implementation programs and monitoring that
can be conducted in parallel with additional
scientific research, without undue delays
inherent in waiting for the research.
The question of which pollutant
components to regulate has been an issue
since the inception of the first PM standards.
Other ambient pollutants (e.g., NO2 or CO)
are uniquely defined as individual chemicals,
whether or not they serve as proxies for a
larger class of substances (e.g., ozone as an
index of photochemical oxidants). Regulating
general PM, as opposed to multiple chemical
components of PM, raises the spectre of a
host of particulate materials of varying
composition, size, and other physicochemical
properties, not all of which are likely to
produce identical effects.
Both EPA's past and present regulatory
experience with PM control programs and its
successive reviews of the standards have
reaffirmed the wisdom of retaining standards
that control particles as a group, rather than
eliminating such standards and waiting for
scientific research to develop information
needed to identify more precise limits for the
literally thousands of particle components.
Each such decision recognized the possibility
that potentially less harmful particles might
be included in the mix that was regulated, but
concluded that the need to provide protection
against serious health effects nonetheless
required action under section 109 of the Act.
The success of this approach is evident in
early U.S. control programs that dramatically
reduced "smoke" and "TSP" in major cities
in the 1960's and 1970's and in the continued
improvement in air quality through the
current PM standards. The major refinements
that have been recommended through the
course of reviews of PM standards have been
to improve the focus of control efforts by
defining scientifically based size classes (i.e.,
moving from TSP to PMio and now, PM2.5)
that will permit more effective and efficient
regulation of those fractions most likely to
present significant risks to health and the
environment.
As discussed in Unit II.C. of this preamble,
the current review has examined the available
evidence to determine whether it would tend
to support inclusion or exclusion of any
physical or chemical classes of PM, for
example sulfates, nitrates, or ultra-fine
particles. That examination concludes that,
while both fine and coarse particles can
produce health effects, the fine fraction
appears to contain more of the reactive
substances potentially linked to the kinds of
effects observed in the recent epidemiological
studies (U.S. EPA 1996b, section V.F.).
However, the available scientific information
does not rule out any one of these
components as contributing to fine particle
effects. Indeed, it is reasonable to anticipate
that no single component will prove to be
responsible for all of the effects of PM.
EPA recognizes that whether the standards
are set for PMio only or also for fine
particles, there are uncertainties with respect
to the relative risk presented by various
components of PM. In this regard, the
Administrator places greater weight on the
concern that by failing to act now, the PM
NAAQS would not control adequately those
components of air pollution that are most
responsible for serious effects, than on the
possibility they might also control some
component that is not. EPA believes that
moving simultaneously to establish standards
based on the best available scientific evidence
and to conduct an aggressive monitoring and
scientific research program designed to help
resolve current uncertainties is a prudent and
responsible approach for addressing both the
risks and the uncertainties inherent in this
important public health issue.
In summary, given the evidence that PM-
related health effects appear likely to occur
at levels below the current standards, the
serious nature and potential magnitude of the
public health risks involved, and the need to
consider the fine and coarse fractions as
distinct classes of particles, the Staff Paper
and the CASAC (Wolff, 1996b) concluded
that revision of the current standards is
clearly appropriate. Moreover, at their May
1996 public meeting (U.S. EPA, 1996c), and
in separate written comments (including
Fippmann et al., 1996), a majority of CASAC
panel members recommended revisions that
would strengthen the health protection
provided by the current PM standards. Based
on the rationale and recommendations
contained in the Staff Paper and the advice
of CASAC, and taking into account public
comments, the Administrator concludes that it
is appropriate at this time to revise the current
PM standards to increase the public health
protection provided against the known and
potential effects of PM identified in the air
quality criteria.
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
15
C. Indicators ofPM
In establishing adequately protective,
effective, and efficient PM standards, it is
necessary to specify the fraction of particles
found in the ambient air that should be used
as the indicators) for the standards. In this
regard, EPA concludes that the most recent
assessment of scientific information in the
Criteria Document, summarized in chapters
IV and V of the Staff Paper, continues to
support past staff and CASAC
recommendations regarding the selection of
size-specific indicators for PM standards.
More specifically, EPA continues to find that
the following conclusions reached in the Staff
Paper and in the 1987 review remain valid:
(1) Health risks posed by inhaled particles
are influenced both by the penetration and
deposition of particles in the various regions
of the respiratory tract and by the biological
responses to these deposited materials.
(2) The risks of adverse health effects
associated with deposition of ambient fine
and coarse fraction particles in the thoracic
(tracheobronchial and alveolar) regions of the
respiratory tract are markedly greater than for
deposition in the extrathoracic (head) region.
Maximum particle penetration to the thoracic
region occurs during oronasal or mouth
breathing.
(3) The risks of adverse health effects from
extrathoracic deposition of general ambient
PM are sufficiently low that particles which
deposit only in that region can safely be
excluded from the standard indicator.
(4) The size-specific indicator(s) should
represent those particles capable of
penetrating to the thoracic region, including
both the tracheobronchial and alveolar
regions.
These conclusions, together with
information on the dosimetry of particles in
humans, were the basis for the promulgation
in 1987 of a new size-specific indicator for
the PM NAAQS, PMio, that includes
particles with an aerodynamic diameter
smaller than or equal to a nominal 10 |im.
The recent information on human particle
dosimetry contained in the Criteria Document
provides no basis for changing 10 |im as the
appropriate cut point for particles capable of
penetrating to the thoracic regions.
As noted in Unit II.B. of this preamble,
however, the Staff Paper concludes that
continued use of PMi0 as the sole indicator
for the PM standards would not provide the
most effective and efficient protection from
the health effects of PM (U.S. EPA, 1996b,
pp. VII-4 to VII-11). Based on the recent
health effects evidence and the fundamental
physical and chemical differences between
fine and coarse fraction particles, the Criteria
Document and Staff Paper conclude that fine
and coarse fractions of PMi0 should be
considered separately (U.S. EPA, 1996a, p.
13-93; 1996b, p. VII-18). Taking into account
such information, CASAC found sufficient
scientific and technical bases to support
establishment of separate standards relating to
these two fractions of PMi0. Specifically,
CASAC advised the Administrator that
"there is a consensus that retaining an annual
PMio NAAQS * * * is reasonable at this
time" and that there is "also a consensus that
a new PM2 5 NAAQS be established" (Wolff,
1996b).
Some commenters have noted that it is
often difficult to distinguish the effects of
either fine or coarse fraction particles from
those of PMio; this is to be expected because
both fractions are themselves components of
PMio, and hence not fully independent. EPA
believes that it is more meaningful to
examine comparisons between the fine and
coarse fraction components. Such
comparisons presented in the Staff Paper
suggest that fine particles are a better
surrogate for those components of PM that
are linked to mortality and morbidity effects
at levels below the current standards (U.S.
EPA, 1996b, p. VII-18). Moreover, a
regulatory focus on fine particles would likely
also result in controls on gaseous precursors
of fine particles (e.g., SOX, NOX, VOC),
which are all components of the complex
mixture of air pollution that has most
generally been associated with mortality and
morbidity effects. The Staff Paper concludes
that, in contrast to fine particles, coarse
fraction particles are more clearly linked with
certain morbidity effects at levels above those
allowed by the current 24-hour standard.
Public comments received on the proposed
indicators were overwhelmingly in favor of
EPA's proposal to maintain PMio as an
indicator for PM, whether as an indicator of
coarse particles in conjunction with a fine PM
standard, or as the sole PM indicator. This
near unanimity shows strong support for
retaining general PM standards. While a
substantial number of commenters supported
EPA's proposal to add an indicator for fine
PM, a number of other commenters objected
to any standard revisions, including addition
of a fine PM indicator. Beyond the general
points about the basis for any revisions
discussed in Unit II.B. of this preamble, these
commenters argued either that the available
epidemiological data did not provide a basis
for separating fine and coarse fraction
particles, or that there were not enough fine
particle studies to support selecting standard
levels. Most of these commenters also
expressed concerns that there were
insufficient ambient fine particle data by
which to evaluate the relative protection
afforded by new standards.
EPA notes that issues relating to the basis
for separating PMio fractions were addressed
in the Criteria Document and/or Staff Paper
assessments, and these perspectives were also
available for CASAC consideration in
developing its recommendations. The
proposal states that the main basis for
separating the fine and coarse fractions of
PMio is that, because they are fundamentally
different PM components with significantly
different physico-chemical properties and
origins (U.S. EPA 1996b, section V.D),
separate standards would permit more
effective and efficient regulation of PM.
While the difficulty in separating these
classes in the epidemiological studies is noted
above, the preponderance of the available
evidence suggests that strategies to control
fine particles will more effectively reduce
population exposure to substances associated
with health effects in the recent
epidemiological studies. Although the number
of studies using fine PM indicators is more
limited than for PMio, there are more than 20
community studies showing significant
associations for a consistent set of mortality
and morbidity effects. A substantial subset of
these studies (Tables V-12 to V-13; U.S.
EPA, 1996b) provides a sufficient
quantitative basis for selecting standard
levels, without the need to rely on estimates
based on PM2.5/PMio ratios.
Having considered the public comments on
this issue, the Administrator concurs with
staff and CASAC recommendations to control
particles of health concern (i.e., PMio)
through separate standards for fine and coarse
fraction particles. The following units outline
the basis for the Administrator's decision on
specific indicators for fine and coarse fraction
particle standards.
1. Indicators for the fine fraction of PMio-
The Administrator continues to conclude that
it is appropriate to control fine particles as a
group, as opposed to singling out particular
components or classes of fine particles. The
more qualitative scientific literature,
evaluated in Chapter 11 of the Criteria
Document and summarized in section V.C of
the Staff Paper, has reported various health
effects associated with high concentrations of
a number of fine particle components (e.g.,
sulfates, nitrates, organics, transition metals),
alone or in some cases in combination with
gases. Community epidemiolgical studies
have found significant associations between
fine particles or PMio and health effects in
various areas across the U.S. where such fine
particle components correlate significantly
with particle mass. As noted above in this
unit, it is not possible to rule out any one of
these components as contributing to fine
particle effects.26 Thus, the Administrator
finds that the present data more readily
support a standard based on the total mass of
fine particles. EPA will conduct additional
research, in cooperation with other Federal
26 As discussed above, a number of commenters
expressed concerns that various portions of fine particles
might not be responsible for any observed effects. One
group (PG&E, 1997) recommended that nitrates should be
excluded from fine PM mass collected on the basis of their
assessment of available effects literature on particulate and
gas phase inorganic nitrates. Based on an examination of
this information as well as the earlier staff assessment,
EPA maintains its conclusion that the available evidence
is not sufficient to exclude nitrates or any other class of
fine particles that are collected by PM monitors
comparable to those used in the recent epidemiological
studies.
-------
16
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
agencies and in partnership with State and
local agencies and the private sector, to better
identify which species are of concern for
human health, and the sources and relative
magnitude of such species.
In specifying a precise size range for a fine
particle standard, both the staff and CASAC
recommended PM2.5 as the indicator of fine
particles (Wolff, 1996b). The particle
diameter reflecting the mass minimum
between the fine and coarse modes typically
lies between 1 and 3 |im, and the scientific
data support a sampling "cut point" to
delineate fine particles somewhere in this
range. Because of the potential overlap of fine
and coarse particle mass in this intermodal
region, EPA recognizes that any specific
sampling cut point would result in only an
approximation of the actual fine-mode
particle mass. Thus, the choice of a specific
diameter within this size range is largely a
policy judgment. The staff and CASAC
recommendations for a 2.5 |im sampling cut
point were based on considerations of
consistency with the community health
studies, the limited potential for intrusion of
coarse fraction particles into the fine fraction,
and availability of monitoring technology.27
PM2.5 encompasses all of the potential agents
of concern in the fine fraction, including most
sulfates, acids, fine particle transition metals,
organics, and ultrafine particles, and includes
most of the aggregate surface area and
particle number in the entire distribution of
atmospheric particles.
The Administrator concurs with the staff
and CASAC recommendations and concludes
that PM2.5 is the appropriate indicator for fine
particle standards. As discussed in Unit VI.B.
of this preamble, technical details of how
PM2.5 is to be measured in the ambient air
are specified in the Federal Reference Method
(40 CFR part 50, Appendix L).
2. Indicators for the coarse fraction of
PMio. The Criteria Document and Staff Paper
conclude that epidemiological information,
together with dosimetry and toxicological
information, support the need for a particle
indicator that addresses the health effects
associated with coarse fraction particles
within PMio (i.e., PMio-2.s). As noted above,
27 The National Mining Association (NMA) and related
companies submitted comments favoring ultimate selection
of a smaller cutpoint of 1 (Im (PMi) to further reduce
coarse particle intrusion. EPA considered this approach in
developing the Staff Paper and proposal. PMi has not been
used in health studies, although in most cases collected
mass should be similar to those for cutpoints of 2.1 or 2.5
(Im. While a PMi indicator could reduce intrusion of
coarse particles, it might also omit portions of hygroscopic
PM components such as acid sulfates, nitrates, and some
organic compounds in higher humidity environments
picked up by PM2.5 measurements. PMi sampling
technologies have been developed, but have not been
widely used in the field to date; there are some concerns
about loss of certain organic materials in available models
relative to an instrument with a larger size cut. NMA has
also recommended consideration of a methodology that
could subtract coarse mass from PM2.5 measurements
where undue coarse particle intrusion resulted in fine
standard violations. EPA will evaluate this
recommendation in the context of implementation policies.
coarse fraction particles can deposit in those
sensitive regions of the lung of most concern.
Although the role of coarse fraction particles
in much of the recent epidemiological results
is unclear, limited evidence from studies
where coarse fraction particles are the
dominant fraction of PMio suggest that
significant short-term effects related to coarse
fraction particles include aggravation of
asthma and increased upper respiratory
illness. In addition, qualitative evidence
suggests that potential chronic effects may be
associated with long-term exposure to high
concentrations of coarse fraction particles.
In selecting an indicator for coarse fraction
particles, the Administrator took into account
the views of several CASAC panel members
who suggested using the coarse fraction
directly (i.e., PMio-2.s) as the indicator.
However, the Administrator notes that the
existing ambient data base for coarse fraction
particles is smaller than that for fine particles,
and that the only studies of clear quantitative
relevance to effects most likely associated
with coarse fraction particles have used
undifferentiated PMi0. In fact, it was the
consensus of CASAC that it is reasonable to
consider PMio itself as a surrogate for coarse
fraction particles, when used together with
PM2.5 standards. The monitoring network
already in place for PMio is large. Therefore,
in conjunction with the decision to have
separate standards for PM2.5, the
Administrator concludes, consistent with
CASAC recommendations and public
comments, that it is appropriate to retain
PMio as the indicator for PM standards
intended to protect against the effects most
likely associated with coarse fraction
particles.
D. Averaging Time ofPM^.s Standards
As discussed above in this unit, the
Administrator has concluded that PM2.5 is an
appropriate indicator for standards intended to
provide protection from effects associated
primarily with fine particles. The recent
health effects information includes reported
associations with both short-term (from less
than 1 day to up to 5 days) and long-term
(from a year to several years) measures of
PM.
On the basis of this information,
summarized in chapter V of the Staff Paper
and in the rationale presented in the proposal,
the Administrator has considered both short-
and long-term PM2 5 standards.
1. Short-term PM2.5 standard. The current
24-hour averaging time is consistent with the
majority of community epidemiological
studies, which have reported associations of
health effects with 24-hour concentrations of
various PM indicators such as PMio, fine
particles, and TSP. Such health effects,
including premature mortality and increased
hospital admissions, have generally been
reported with same-day, previous day, or
longer lagged single-day concentrations,
although some studies have reported stronger
associations with multiple-day average
concentrations. In any case, the Administrator
recognizes that a 24-hour PM2 5 standard can
effectively protect against episodes lasting
several days, since attainment of such a
standard would provide protection on each
day of a multi-day episode, while also
protecting sensitive individuals who may
experience effects after even a single day of
exposure.
Although most reported effects have been
associated with daily or longer measures of
PM, evidence also suggests that some effects
may be associated with PM exposures of
shorter durations. For example, controlled
human and animal exposures to specific
components of fine particles, such as acid
aerosols, suggest that bronchoconstriction can
occur after exposures of minutes to hours.
Some epidemiological studies of exposures to
acid aerosols have also found changes in
respiratory symptoms in children using
averaging times less than 24 hours. However,
such reported results do not provide a
satisfactory quantitative basis for setting a
fine particle standard with an averaging time
of less than 24 hours, nor do current
gravimetric mass monitoring devices make
such shorter durations generally practical at
present. Further, the Administrator recognizes
that a 24-hour average PM2.5 standard which
leads to reductions in 24-hour average
concentrations is likely to lead as well to
reductions in shorter-term average
concentrations in most urban atmospheres,
thus providing some degree of protection
from potential effects associated with shorter
duration exposures.
2. Long-term PM2.5 standard. Community
epidemiological studies have reported
associations of annual and multi-year average
concentrations of PMio, PM2.5, sulfates, and
TSP with an array of health effects, notably
premature mortality, increased respiratory
symptoms and illness (e.g., bronchitis and
cough in children), and reduced lung
function. The relative risks associated with
such measures of long-term exposures,
although highly uncertain, appear to be larger
than those associated with short-term
exposures. Based on the available
epidemiology, and consistent with the limited
relevant toxicological and dosimetric
information, the Administrator concludes that
significant, and potentially independent,
health consequences are likely associated
with long-term PM exposures.
The Administrator has considered this
evidence, which suggests that some health
endpoints reflect the cumulative effects of
PM exposures over a number of years. In
such cases, an annual standard would provide
effective protection against persistent long-
term (several years) exposures to PM.
Requiring a much longer averaging time
would also complicate and unnecessarily
delay control strategies and attainment
decisions.
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
17
The Administrator has also considered the
seasonality of emissions of fine particles and
their precursors in some areas (e.g.,
wintertime smoke from residential wood
combustion, summertime regional acid sulfate
and ozone formation), which suggests that
some effects associated with annual average
concentrations might be the result of repeated
seasonally high exposures. However, different
seasons are likely of concern in different parts
of the country, and the current evidence does
not provide a satisfactory quantitative basis
for setting a national fine particle standard in
terms of a seasonal averaging time.
In addition, the Administrator recognizes
that an annual standard would have the effect
of improving air quality broadly across the
entire annual distribution of 24-hour PM2.5
concentrations, although such a standard
would not as effectively limit peak 24-hour
concentrations as would a 24-hour standard.
The risk assessment summarized above found
that because such 24-hour peaks contribute
much less to the total health risk over a year
than the more numerous low- to mid-range
PM2.5 levels, an annual standard could also
provide effective protection from health
effects associated with short-term exposures
to PM2.5 as well as those associated with
long-term exposures (see figure 2; 61 FR
65652-65653, December 13, 1996).
3. Combined effect of annual and 24-hour
standards. For the reasons outlined in Units
II.C.l. and 2. of this preamble, the
Administrator concluded in the proposal that
a short-term PM2.5 standard with a 24-hour
averaging time can serve to control short-term
ambient PM2.5 concentrations, thus providing
protection from health effects associated with
short-term (from less than 1-day to up to 5-
day) exposures to PM2.5. Further, a long-term
PM2.5 standard with an annual averaging time
can serve to control both long- and short-term
ambient PM2.5 concentrations, thus providing
protection from health effects associated with
long-term (seasonal to several years) and, to
some degree, short-term exposures to PM2.5.
EPA received comparatively few public
comments on these proposed averaging times.
Those supporting PM2.5 standards also
strongly supported adopting both annual and
24-hour averaging times. Many of those
opposing PM2.5 standards, for the reasons
discussed in Unit II.B. of this preamble,
provided contingent comments that variously
supported both averaging times for PM2.5
standards in the event the Administrator
disagreed with their overall
recommendations. Other opponents of PM2.5
standards disagreed with having two
standards on administrative grounds, or
because some CASAC members did not
support both averaging times.
The relationship between standards for the
two averaging times is discussed below in
this unit. In essence, based on its examination
of the effects data and air quality
relationships, EPA believes that a single
PM2.5 standard (24-hour or annual) either
would not provide adequate protection against
effects of concern for all averaging times, or
would be inefficient in the sense that it was
more stringent than necessary for at least one
averaging time. Contrary to commenters who
focused on minority CASAC opinions, EPA
notes that a clear majority of CASAC
supported both 24-hour and annual
standards28. After considering public
comments on averaging time and the rationale
outlined above, the Administrator has
concluded that both 24-hour and annual
PM2.5 standards are appropriate.
The Administrator next considered the
potential combined effects of such standards
on PM concentration levels and distributions.
The existing health effects evidence could, of
course, be used to assess the form and level
of each standard independently, with short-
term exposure health effects evidence being
used as the basis for a 24-hour standard and
the long-term exposure health effects
evidence as the entire basis for an annual
standard. Some CASAC panel members
apparently used this approach as a basis for
their views on appropriate averaging times
and standard levels. In particular, a few
members focused only on a 24-hour PM2.5
standard in light of the relative strength of the
short-term exposure studies. On the other
hand, two members focused only on an
annual standard, recognizing that strategies to
meet an annual standard would provide
protection against effects of both short- and
long-term exposures.
As noted above in this unit, attempting to
provide protection for all of the effects
identified in long- and short-term PM
exposure studies with a single averaging time
would result in either inadequate protection
for some effects, or unnecessarily stringent
control for others. The Administrator has,
instead, emphasized a policy approach that
considers the consistency and coherence, as
well as the limitations, of the body of
evidence as a whole, and recognizes that there
are various ways to combine two standards
to achieve an appropriate degree of public
health protection. Such an approach to
standard setting, which integrates the body of
health effects evidence and air quality
analyses, and considers the combined effect
of the standards, has the potential to result in
a more effective and efficient suite of
standards than an approach that only
considers short- and long-term exposure
evidence, analyses, and standards
independently.
In considering the combined effect of such
standards, the Administrator notes that while
an annual standard would focus control
programs on annual average PM2 5
concentrations, it would also result in fewer
and lower 24-hour peak concentrations.
Alternatively, a 24-hour standard that focuses
controls on peak concentrations could also
result in lower annual average concentrations.
Thus, either standard could be viewed as
providing both short- and long-term
protection, with the other standard serving to
address situations where the daily peaks and
annual averages are not consistently
correlated.
The Administrator proposed that the suite
of PM2.5 standards could most effectively and
efficiently be defined by treating the annual
standard as the generally controlling standard
for lowering both short- and long-term PM2 5
concentrations. In conjunction with the annual
standard, the 24-hour standard would serve to
provide protection against days with high
peakPM25 concentrations, localized "hot
spots," and risks arising from seasonal
emissions that would not be well controlled
by a national annual standard.
Relatively few public comments were
addressed specifically to the proposal that the
annual standard be directed toward
controlling both 24-hour and annual levels
(thereby basing the annual standard on an
evaluation of both the short- and long-term
health effects information), with the 24-hour
standard being used to address more localized
short-term peaks. A number of commenters,
notably some among the groups opposing any
revised PM standards, appeared to have
ignored this fundamental aspect of the
proposal, judging by their assertions that the
sole basis for EPA's proposed annual
standards was two long-term exposure studies
(Dockery et al., 1993; Pope et al. 1995). This
is incorrect; as the proposal states, EPA based
the proposed annual standard level on a wider
range of short- and long-term exposure
studies. Other commenters, including some
environmental groups, reserved comment on
this specific issue, but expressed concerns
that the specific levels for both standards
were not stringent enough, regardless of
which standard is intended to be controlling.
Issues regarding specific levels are discussed
below in Unit II.F. of this preamble.
Some commenters, however, disagreed
with the proposition that EPA's proposed
approach would necessarily provide the most
effective and efficient standards. In the view
of some who opposed PM2.5 standards, the
likelihood that there are thresholds below
which no effects occur means that a 24-hour
standard would be more efficient than an
annual standard. In this view, the reductions
made on days that were below the threshold
would provide no protection.29 Some
28 Of the 19 panel members who joined in the consensus
for PM2.5 standards, 17 (90 percent) recommended a 24-
hour standard and 13 (70 percent) recommended an annual
standard (Wolff, 1996b).
29 A related comment criticized the risk assessment
conclusion that peak 24-hour concentrations contribute
much less to the total risk over a year as inconsistent with
the experience in historic air pollution episodes. EPA
disagrees. While the historic London episodes were
quantitatively different from those assumed in the risk
assessment, the record over 14 London winters indicates
a continuum of effects down to the lowest levels. It is
therefore likely that the cumulative increase in mortality
calculated for all the days in the whole 14-year period
Continued
-------
18
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
commenters also noted that while a majority
of CASAC members favored both annual and
24-hour standards, more recommended 24-
hour standards.
While the available epidemiological studies
provide strong evidence suggesting that PM
causes or contributes to health effects at
levels below the current standards, EPA
agrees, as stated previously, that uncertainties
increase markedly at lower concentrations.
Nevertheless, the level or even existence of
population thresholds below which no effects
occur cannot be reliably determined by an
examination of the results from the available
studies. Analyses have placed some limits,
however, and EPA has considered
hypothetical thresholds in its risk assessment.
As noted in Unit II. A. of this preamble, even
assuming an example threshold of 18 |ig/m3,
the risk assessment (see Figure 2c; 61 FR
65653, December 13, 1996) finds that most
of the annual aggregate risk associated with
short-term exposures still results from the
large number of days at lower to mid-range
values above the mean. Given that neither the
Criteria Document nor commenters have
provided quantitative evidence regarding the
likelihood of a threshold at levels much
higher than the above example, EPA believes
that the evidence provided in the risk
assessment does not support the commenters'
position. As noted above, EPA believes that
most CASAC opinions on averaging time
reflect panelists' judgments on the relative
strength of the short-term exposure
epidemiological studies, a judgment that EPA
shares. Although most CASAC panel
members did not offer an opinion on the use
of short-term exposure studies in specifying
annual standards, two panelists did support
this notion. EPA therefore believes this
approach is neither inconsistent with the
underlying science nor discordant with the
advice of CASAC.
Another concern was raised by some air
pollution control officials who otherwise
supported revised PM standards. These
commenters state that, from an
implementation perspective, it is often easier
to design control strategies for single short-
term events than for annual averages. Aside
from whether this is a proper consideration
in establishing NAAQS, the point in fact
highlights one of the important strengths of
an annual standard in addressing short-term
risks associated with PM2.5. As noted by the
commenters, risk management for a short-
term standard focuses on a characteristic
"design value" episode responsible for peak
concentrations. For PM, such peak values can
be associated with single source
contributions. Meteorology, relative source
contributions, and resulting particle
composition for that day may or may not be
typical for the area or for the year. Yet the
short-term exposure epidemiological results
are largely drawn from studies that associated
variations in area-wide effects with
monitor(s) that gauged the variation in daily
levels over the course of up to 8 years. The
strength of the associations in these data is
demonstrably in the numerous "typical" days
in the upper to middle portion of the annual
distribution, not on the peak days.30 For these
reasons, strategies that focus only on reducing
peak days are less likely to achieve reduction
of the mix and sources of urban and regional-
scale PM pollution most strongly associated
with health effects. Although designing
control strategies to reduce annual levels may
be more difficult than for 24-hour standards,
the available short- and long-term
epidemiological data suggest it is also likely
to result in a greater reduction in area-wide
population exposure and risk.
The Administrator concludes that the most
effective and efficient approach to
establishing PM2.5 standards is to treat the
annual standard as the generally controlling
standard for lowering both short- and long-
term PM2.5 concentrations, while the 24-hour
standard would serve to provide protection
against days with high peak PM2.5
concentrations, localized "hot spots," and
risks arising from seasonal emissions that
would not be well controlled by a national
annual standard. In reaching this view, the
Administrator took into account the public
comments and the factors discussed below in
this unit.
(1) Based on one of the key observations
from the quantitative risk assessment
summarized above (see Figures 2a,b,c; 61 FR
65652-65653, December 13, 1996), the
Administrator notes that much if not most of
the aggregate annual risk associated with
short-term exposures results from the large
number of days during which the 24-hour
average concentrations are in the low- to mid-
range, below the peak 24-hour concentrations.
As a result, lowering a wide range of ambient
24-hour PM2.5 concentrations, as opposed to
focusing on control of peak 24-hour
concentrations, is the most effective and
efficient way to reduce total population risk.
Further, there is no evidence suggesting that
risks associated with long-term exposures are
likely to be disproportionately driven by peak
24-hour concentrations. Thus, an annual
standard that controls an area's attainment
status is likely to reduce aggregate risks
associated with both short- and long-term
exposures with more certainty than a 24-hour
standard.
(2) The consistency and coherence of the
health effects data base are, therefore, more
directly related to the more frequently
occurring PM exposures reflected in study
period mean measures of air quality (e.g., the
annual distributions of 24-hour PM
would not be dominated by the more limited number of
episode days.
30 This point is buttressed by studies that have taken
out a limited number of higher PM concentration days with
little effect on the effects estimates or significance of the
association (e.g., Schwartz et al., 1996; Pope and Dockery,
1992).
concentrations), than to the potentially site-
specific and/or otherwise infrequent PM
exposures reflected in a limited number of
peak 24-hour concentrations. More
specifically, judgments about the quantitative
consistency of the large number of short-term
exposure studies reporting associations with
24-hour concentrations arise from comparing
the relative risk results per PM increment as
derived from analyzing the associations
across the entire duration of the studies.
These studies typically spanned at least an
annual time frame and the reported
associations are most strongly influenced by
the large number of days toward the middle
of the distribution.
(3) An annual average measure of air
quality is more stable over time than are 24-
hour measures. Thus, a controlling annual
standard is likely to result in the development
of more consistent risk reduction strategies
over time, since an area's attainment status
will be less likely to change due solely to
year-to-year variations in meteorological
conditions that affect the formation of fine
particles, than under a controlling 24-hour
standard.
Under this policy approach, the annual
PM2.5 standard would serve in most areas as
the target for control programs designed to be
effective in lowering the broad distribution of
PM2.5 concentrations, thus protecting not only
against long-term effects but also short-term
effects as well. In combination with such an
annual standard, the 24-hour PM2.5 standard
would be set so as to protect against the
occurrence of peak 24-hour concentrations,
particularly peak concentrations that present
localized or seasonal exposures of concern in
areas where the highest 24-hour-to-annual
mean PM2.5 ratios are appreciably above the
national average.
E. Form ofPM2.5 Standards
1. Annual standard. As discussed in some
detail during the last review of the PM
NAAQS (see 49 FR 10408, March 20, 1984;
52 FR 24634, July 1, 1987) and in the
December 13, 1996 proposal, the annual
arithmetic mean form of the current annual
PMio standard (i.e., the annual arithmetic
mean averaged over 3 years) is a relatively
stable measure of air quality that reflects the
total cumulative dose of PM to which an
individual or population is exposed. Short-
term peaks have an influence on the
arithmetic mean that is proportional to their
frequency, magnitude, and duration, and,
thus, their contribution to cumulative
exposure and risk. As a result, the annual
arithmetic mean form of an annual standard
provides protection across a wide range of the
air quality distribution contributing to
exposure and risk, in contrast to other forms,
such as the geometric mean, that de-
emphasize the effects of short-term peak
concentrations.
While almost no commenters took specific
issue with use of an annual arithmetic mean,
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
19
a number of commenters disagreed with
averaging over 3 years for both the annual
and 24-hour standards because of their desire
for quick action in the initial implementation
ofPM2.5 controls. The Administrator
recognizes the importance of promptly
implementing appropriate control programs,
but she does not believe that implementation
start-up concerns are an adequate basis for
adopting a form (e.g., a single year annual
average) that would provide less stable risk
reduction in the long-run. Therefore, the
Administrator continues to concur with the
Staff Paper recommendation, supported by
CASAC, to use the annual arithmetic mean,
averaged over 3 years, as the form for an
annual PM2.5 standard consistent with the
current form of the annual PMi0 standard.
Nevertheless, EPA intends to address the
concerns of those who commented that the 3-
year form might prevent the public from
being informed about the air quality status of
their communities. As outlined in Unit II.H.
of this preamble, EPA plans to issue revised
Pollutant Standard Index criteria for PM2.5, to
ensure the public is informed promptly about
air quality status.
The Staff Paper and some CASAC panel
members also recommended that
consideration be given to calculating the
PM2.5 annual arithmetic mean for an area by
averaging the annual arithmetic means
derived from multiple monitoring sites within
a monitoring planning area. In proposing a
calculation method for annual arithmetic
averages that involves spatial averaging of
monitoring data, the Administrator reasoned
as follows:
(1) Many of the community-based
epidemiological studies examined in this
review used spatial averages, when multiple
monitoring sites were available, to
characterize area-wide PM exposure levels
and the associated population health risk. In
those studies that used only one monitoring
location, the selected site was chosen to
represent community-wide exposures, not the
highest value likely to be experienced within
the community. Thus, spatial averages are
most directly related to the epidemiological
studies used as the basis for the proposed
revisions to the PM NAAQS.
(2) As a part of the overall policy approach
discussed in Unit II.D. of this preamble, the
annual PM2.5 standard would be intended to
reduce aggregate population risk from both
long- and short-term exposures by lowering
the broad distribution of PM2.5 concentrations
across the community. An annual standard
based on spatially averaged concentrations
would better reflect area-wide PM exposure
levels than would a standard based on
concentrations from a single monitor with the
highest measured values.
(3) Under this policy approach, the 24-hour
PM2.5 standard would be intended to work in
conjunction with a spatially averaged annual
PM2 5 standard by providing protection
against peak 24-hour concentrations, localized
"hot spots," and higher PM2.5 concentrations
arising from seasonal emissions and
meteorology that would not be as well
controlled by an annual standard.
Accordingly, the 24-hour PM2.5 standard
should be based on the single population-
oriented monitoring site within the
monitoring planning area with the highest
measured values.
Based on these considerations, the
Administrator proposed that the form of an
annual PM2 5 standard be expressed as the
annual arithmetic mean, temporally averaged
over 3 years and spatially averaged over all
designated monitoring sites,31 which, in
conjunction with a 24-hour PM2.5 standard,
was intended to provide the most appropriate
target for reducing area-wide population
exposure to fine particle pollution.
Recognizing the complexities that spatial
averaging might introduce into risk
management programs, in the proposal the
Administrator also requested comment on the
alternative of basing the annual standard for
PM2.5 solely on the single population-oriented
monitor site within the monitoring planning
area with the highest 3-year average annual
mean.
The proposed approach to designating sites
that are appropriate for spatial averaging was
based on criteria and constraints contained in
the proposed revision to the monitoring siting
and network planning requirements in 40
CFR part 58. In proposing this approach, the
Administrator noted concerns regarding the
development and implementation of
appropriate and effective criteria for the
selection of sites and designations of areas for
spatial averaging.
A number of commenters who otherwise
favored setting PM2.5 standards objected to
the concept of population-oriented monitors
and expressed the view that any monitor
regardless of where it was sited should be
eligible for comparison to the annual PM2.5
standard. They further maintained that the
proposed provisions for spatial averaging
would fail to provide adequate health
protection because "clean areas" and "dirty
areas" would be averaged together. Some
commenters expressed concern that the
proposed constraints on spatial average would
not be sufficient to prevent use of such
averaging to avoid pollution abatement.
Others may not have fully understood the
implications of the specific constraints and
siting requirements discussed in the proposed
revisions to 40 CFR part 58, which were
intended to ensure that the population-
oriented monitors used for the annual
standard were actually reflective of
community-wide exposures and that the
spatial averages did not include non-
representative monitored values from either
"clean areas" or "dirty areas."32 In order to
clarify the intent that the spatially averaged
annual standard protect those in smaller
communities, as well as those in larger
population centers, the final revisions to 40
CFR part 58 adopt the term "community-
oriented" monitors.
Other commenters, who supported PM2.5
annual standards, endorsed the concept of
spatial averaging as being more reflective of
the air quality data used in the underlying
health studies and because there is general
uniformity of fine particle concentrations
across an area. Opponents of the PM2.5
standards expressed contingent support for
spatial averaging in concept, again citing the
linkage to the underlying health studies.
Indeed, they advocated the extension of
spatial averaging to the daily form of the
standard, and/or recommended less
constrained spatial averaging to allow for
averaging across entire metropolitan areas.
The Administrator, of course, shares
commenters' concerns that the form of the
standards, in conjunction with other
components of the standards, must protect
public health adequately against risks
associated with PM. It was for this reason that
EPA proposed a policy approach providing
for greatest overall risk reduction for all
citizens in the community from exposures to
the mix of urban and regional scale PM
pollution most strongly associated with health
effects. In specifically considering whether to
allow for the use of spatial averaging, the
Administrator placed great weight on
consistency with the underlying body of
health effects evidence. The Administrator is
mindful that some community studies relied
inherently on exposure and effects estimates
that reflect comparatively broad spatial scales,
as highlighted by those commenters desiring
to extend permissible averaging; however,
this type of exposure characterization may not
be appropriate for all circumstances and
might leave some areas without adequate
protection.33
31 The notice of proposed revisions to 40 CFR part 58
recognized that a single appropriately sited monitor could
suffice for an area in place of an average of multiple
monitors.
32 The 40 CFR part 58 proposed rule identified the
proposed criteria for monitors to be averaged; namely,
monitors must be properly sited to reflect population-
orientation, primarily influenced by similar sources, and
within +/-20 percent of the average levels and a specific
degree of correlation (or meet a "homogeneity"
constraint). Additional criteria include demonstrations that
the monitors to be averaged are influenced primarily by
similar sources (e.g., to prevent the placement of monitors
upwind in unrepresentative locations), EPA oversight of
the monitoring program which includes regular review and
approval of the State PM monitoring network design, and
other criteria to ensure proper monitor siting. The final rule
includes the addition of provisions that the State PM
monitoring network design be available for public
inspection.
33 Daily mortality studies generally use urban or metro-
areawide effects statistics in conjunction with single or
multiple monitors that index day-to-day pollution changes
across the area. Ito et al. (1995) found that spatial averages
from multiple PM monitors in Chicago were better
correlated with daily mortality than were most single
monitors, but that single monitors were also associated. A
number of morbidity studies (e.g., Schwartz et al., 1994;
Continued
-------
20
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
For these reasons, the 40 CFR part 58
proposal package contained criteria and
constraints on spatial averaging. These
criteria and constraints were intended to
ensure that spatial averaging would not result
in inequities in the level of protection
provided by the PM standards. The
Administrator again recognizes that either a
single properly sited community-oriented
monitor, or an average of more than one such
monitors, are both appropriate indices of area-
wide population exposures. Both are
consistent with monitoring approaches used
in community epidemiological studies upon
which the standards are based. On the other
hand, comparing the annual PM2.5 standard to
the maximum concentrations at a site that is
not representative of community exposures,
as some have suggested, would be
inconsistent with the Administrator's goal of
using the annual standard to reduce urban and
regional scale exposures and risks. Further,
the Administrator believes that the criteria
and, siting requirements contained in 40 CFR
part 58, provide adequate safeguards against
inappropriate application of spatial averaging.
Therefore, the Administrator continues to
believe that an annual PM2.5 standard
reflective of area-wide exposures, in
conjunction with a 24-hour standard designed
to provide adequate protection against
localized peak or seasonal PM2.5 levels,
reflects the most appropriate approach for
public health against the effects of PM
reported in the scientific literature.34
The majority of comments from States
stressed the need for flexibility in specifying
network designs and spatial averaging, given
that the nature and sources of particle
pollution vary from one area to another. One
State agency specifically requested the
flexibility to choose whether to use a single
community-oriented monitor or a spatial
average of several of such monitors, arguing
that it is appropriate to provide this flexibility
as PM2.5 monitoring networks evolve and to
address the diversity of local conditions.
As a result of EPA's evaluation of these
comments, the requirements of 40 CFR part
50, Appendix K, and 40 CFR part 58 have
been revised to clarify that the implementing
agencies have the flexibility to compare the
annual PM2.5 standard either to the measured
value at a single representative community-
oriented monitoring site, or to the value
resulting from an average of community-
oriented monitoring sites that meet the
revised criteria and constraints enumerated in
the 40 CFR part 58 final rule.
In the Administrator's view, the final
criteria and siting requirements contained in
40 CFR part 58 and in the new 40 CFR part
Neas et al., 1995; Raizenne et al.; 1996) used community
scale monitors and effects information from a defined
group of subjects from the community, who were more
closely represented by the monitor.
34 Because the 24-hour standard is designed to address
localized peaks, it would be inappropriate to extend spatial
averaging forms to this standard.
50, Appendix N, address the concerns raised
by these commenters about the protection
afforded by the form of the annual standard.
Therefore, the Administrator continues to
believe that the form of a PM2.5 annual
standard should be expressed as an annual
arithmetic mean, averaged over 3 years, from
single or multiple community-oriented
monitors, in accordance with 40 CFR part 50,
Appendix N and 40 CFR part 58. In her
judgment, an annual standard expressed in
this manner and set at an appropriate level,
in conjunction with a 24-hour PM2.5 standard,
will adequately protect public health.
2. 24-hour standard. The current 24-hour
PMio standard is expressed in a "1-expected-
exceedance" form. That is, the standard is
formulated on the basis of the expected
number of days per year (averaged over 3
years) on which the level of the standard will
be exceeded. The test for determining
attainment of the current 24-hour standard is
presented in Appendix K to 40 CFR part 50.
As discussed in the proposal, since
promulgation of the current 24-hour PMio
standard in 1987, a number of concerns have
been raised about the 1-expected-exceedance
form. These include, in particular, the year-
to-year stability of the number of
exceedances, the stability of the attainment
status of an area, and the complex data
handling conventions specified in 40 CFR
part 50, Appendix K, including the
procedures for making adjustments for
missing data and less-than-every-day
monitoring.
In light of these concerns, the Staff Paper
and several CASAC panel members (Wolff,
1996b) recommended that consideration be
given to adoption of a more stable and robust
form for 24-hour standards. In considering
this recommendation for the proposal, the
Administrator noted that the use of a
concentration-based percentile form would
have several advantages over the current 1-
expected-exceedance form:
(1) Such a concentration-based form would
be more directly related to the ambient PM
concentrations that are associated with health
effects. Given that there is a continuum of
effects associated with exposures to varying
levels of PM, the extent to which public
health is affected by exposure to ambient PM
is related to the actual magnitude of the
concentration, not just whether the
concentration is above a specified level. With
an exceedance-based form, days on which the
ambient concentration is well above the level
of the standard are given equal weight to
those days on which the concentration is just
above the standard (i.e., each day is counted
as one exceedance), even though the public
health impact on the 2 days is significantly
different. With a concentration-based form,
days on which higher concentrations occur
would weigh proportionally more than days
with lower concentrations for the design
value, since the actual concentrations would
be used directly in determining whether the
standard is attained.
(2) A concentration-based percentile form
would also compensate for missing data and
less-than-every-day monitoring, thereby
reducing or eliminating the need for complex
data handling procedures in the 40 CFR part
50, Appendix K test for attainment. As a
result, an area's attainment status would be
based directly on monitoring data rather than
on a calculated value adjusted for missing
data or less-than-every-day monitoring.
(3) Further, a concentration-based form,
averaged over 3 years, would also have
greater stability than the expected exceedance
form and, thus, would facilitate the
development of more stable implementation
programs by the States.
The proposal discussed various specific
percentile values for such a form (e.g., 90th
to 99th percentiles), taking into account two
factors. First, the 24-hour PM2.5 standard is
intended to supplement the annual PM2.5
standard by providing additional protection
against extremely high peak days, localized
"hot spots," and risks arising from seasonal
emissions. Second, given an appropriate level
of health protection, the form of the 24-hour
PM2.5 standard should provide an appropriate
degree of increased stability relative to the
current form. The Administrator noted in the
proposal that a more stable statistic would
reduce the impact of a single high exposure
event that may be due to unusual
meteorological conditions alone, and thus
would provide a more stable basis upon
which to design effective control programs.
With these purposes in mind, the
Administrator observed in the proposal that
while a percentile value such as the 90th or
95th would provide substantially increased
stability when compared to a more extreme
air quality statistic (e.g., the current 1-
expected-exceedance form), it would likely
not serve as an effective supplement to the
annual standard, because it would allow a
large number of days with peak PM2.5
concentrations above the standard level. For
example, in a 365-day data base, the 90th and
95th percentiles would equal the 37th and 19th
highest 24-hour concentrations, respectively.
On the other hand, a percentile value selected
much closer to the tail of the air quality
distribution (e.g. a 99th or greater percentile)
would not likely provide significantly more
health protection or significantly increased
stability as compared to a 1-exceedance form.
In balancing these issues in the proposal, the
Administrator ultimately proposed a 98th
percentile value form of the standard.
Some commenters maintained that EPA
should retain the current 1-expected-
exceedance form for the 24-hour PM2.5
standard to limit the number of days per year
that the standard is exceeded. These
commenters apparently gave little weight to
EPA's rationale that a concentration-based
form is more directly related to ambient PM
concentrations that are associated with health
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
21
effects because it takes into account the
magnitude of PM concentrations, not just
whether the concentrations are above a
specific level. These commenters also
discounted the other advantages of a
concentration-based percentile form outlined
above in this unit. A number of other
commenters supported the concentration-
based percentile form for the reasons outlined
in the proposal but, as discussed below in this
unit, argued for alternative percentile values
that were higher or lower than the proposed
98th percentile value.
EPA continues to believe that a
concentration-based percentile form is more
reflective of the health risk posed by elevated
PM concentrations, because it gives
proportionally greater weight to days when
concentrations are well above the level of the
standard than to days when the concentrations
are just above the standard. This factor,
coupled with the other advantages outlined
above in this unit, leads EPA to conclude that
a concentration-based percentile form will
provide for more effective health protection
than a 1-expected-exceedance form.
Some commenters supporting a single
exceedance form or a more restrictive
concentration-based percentile form (e.g. a
99th percentile) expressed concern that the
proposed 98th percentile form could allow too
many high concentration excursions, and thus
fail to provide adequate protection against
seasonal emissions problems or localized
peaks. In particular, some commenters
expressed concerns that in areas with strongly
seasonal emissions, such as western areas
with winter inversions, over a three year
period an area could experience several
excursions in which levels could reach as
high as 250 |J.g/m3 and still comply with both
the annual and daily standards if the
remainder of the days had low levels (e.g.,
10 |ig/m3). Although this combination of
events is theoretically possible, EPA believes
it is unlikely. Moreover, if such episodic
events did occur, the Act provides for
emergency State or Federal action to address
them.35 In view of the limits on truely
episodic peak concentrations, EPA believes
that an appropriately selected 24-hour
standard with a concentration-based 98th
percentile form can provide a stable and
adequately protective supplement to the
annual standard in areas with periodic peak
concentrations.
Other commenters who were also
concerned with monitoring requirements
associated with spatial averaging in the
annual standard, argued that a 98th percentile
35 See sections 303, 110(a)(2)(y); 40 CFR part 51. EPA
intends to establish a significant harm level for PM2.5 and
associated guidance so States can develop appropriate
emergency episode plans. The significant harm and
episode criteria will be included in forthcoming proposed
revisions to 40 CFR part 51 and 40 CFR part 58
implementation guidance. In the interim, existing PMio
emergency episode plans should be triggered by events of
this magnitude.
form, coupled with the proposed monitoring
requirements that would limit compliance
monitors for the 24-hour standard to
population-oriented sites, would not protect
people residing in or near localized "hot
spots" in some areas.36 The Administrator
believes that the siting requirements as
proposed and finalized in 40 CFR part 58 for
population-oriented sites will provide
adequate safeguards for such residential areas.
Other commenters, who otherwise opposed
setting PM2.5 standards, recommended that
alternative lower percentiles (e.g., 95th
percentiles) be used, if EPA proceeds to set
such standards. As discussed above in this
unit, however, EPA continues to hold the
view that a 90th to 95th percentile form would
not provide an adequate limit against periodic
peak values in areas with low annual values
and periodic high seasonal or source-oriented
peaks.
After carefully assessing the comments
received, the Administrator is persuaded that
the adoption of a 98th percentile form for the
24-hour PM2.5 standard measured at each
population-oriented monitoring site in an area
would provide an effective supplement to the
annual PM2.5 standard. This form will
provide adequate protection against 24-hour
peak PM2.5 levels in locations dominated by
single point sources, as well as in areas
dominated by seasonal emissions. The
Administrator also believes that a 98th
percentile form, with more frequent sampling
and averaged over 3 years, will provide
increased stability and robustness as
recommended by several members of the
CASAC panel. For these reasons, the
Administrator has decided to adopt the 98th
percentile form for the final PM2.5 24-hour
standard. The 24-hour PM2.5 standard would
be attained when the 3-year average of the
98th percentile of 24-hour concentrations at
each populated oriented monitor within an
area is less than or equal to the level of the
standard. Further details regarding the
interpretation of the form, as well as
associated calculations and other data
handling conventions are specified in the new
40 CFR part 50, Appendix N.
F. Levels for the Annual and 24-Hour PM^.^
Standards
As discussed in Unit II.D. of this preamble,
the Administrator believes that an annual
PM2.5 standard can provide the requisite
reduction in risk associated with both annual
and 24-hour averaging times in most areas of
the United States. Under this approach, the
24-hour standard would be intended to
provide supplemental protection against
extreme peak fine particle levels that may
occur in some localized situations or in areas
with distinct variations in seasonal fine
particle levels. In reaching judgments as to
36 The 40 CFR part 58 monitoring rule proposed to limit
sites that would be eligible for comparisons to the 24-hour
standard to population-oriented monitoring sites.
appropriate levels to propose for both the
annual and 24-hour PM2 5 standards, the
Administrator has considered the combined
protection afforded by both the annual and
24-hour standards, taking into account the
forms discussed in Unit II.E. of this preamble.
With this approach in mind, the
Administrator has considered the available
health effects evidence and related air quality
information presented in the Criteria
Document and summarized in chapters IV—
VII of the Staff Paper, which provides the
basis for decisions on standard levels that
would reduce risk sufficiently to protect
public health with an adequate margin of
safety, recognizing that such standards will
not be risk-free. In so doing, the
Administrator has considered both the
strengths and the limitations of the available
evidence and information, as well as
alternative interpretations of the scientific
evidence advanced by various CASAC panel
members (Wolff, 1996b; Lippmann et al.,
1996) and public commenters, arising
primarily from the inherent uncertainties and
limitations in the health effects studies.
Beyond those factors, but clearly related to
them, a range of views have been expressed
by CASAC panel members and the public as
to the appropriate policy response to the
available health effects evidence and related
air quality information. Toward one end of
the spectrum, the view has been expressed
that only a very limited policy response is
appropriate in light of the many key
uncertainties and unanswered questions that,
taken together, call into question the
fundamental issue of causality in the reported
associations between ambient levels of PM2.5
and mortality and other serious health effects.
Toward the other end, the view has been
expressed that the consistency and coherence
of the epidemic logical evidence should be
interpreted as demonstrating causality in the
relationships between PM2.5 and health
endpoints that are clearly adverse, and that
uncertainties in the underlying health effects
information should be treated, regardless of
their nature, as warranting a maximally
precautionary policy response. A third view
would suggest an alternative policy response,
taking into account not only the consistency
and coherence of the health effects evidence,
but also the recognition of key uncertainties
and unanswered questions that increasingly
call into question the likelihood of PM-related
effects as PM2.5 concentrations decrease
below the mean values in areas where effects
have been observed and/or as such
concentrations approach background levels.
Reflecting these divergent views, both of
the science itself and of how the science
should be used in making policy decisions on
proposed standards, the Administrator
considered three alternative approaches to
selecting appropriate standard levels, as
described in the proposal, ultimately deciding
to propose standards based on a balanced
view of the strengths and uncertainties of the
-------
22
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
scientific information that reflects the
intermediate approach.
Judging by the public comments received,
EPA accurately reflected the bases for
divergent views. A substantial body of public
comments supported revising the PM
standards by adding PM2.5 standards with
levels at least as stringent as those proposed
by the Administrator. In general, however,
comments on levels for PM2.5 standards
revealed a strong dichotomy between those
who recommended even stronger standards
than proposed, and those who counseled
against revising the standards at all. As noted
above in this unit, many in this latter group
made contingent recommendations with
respect to the levels and other aspects of
PM2.5 standards, if the Administrator
concluded that any revisions were
appropriate.
This latter group of "contingent"
commenters recommended levels well above
those proposed by the Administrator. These
commenters placed great weight on factors
outlined in Units KB. and II.C. of this
preamble that led them to oppose any
revisions to the PM standards, including the
uncertainties and limitations in the available
health effects studies considered individually,
such as the possible existence of effects
thresholds and unanswered questions
regarding the causal agent(s) responsible for
the reported health effects. Further, they
emphasized the limited amount of research
currently available that has measured PM2.5
directly. A substantial group recommended
that PM2.5 standards be selected so as to be
equivalent or close in stringency to the
current PMio standards, and cited the
opinions of some CASAC PM panel members
as support. Some of these commenters
provided supplemental analyses of air quality
data, arguing that they demonstrate that
"equivalent" standards would be at PM2.5
levels as high as approximately 95 |J.g/m3 24-
hour average and 27 |J.g/m3 annual average.
Having evaluated these comments, the
Administrator rejects both their underlying
rationale and the specific recommendations
for PM2.5 standard levels that result in similar
or only marginally more protection than that
afforded by the current PMio standards. Aside
from technical problems in the commenters'
supporting analyses on the issue of defining
"equivalent" standards,37 the Administrator
finds this approach inconsistent with her
conclusions regarding the adequacy of the
current standards and the need to provide
37 Nationwide PM2.5 estimates have been derived from
the current PM air quality data base, but reflect a
significant degree of uncertainty due to the highly variable
relationship between PM2.5 and PMio air quality values
across locations and seasons (Fitz-Simons et al., 1996).
The American Iron and Steel Institute (AISI) submitted a
useful data base (Cooper Associates, 1997) on PM2.5/PMio
relationships that examines both these predictions and the
issue of equivalence. An EPA examination of this material,
which found some problems with the analysis and with
commenters' conclusions that appear inconsistent with the
Cooper report, is included in the Response to Comments.
additional protection as articulated in Unit
KB. of this preamble. The Administrator
believes that, despite well recognized
uncertainties, the consistency and coherence
of the epidemiological evidence and the
seriousness of the health effects require a
more protective response than provided by
"equivalence" or a marginal strengthening of
the standards. Moreover, EPA believes that
the standard levels should be based on the
most recent assessment of the scientific
criteria for PM, not on applying uncertain
ratios to standard decisions based on much
more limited evidence in 1987. The
Administrator also rejects the premise of
some38 who suggest that adopting a standard
that prompts little or no additional control
would cause no delay in risk reduction as
compared to conducting monitoring and
research now and setting a more stringent
standard after the next review. These
comments do not consider the realities of
implementing air quality standards, which
ensure that such an approach would add
several years to the risk reduction process.
Thus, aside from her obligations under the
statute,39 the Administrator believes that the
most prudent and appropriate course is to
establish appropriately protective standards
now that put into motion monitoring and
strategy development programs, while at the
same time pursuing an expanded research
program to improve implementation and to
inform the next periodic review of the criteria
and standards.
In sharp contrast to the commenters
discussed immediately above, a number of
other commenters strongly supported standard
levels more stringent than those proposed by
EPA. These commenters supported EPA's
conclusions regarding the epidemiological
studies, but would place much less weight on
uncertainties related to the concentration-
response relationships for PM2.5 as a
surrogate for PM and the relative importance
of various PM components. Based on their
evaluation of the information, and citing the
support of some CASAC panel members,
these commenters variously recommended
24-hour PM2.5 standards as low as 18 to 20
|ig/m3 and annual standards of 10 to 12 jig/
m3.40
EPA notes that setting such standards
would result in commensurate reductions in
38 Some commenters suggest that CASAC and EPA
support for PM2.5 standards is based on the need to
stimulate additional monitoring and research. While the
Administrator agrees that the additional monitoring and
research that would accompany establishment of
equivalent or marginally tighter PM2.5 standards are very
important goals, they do not form an adequate rationale
for establishing air quality standards.
39 As stated previously, section 109(d) of the Act
requires that, after reviewing the existing criteria and
standards for PM, the Administrator make such revisions
in the standards and promulgate such new standards as are
appropriate under section 109(b) of the Act.
40 This range of levels for a 24-hour PM2.5 standard is
close to the lower bound levels recommended by four
CASAC panel members (20 (Ig/m3); no member supported
an annual PM2.5 standard as low as 10 to 12 (Ig/m3.
health risks only if, in fact, there is a
continuum of health risks down to the lower
end of the ranges of air quality observed in
the key epidemiological studies, and only if
the reported associations are, in fact, causally
related to PM2.5 at the lowest concentrations
measured. Setting standards at low levels
where the possibility of effects thresholds is
greater, and where there is greater potential
that other elements in the air pollution mix
(or some subset of particles within the fine
fraction) become more responsible for (or
modify) the effects being causally attributed
to PM2.5, might result in regulatory programs
that go beyond those that are needed to
effectively reduce risks to public health.
While placing substantial weight on the
results of the key health studies in the higher
range of concentrations observed, EPA is
persuaded that the inherent scientific
uncertainties are too great to support
standards based on the lowest concentrations
measured in such studies, which approach the
maximum range of PM2.5 values estimated
for short-term background conditions.
Having considered the comments reflecting
the two contrasting views summarized above
in this unit, the Administrator concludes that
the approach she set forth in the proposal is
the most appropriate for selecting levels for
annual and 24-hour PM2.5 standards. This
approach focuses primarily on standard levels
designed to limit annual PM2.5 concentrations
to somewhat below those where the body of
epidemiological evidence is most consistent
and coherent, in recognition of both the
strengths and the limitations of the full range
of scientific and technical information on the
health effects of PM, as well as associated
uncertainties, as interpreted by the Criteria
Document, Staff Paper, and CASAC. The
Administrator believes that this approach
appropriately reflects the weight of the
evidence as a whole.
In identifying PM2.5 standard levels
consistent with this overall approach, the
Administrator has placed greatest weight on
those epidemiological studies reporting
associations between health effects and direct
measures of fine particles, most notably those
recent studies conducted in North America
(summarized in Tables V-12 and V-13 of the
Staff Paper).41 Key considerations and study
41 Some confusion is apparent in comments regarding
the basis on which the Administrator selected levels for
the proposed PM2.5 standards, with some commenters
suggesting two or at most three studies were used, and
others suggesting that EPA relied extensively on uncertain
conversion factors to estimate levels for the standards.
These comments are in error. To clarify, as stated in the
proposal, the Administrator is basing her decision to revise
the standards on the full range of PM health effects studies
summarized in the Criteria Document and Staff Paper, but
in selecting specific levels for PM2.5 standards, is relying
chiefly on U.S. and Canadian studies, listed in Tables V-
12 and V-13 of the Staff Paper, that measured fine PM
levels. To ease identification and use of these key studies,
the short-term exposure studies and key PM air quality
statistics are cited in Koman (1996) and all long-term
exposure studies are cited in this preamble. The referenced
memorandum (Koman, 1996) has been updated (Koman,
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
23
results upon which this approach is based are
presented as follows.
As previously discussed, the Administrator
has concluded that it is appropriate to select
the level of the annual standard so as to
protect against the range of effects associated
with both short- and long-term exposures to
PM, with the 24-hour standard level selected
to provide supplemental protection against
peak concentrations that might occur over
limited areas and/or for limited time periods.
In selecting the level for the annual standard,
therefore, the Administrator has considered
both short- and long-term exposure studies.
In accordance with EPA staff and CASAC
views on the relative strengths of the
epidemiological studies, the Administrator
has placed greater emphasis on the short-term
exposure studies in selecting the level of the
annual standard. The approach she took to
this issue consisted of determining a
provisional level based on the short-term
exposure studies, and then determining
whether the long-term exposure studies are
consistent with that level or, instead, suggest
the need for a lower level. The effects
estimates from the short-term exposure
studies (in Table V-12 of the Staff Paper) are
based on analyses of daily PM2.5
concentrations that occurred over the course
of the study period. While effects may occur
over the full range of concentrations observed
in the studies, consistent with the discussion
of this issue in Unit II.D. of this preamble,
the strongest evidence for short-term PM2.5
effects occurs at concentrations near the long-
term (e.g., annual) average. More specifically,
the strength of the evidence of effects
increases for concentrations that are at or
above the long-term (e.g., annual) mean
levels reported for these studies.42 Given the
serious nature of the potential effects, the
Administrator believes it is both prudent and
appropriate to select a level for an annual
standard at or below such concentrations. An
examination of the long-term means from the
combined six city analyses of daily mortality
(Schwartz et al., 1996a) and morbidity
(Schwartz et al., 1994), together with those
from studies in individual cities for which
statistically significant PM-effects
associations are reported (from Table V-12 in
the Staff Paper), finds mean concentrations
ranging from about 16 to about 21 |ig/m3
(Roman, 1996; 1997). In addition, the mean
concentrations in cities where short-term
1997) to clarify key aspects of the studies cited and
relevant air quality statistics. In accordance with EPA and
CASAC views on the relative strength of these studies,
greater weight is placed on short-term exposure studies
than on long-term exposure studies. Where studies found
statistically significant associations with PM2.5 components
(e.g., sulfates and/or acids, in Thurston et al., 1994;
Dockery et al., 1996), the corresponding PM2.5 or PM2.i
values from the study are cited. No conversions were made
from the original measurements used in these studies.
42 As discussed in the proposal and Appendix E of the
Staff Paper (U.S. EPA, 1996b, p. E-4), there is generally
greatest statistical confidence in observed associations for
levels at and above the mean concentration.
exposure associations are characterized in the
Criteria Document as nearly statistically
significant (U.S. EPA, 1996a, p. 13-40) range
from about 11 u.g/m3 to 30 u.g/m3. Taken
together, and placing greatest weight on those
studies that were clearly statistically
significant, this evidence suggests that an
annual standard level of 15 |ig/m3 is
appropriate to reduce the risk of effects from
short-term exposure to fine particles.
Before reaching a final conclusion, the
Administrator also examined this level in
light of the effects reported in
epidemiological studies of long-term
exposures to fine particles (Table V-13 in the
Staff Paper), which may reflect the
accumulation of daily effects over time as
well as potential effects uniquely associated
with long-term exposures. Even though
subject to additional uncertainties, the long-
term exposure studies provide important
insights with respect to the overall protection
afforded by an annual standard. These studies
were examined for general consistency and
support for the levels derived from the short-
term exposure studies, and to determine
whether they provide evidence that a more
stringent level is needed.
The most direct comparison with the daily
fine particle mortality studies is provided by
two long-term prospective cohort studies
(Dockery et al., 1993; Pope et al., 1995). The
annual mean PM2.5 concentration for the
multiple cities included in these studies (6
and 50 cities, respectively) was 18 |J.g/m3
(Dockery et al., 1993), and about 21-22 jag/
m3 for the larger Pope et al. (1995) study.43
The Staff Paper assessment of the
concentration-response results from Dockery
et al. (1993) concluded that the evidence for
increased risk was more apparent at annual
concentrations at or above 15 |ig/m3 (Table
E-3; U.S. EPA; 1996b).44 EPA notes that the
estimated mean values for most of the cities
in Pope et al. (1995) are above 15 l-ig/m3. As
noted in the Staff Paper and the Criteria
Document, the estimated magnitude of effects
in both long-term exposure mortality studies
may be related to higher historical
concentrations than the affected communities
experienced during the time period of the
studies; this consideration suggests that a
level of 15 |ig/m3 would incorporate a margin
43 Based on a public comment, EPA found that the mean
of 18 (Jg/m3 in Pope et al. (1995) reported in the Criteria
Document and elsewhere was actually the mean of median
values. Based on typical air quality relationships, the
conventional arithmetic mean would be approximately 21
to 22 |lg/m3 (Freas, 1997). The lowest median
concentration measured in this study (9 (Ig/m3), which was
relied upon by some commenters as a basis for annual
standards of 10 (Ig/m3, is about 11 to 12 (Ig/m3 as an
arithmetic mean.
44 Based on public comments and a further evaluation
of the underlying study, EPA concludes that the
comparable assessment of the concentration-response
function summarized in Table E-3 for Pope et al. (1995)
is not appropriate, because it was based on a supplemental
"ecologic" comparison for these cities and not on the far
more reliable prospective-cohort analysis that was the main
focus of the paper.
of safety. An examination of morbidity
effects and long-term exposures is provided
by the recent "24 city" studies, which found
that reduced lung function and increased
respiratory symptoms in children followed the
gradient in annual mean concentrations of
fine particles and/or acid-sulfate components
of fine particles (Raizenne et al., 1996;
Dockery et al., 1996). The results indicate a
greater likelihood of effects at annual mean
PM2.i levels above about 15 |ig/m3 (U.S.
EPA, 1996b; Figure V-7). In the judgment of
the Administrator, these studies are consistent
with a standard level of 15 |ig/m3. While they
provide some suggestion of risks extending to
lower concentrations, they do not provide a
sufficient basis for establishing a lower
annual standard level.
Taking the epidemiological studies of both
short- and long-term exposures together, the
Administrator believes the concordance of
evidence for PM effects and associated levels
provides clear support for an annual PM2.5
standard level of 15 |ig/m3. This level is
below the range of annual data most strongly
associated with both short- and long-term
exposure effects, and because even small
changes in annual means in this concentration
range can make significant differences in
overall risk reduction and total population
exposures, the Administrator believes it will
provide an adequate margin of safety against
the effects observed in these epidemiological
studies. Moreover, the means in areas where
PM2.5 concentrations were statistically
significantly associated with daily mortality
(about 16 to 21 |ig/m3) reflect a 7 to 9-year
average; thus, the use of a 3-year mean will
provide additional protection. Although the
possibility of effects at lower annual
concentrations cannot be excluded, the
evidence for that possibility is highly
uncertain and, as previously discussed, the
likelihood of significant health risk, if any,
becomes smaller as concentrations approach
the lower end of the range of air quality
observed in the key epidemiological studies
and/or background levels.
The final annual standard will provide
substantial protection against short-term as
well as long-term exposures to particles.
Nevertheless, for the reasons specified above,
a spatially averaged annual standard cannot
be expected to offer an adequate margin of
safety against the effects of all potential
short-term exposures in areas with strong
local or seasonal sources. The broad-based
community studies considered in this review
generally could not evaluate such peak
exposure conditions directly. Given the public
health purposes of the 24-hour standard, the
Administrator believes it should be set at a
level that generally supplements the control
afforded by an annual standard and proposed
an approach based on providing a reasonable
degree of protection against the peak levels
observed or expected in communities where
health effects have been associated with daily
levels of fine particles.
-------
24
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
For the reasons specified in the previous
unit, the Administrator has decided to use a
98th percentile concentration-based form of
the standard. As noted in the proposal, the
98th percentile 24-hour PM2.5 concentrations
in cities with statistically significant or nearly
significant short-term fine particle exposure-
effects associations ranged from 34 |ig/m3 to
as high as 90 |ig/m3 (Roman, 1996, 1997).
Based on an examination of these results,
EPA originally proposed a level for the 24-
hour standard of 50 |ig/m3, and solicited
comments on higher and lower alternative
levels.
In considering comments on alternative
levels for the purpose of making a final
decision on the 24-hour standard, the
Administrator recognizes the significant
uncertainties in identifying the extent of the
incremental risk associated with single peak
exposures to PM2.5 in areas where the annual
standard is met. Clearly, the risks associated
with the 98th percentile air quality data used
in the selecting the proposed level are from
the same study cities that experienced long-
term levels at varying amounts above that
selected for the annual standard. It is unclear
what risks might have been associated with
such peak levels had the long-term averages
in these areas been below that selected for the
annual standard. Regardless of this
uncertainty, it is clear that reducing the
annual concentrations in such areas to that of
the annual standard would reduce the risk
associated with peak days, whatever the
magnitude, as well as that associated with the
far more numerous days with concentrations
near the annual average. Given these
uncertainties and the significant degree of
protection afforded by the annual standard,
the Administrator is persuaded that it is
appropriate to adopt a different approach for
selecting the levels of the 24-hour standard
than the one proposed.
In making a final decision on an
appropriate level for the 24-hour standard, the
Administrator considered several key factors:
the significant protection afforded against
short-term exposures by the annual PM2.5
standard; the role of the 24-hour standard in
providing supplemental protection against
peak exposures not addressed by the annual
standard; the air quality and effects
information in the studies cited above; the
uncertainties in the risks associated with
infrequent and isolated peak exposures in
areas that meet the annual standard; the range
of levels recommended by EPA staff and
CASAC panel members; and the extensive
public comment on the alternative levels
proposed, which ranged between 20 and 65
|-ig/m3. Because of the approach of
establishing the annual standard as the
controlling standard, and, in particular, the
decision to set the level at the lower end of
the annual range, there is no need to consider
levels in the lower portion of the 24-hour
range below the level proposed. Therefore,
the Administrator focused on evaluating the
margin of safety associated with levels
between 50 and 65 |ig/m3.
As has been discussed in previous units,
the extent of total risk over the course of a
year associated solely with a limited number
of peak exposures is uncertain, but it is
considerably smaller than that associated with
the entire air quality distribution. Further, the
risk associated with infrequent peak 24-hour
exposures in otherwise clean areas is not well
enough understood at this time to provide a
basis for selecting the more restrictive levels
in the range of 50 to 65 |ig/m3. On the other
hand, it is clear that any standard level within
this range would provide some margin of
safety. Taking into account the factors
outlined above, the Administrator has
concluded that a 24-hour standard at the level
of 65 |J.g/m3 would provide an effective limit
in the role as a supplement to the annual
standard. This level is at the upper end of the
range recommended by staff and most
CASAC panel members, and below the levels
suggested by some CASAC panel members
and by a number of public commenters.
Although this level is not risk free, the
Administrator believes that it would provide
an appropriate degree of additional protection
over that provided by the annual PM2.5
standard. Accordingly, after weighing these
factors in light of the scientific uncertainties,
the Administrator believes that a 98th
percentile 24-hour PM2.5 standard of 65 |j,g/
m3 would provide an adequate margin of
safety against infrequent or isolated peak
concentrations that could occur in areas that
attain the annual standard of 15 |ig/m3.
In the Administrator's judgment, the
factors discussed above provide ample reason
to believe that both annual and 24-hour PM2.5
standards are appropriate to protect public
health from adverse health effects associated
with short- and long-term exposures to
ambient fine particles. Further, she believes
these factors provide a clear basis for judging
that an annual PM2.5 standard set at 15 |ig/
m3, in combination with a 24-hour standard
set at 65 |j,g/m3, will protect public health
with an adequate margin of safety.
G. Conclusions Regarding the Current PMio
Standards
1. Averaging time and form. In conjunction
with PM2.5 standards, the new function of
PMio standard(s) is to protect against
potential effects associated with coarse
fraction particles in the size range of 2.5 to
10 |im. Coarse fraction particles are plausibly
associated with certain effects from both
long- and short-term exposures (EPA
1996a,b). Based on qualitative considerations,
deposition of coarse fraction particles in the
respiratory system could be expected to
aggravate effects in individuals with asthma.
The Criteria Document and Staff Paper found
support for this expectation in limited
epidemiological evidence on the effects of
coarse fraction particles, suggesting that
aggravation of asthma and respiratory
infections and symptoms may be associated
with daily or episodic increases in PMi0 that
are dominated by coarse fraction particles.
The potential build-up of insoluble coarse
fraction particles in the lung after long-term
exposures to high levels should also be
considered.
Based on assessments of the available
information in the Criteria Document and
Staff Paper, both the staff and CASAC
recommended retention of an annual PMio
standard. The staff, with CASAC
concurrence, recommended retention of the
current annual arithmetic mean form of the
standard, which is the same form being
adopted for the annual PM2.5 standard. As
noted in the staff assessment, the current
annual PMio standard offers substantial
protection against the effects of both long-
and short-term exposure to coarse fraction
particles. Public comment was nearly
unanimous in recommending retention of this
standard. The Administrator therefore has
decided to continue a long-term PMio
standard as an annual arithmetic mean,
averaged over 3 years.
The staff and CASAC also recommended
that consideration be given to retention of a
24-hour standard to provide additional
protection against potential effects of short-
term exposures to coarse fraction particles.
The staff, with CASAC concurrence, also
recommended that if a 24-hour standard is
retained, the form of the standard should be
revised to provide a more robust target for
coarse fraction particle controls. The
Administrator originally proposed a 98th
percentile form for the 24-hour PMio
standard based primarily on the reasons
outlined above in this unit regarding the
proposed form of the 24-hour PM2.5 standard.
The EPA received few comments
supporting elimination of the 24-hour PMio
standard. The main exceptions were some
industries, most notably the mining industry,
which as noted above in this unit, argued that
the available data provide little evidence for
coarse particle effects at current ambient
levels. These groups, who generally opposed
PM2.5 standards, also argued that the daily
PMio standard could be eliminated if PM2.5
standards were set. Based on the potential
aggravation of respiratory symptoms from
short-term exposure to coarse fraction
particles discussed in the Criteria Document
and by numerous commenters, as well as the
recommendations of a majority of CASAC
panelists who also supported PM2.5 standards,
the Administrator concludes it is appropriate
to retain a 24-hour PMio standard.
In general, comments received on the form
of the 24-hour PMio standard paralleled those
on the form of the PM2.5 standard. Substantial
concerns were expressed by environmental
groups, some States, and others that the 98th
percentile would not provide an adequate
limit on the number and magnitude of 24-
hour peak PMio excursions. While a number
of these commenters suggested keeping the
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
25
current 1-expected-exceedance form, EPA
believes that a concentration- based percentile
form offers significant advantages, as outlined
above in this unit, for both PM indicators.
Some air pollution control officials, who were
concerned about the extent to which the 24-
hour PMio standard would be relaxed under
the proposed form, suggested consideration of
a 99th percentile form with increased
monitoring as an appropriately protective
form. Other commenters, particularly some
industry groups and some States, strongly
supported concentration-based percentile
forms, with some recommending
consideration of the 95th percentile form.
The proposal noted that a percentile value
selected closer to the "tail" of the air quality
distribution (e.g., a 99th or greater percentile)
would not significantly increase stability as
compared to the current form. However, an
association of 8 State air pollution agencies
commented that a 99th percentile form could
provide increased stability if combined with
a daily or l-in-3-day sampling frequency and
with greater data capture. In addition, EPA
notes that this concentration-based form is
inherently more stable than the current
exceedance-based form.
Many of these and other commenters were
concerned that the uncertainties in the
available scientific information on the effects
of coarse particles were a reason to be
concerned that, assuming the current standard
level was kept, a 98th percentile form would
represent a significant relaxation in protection
relative to the current standards. Unlike the
situation for the new PM2.5 standards, in the
case of the PMio standards, the 24-hour
standard has generally been the "controlling"
standard, making changes to the form of the
24-hour standard potentially more significant
to the overall national level of protection
afforded. Given the uncertainties in the
available scientific evidence with respect to
the potential health effects of short-term
exposures to coarse fraction particles, the
Administrator is persuaded that the somewhat
more cautious approach with respect to
revising the 24-hour PMio standard
recommended by many commenters is
appropriate. The only approaches available
for increasing the extent of protection for this
standard as compared to that of the proposed
standard involve modifying the form or
reducing the level. For reasons discussed in
the following section, the Administrator
believes it is not appropriate to revise the
level of the standard. In order to provide
adequate protection against the potential risk
associated with multiple short-term peak
exposures to coarse fraction particles, the
Administator accepts commenters'
recommendations to decrease the frequency
of peak values, while still providing for a
more stable control target than afforded by
the current 1-expected-exceedance form.
Therefore, the Administrator concludes that
the 99th percentile concentration-based form,
averaged over 3 years, and combined with
more frequent sampling, would be an
appropriate form for a 24-hour PMio
standard.
2. Levels for the annual and 24-hour PMio
standards—a. Annual PMio standard. As a
result of the more limited information for
coarse fraction particles, the Administrator's
approach for selecting a level of the standard
is directly related to the approach taken in the
last review of the PM NAAQS. In that
review, evidence from limited quantitative
studies was used in conjunction with support
from the qualitative literature in selecting the
level of the current annual PMio standard. In
the current review, the staff assessment of the
major quantitative basis for the level of that
standard (Ware et al., 1986), together with a
more recent related study (Dockery et al.,
1989), recommended the same range of levels
of concern (40 to 50 |ig/m3) as in the 1986
staff paper. The staff concludes that it is
possible, but not certain, that coarse fraction
particles, in combination with fine particles,
may have influenced the observed effects at
these levels. Based on particle deposition
considerations, it is possible that cumulative
deposition of coarse fraction particles could
be of concern in children, who are more
prone to be active outdoors than sensitive
adult populations.
Qualitative evidence of other long-term
coarse particle effects, most notably from
long-term build-up of silica-containing
materials, supports the need for a long-term
standard, but does not provide evidence of
effects below the range of 40 to 50 |ig/m3
(U.S. EPA, 1996a, p. 13-79). The staff
concludes that the qualitative evidence with
respect to biological aerosols also supports
the need to limit coarse materials, but should
not form the major basis for a national
standard (U.S. EPA, 1996a, p. 13-79). In
addition, staff notes that the nature and
distribution of such materials, which vary
from endemic fungi (e.g., valley fever) to
pollens larger than 10 |im, are not
appropriately addressed by traditional air
pollution control programs.
Based on its review of the available
information, CASAC found "a consensus that
retaining an annual PMio NAAQS at the
current level is reasonable at this time"
(Wolff, 1996b). With few exceptions, public
comments supported levels at least as
stringent as the current annual PMio
standard.45 Taking into account these
comments and the above considerations, as
45 Some commenters, including some environmental
groups and the State of California (Gal EPA, 1997),
suggested that the large number of recent studies showing
effects at PMio levels below the current standards provides
a basis for establishing stricter annual and 24-hour PMio
standards, in conjunction with PM2.5 standards. As
discussed in Units II.B. and C. of this preamble, while
these studies could be used either to tighten the PMio
standards or to add standards that tighten control of the
fine fraction of PMio, the weight of evidence from all of
the relevant information more readily supports the
development of additional protection for the PM2.5
fraction.
more fully detailed in the Staff Paper and the
CASAC recommendations, the Administrator
has decided to retain the current annual PMio
standard of 50 |ig/m3 to protect against the
known and potential effects of long-term
exposure to coarse fraction particles.
b. 24-hour PMio standard. As discussed
above in this unit, EPA staff and CASAC also
recommended that consideration be given to
a 24-hour standard for coarse fraction
particles as measured by PMio. Unlike the
case for the annual standard, however, the
staff found that the original quantitative basis
for the level of the current 24-hour PMio
standard (150 |ig/m3) is no longer
appropriate. Instead, the staff found that the
main quantitative basis for a short-term
standard is provided by the two recent
community studies of exposure to fugitive
dust (Gordian et al., 1996; Hefflin et al.,
1994). Because these studies reported
multiple large exceedances of the current 24-
hour standard, and because of limitations in
the studies themselves, the staff concluded
that they provide no basis to lower the level
of the standard below 150 |ig/m3. Moreover,
staff concluded that none of the qualitative
literature regarding the potential effects of
short-term exposure to coarse particles
provides a basis for a lower standard level.
Both EPA staff and CASAC recommended
that if a 24-hour PMio standard is retained,
the level of the standard should be maintained
at 150 |ig/m3, although with a revised form.
Beyond the comments summarized above
recommending elimination of the 24-hour
standard, no commenters recommended a less
stringent level, while some others, as
summarized above in this unit, recommended
more stringent levels. Most comments
favored the current level.
Having considered these factors and the
public comments, the Administrator judges
that, retention of a 24-hour PMio standard at
the level of 150 |i/m3 with a 99th percentile
form is appropriate and will provide adequate
protection against the known and potential
effects of short-term coarse fraction particle
exposures that have been identified to date in
the scientific literature.
H. Final Decisions on Primary PM Standards
For the reasons discussed above in this
unit, and taking into account the information
and assessments presented in the Criteria
Document and the Staff Paper, the advice and
recommendations of CASAC, and public
comments received on the proposal, the
Administrator is revising the current PM
NAAQS by adding new PM2.5 standards and
by revising the form of the current 24-hour
PMio standard. Specifically, the
Administrator is making the following
revisions:
(1) The suite of PM standards is revised to
include an annual primary PM2.5 standard and
a 24-hour PM2.5 standard.
(2) The annual PM2.5 standard is met when
the 3-year average of the annual arithmetic
-------
26
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
mean PM2.5 concentrations, from single or
multiple community-oriented monitors (in
accordance with EPA's final rule on
monitoring siting guidance, 40 CFR part 58,
published in a separate document elsewhere
in this issue of the Federal Register) is less
than or equal to 15 |ig/m3, with fractional
parts of 0.05 or greater rounding up.
(3) The 24-hour PM2.5 standard is met
when the 3-year average of the 98th percentile
of 24-hour PM2.5 concentrations at each
population-oriented monitor within an area is
less than or equal to 65 |ig/m3, with fractional
parts of 0.5 or greater rounding up.
(4) The form of the current 24-hour PMio
standard is revised to be based on the 3-year
average of the 99th percentile of 24-hour
PMio concentrations at each monitor within
an area.
In addition, the Administrator is retaining the
current annual PMio standard at the level of
50 |ig/m3, which is met when the 3-year
average of the annual arithmetic mean PMio
concentrations at each monitor within an area
is less than or equal to 50 |ig/m3, with
fractional parts of 0.5 or greater rounding up.
As discussed below in Units V. and VI. of
this preamble, data handling conventions and
completeness criteria for the revised standards
are being established (40 CFR part 50,
Appendix N). The reference method for
monitoring PM as PMio for the revised
standards has been established (40 CFR part
50, Appendix M). A new reference method
is being established for monitoring PM as
PM2.5 (40 CFR part 50, Appendix L). In a
separate document published elsewhere in
this issue of the Federal Register, EPA is
providing opportunity for public comment on
supplemental information relating to the new
reference method for monitoring PM as PM2.5
(40 CFR part 50, Appendix L).
As indicated previously, EPA plans to
propose related revisions to the Pollutant
Standards Index for PM (40 CFR 58.50) and
the significant harm level program (40 CFR
51.66) at a later date.
HL Rationale for the Secondary Standards
The Criteria Document and Staff Paper
examined the effects of PM on such aspects
of public welfare as visibility, materials
damage, and soiling. The following
discussion of the rationale for revising the
secondary standards for PM focuses on those
considerations most influential in the
Administrator's decision.
A. Need for Revision of the Current
Secondary Standards
1. Visibility impairment. This unit of the
document presents the Administrator's
decision to address the welfare effects of PM
on visibility by setting secondary standards
identical to the suite of PM2.5 primary
standards, in conjunction with the
establishment of a regional haze program
under section 169A of the Act.46 In the
Administrator's judgment, this approach is
the most effective way to address visibility
impairment given the regional variations in
concentrations of non-anthropogenic PM as
well as other regional factors that affect
visibility, such as humidity. By augmenting
the protection provided by secondary
standards set identical to the suite of PM2.5
primary standards with a regional haze
program, the Administrator believes that an
appropriate degree of visibility protection can
be achieved in the various regions of the
country.
In coming to this decision, the
Administrator took into account several
factors, including: The pertinent scientific and
technical information in the Criteria
Document and Staff Paper, difficulties
inherent in attempting to establish national
secondary standards to address visibility
impairment, the degree of visibility
improvement expected through attainment of
secondary standards equivalent to the suite of
PM2.5 primary standards, the effectiveness of
addressing the welfare effects of PM on
visibility through the combination of a
regional haze program and secondary
standards for PM2.5 equivalent to the suite of
primary standards, and comments received
during the public comment period. The
Administrator's consideration of each of these
factors is discussed below in this unit.
The Administrator first concluded, based
on information presented and referenced in
the Criteria Document and Staff Paper, that
particulate matter can and does produce
adverse effects on visibility in various
locations, depending on the PM
concentrations involved and other factors
discussed below. It has been demonstrated
that impairment of visibility is an important
effect of PM on public welfare, and that it
is experienced throughout the United States,
in multi-state regions, urban areas, and
remote mandatory Class I Federal areas47
alike. Visibility is an important welfare effect
because it has direct significance to people's
enjoyment of daily activities in all parts of the
country. Individuals value good visibility for
the well-being it provides them directly, both
where they live and work, and in places
where they enjoy recreational opportunities.
Visibility is highly valued in significant
natural areas, such as national parks and
wilderness areas, because of the special
emphasis given to protecting these lands now
and for future generations. The Criteria
Document cites many studies designed to
quantify the benefits associated with
improvements in visibility.
The Administrator considered information
from the Staff Paper and Criteria Document
regarding the effect of the composition of
particulate matter on visibility. Visibility
conditions are determined by the scattering
and absorption of light by particles and gases,
from both natural and anthropogenic sources.
Visibility can be described in terms of visual
range, light extinction, or deciview48. The
classes of fine particles principally
responsible for visibility impairment are
sulfates, nitrates, organic matter, elemental
carbon (soot), and soil dust. Fine particles are
more efficient per unit mass at scattering light
than coarse particles. The scattering
efficiency of certain classes of fine particles,
such as sulfates, nitrates, and some organics,
increases as relative humidity rises because
these particles can absorb water and grow to
sizes comparable to the wavelength of visible
light. In addition to limiting the distance that
one can see, the scattering and absorption of
light caused by air pollution can also degrade
the color, clarity, and contrast of scenes.
The Administrator next considered what
would be an appropriate level for a secondary
standard to address adverse effects of
particulate matter on visibility. The
determination of a single national level is
complicated by regional differences in
visibility impairment due to several factors,
including background and current levels of
PM, composition of particulate matter, and
average relative humidity.
The Criteria Document and Staff Paper
describe estimated background levels of PM
and natural light extinction. In the United
States, estimated annual mean background
levels of PM2.5 are significantly lower in the
West than in the East. Based on estimated
background fine particle and light extinction
levels summarized in Table VIII-2 of the
Staff Paper, naturally occurring visual range
in the East is approximately 105 to 195
kilometers, whereas in the West it is
approximately 190 to 270 kilometers. This
significant regional difference in estimated
background conditions results from two main
factors. First, in the western United States,
visibility is more sensitive to an additional 1-
2 |-ig/m3 of PM2.5 in the atmosphere than in
the eastern United States. Secondly, light
scattering is increased for certain particles
(e.g., sulfates, nitrates, and some organics)
46 Congress adopted section 169A of the Act because
of concern that the NAAQS and Prevention of Significant
Deterioration programs might not provide adequate
visibility protection nationally, particularly for "areas of
great scenic importance." See H.R. Rep. No. 95-294,at
203-205 (1977).
47 There are 156 mandatory Class I Federal areas
protected by the visibility provisions in sections 169A and
169B of the Act. These areas are defined in section 162
of the Act as those national parks exceeding 6,000 acres,
wilderness areas and memorial parks exceeding 5,000
acres, and all international parks which were in existence
on August 7, 1977.
48 Visual range can be defined as the maximum distance
at which one can identify a black object against the horizon
sky. It is typically described in miles or kilometers. Light
extinction is the sum of light scattering and absorption by
particles and gases in the atmosphere. It is typically
expressed in terms of inverse megameters (Mm"1), with
larger values representing poorer visibility. The deciview
metric describes perceived visual changes in a linear
fashion over its entire range, analogous to the decibel scale
for sound. A deciview of 0 represents pristine conditions.
Under many scenic conditions, a change of 1 deciview is
considered perceptible by the average person.
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
27
due to higher average relative humidity in the
East.
The combination of naturally occurring and
manmade emissions also leads to significant
differences in current visibility conditions
between the eastern United States, 23-39
kilometers average visual range, and western
United States, 55-150 kilometers average
visual range. Table VIII-4 of the Staff Paper
indicates that the current level of annual
average light extinction in several western
locations, such as the Colorado Plateau, is
about equal to the level of background light
extinction, i.e., the level generally regarded as
representing the absence of anthropogenic
emissions in North America, in the East. This
regional difference is due to higher
background particle concentrations in the
East, a composition of fine particles in the
East that, in association with higher eastern
humidity levels, is more efficient at light
scattering, and significantly lower
concentrations of anthropogenic PM in
remote western locations as compared with
remote eastern sites.
Because of these regional differences, it is
the Administrator's judgment that a national
secondary standard intended to maintain or
improve visibility conditions on the Colorado
Plateau or other parts of the West would have
to be set at or even below natural background
levels in the East, which would effectively
require elimination of all eastern
anthropogenic emissions. Conversely, a
national secondary standard that would
achieve an appropriate degree of visibility
improvement in the East would permit further
degradation in the West. Due to this regional
variability in visibility conditions created by
differing background fine particle levels, fine
particle composition, and humidity effects,
the Administrator finds that addressing
visibility solely through setting more stringent
national secondary standards would not be an
appropriate means to protect the public
welfare from adverse impacts of PM on
visibility in all parts of the country.49 Aside
from the problem of regional variability, the
Administrator has also determined that the
Agency currently lacks sufficient information
to establish a level for a national secondary
standard that would represent a threshold
above which visibility conditions would
always be adverse and below which visibility
conditions would always be acceptable.
Because visibility varies not only with PM
concentration, but also with PM composition
and humidity levels, attaining even a low
concentration of fine particles might or might
not provide adequate protection, depending
on these factors.
The Administrator next assessed potential
visibility improvements50 that would result
from attainment of the new primary standards
for PM2.5. The spatially averaged form of the
annual standard is well suited to the
protection of visibility, which involves effects
of PM throughout an extended viewing
distance across an urban area. Indeed, as the
generally controlling standard focused on
reducing urban and regional scale fine
particle levels, most of the visibility
protection provided by the PM2.5 primary
standards would be derived from the annual
standard. In many cities having annual mean
PM2.5 concentrations exceeding 17 |j,g/m3,
improvements in annual average visibility
resulting from attainment of the new annual
PM2.5 primary standard are expected to be
perceptible (i.e., to exceed 1 deciview). Based
on annual mean PM2.5 data reported in Table
12-2 of the Criteria Document and Table V-
12 in the Staff Paper, many cities in the
Northeast, Midwest, and Southeast, as well as
Los Angeles, would be expected to see
perceptible improvement in visibility if the
annual PM2.5 primary standard is attained.
In Washington, DC, for example, where the
IMPROVE network51 shows annual mean
PM2.5 concentrations at about 19 |ig/m3
during 1992-1995, approximate annual
average visibility would be expected to
improve from 21 km visual range (29
deciview) to 27 km (27 deciview). Annual
average visibility in Philadelphia, where
annual PM2.5 levels have been recently
measured at 17 |ig/m3, would be expected to
change from about 24 to 27 km, an
improvement of about 1 deciview. In Los
Angeles, where recent data shows annual
mean PM2 5 concentrations at approximately
30 |ig/m3, visibility would be expected to
improve from about 19 to 34 km (30 to 24
deciview) if the new annual primary PM2 5
standard is attained.
It is important to note that some urban
areas, many in the eastern United States,
would be expected to have annual mean
PM2.5 concentrations reduced below the
primary standard level of 15 |ig/m3 when
implementation of regional control strategies
for PM and other air quality programs, such
as those addressing acid rain and mobile
sources, are taken into account together. On
49 Congress adopted a visibility protection program in
section 169A of the Act because it recognized the
impracticability of revising the NAAQS to protect
visibility in all areas of the country: "It would be
impracticable to require a major city such as New York
or Los Angeles to meet the same visibility standards as
the Grand Canyon and Yellowstone Park." See H.R. Rep.
No. 95-294 at 205. (1977)
50 Estimates of annual average visibility improvements
assume that, on a percentage basis, the reduction for each
fine particle component is equal to the % reduction in the
mass of fine particles, and that the overall light extinction
efficiency of the fine particle pollutant mix does not
change. Further, for the estimates presented here, the
reductions in fine mass at monitored locations are assumed
to reflect the spatial average concentrations through the
viewing distance. (Damberg and Polkowsky, 1996.)
51 IMPROVE (Interagency Monitoring of PROtected
Visual Environments) is a visibility monitoring network
managed cooperatively by EPA, Federal land management
agencies, and State representatives. An analysis of
IMPROVE data for 1992-1995 is found in Sisler et al.
(1996).
the other hand, some urban areas with annual
PM2 5 levels at or below the 15 |ig/m3 level
would be expected to see little, if any,
improvement in annual average visibility.
This may be particularly true of certain
western urban areas that are dominated by
coarse rather than fine particles.
The Administrator also considered the
potential effect on urban visibility if the 24-
hour 98th percentile PM2 5 standard of 65 m3
is attained. In areas with violations caused by
localized hot spots, the 24-hour standard
might have little effect other than on visible
source emissions. In other areas, for example,
with seasonally high woodsmoke, a more
areawide improvement is possible. In such
urban areas, attainment of the 24-hour
standard would be expected to reduce, to
some degree, the number and intensity of
"bad visibility" days, i.e., the 20% of days
having the greatest impairment over the
course of a year. For example, maximum 24-
hour PM2.5 concentrations have been
recorded in recent years at over 140 |J.g/m3
at several California locations. If the level and
frequency of peak PM concentrations are
reduced, improvements would be expected in
those days where visibility is worst, even in
urban areas having annual averages below the
annual PM2 5 primary standard.
Having concluded that attainment of the
annual and 24-hour PM2.5 primary standards
would lead to visibility improvements in
many eastern and some western urban areas,
the Administrator also considered potential
improvements to visibility on a regional scale.
In the rural East, attainment of the PM2.5
primary standards could result in regional
visibility improvement, e.g., in certain
mandatory Class I Federal areas such as
Shenandoah and Great Smoky Mountains
National Park, if regional control strategies
are adopted and carried out in order to reduce
the impact of long-range transport of fine
particles such as sulfates. Fine particle
emission reductions achieved by other air
quality programs, such as those to reduce acid
rain or mobile source emissions, are also
expected to improve Eastern regional
visibility conditions (U.S. EPA, 1993). In the
West, strategies to attain the primary PM2.5
standards are less likely to significantly
improve visibility on a regional basis.
However, areas downwind from large urban
areas, such as Southern California, would
likely see some improvement in annual
average visibility.
Based on the foregoing, the Administrator
concludes that attainment of PM2.5 secondary
standards set at the level of the primary
standards for PM2.5 would be expected to
result in visibility improvements in the
eastern United States at both urban and
regional scales, but little or no change in the
western United States except in and near
certain urban areas. Additionally, the
Administrator determined that attainment of
secondary standards equivalent to the suite of
PM2.5 primary standards for particulate matter
-------
28
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
would address some but not all of the effects
of particulate matter on visibility. The extent
to which these effects would be addressed is
expected to vary regionally.
The Administrator then considered the
potential effectiveness of a regional haze
program to address the remaining effects of
particulate matter on visibility (i.e., those that
would not be addressed through attainment of
secondary standards identical to the suite of
PM2.5 primary standards). A program to
address the widespread, regionally uniform
type of haze caused by a multitude of sources
is required by sections 169A and 169B of the
Act. In 1977, Congress established as a
national goal' 'the prevention of any future,
and the remedying of any existing,
impairment of visibility in mandatory Class
I Federal areas which impairment results from
manmade air pollution", section 169A(a)(l)
of the Act. The EPA is required by section
169A(a)(4) of the Act to promulgate
regulations to ensure that "reasonable
progress" is achieved toward meeting the
national goal. EPA originally deferred
establishment of a program to address
regional haze in 1980 due to the need for
greater scientific and technical knowledge,
but the current Criteria Document and Staff
Paper cite information supporting the
Administrator's conclusion that the scientific
state of understanding and analytical tools are
now adequate to develop such a program.
Because regional emission reductions are
needed to make visibility improvements in
mandatory Class I Federal areas, the structure
and requirements of sections 169A and 169B
of the Act, provide for visibility protection
programs that can be more responsive to the
factors contributing to regional differences in
visibility than can programs addressing a
nationally applicable secondary NAAQS. The
visibility goal is more protective than a
secondary NAAQS since the goal addresses
any man-made impairment rather than just
impairment at levels determined to be
adverse.
Thus, an important factor considered in this
review is whether a regional haze program,
in conjunction with secondary standards set
identical to the suite of PM2.5 primary
standards, would provide appropriate
protection for visibility in non-Class I areas.
The Administrator continues to believe that
the two programs and associated control
strategies should provide such protection due
to the regional approaches needed to manage
emissions of pollutants that impair visibility
in many of these areas. Regional strategies
implemented to attain the NAAQS, meet
other air program goals, and make reasonable
progress toward the national visibility goal in
mandatory Class I Federal areas are expected
to improve visibility in many urban and non-
Class I areas as well. The following
recommendation from the 1993 report of the
National Research Council, Protecting
Visibility in National Parks and Wilderness
Areas, addresses this point:
Efforts to improve visibility in Class I areas also
would benefit visibility outside these areas.
Because most visibility impairment is regional in
scale, the same haze that degrades visibility within
or looking out from a national park also degrades
visibility outside it. Class I areas cannot be
regarded as potential islands of clean air in a
polluted sea.
Before making a final decisions on the
secondary standards, the Administrator also
considered a number of public comments that
addressed this aspect of the proposal. Some
commenters suggested setting secondary
standards for PM2.5 more stringent than the
proposed primary standards for the purpose of
addressing visibility impairment and other
environmental effects. For the reasons
discussed above in this unit, however, the
Administrator has concluded that this may not
be an effective and would not be an
appropriate means of protecting against
visibility impairment in all parts of the
country. Other commenters raised the
possibility of establishing a nationally
applicable secondary standard defined as a
"floor," or increment, above regionally
specific background levels of PM2.5 or
associated visibility. Although this idea is of
interest and may warrant further study, the
Administrator determined that it was not
appropriate to pursue such an approach at this
time for two principal reasons. First, the
Agency does not currently have adequate
scientific information to establish a specific
floor or increment level that would protect
against adverse effects nationally, nor is it
clear as a conceptual matter whether further
information would support selection of a
single, uniform increment as providing an
appropriate degree of protection in all areas
of the country. Second, there are serious,
unresolved questions about whether such an
approach is consistent with the statutory
language and purposes of section 109 of the
Act.
Other commenters argued that national
secondary standards equivalent to the
proposed PM2.5 primary standards are not
necessary or not supported by the
Administrator's findings. As noted earlier,
however, it is clear that coarse and fine
particles can cause adverse effects on
visibility and significant quantitative data
exist to demonstrate that visibility impairment
occurs at small concentrations of PM2.5.
Substantial efforts have been put forth to
assess the effects of PM on visibility. For
example, the Grand Canyon Visibility
Transport Commission52 spent several years
and significant effort studying the effects of
pollution on 16 mandatory Class I Federal
areas on the Colorado plateau and has made
recommendations to the Administrator for
actions to improve visibility in these areas
(GCVTC, 1996). All of the mandatory Class
I Federal areas studied by the GCVTC with
monitoring data have annual mean PM2.5
concentrations below 5 |J.g/m3 (Sisler, 1996)
while also documenting anthropogenic
visibility impairment. The Southern
Appalachian Mountain Initiative53 is
currently assessing air pollution impacts on
visibility, terrestrial resources, and aquatic
resources in the southeastern U.S. in order to
recommend measures to remedy existing and
prevent future adverse effects on these air
quality related values. The IMPROVE
network shows that all of the mandatory
Class I Federal areas in the SAMI region
have annual mean PM2 5 concentrations for
1992-95 between 11.0-13.5 |ig/m3 (Sisler,
1996). The inclusion in section 169A of the
Act of a national visibility goal of no
manmade impairment also places significant
value on reducing PM concentrations and
resulting visibility impairment to low
levels.54 The differences between the fine
particle levels associated with visibility
impairment in eastern and western mandatory
Class I Federal areas provide further impetus
to act under the provisions of sections 169A
and 169B enabling the Administrator to
establish a regionally-tailored visibility
program to address impairment of visibility in
mandatory Class I Federal areas. For these
reasons, the Administrator has concluded that
a national regional haze program allowing for
regional approaches to addressing fine
particle pollution, combined with a nationally
applicable level of protection achieved
through secondary PM2 5 standards set equal
to the suite of primary standards, would be
more effective in addressing regional
variations in the adverse effects of PM2 5 on
visibility than establishing national secondary
standards for particulate matter that are lower
than the suite of PM2 5 primary standards.
The Administrator emphasizes that in order to
appropriately address the regional differences
in adverse effects of particulate matter on
visibility, it is essential to establish secondary
standards for PM2.5 equivalent to the primary
standards and an effective new regional haze
program. A regional haze program will be
particularly important in those areas of the
country that do not exceed any of the primary
standards for PM2.5, yet still experience
significant visibility impairment due to
particulate matter. The EPA will propose a
regional haze regulation in the near future.
52 EPA established the Grand Canyon Visibility
Transport Commission (GCVTC) in 1991 under section
169B of the Act. Section 169B(d) requires visibility
transport commissions to assess the "adverse impacts on
visibility from potential or projected growth in emissions"
and to recommend to EPA measures to remedy such
adverse impacts. The Commission issued its final report
in June 1996.
53 The Southern Appalachian Mountain Initiative is a
voluntary effort begun in 1993. Participants include eight
southeastern States, Federal land managers, EPA, and
representatives from industry and environmental groups. A
final report has not been issued to date.
54 Indeed, Congress recognized when it adopted section
169A that the 'Visibility problem is caused primarily by
emission into the atmosphere of sulfur dioxide, oxides of
nitrogen and particulate matter, especially fine particulate
matter, from inadequately controlled sources." H.R. Rep.
No. 95-294 at 204 (1977).
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
29
In addition to providing a more regionally
tailored approach than establishing a more
stringent national secondary standard, an
effective regional haze program will also
fulfill the Administrator's regulatory
responsibility under sections 169A and 169B
of the Act to address both reasonably
attributable impairment and regional haze
impairment in mandatory Class I Federal
areas. Indeed, regional haze has been shown
to be the principal cause of visibility
impairment in mandatory Class I Federal
areas today. Thus, the promulgation of a
regional haze program in conjunction with
secondary standards for PM2.5 equivalent to
the suite of primary standards will serve as
an appropriate approach for addressing
adverse effects of visibility that vary
regionally, and it will also establish a
comprehensive program for making
reasonable progress toward the national
visibility goal in mandatory Class I Federal
areas by addressing visibility impairment in
the form of both source-specific impacts and
regional haze. Further, the regional haze
rulemaking will fulfill the Administrator's
responsibilities to address the visibility
protection recommendations of the Grand
Canyon Visibility Transport Commission,
pursuant to section 169B(e) of the Act.
The Administrator recognizes that people
living in certain urban areas may place a high
value on unique scenic resources in or near
these areas, and as a result might experience
visibility problems attributable to sources that
would not necessarily be addressed by the
combined effects of a regional haze program
and secondary standards identical to the suite
of primary standards for PM2.5. Commenters
from certain western cities and States raised
this issue. In the Administrator's judgment,
State or local regulatory approaches, such as
past action in Colorado to establish a local
visibility standard for the City of Denver,
would be more appropriate and effective in
addressing these special situations because of
the localized and unique characteristics of the
problems involved. Visibility in an urban area
located near a mandatory Class I Federal area
can also be improved through State
implementation of the current visibility
regulations, by which emission limitations
can be imposed on a source or group of
sources found to be contributing to
"reasonably attributable" impairment in the
mandatory Class I Federal area. EPA also
intends to pursue opportunities to obtain
information on urban and non-Class I area
visibility through examination of available
fine particle monitoring data. Current or
planned monitoring networks and initiatives,
such as monitoring and chemical analysis of
PM2.5 in urban and background sites, efforts
to better characterize real-time environmental
conditions in major populations centers, and
new automated airport visibility monitoring
networks should provide data needed to
evaluate trends in these areas. This
information should help to better characterize
the nature and spatial extent of urban and
non-Class I visibility problems and thus serve
to inform future decisions on NAAQS
revisions or other appropriate measures.
Based on all of the considerations
discussed, the Administrator has decided to
establish secondary standards identical to the
suite of PM2.5 primary standards, in
conjunction with a regional haze program
under sections 169A and 169B of the Act, as
the most appropriate and effective means of
addressing the welfare effects associated with
visibility impairment. Together, the two
programs and associated control strategies
should provide appropriate protection against
the effects of PM on visibility and enable all
regions of the country to make reasonable
progress toward the national visibility goal.
2. Materials damage and soiling effects.
Annual and 24-hour secondary standards for
materials damage and soiling effects of PM
were established in 1987 at levels equal in all
respects to the primary standards. As
discussed in the Criteria Document and Staff
Paper, particles affect materials by promoting
and accelerating the corrosion of metals, by
degrading paints, and by deteriorating
building materials such as concrete and
limestone. Soiling is found to reduce the
aesthetic quality of buildings and objects of
historical or social interest. Past studies have
found that residential properties in highly
polluted areas typically have lower values
than those in less polluted areas. Thus, at high
enough concentrations, particles become a
nuisance and result in increased cost and
decreased enjoyment of the environment.
In the proposal, EPA proposed to establish
secondary standards for PMio and PM2.5
identical to the suite of proposed primary
standards. Several comments recommended
setting secondary standards at levels more
stringent than the proposed primary standards
in order to address various welfare effects of
PM, including soiling and materials damage,
acid deposition, and visibility. Some
commenters specifically suggested changing
the form or level of the proposed 24-hour,
98th percentile PM standards to better protect
against elevated PM episodes and associated
soiling, materials damage, and visibility
effects.
After reviewing the extent of relevant
studies and other information provided since
the 1987 review of the PM standards, the
Administrator concurs with staff and CASAC
conclusions that the available data do not
provide a sufficient basis for establishing a
separate secondary standard based on soiling
or materials damage alone. In the
Administrator's judgment, however, setting
secondary standards identical to the suite of
PM2.5 and PMio primary standards would
provide increased protection against the
effects of fine particles and retain an
appropriate degree of control on coarse
particles. Accordingly, the Administrator
establishes the secondary standards for PM2.5
identical to the suite of primary standards to
protect against materials damage and soiling
effects of PM.
B. Decision on the Secondary Standards
The Administrator establishes secondary
standards identical to the suite of primary
standards. In the Administrator's judgment,
the establishment of these standards, in
conjunction with implementation of a
regional haze program, will provide
appropriate protection against the welfare
effects associated with particle pollution.
IV. Other Issues
Commenters have raised a number of legal
and procedural issues that are discussed in
this unit. These includd:
(1) Whether EPA must give consideration
to costs and similar factors in setting
NAAQS.
(2) Whether EPA erred in its selection of
a methodology for determining the level of
a NAAQS that protects public health with an
adequate margin of safety.
(3) Whether EPA committed a procedural
error by not entering into the rulemaking
docket underlying data from certain
epidemiological studies.
(4) Whether the 1990 amendments to the
Act preclude EPA from revising the PM
NAAQS to establish a new PM2 5 indicator.
Responses to other legal and procedural
issues are included in the Response-to-
Comments Document.
A. Consideration of Costs
For more than a quarter of a century, EPA
has interpreted section 109 of the Act as
precluding consideration of the economic
costs or technical feasibility of implementing
NAAQS in setting them. As indicated in the
proposal, a number of judicial decisions have
confirmed this interpretation. Natural
Resources Defense Council v. Administrator,
902 F.2d 962, 972-973 (D.C. Cir. 1990)(PM
NAAQS)("PM10"), vacated, in part,
dismissed, 921 F.2d 326 (D.C. Cir.), certs.
dismissed, 498 U.S. 1075, and cert, denied,
498 U.S. 1082 (1991); Natural Resources
Defense Council v. EPA, 824 F.2d 1146,
1157-1159 (D.C. Cir. 1987)(enbanc)(CAA
section 112 standards for vinyl
chloride)("Vinyl Chloride"); American
Petroleum Institute v. Costle, 665 F.2d 1176,
1185-1186 (D.C. Cir. 1981)(ozone
NAAQS)("Ozone"), cert, denied, 455 U.S.
1034 (1982); Lead Industries Ass 'n v. EPA,
647F.2d 1130, 1148-1151 (D.C. Cir.)(lead
NAAQS)(Lead Industries), cert, denied, 449
U.S. 1042(1980).
Some commenters have argued that costs
and similar factors should, nonetheless, be
considered, both in this rulemaking and in the
rulemaking on proposed revisions to the
NAAQS for ozone. Although most of the
commenters' arguments are inconsistent with
the judicial decisions cited in this unit, several
commenters have argued that those decisions
-------
30
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
are not dispositive. For reasons discussed in
this unit and in the Response-to-Comments
Document, EPA disagrees with these
comments and maintains its longstanding
interpretation of the Act as precluding
consideration of costs and similar factors in
setting NAAQS.
1. Background. Given the nature of the
points raised, a brief review of the issue
seems useful before addressing the comments.
The requirement that EPA establish national
ambient air quality standards for certain
pollutants, to be implemented by the States,
was enacted in 1970 as part of a set of
comprehensive amendments that established
the basic framework for Federal, State, and
local air pollution control. When EPA
promulgated the original NAAQS in 1971, its
first Administrator, William D. Ruckelshaus,
concluded that costs and similar factors could
not be considered in that decision.55 This
conclusion was not challenged in litigation on
the original NAAQS. It has been confirmed
since then, however, by every judicial
decision that has considered the issue.
As discussed in this unit, EPA's
interpretation rests primarily on the language,
structure, and legislative history of the
statutory scheme adopted in 1970. It is also
supported by the judicial decisions cited in
this unit, as well as by legislative
developments since 1970 that reaffirm
Congress' original approach to the issue.
Without cataloguing all relevant aspects of
the 1970 amendments and their legislative
history, several basic points should be noted.
Under section 109(b) of the Act, NAAQS are
to be "based on" the air quality criteria
issued under section 108 of the Act. Under
section 108(a)(2) of the Act, the kind of
information EPA is required to include in
criteria documents is limited to information
about health and welfare effects ' 'which may
be expected from the presence of [a] pollutant
in the ambient air * * * ." There is no
mention of the costs or difficulty of
implementing the NAAQS, nor of "effects"
that might result from implementing the
NAAQS (as opposed to effects of pollution
in the air).56 By contrast, Congress explicitly
provided for consideration of costs and
similar factors in decisions under other
sections of the Act.57 Moreover, States were
permitted to consider economic and
technological feasibility in developing plans
55 36 FR 8186, April 30, 1971. EPA has maintained this
interpretation consistently since then.
56 That consideration of such factors was not intended
in NAAQS decisions is also supported by section 109(a)(l)
of the Act. For pollutants for which air quality criteria had
been issued prior to the 1970 amendments, that provision
required EPA to propose NAAQS within 30 days after
enactment and to take final action 90 days later. The
criteria issued previously did not include information on
costs and similar factors, and it would have been difficult
if not impossible for EPA to supplement them in time to
include meaningful consideration of such factors in
NAAQS proposed 30 days after enactment.
57 See, e.g., sections 110(e)(l), lll(a)(l), 231 (b) of the
1970 Act; see also, e.g., sections 113(d)(4)(C)(ii),
125(a)(3), 202(a)(3)(C), 317 of the 1977 Act.
to implement the NAAQS to the extent such
consideration did not interfere with meeting
statutory deadlines for attainment of the
standards.58 Finally, the legislative history
indicated that Congress had considered the
issue and had deliberately chosen to mandate
NAAQS that would protect health regardless
of concerns about feasibility.59
The first judicial decision on the issue
came in the Lead Industries case. An industry
petitioner argued that EPA should have
considered economic and technological
feasibility in allowing a "margin of safety"
in setting primary standards for lead. Based
on a detailed review of the language,
structure, and legislative history of the
statutory scheme, the U.S. Court of Appeals
for the District of Columbia Circuit
concluded that:
This argument is totally without merit. [The
petitioner] is unable to point to anything in either
the language of the Act or its legislative history that
offers any support for its claim * * * . To the
contrary, the statute and its legislative history make
clear that economic considerations play no part in
the promulgation of ambient air quality standards
under section 109.
647F.2datll48.
The Court cited a number of reasons for
this conclusion. Id. at 1148-1150. Among
other things, it noted the contrast between
section 109(b) of the Act and other provisions
in which Congress had explicitly provided for
consideration of economic and technological
feasibility, as well as the requirement that
NAAQS be based on air quality criteria
defined without reference to such factors. Id.
at 1148-1149 and n.37. The Court also noted
that, in developing plans to implement
NAAQS, States may consider economic and
technological feasibility only to the extent
that this does not interfere with meeting the
statutory deadlines for attainment of the
standards; and that EPA may not consider
such factors at all in deciding whether to
approve State implementation plans. Id. at
1149 n.37 (citing Union Electric Co. v. EPA,
427 U.S. 246, 257-258, 266 (1976)).«°
As to the legislative history of the 1970
amendments, the Court observed that:
58 Union Electric Co. v. EPA, 427 U.S. 246, 257-58
(1976).
59 The Senate report on the 1970 amendments stated:
"In the Committee discussions, considerable concern was
expressed regarding the use of the concept of technical
feasibility as the basis of ambient air standards. The
Committee determined that (1) the health of people is more
important than the question of whether the early
achievement of ambient air quality standards protective of
health is technically feasible; and, (2) the growth of
pollution load in many areas, even with application of
available technology, would still be deleterious to public
health."
'' Therefore, the Committee determined that existing
sources of pollutants either should meet the standard of
the law or be closed down * * * ."
S. Rep. No. 91-1196, at 2-3 (1970).
60 These limitations would, of course, make little sense
if such factors could be considered in setting the NAAQS
themselves.
[T]he absence of any provision requiring
consideration of these factors was no accident; it
was the result of a deliberate decision by Congress
to subordinate such concerns to the achievement of
health goals.
Id. at 1149. Citing several leading Supreme
Court decisions, as well as the Senate report
quoted in this unit, the Court noted that
Congress had intended a drastic change in
approach toward the control of air pollution
in the 1970 amendments and was well aware
that sections 108-110 of the Act imposed
requirements of a "technology-forcing"
character. Id.61
The Court also noted that Congress had
already acted, in further amendments adopted
in 1977, to relieve some of the burdens
imposed by the 1970 amendments. Id. at 1150
n.38. Observing that Congress had, however,
declined to amend section 109(b) of the Act
to provide for consideration of costs and
similar factors as requested by industrial
interests, Id. n.39, the Court concluded:
A policy choice such as this is one which only
Congress, not the courts and not EPA, can make.
Indeed, the debates on the [1970 amendments]
indicate that Congress was quite conscious of this
fact * * * .
* * * [I]f there is a problem with the economic
or technological feasibility of the lead standards,
[the petitioner], or any other party affected by the
standards, must take its case to Congress, the only
institution with the authority to remedy the
problem.
Mat 1150.
After the decision in Lead Industries,
Supreme Court review was sought on the
question whether costs and similar factors
could be considered in setting NAAQS,
among other issues. The Supreme Court
declined to review the decision. Lead
Industries Ass 'n v. EPA, 449 U.S. 1042
(1980). The subsequent decisions in Ozone,
Vinyl Chloride, and PMio, cited in this unit,
strongly reaffirmed the interpretation adopted
in Lead Industries.62 Supreme Court review
of the Ozone and PMio decisions was sought
but denied. American Petroleum Institute v.
Gorsuch, 455 U.S. 1034 (1984); American
Iron and Steel Institute v. EPA, 498 U.S.
1082(1991).
61 Such requirements "are expressly designed to force
regulated sources to develop pollution control devices that
might at the time appear to be economically or
technologically infeasible.'" Id. (quoting Union Electric
Co. v. EPA, 427 U.S. at 257).
62 In the PMio case, for example, the Court considered
an argument that EPA should have considered potential
health consequences of unemployment that might result
from revision of the primary NAAQS for PM:
"This claim is entirely without merit. In three previous
cases, this court has emphatically stated that § 109 does
not permit EPA to consider such costs in promulgating
national ambient air quality standards * * * . It is only
health effects relating to pollutants in the air that EPA
may consider * * * . Consideration of costs associated with
alleged health risks from unemployment would be flatly
inconsistent with the statute, legislative history and case
law on this point."
902 F.2d at 973 (emphasis in original; citations omitted).
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
31
The Lead Industries opinion focused
largely, though not exclusively, on the 1970
amendments and their legislative history.
Perhaps as a result, it did not canvass all the
factors that, in fact, supported its conclusions
at the time. For example, when Congress
enacted major amendments to the Act in
1977, it was clearly aware that some areas of
the country had experienced difficulty in
attempting to attain some of the NAAQS.63
It was also aware that there might be no
health-effects thresholds for the pollutants
involved, and that significant uncertainties are
inherent in setting health-based standards
under the Act.64 In response, Congress made
significant changes in the provisions for
implementation of the NAAQS, including
changes intended to ease the burdens of
attainment. It also amended sections 108 and
109 of the Act in several ways; for example,
by requiring periodic review and, if
appropriate, revision of air quality criteria and
NAAQS and by establishing a special
scientific advisory committee (CASAC) to
advise EPA on such reviews. Notably,
Congress recognized that implementation of
NAAQS could cause "adverse public health,
welfare, social, economic, or energy effects"
and charged CASAC with advising EPA on
such matters.65 Yet it made no changes in
section 109(b) or section 108(a)(2) of the Act;
that is, in the substantive criteria for setting
or revising NAAQS. In other words,
Congress chose to address economic and
other difficulties associated with attainment
of the NAAQS by adjusting the scheme for
their implementation, rather than by changing
the instructions for setting them.66
Congress enacted major amendments to the
Act again in 1990, well after the Lead
Industries and Ozone decisions that
interpreted section 109 of the Act as
precluding consideration of costs in NAAQS
63 See, e.g., H.R. Rep. No. 95-294, at 207-217 (1977).
« See, e.g., Id at 110-112; Id at 43-51.
65 Section 109(d)(2)(C)(iv) of the Act. Some
commenters have argued that this provision requires EPA
to consider such effects in setting NAAQS. From the
language and structure of section 109(d) of the Act,
however, it is clear that CASAC's responsibility to advise
on these factors is separate from its responsibility to
review and recommend revision of air quality criteria and
NAAQS, and that the advice pertains to the
implementation of NAAQS rather than to setting them.
The legislative history confirms this view, indicating that
the advice was intended for the benefit of the States and
Congress. See H.R. Rep. No. 95-294, at 183 (1977).
66 The 1977 amendments also required EPA to prepare
economic impact assessments for specified actions but
limited the requirement to non-health-based standards,
excluding decisions under sections 109 and 112 of the Act.
Section 317; H.R. Rep. No. 95-294, at 51-52 (1977). In
this and other respects, Congress continued the approach
it took in the 1970 amendments, making careful choices
as to when consideration of costs and similar factors would
be required and giving paramount priority to protection of
health. See 123 Cong. Rec. H8993 (daily ed. Aug. 4, 1977)
(Clean Air Conference Report (1977); Statement of Intent;
Clarification of Select Provisions), reprinted in 3 Senate
Committee on Environment and Public Works, 95th Cong.,
A Legislative History of the Clean Air Act Amendments
of 1977, at 319 (1978).
decisions.67 In doing so, Congress was clearly
aware of intervening developments such as
EPA's decision to revise the PM NAAQS in
1987—the result of an elaborate review in
which the Administrator strongly underscored
the scientific uncertainties involved68—and
the Vinyl Chloride case drawing a sharp
distinction between sections 109 and 112 of
the Act with regard to consideration of costs
and similar factors.69 Indeed, the legislative
history of the 1990 amendments reflects
Congress' understanding that primary
NAAQS were to be based on protection of
health "without regard to the economic or
technical feasibility of attainment."70 Again,
however, Congress chose to respond to
severe, widespread, and persistent problems
with attaining the NAAQS by adjusting the
scheme for their implementation rather than
by changing the basis for setting them. See,
e.g., sections 181-192 of the Act.
2. Public comments. As noted previously,
a number of commenters have argued that
costs and similar factors should be considered
in EPA's final decisions on revision of both
the particulate and ozone NAAQS. Aside
from arguments that are simply inconsistent
with the judicial decisions cited in this unit,
some of the commenters argue that those
decisions are not dispositive for a variety of
reasons. One commenter submitted a
particularly comprehensive version of this
argument; the following discussion focuses
67 In the interim, the National Commission on Air
Quality had also submitted its report to Congress as
required by a provision of the 1977 amendments. Among
other things, the Commission recommended that the
statutory approach of requiring NAAQS to be set at levels
necessary to protect public health, without consideration
of economic factors, be continued without change.
National Commission on Air Quality, To Breathe Clean
Air 55 (1981).
68 As the Administrator indicated in EPA's proposal to
revise the PM standards:
'' [T]hat review has revealed a highly limited data
base—particularly where quantitative studies are
concerned—and a wide range of views among qualified
professionals about the exact pollution levels at which
health effects are likely to occur. The setting of an
'adequate margin of safety' below these levels calls for
a further judgment—in an area for which the scientific data
base is even more sparse and uncertain * * * ."
"* * * [L]ong and expert review of public health issues
has to date revealed no scientific method of assessing
exactly what level of standards public health requires. The
scientific review indicates substantial uncertainties
concerning the health risks associated with lower levels of
particulate matter." (49 FR 10408, 10409, March 20, 1984)
69 Congress was clearly aware of the 1987 decision to
revise the PM NAAQS, which among other things
involved changing the indicator for particulate matter from
"total suspended particulate" to PMio, because it enacted
special nonattainment provisions, as well as provisions for
PSD increments, applicable to PMM. Sections 188-190 of
the Act; section 166(f) of the Act. It was clearly aware
of the Vinyl Chloride decision because it amended section
112 of the Act in response to that decision, essentially
creating a new scheme for setting emission standards for
hazardous pollutants.
70 H.R. Rep. No. 101^490, pt. 1, at 145 (1990). See
also S. Rep. No. 101-228, at 5 (1989).
primarily on points raised by that commenter,
among others.71
As a general matter, the commenter
acknowledges that Congress intended to
preclude consideration of economic costs and
similar factors in setting NAAQS. The
commenter argues, however, that this is so
only when the scientific basis for NAAQS is
"clear and compelling" or "unambiguous."
From that premise, the commenter advances
three key assertions:
a. Where non-threshold pollutants are
involved and the health evidence is
ambiguous, section 109 of the Act must be
interpreted to allow consideration of all
relevant factors, including the practical
consequences of EPA's decisions.
b. To the extent the judicial decisions cited
in this unit are read as precluding this, they
rest on a faulty analysis that pre-dates and
cannot survive scrutiny under Chevron,
U.S.A. v. Natural Resources Defense Council,
467 U.S. 837(1984).72
c. Because EPA has discretion to consider
costs and similar factors where the health
evidence is ambiguous, it must do so in light
of Executive Order 12866 (58 FR 51735,
October 4, 1993), and two recent statutes, the
Unfunded Mandate Reform Act of 1995, 2
U.S.C. 1501-1571 (UMRA), and the Small
Business Regulatory Enforcement Fairness
Act of 1996, Pub. L. 104-121, 110 Stat. 857
(SBREFA), which in part amended the
Regulatory Flexibility Act, 5 U.S.C. 601-808.
EPA believes all three assertions are
clearly incorrect. Regarding the first point, it
should be evident, both from previous
NAAQS decisions and from the court
opinions upholding them, that the scientific
basis for NAAQS decisions has never pointed
clearly and unambiguously to a single "right
answer."73 This is inherent in the statutory
scheme for the establishment and revision of
NAAQS, which in effect requires them to be
based on the "latest scientific knowledge" on
potential health and welfare effects of the
71 Additional responses to points raised by this
commenter and others are included, as appropriate, in the
Response-to-Comments Document.
72 Several other commenters argue that the cited
decisions are not dispositive because they held only that
EPA is not required to consider costs and similar factors
in setting NAAQS. As discussed in this unit in connection
with Chevron, however, the decisions clearly concluded
that Congress intended to preclude consideration of such
factors, and that EPA is not free to alter that congressional
choice. Although these conclusions are technically dicta,
nothing in the Court's opinions suggests that it would have
interpreted section 109 of the Act differently had EPA
claimed authority to consider costs and similar factors in
NAAQS decisions. Indeed, the tone of the opinions argues
to the contrary. See, e.g., PMio, 902 F.2d at 973. Cf Ethyl
Corp. v. EPA, 51 F.3d 1053 (D.C. Cir. 1995).
73 See, e.g., Lead Industries, 647 F.2d at 1146-1147,
1153-1156, 1160-1161, 1167n.l06. In enacting the 1970
amendments, Congress was aware that there were gaps in
the scientific information available then as a basis for
establishing the original NAAQS. See, e.g., S. Rep. No.
91 -1196, at 9-11 (1970). If anything, Congress had an
even greater understanding of the point when it enacted
the 1977 amendments without changing the substantive
criteria for setting NAAQS. See H.R. Rep. No. 95-294,
at 43-51, 181-182(1977).
-------
32
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
pollutant in question. See sections 109(b) and
108(a)(2) of the Act. Although advances in
science increase our understanding of such
effects, they also raise new questions. For this
reason, the key studies for any given decision
on revision of a NAAQS are, almost by
definition, "at the very 'frontiers of scientific
knowledge.'"73 That is, studies that call into
question the adequacy of a standard are
always those that go beyond previous
studies—by reporting new kinds of effects,
for example, or effects at lower
concentrations than those at which effects
have been reported previously.
As with pioneering work in other fields,
such studies may have a variety of strengths
and limitations.875 As a result, the validity
and implications of such studies may be both
uncertain and highly controversial. Given the
precautionary nature of section 109 of the
Act,76 however, it is precisely these kinds of
studies that the Administrator must grapple
with when advances in science suggest that
revision of a NAAQS is appropriate.
As a result, the EPA staff typically
recommends for consideration, and the
Administrator may propose for comment, a
range of alternatives based on what the
commenter would call "ambiguous" science.
In this respect, the current reviews of the
NAAQS for ozone and particulate matter are
not unusual and do not differ, for example,
from the review that led to adoption of the
PM10 NAAQS in 1987.77 Indeed, the
NAAQS that were upheld in the Lead
Industries, Ozone, and PMi0 decisions were
all based on highly controversial health
evidence; the Lead Industries decision took
note of congressional statements recognizing
that there may be no thresholds for criteria
pollutants; and the Ozone and PMio decisions
noted the Administrator's findings that clear
thresholds could not be identified for ozone
and particulate matter, respectively.78 Thus,
the present decisions on revision of the
NAAQS for ozone and particulate matter
cannot be distinguished from those past
74 Lead Industries, 647 F.2d at 1147 (quoting Ethyl
Corp. v. EPA, 541 F.2d 1, 24-27 (D.C. Cir.) (en bane),
cert, denied, 426 U.S. 941 (1976)).
75 They may have methodological flaws, for example,
but nonetheless report effects that are of serious medical
significance; or they may be of impeccable quality but
involve effects of uncertain significance. Others may
involve results that are striking but hard to explain in terms
of previous knowledge, or results that seem plausible and
important but are not yet replicated by other studies.
76 See, e.g., Lead Industries, 647 F.2d at 1155-1156;
H.R. Rep. No. 94-295, at 43-51 (1977).
77 As previously discussed, the Administrator strongly
emphasized the uncertainties involved in that review. As
a result of the uncertainties, he proposed "relatively
broad'' ranges for comment, though he focused on lower
levels within the ranges as providing greater margins of
safety against the health risks involved. See 49 FR 10408,
10409, March 20, 1984.
78 See, e.g., Lead Industries, 647 F.2d at 1152-53 and
n. 43, 1159-60; Ozone, 665 F.2d at 1185, 1187;PM10,
902 F.2d at 969-71, 972.
decisions in terms of the nature of the health
evidence or pollutants involved.78
Regarding the second of the commenter's
key assertions, EPA determines it is clear that
the judicial decisions cited in this unit were
correctly decided and continue to be good law
under Chevron. In Chevron, the Supreme
Court essentially reaffirmed the principle that
courts must defer to reasonable agency
interpretations of the statutes they administer
where Congress has delegated authority to
them to elucidate particular statutory
provisions. Where the intent of Congress on
an issue is clear, however, it must be given
effect by the agency and the courts. See 467
U.S. at 842^15. Thus, the first question on
review of an agency's interpretation under
Chevron is "whether Congress has directly
spoken to the precise question at issue." If
the court determines that it has not, the
remaining question for the court is ' 'whether
the agency's answer is based on a permissible
construction of the statute." 467 U.S. at 842-
843 (footnote omitted). In determining
whether Congress "had an intention on the
precise question at issue," a court employs
"traditional tools of statutory construction."
Id. at 843 n.9.80
In essence, the commenter's argument here
is that the Lead Industries decision did not
address whether Congress had "spoken
directly" to the precise issue posed by the
commenter; that is, whether section 109 of
the Act must be interpreted differently for
NAAQS decisions involving non-threshold
pollutants and "ambiguous" health evidence.
The Lead Industries opinion, which pre-dated
Chevron, did not pose the question in those
terms. Its focus, however, was clearly on
what Congress intended to be the basis for
NAAQS decisions, in a context the Court
understood to involve considerable
uncertainty and debate about the health
evidence, as well as the possibility that there
was no threshold for health effects of the
pollutant.81 In short, the health evidence was
hardly "unambiguous," yet the Court
interpreted section 109 of this Act as
precluding consideration of costs and similar
factors even in allowing a margin of safety.
Nothing in the Lead Industries decision or in
the subsequent cases suggests in any way that
section 109 of the Act should be interpreted
differently based on the nature of the
pollutants or health evidence involved, and
the Court's findings on congressional intent
admit of no exceptions:
* * * [T]he statute and its legislative history
make clear that economic considerations play no
part in the promulgation of ambient air quality
standards under Section 109.
647F.2datll48.
Alternatively, the commenter argues that
the Lead Industries case decided the issue
incorrectly in light of the principles
announced subsequently in Chevron. In this
context, the commenter essentially argues that
the Lead Industries decision rested on two
factors that are no longer probative:
(1) That there was no indication that
Congress meant to allow consideration of
costs in NAAQS decisions, and
(2) That Congress specifically provided for
such consideration in other sections of the
Act but not in section 109.
On the first point, the commenter argues
that EPA is free under Chevron to consider
costs and similar factors (by reinterpreting
section 109 of the Act) unless there is
evidence that Congress intended to restrict its
discretion. As to the second point, the
commenter argues that similar reasoning was
rejected in Vinyl Chloride.
In Vinyl Chloride, however, an en bane
decision that post-dated Chevron, the Court
essentially underscored the point that such
issues cannot be decided mechanically but
must turn, instead, on more analytical
attention to relevant indicia of congressional
intent. See, e.g., 824 F.2d at 1157 n.4; Id. at
1157-1163. With reference to NAAQS
decisions in particular, the Court concluded
that there were concrete indications of
congressional intent to preclude consideration
of costs and similar factors; for example, the
fact that section 108 of the Act "enumeratefs]
specific factors to consider and pointedly
exclude[s] feasibility." 824 F.2d at 1159. In
a later case, moreover, the same Court held
that EPA could not consider certain factors,
in decisions under section 211(f)(4) of the
Act, for reasons exactly parallel to those that
the commenter criticizes in Lead Industries.
See Ethyl Corp. v. EPA, 51 F.3d 1053, 1057-
1063 (D.C. Cir. 1995).
Beyond this, the commenter's
characterization of the Lead Industries
decision ignores or discounts much of the key
evidence cited by the Court, including the
language, structure, and legislative history of
the statutory scheme established in 1970, for
its conclusion that Congress intended to
preclude consideration of costs and similar
factors in NAAQS decisions.82 As indicated
79 Indeed, the present decisions on the NAAQS for PM
and ozone are based on some of the best scientific
information the Agency has ever been able to rely on in
NAAQS decision-making. In particular, the science
underlying these decisions is much more extensive and of
much better quality than the science underlying the
existing NAAQS for PM and ozone.
80 In practice, analysis of this question is sometimes
referred to as a "Chevron step one" analysis.
si See, e.g., 647 F.2d at 1148-51, 1152-53 and n.43,
1160-61.
82 See 647 F.2d at 1148-51. By contrast, the
commenter's argument that Congress actually intended
EPA to consider such factors relies heavily on statements
made in subsequent legislative history, most of which were
made in floor debate, that sought to justify controversial
amendments to establish a different program than the
NAAQS and did not involve any proposed changes in
section 109 of the Act or related provisions; and statements
in early judicial decisions involving programs under other
statutory provisions. In context, EPA determines these and
other statements cited by the commenter are consistent
with and do not alter the conclusion that Congress intended
to preclude consideration of costs and similar factors under
section 109 of the Act.
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
33
in this unit, the Vinyl Chloride and PMio
cases, both of which post-dated Chevron,
reached the same conclusion.
Moreover, this series of decisions went far
beyond mere deference to an agency
interpretation. As indicated in the Vinyl
Chloride case, the Lead Industries court found
"clear evidence" of Congressional intent,
which was to limit the factors EPA may
consider under section 109 of the Act. 824
F.2d 1159. Consistent with Chevron, these
findings were based on traditional tools of
statutory construction. See Id. at 1157-1159;
Lead Industries, 647 F.2d at 1148-1151. In
terms of the analytical framework later
established by Chevron, these were Chevron
step one findings, meaning that the statute
spoke directly to the issue and that the courts,
as well as the agency, must give effect to
Congress' intent as so ascertained. See 467
U.S. at 842-843.83 Thus, absent a more
recent legislative enactment overriding that
intent, EPA has no discretion to alter its
longstanding interpretation that consideration
of costs and similar factors is precluded in
NAAQS decisions under section 109 of the
Act.84
As to the commenter's third key assertion,
Executive Order 12866, UMRA sections 202
and 205, and the Regulatory Flexibility Act
(RFA), as amended by SBREFA, do not
conflict with this interpretation or require a
different result. Basically, the commenter
argues that the Executive Order, UMRA, and
the RFA (as amended by SBREFA) require
83 The commenter argues that the post-Chevron cases
accepted the Lead Industries analysis uncritically rather
than re-examining it under Chevron. Clearly, this elevates
form over substance. It is true that neither case referred
to Chevron in discussing the point at issue. In Vinyl
Chloride, however, the Court retraced the steps in the Lead
Industries analysis in some detail, characterized some of
the key evidence reviewed in that analysis in terms going
beyond mere rote repetition (e.g., "a far clearer statement
than anything in the present case that Congress considered
the alternatives"), and used Chevron-like language in
discussing the significance of that evidence; that is, that
it demonstrated congressional intention on the point at
issue. E.g., 824 F.2d at 1159. Given that the Vinyl
Chloride case was decided three years after Chevron, that
it was an en bane decision of the D.C. Circuit involving
interpretation of statutory language very similar to that in
Lead Industries, and that the Court cited Chevron twice
in analyzing the language and history of section 112 of
the Act, it seems highly unlikely that the Court was
unmindful of Chevron principles in concluding that
Congress intended to preclude consideration of costs under
section 109 of the Act but not under section 112 of the
Act.
In the PMio decision, the Court confirmed the sharp
distinction it had drawn, based on such evidence of
congressional intent, between sections 109 and 112 of the
Act in Vinyl Chloride. 902 F.2d at 972-973. Although
discussion of the point was brief and did not mention
Chevron, the industry petitioner raising the point had cited
Chevron in arguing that the Lead Industries interpretation
was not binding, and that EPA's decision on the PMio
standards should be reversed on the ground that it rested
on a legal position that EPA unjustifiably believed was
mandated by Congress. Reply Brief of the American Iron
and Steel Institute at 11 and n. 10, Natural Resources
Defense Council v. Administrator, 902 F.2d 962 (D.C. Cir.
1990) (Nos. 87-1438 et al.). Thus, Chevron issues were
properly before the Court and were brought squarely to
its attention.
84 See also 52 FR 24854, July 1, 1987.
agencies to use cost (or similar factors) as a
decisional criterion in making regulatory
decisions, and that this modifies the Clean Air
Act's directive that EPA is precluded from
considering costs when setting a NAAQS.
The commenter's argument is flawed on a
number of grounds. First, UMRA and the
RFA (as amended by SBREFA) do not
conflict with section 109 of the Act because
they do not apply to this decision, as
discussed in Unit VIII. of this preamble.
Second, the Executive Order and both statutes
are quite clear that they do not override the
substantive provisions in an authorizing
statute. Third, the commenter's premise that
UMRA and the RFA (as amended by
SBREFA) establish substantive decisional
criteria that agencies are required to follow
is wrong.
As a matter of law, the Executive Order
cannot (and does not purport to) override the
Clean Air Act. The Executive Order does not
conflict with section 109 of the Act because
the requirement that agencies "select
approaches that maximize net benefits'' does
not apply if a "statute requires another
regulatory approach." Executive Order
12866, section (l)(a), (58 FR 51735, October
4, 1993). More generally, the Executive Order
provides that agencies are to adhere to its
regulatory principles only "to the extent
permitted by law." Id., section (l)(b).
UMRA sections 202 and 205 do not apply
to this decision, as discussed in Unit VIII. of
this preamble. Even when they do apply to
a regulatory action, they do not establish
decisional criteria that an agency must follow,
much less override decisional criteria
established in the statute authorizing the
regulatory action. UMRA does not require an
agency to select any particular alternative.
Rather, an agency can select an alternative
that is not the least costly, most cost-effective
or least burdensome if the agency explains
why. Section 205(b)(l) of UMRA. Such an
explanation is not required if the least costly,
most cost-effective or least burdensome
alternative would have been "inconsistent
with law," section 205(b)(2) of UMRA, and
the only alternatives that an agency should
consider are ones that "achievef] the
objectives of the rule," section 205(a) of
UMRA. The UMRA Conference Report
confirms that UMRA does not override the
authorizing statute. "This section [202] does
not require the preparation of any estimate or
analysis if the agency is prohibited by law
from considering the estimate or analysis in
adopting the rule." 141 Cong. Rec. H3063
(daily ed. March 13, 1995).
The RFA (as amended by SBREFA) also
does not apply to this decision, as discussed
in Unit VIII. of this preamble. As is the case
with UMRA, even when the RFA (as
amended by SBREFA) does apply to a
regulatory action, it does not establish
decisional criteria that an agency must follow,
much less override the underlying substantive
statute. When the RFA was adopted in 1980,
Congress made clear that it did not alter the
substantive standards contained in authorizing
statutes: "The requirements of section 603
and 604 of this title [to prepare initial and
final regulatory flexibility analyses] do not
alter in any manner standards otherwise
applicable by law to agency action." Section
606 of the RFA. The legislative history
further explains that section 606 "succinctly
states that this bill does not alter the
substantive standard contained in underlying
statutes which defines the agency's
mandate."85 When Congress passed
SBREFA in 1996 and amended parts of the
RFA, it did not amend section 606.
Even when a regulatory decision is subject
to sections 603 and 604 of the RFA and an
agency is therefore required to analyze
alternatives that minimize significant
economic impacts on small entities, the RFA
(as amended by SBREFA) does not establish
decisional criteria that an agency is required
to follow. Both section 603 and 604 of the
RFA provide that the alternatives an agency
should consider are to be "consistent with the
stated objectives of applicable statutes."
Section 603(c) and 604(a)(5) of the RFA.
Furthermore, although the RFA (as amended
by SBREFA) requires agencies to consider
alternatives that minimize impacts on small
entities subject to the rules' requirements and
to explain their choice of regulatory
alternatives, it does not require agencies to
select such alternatives. For these reasons, the
RFA (as amended by SBREFA) does not
conflict with or override the Clean Air Act's
preclusion of considering costs and similar
factors in setting NAAQS.
3. Conclusion. In summary, EPA
determines that the judicial decisions cited in
this unit are both correct and dispositive on
the question of considering costs in setting
NAAQS, and that the Agency is not free to
reinterpret the Act on that question.
B. Margin of Safety
Several commenters questioned the
approach used by the Administrator in
specifying PM standards that protect public
health with an adequate margin of safety.
Rather than the integrative approach applied
by the Administrator, these commenters
maintained that EPA must employ a two-step
process. One line of argument was that the
Administrator must first determine a "safe
level'' and then apply a margin of safety
taking into account costs and societal impacts.
It was argued that this was the only approach
that would enable the Administrator to reach
a reasoned decision on a standard level that
protects public health against unacceptable
risk of harm, such that any remaining risk
was "acceptable." In effect, these
commenters adopted the two-step
methodology endorsed by Vinyl Chloride,
85 126 Cong. Rec. 21452, 21455 (1980) (Description of
Major Issues and Section-By-Section Analysis of
Substitute for S. 299).
-------
34
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
824 F.2d 1146, for setting hazardous air
pollutant standards under section 112 of the
Act. Another commenter also maintained that
the Administrator must apply a two-step
process but from a different perspective. It
was argued that EPA should first identify the
lowest observed effect level and then apply
a margin of safety to address uncertainties
and to protect the most sensitive individuals
within the at-risk population(s). This
commenter also maintained that the use of
risk assessment in establishing a NAAQS was
a departure from past practice, and that this
departure was not adequately explained.
In recognition of the complexities facing
the Administrator in determining a standard
that protects public health with an adequate
margin of safety, the courts have declined to
impose any specific requirements on the
Administrator's methodological approach.
Thus, in Lead Industries the court held that
the selection of any particular approach to
providing an adequate margin of safety "is
a policy choice of the type Congress
specifically left to the Administrator's
judgment. This court must allow him the
discretion to determine which approach will
best fulfill the goals of the Act." 647 F.2d
at 1161-1162. As a result, the Administrator
is not limited to any single approach to
determining an adequate margin of safety
and, in the exercise of her judgment, may
choose an integrative approach, a two-step
approach, or perhaps some other approach,
depending on the particular circumstances
confronting her in a given NAAQS review.
With respect to the approaches advanced in
comment, the PMio case made clear that the
two-step process endorsed in Vinyl Chloride
was necessary because of the need under
section 112 of the Act to "sever
determinations that must be based solely on
health considerations from those that may
include economic and technical
considerations." 902 F.2d at 973. Because the
Administrator may not consider cost and
technological feasibility under section 109 of
the Act, however, the Court concluded that
"the rationale for parsing the Administrator's
determination into two steps is inapposite."
Id.
The claim that EPA must follow a two-step
process of first identifying the lowest
observed effects level and then applying a
margin of safety has also been rejected by the
courts. In Lead Industries, the Court
specifically held that the Administrator need
not apply a margin of safety at the end of the
analytical process but may take into account
margin of safety considerations throughout
the process as long as such considerations are
fully explained and supported by the record.
647 F.2d 1161-1162. Accord, PM10, 902 F.2d
at 973-974.
Because such factors as the nature and
severity of the health effects involved, the
size of the sensitive population(s) at risk, the
types of health information available, and the
kind and degree of uncertainties that must be
addressed will vary from one pollutant to
another, the most appropriate approach to
establishing a NAAQS with an adequate
margin of safety may be different for each
standard under review. Thus, no generalized
paradigm such as that imbedded in EPA's
cancer risk policy can substitute for the
Administrator's careful and reasoned
assessment of all relevant health factors in
reaching such a judgment. As noted in this
unit, both Congress and the courts have left
to the Administrator's discretion the choice of
analytical approaches and tools, including risk
assessments, rather than prescribing a
particular formula for reaching such
determinations.86 Because of the inherent
uncertainties that the Administrator must
address in margin of safety determinations,
they are largely judgmental in nature,
particularly with respect to non-threshold
pollutants, and may not be amenable to
quantification in terms of what risk is
"acceptable" or any other metric. In view of
these considerations, the task of the
Administrator is to select an approach that
best takes into account the nature of the
health effects and other information assessed
in the air quality criteria for the pollutant in
question and to apply appropriate and
reasoned analysis to ensure that scientific
uncertainties are taken into account in an
appropriate manner.
In mis instance, the Administrator has
clearly articulated the factors she has
considered, the judgments she has had to
make in the face of uncertain and incomplete
information, and alternative views as to how
such information should be interpreted, in
reaching her decision on standard
specifications that will protect public health
with an adequate margin of safety. See Unit
II. of this preamble. Her conclusions on these
matters are fully supported by the record.
C. Data Availability
Several commenters questioned EPA's
ability to rely on studies demonstrating an
association between PM and excess mortality
without obtaining and disclosing the raw
"data" underlying these studies for public
review and comment. In particular, a number
of commenters cited Dockery, D.W., et al.
1993 and Pope, C.A. Ill, et al., 1995, as
studies upon which EPA relied without
obtaining and disclosing the underlying raw
data. One commenter also cited J. Schwartz
et al., 1996 in the same context.87 According
to the commenters, without the underlying
data used in these studies, the reliability of
these studies cannot be assessed accurately.
These commenters requested that EPA obtain
the relevant data and make it available for
public review. In light of the court-ordered
requirement that EPA publish its rule by July
19, 1997, the commenters argued that EPA
must retain the current PMi0 NAAQS
pending additional review of the raw data and
the studies at issue. One commenter, the
American Petroleum Institute (API) requested
that EPA remove the studies from the docket,
unless the underlying data was also included
in the docket.88
A few commenters argued that section
307(d) of the Act requires that EPA obtain
the raw data underlying these studies and that
a failure to do so contradicts the plain
language of section 307(d)(3) of the Act,
which requires EPA to place in the docket
any "factual data on which the proposed rule
is based." Other commenters argued that
under section 307(d)(8) of the Act, a failure
to obtain and disclose the underlying raw data
used in the studies would constitute an error
"so serious and related to matters of such
central relevance to the rule that there is a
substantial likelihood that the rule would have
been significantly changed if such errors had
not been made." Id. According to one
commenter, without the raw data and an
opportunity for an analysis of it, "EPA has
no legal alternative other than to conclude
that no new air quality standard would be
appropriate within the meaning of CAA
section 109(a)(l)(B)." Finally, a number of
commenters have argued that recent caselaw
under the Clean Air Act and other statutes
makes clear that EPA has a legal obligation
to obtain and disclose the data used in these
studies.89
In developing the proposed revisions to the
PM NAAQS, the Administrator relied on the
scientific studies cited in the rulemaking
record, rather than on the raw data underlying
86 Contrary to one of the comments received, EPA's use
of risk assessment in this rulemaking is by no means a
departure from past practice. The EPA first considered and
began applying risk assessment methods in the late 1970's
(44 FR 8210, 8211, February 8, 1979).
87 Contrary to this commenter's assertion, both the
health and air quality data used in the 1996 Schwartz study
are available to interested parties. EPA's Office of
Research and Development maintains a copy of the air
pollution database used in the Schwartz study and it has
previously been made available in response to Freedom of
Information Act requests from interested parties, such as
the American Iron and Steel Institute (AISI). The Harvard
School of Public Health has also made this data available
to several collaborators and to the Health Effects Institute.
With regard to the health data underlying the Schwartz
study, that mortality data was compiled by the National
Center for Health Statistics (NCHS) and can be purchased
from the NCHS by interested parties. Thus, there is no real
data availability concern with regard to the 1996 Schwartz
study. However, even were this not the case, for the
reasons discussed more fully in this unit and elsewhere in
the preamble, EPA believes it would be entitled to rely
upon this study and other studies, including the Dockery
and Pope studies, regardless of the availability of the
underlying health data.
88 API's letter stated that "API petitions EPA to identify
all studies that rely, in any way, on data not available for
public review as part of the rulemaking process and
remove those studies from the record." To the extent this
letter constitutes a "petition" for EPA action, EPA hereby
denies the "petition" for the reasons stated in this unit
and elsewhere in this preamble.
89 One commenter argued that the failure to obtain and
disclose the underlying data was a violation of the
Administrative Procedure Act (APA). The NAAQS
rulemaking is promulgated under section 307(d) of the
Act; the APA generally does not apply to such
rulemakings. See section 307(d)(l) of the Act.
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
35
them.90 In this case, the raw data consists of
responses to health questionnaires based on
information supplied by individual citizens, or
computer tabulations of this information,
which remains confidential, and air quality
and monitoring data, most of which is now
publicly available. EPA does not generally
undertake evaluations of raw, unanalyzed
scientific data as part of its public health
standard setting process. Only in extreme
cases—for example where there are credible
allegations of fraud, abuse or misconduct—
would a review of raw data be warranted. It
would be impractical and unnecessary for
EPA to review underlying data for every
study upon which it relies as support for
every proposed rule or standard. If EPA and
other governmental agencies could not rely
on published studies without conducting an
independent analysis of the enormous volume
of raw data underlying them, then much
plainly relevant scientific information would
become unavailable to EPA for use in setting
standards to protect public health and the
environment. In addition, such data are often
the property of scientific investigators and are
often not readily available because of the
proprietary interests of the investigators or
because of arrangements made to maintain
confidentiality regarding personal health
status and lifestyle information of individuals
included in such data. Without provisions of
confidentiality, the possibility of conducting
such studies could be severely
compromised.91
90 It is important to note that while EPA did use the
Dockery and Pope studies to confirm its conclusions
regarding the health effects of fine particulate air pollution
and thus as support for its decision to revise the PM
standard, these studies do not provide the sole (or even
primary) basis for EPA's decision regarding PM2.5, despite
the assertions of numerous commenters. The proposed
standards are based on a consideration of a large body of
epidemiological studies, a clear majority of which suggest
PM is strongly linked to mortality and other serious health
effects at concentrations permitted under the current
standards. Although the specific levels of the PM2.5
standards are based on a more limited number of studies
that actually measured fine particles and/or components of
fine particles, the Dockery and Pope studies were not used
in initially selecting the annual fine particle standard level,
which was principally based on examination of other daily
mortality and respiratory effects studies (Koman, 1996,
1997) that found significant associations between fine PM
and effects in cities with annual average PM2.5
concentrations of about 16 to 21 (Ig/m3. Only then were
the long-term Dockery and Pope studies examined and
used to help corroborate this result; in the opinion of the
Administrator, neither study alone (or together) provided
sufficient evidence to support more stringent levels below
those identified from the daily studies. Thus, removal of
the Dockery and Pope studies would not affect the
conclusions about the significance of the risks and
therefore, while these long-term studies tend to strengthen
the need for fine particle control and provide important
insights into the nature of PM effects, removal of these
two studies from consideration would not have changed
the selected standard level.
91 Some commenters noted that with regard to the health
data underlying the 1993 Dockery and 1995 Pope studies,
since EPA provided partial funding for these studies, EPA
has access to this data and cannot shield itself from the
duty to obtain this data by claiming that it is not in its
possession. Although a legal argument potentially exists
that EPA may obtain access to such data, this legal
In this case, the merits of the studies
considered and used in developing the PM2.5
standard have been discussed and debated
extensively over the past several years, both
as part of the EPA review of the pertinent
science and in a number of other public
forums. The studies at issue were critically
evaluated by the Agency's Office of Research
and Development (ORD) and by the EPA's
independent Clean Air Scientific Advisory
Committee (CASAC), in a multi-year process
for assessment of the science at issue. As with
other studies on which EPA relied, particular
attention was given to the strengths and
limitations of the Dockery, Schwartz and
Pope studies during this process, which
involved numerous opportunities for public
participation and extensive input from
interested parties. The results of these studies
are not only consistent with each other, but
they are also consistent with the results of
other studies demonstrating significant
associations between long-term exposure to
fine particle indicators and mortality. See
U.S. EPA, 1996b, p. V-62. The CASAC
concluded that EPA's assessments of the
pertinent science properly characterized both
the current state of knowledge and the range
of policy options for revising the standards.
In fact, many peer reviewed studies have
reported associations between PM and
premature death; the Dockery, Schwartz and
Pope studies are among the most recent
studies to corroborate this association. In the
early 1990s, several studies were published
showing associations at levels below the
current PM standards. Some critics began
raising questions about the extent to which
the results could be reproduced and the
unavailability of underlying data. In response,
an independent group of investigators under
the auspices of the Health Effects Institute
(HEI), a highly respected research
organization jointly funded by EPA and
several motor vehicle manufacturers,
undertook a reanalysis of several such studies.
The original investigators of several studies,
including studies conducted at Harvard
University, Brigham Young University, and
the San Francisco Bay Area Air Quality
Management District provided their raw air
quality data sets to the HEI investigation team
for reanalysis. HEI's reanalysis produced
numerical results from the data sets for all six
cities that closely agree with and, in general,
confirm the results of the original
investigators. Thus, as noted in Unit II. of this
preamble, these reanalyses by respected
independent scientists confirmed the
reliability and reproduceability of prior work
of the original investigators, including work
by Dockery et al. (1992), Pope et al. (1992),
argument has not been tested in the courts. More
importantly, EPA's ability to rely on studies without
reviewing the raw data should not depend on whether
some Agency of the Federal government funded the
science.
Schwartz and Dockery (1992a), and Schwartz
(1993).
Thus, the 1993 Dockery and 1995 Pope
studies build upon previous studies done by
a number of different researchers and have
been subject to an extensive peer review
process by EPA's ORD, CASAC and HEI.
They also underwent a peer review process
at the time of their publication in reputable
scientific journals. Given the consistency and
coherence of the scientific evidence and the
scrutiny the studies have received in peer
review and in the extensive scientific review
process described in this unit, EPA does not
agree that review of the underlying data for
these studies is also necessary. Considering
the various reviews described in this unit and
the fact that EPA has received no specific and
substantiated reason, such as plausible
allegations of fraud or scientific abuse, to
doubt the overall validity of their conclusions,
EPA agrees with CASAC that revision of the
standard is appropriate, based on these and
other studies.
In spite of EPA and CASAC's conclusion
that it is appropriate to rely on the Pope,
Dockery and other studies to establish a
PM2.5 NAAQS, EPA also believes in public
disclosure and supports efforts to seek
appropriate release of data underlying the
studies in question. On January 31, 1997,
EPA wrote to the principal scientific
investigators at the Harvard School of Public
Health and at Brigham Young University and
urged them to make the data associated with
their studies available to interested parties.
Studies conducted by these investigators
relied on data compiled as part of the Harvard
Six-Cities Study and data compiled by the
American Cancer Society (ACS) as part of
the Cancer Prevention Study II.
The studies in question combined health
data on individuals with air pollution data.
The air pollution data are publicly available.
The health data consist of personal and
confidential information, e.g. age, sex,
weight, eduction level, smoking history,
occupational exposures, medical history.
These data are not publicly available. In
compiling these data, researchers have
promised study participants that private,
personal information would be kept
confidential under signed assurances of
confidentiality. Data-sharing arrangements
with outside parties must, therefore,
accommodate interests both in making data
accessible and in protecting the
confidentiality of the information contained
within them.
Both the Harvard School of Public Health
and the American Cancer Society have made
such arrangements. Both have processes
which allow ouside scientists, in collaboration
with Harvard and ACS researchers, to access
their databases for the conduct of legitimate
scientific research. Scientists from all over the
world have applied for and have been granted
such access and numerous studies have been
conducted and published using the databases.
-------
36
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
Because of increased interest resulting
from EPA's rulemaking on PM standards and
at the request of the Harvard School of Public
Health, HEI is taking additional steps to
provide a forum for outside researchers to
access health data associated with the
Harvard-Six Cities Study and perhaps others.
HEI has convened an expert panel of
esteemed scientists to access underlying data
and to conduct additional reanalyses. This
arrangement appears to provide a constructive
venue for testing legitimate scientific
hypotheses while protecting the
confidentiality of the underlying data.
Nevertheless, as noted previously, EPA has
full confidence in the scientific integrity of
the Dockery, Schwartz, and Pope studies and
their suitability for use in the Agency's
rulemaking on PM, without undertaking a
separate or additional review and analysis of
the underlying raw data. The decision to
propose revisions of the current PM standards
was based on careful assessment of the
scientific and technical information presented
in the PM Criteria Document and Staff Paper.
The decision was also consistent with the
consensus of CASAC that "although an
understanding of health effects of PM is far
from complete, the Staff Paper, when revised,
will provide an adequate summary of our
present understanding of the scientific basis
for making regulatory decisions concerning
PM standards." The extensive PM
epidemiological data base provides evidence
that serious adverse health effects, e.g.,
mortality, exacerbation of chronic disease,
increased hospital admissions, respiratory
symptoms, and pulmonary function
decrements, in sensitive subpopulations, e.g.,
the elderly, individuals with cardiopulmonary
disease and children, are attributable to PM
at levels below the current standards. The
increase in risk is significant from an overall
public health perspective because of the large
number of individuals in sensitive
subpopulations that are exposed to ambient
PM and the significance of the health effects.
These considerations, as well as others
discussed in the proposal and Staff Paper,
such as the need to consider fine and coarse
particles as distinct classes, led both the
Administrator and CASAC to conclude that
revision of the current standards is clearly
appropriate. This conclusion remains
unchanged despite the fact that EPA is
without the actual raw and unanalyzed health
data underlying the studies.
A number of commenters cited section
307(d) of the Act in support of their position
that EPA is required to obtain and disclose
the underlying raw data. Under section
307(d)(3) of the Act, EPA is required to issue
a notice of proposed rulemaking in the
Federal Register that is accompanied by a
"statement of basis and purpose" that
includes "a summary" of:
(A) The factual data on which the proposed
rule is based.
(B) The methodology used in obtaining the
data and in analyzing the data.
Thus, it is clear from the language of
section 307(d) of the Act that where EPA
relies on any "data" as support in its
rulemakings under the Clean Air Act, it has
an obligation to include such data or
information in the rulemaking docket that is
open to the public. Where EPA fails to do
so and the error is "so serious and related to
matters of such central relevance to the rule
that there is a substantial likelihood that the
rule would have been significantly changed if
such errors had not been made," a reviewing
court may overturn the rule.
In this case, as noted previously, EPA did
not rely upon the raw health data supporting
the Dockery and Pope studies; it relied
instead upon the studies themselves. These
studies may properly be considered "data."
The EPA has never had the raw health data
in its possession; thus EPA has neither
reviewed it nor had an opportunity to place
it in the docket. The EPA did rely on the
studies and these studies are included in the
docket and are available for public review.
Because EPA neither reviewed nor relied
upon the raw data, there is no obligation to
obtain it or to make it available.
Some commenters argued that the language
of section 307(d) of the Act, which refers to
the "factual data" and which also discusses
the ' 'methodology used in obtaining and
analyzing the data" distinguishes between
raw data and studies. In the view of these and
other commenters, the plain language of
section 307(d) of the Act requires that EPA
obtain and disclose the raw data used in the
Dockery and Pope studies. According to these
commenters, without such raw "data," EPA
cannot legally promulgate its rule.
The EPA disagrees with this narrow
interpretation of the word "data" and of
section 307(d) of the Act. Data can take many
forms, including studies, reports, tabulations,
graphs and summaries, as well as more raw
forms, such as questionnaire responses, test
results and even actual physical specimens.
The "factual data" called for by section
307(d) of the Act may clearly include peer-
reviewed scientific studies. Nor does section
307(d) of the Act prohibit EPA from relying
on a study for standard setting without
obtaining the raw, underlying data supporting
a study. Indeed, as noted in the legislative
history to section 307(d) of the Act,
* * * [t]he [House Commerce] Committee
recognizes that the factual support needed for a rule
may vary greatly according to the subject being
addressed and that rules on some subjects, such as
procedures, may not require any factual basis at all.
There is no intention to increase the amount of
'factual' support now required to support 'policy
judgments where no factual certainties exist or
where facts alone do not provide the answer,'
Industrial Union Department, AFL-CIO v.
Hodgson, 499 F.2d 467, 476 (D.C. Cir. 1974). Nor
is there any intent to diminish the Administrator's
authority to adopt precautionary regulations based
on a showing of risk * * * .
H.R. Rep. No. 95-294, at 323 (1977)
(footnote omitted). As this legislative history
makes clear, the language in section 307(d)
of the Act is not intended to require EPA to
change the amount of "factual support" that
EPA must assemble in order to promulgate
a rule and EPA may adopt "precautionary"
regulations ' 'where no factual certainties
exist." Given this clarification in the
legislative history, it is evident that EPA is
entitled under section 307(d) of the Act to
rely on studies rather than raw data in
developing its Clean Air Act rules, despite the
arguably ambiguous use of the term
"data."92
Moreover, EPA has relied on studies in the
past (including studies using the undisclosed
Six Cities data) without obtaining or
disclosing the underlying raw data, and
EPA's reliance on such studies to set Clean
Air Act standards has been upheld in court.
In NRDC v. EPA, 902 F. 2d 962 (D.C. Cir.
1990), the D.C. Circuit declined to delay its
review of the PMio NAAQS rulemaking due
to concerns raised by the American Iron and
Steel Institute about the integrity of the Six
Cities data base. 902 F.2d at 974. In that case,
EPA had relied upon an earlier Dockery study
based on the Six Cities data base. Although
the National Institutes of Health (NIH)
undertook a review of the allegations
regarding the Six Cities database, the court
nevertheless upheld EPA's reliance on that
Dockery study without waiting for the results
of the NIH review. NIH eventually concluded
that the allegations were without merit.
According to the court in the NRDC case:
AISI claims that the EPA relied too much on the
Six Cities Study, which is comprised of the
Dockery study and the Ware study * * * . We do
not agree that the Administrator's selection of the
twenty-four hour standard lacks the necessary
reasoned analysis and supportive evidence * * * .
After carefully reviewing the record, we find EPA's
selection of the twenty four hour standard
reasonable in light of the divergent results in the
studies and the agency's mandate to provide an
adequate margin of safety. Studies contained in the
record provided evidence of adverse health effects
at levels below 250 |lg/m3.
902 F.2d at 969 (footnotes omitted; emphasis
in original). The court also stated that:
In setting a standard under section 109 of the
Act, the Administrator must' 'take into account all
the relevant studies revealed in the record" and
' 'make an informed judgment based on available
92 EPA also does not agree that because the language
of section 307(d) of the Act mentions "factual data" as
well as "the methodology used in obtaining and analyzing
the data," EPA cannot rely on a study alone. In this case,
the study is the "factual data" and EPA's methodology
used in obtaining and analyzing the "factual data" is the
method that EPA used to review and rely upon the studies.
This methodology is discussed extensively in the staff
paper and summarized in some detail elsewhere in this
preamble. In fact, as is clear from the overall structure of
section 307(d) of the Act, as well as the legislative history
cited in this unit, section 307(d) of the Act merely requires
that EPA summarize and disclose the information and
methodology that it relied upon in developing its rule. It
leaves unchanged the "level" of support that an agency
must bring to bear in drafting a proposed rule.
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
37
evidence." American Petroleum Institute v. Costle,
665 F.2d at 1187. The record shows that the
Administrator did so. The Administrator relied on
studies which showed adverse effects at and below
the 250 Ug/m3 level. AISI essentially asks this
court to give different weight to the studies than
did the Administrator. We must decline. It is
simply not the court's role to "second-guess the
scientific judgments of the EPA. * * * [T]he
Administrator did not act arbitrarily in drawing
conclusions from the uncertain and conflicting data.
The Administrator may reasonably apply his
expertise to draw conclusions from "imperfect
data," Ethyl Corp., 541 F.2d at 28, as he did here.
Mat 971.
As this language makes plain, the term
"data" may include a study relied upon by
EPA. It should be equally plain that EPA may
properly rely on such a study in setting a
standard despite the fact that such "data"
may be "imperfect," "conflicting," and
"uncertain." There are numerous other cases
in which EPA has relied on studies in setting
standards under the Clean Air Act. See, e.g.,
Engine Manufacturers Association v. EPA, 88
F. 3d 1075, 1099 (B.C. Cir. 1996)(upholding
EPA's use of the 1993 Dockery study for
setting mobile source standards); API v.
Costle, 665 F.2d 1176, 1185 (B.C. Cir.
1981)(Administrator's conclusion that normal
body functions are disrupted by ozone is
"supported by the studies").
A number of commenters cited
Endangered Species Committee v. Babbitt,
852 F. Supp. 32 (D.D.C. 1994) (hereafter
"Gnatcatcher") in support of the proposition
that EPA must obtain and disclose the raw
data underlying the Dockery and Pope
studies. Relying on cases such as Connecticut
Light and Power Co. v. NRC, 673 F.2d 525
(B.C. Cir. 1982), Portland Cement v.
Ruckelshaus, 486 F.2d 375 (B.C. Cir. 1973),
and United States v. Nova Scotia Food
Processing Corp, 568 F.2d 240 (2nd Cir.
1977), these commenters suggest that "a
body of legal decisions is emerging whereby
federal courts are increasingly dubious of
final regulatory decisions that are being made
absent public scrutiny of the data underlying
and purportedly supporting such decisions."
According to these commenters, based on
Gnatcatcher and other cases, failure by EPA
to obtain and place in the docket the raw
unanalyzed data used in the Bockery and
Pope studies constitutes serious procedural
error under the Clean Air Act.
Under Connecticut Light and Power,
agencies must make available technical
studies and data that have been relied upon
during the rulemaking process in order for the
public to have an adequate opportunity for
notice and comment. There is no question that
EPA has done this with regard to the Bockery
and Pope studies, which are included in the
rulemaking docket. The Portland Cement case
makes clear that where an agency actually
relies on factual data it cannot' 'promulgate
rules on the basis of inadequate data, or on
data that, [to a] critical degree, is known only
to the agency." 486 F.2d at 393. See also,
Nova Scotia, 568 F.2d 240, at 251 (where all
of the research was collected by the agency,
and none of it was disclosed "as the material
upon which the proposed rule would be
fashioned," error resulted); CMA v. EPA, 870
F.2d 177, 200 (5th Cir. 1989) ("fairness
requires that the agency afford interested
parties an opportunity to challenge the
underlying factual data relied on by the
agency").
However, in this case, EPA did not rely on,
nor did it ever have or review, the underlying
data used in the Bockery and Pope studies.
Instead, it relied upon the studies themselves.
Thus, the cases cited in this unit are
inapposite. They stand only for the
proposition that where an agency actually
reviews and relies on "data," which may be
raw data, a study or a variety of other forms
of information, it must make these data
available. They do not and cannot stand for
the proposition that an agency may not rely
on a study alone and must always obtain the
raw and unanalyzed data underlying a study.
Indeed, as one B.C. Circuit case noted:
"Portland Cement and Nova Scotia simply
cannot be twisted so as to require notices of
proposed or interim rules to contain elaborate
reproductions of underlying studies." Petry v.
Block, 737F.2d 1193, 1198 (B.C. Cir. 1984).
Requiring EPA to obtain, analyze and
disclose the data underlying the Pope and
Bockery studies, which EPA neither reviewed
nor relied upon, would be to require EPA to
attempt such an "elaborate reproduction."
Such a step is not required under the law and
would make it extremely difficult, if not
impossible, for EPA to regulate in complex,
technical areas "at the frontiers of science."
Baltimore Gas and Electric Co. v. NRC, 462
U.S. 87(1983).
The district court's decision in the
Gnatcatcher case is similarly inapposite. That
case concerned a scientific study regarding
the range of the California Gnatcatcher, a
small insectivorous songbird. As the
Gnatcatcher opinion itself notes, "courts have
generally allowed agencies to rely on
scientific reports." Gnatcatcher, 852 F.Supp.
at 37. Thus, the question at issue in
Gnatcatcher was whether specific
circumstances exist in which an agency may
not be entitled to rely on studies alone. In the
Gnatcatcher case, a single author had
published two directly contradictory studies
on the same issue, while relying on the same
data. In light of this clear contradiction,
commenters in that rulemaking argued that
without the underlying data it was impossible
to determine whether the conclusions in either
study were correct. The district court noted
that:
The Secretary had before him a report by an
author who, two years before had analyzed the
same data and come to an opposite conclusion. It
is the disputed nature of this report that
distinguishes this from other cases where a
scientific report alone has been considered
sufficient for ESA purposes.
Id. Thus, according to the court: ' 'While
courts have generally allowed agencies to rely
on scientific reports * * * this is not sufficient
in this case because the report itself is under
serious question." Id.
The EPA's current reliance on the Bockery
and Pope studies bears no resemblance to the
circumstances present in the Gnatcatcher
decision. As noted previously, these studies
have been subject to extensive peer review
and scrutiny, and neither researcher has
published a contradictory study on the same
issue, much less using the same data base.
The EPA is not aware of, nor have any of
the commenters raised any particular issues
relating to either gross error, fraud or
scientific abuse arising from the data. Indeed,
as noted previously, the prior work of these
particular researchers has been subject to
extensive independent scrutiny and reanalysis,
which has confirmed, rather than called into
question, the underlying validity of their
conclusions and the integrity of their research
methods. Reading Gnatcatcher to suggest that
EPA cannot rely on such a study, where the
study and its methods have been subject to
extensive peer review, would place the
district court's rationale in Gnatcatcher in
conflict with applicable B.C. Circuit
precedent that makes evident the right of
agencies to rely on studies alone. See, e.g.,
Engine Manufacturers Association v. EPA, 88
F.3d 1075, 1099 (B.C. Cir 1996); AP/v.
Costle, 665 F.2d 1176, 1185 (B.C. Cir. 1981),
"studies discussed in the Criteria Bocument
constitute a rational basis for the finding that
adverse health effects occur at ozone levels
of 0.15-0.25 ppm for sensitive individuals";
see also, NRDC v. Thomas, 805 F.2d 410,
418 (B.C. Cir. 1986)(EPA use of a summary
of confidential data that was not disclosed
provides "a reasoned explanation for moving
from a 4.0 to 5.0 long term NOx standard").
In addition, to require EPA to obtain and
analyze the data prior to revising the standard
would also contradict the "common sense
notion that Congress, in providing for notice
and comment under the APA, could not have
intended to subject the agencies—and the
public on whose behalf they regulate—to [a]
sort of interminable back and forth."
International Fabricare Institute v. EPA, 972
F.2d 384, 399 (B.C. Cir. 1992). In the view
of some commenters, EPA has no choice but
to either postpone its decision for a year or
more awaiting a review of data or choose to
retain the current standard. Yet were EPA to
adopt such an approach, these commenters
would undoubtedly insist that EPA be
required to include an analysis of the data in
the docket; further questions would likely be
raised regarding the re-analysis and once
again EPA might find itself unable to
promulgate its rule pending review of further
hypothetical questions. This type of unending
inquiry is not required under the law. As the
B.C. Circuit has noted:
-------
38
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
* * * [Disagreement among the experts is
inevitable when the issues involved are at the
"very frontiers of scientific knowledge," and such
disagreement does not prevent us from finding that
the Administrator's decisions are adequately
supported by the evidence in the record * * * . It
is not our function to resolve disagreement among
the experts or to judge the merits of competing
expert views * * * . Cf. Hercules, Inc. v. EPA, 598
F.2d 91,115 (D.C. Cir. 1978) ("[c]hoice among
scientific test data is precisely the type of judgment
that must be made by EPA, not this court").
Lead Industries Association v. EPA, 647 F.2d
1130, 1160 (D.C. Cir. 1980).
Neither Gnatcatcher, nor any other case can
fairly be read to suggest that EPA has an
obligation to respond to all possible questions
that might be raised regarding its scientific
conclusions or that where EPA relies on a
study, it must engage in a multi-phased and
possibly unending re-examination of the data
supporting such a study until all commenters
are satisfied in full with the details of the
underlying science. Even assuming that EPA
could obtain the confidential Six Cities data
through litigation, a substantial delay of many
months, if not years, would likely result, in
order for both EPA and industry to reanalyze
the data. In the meantime, some tens of
thousands of premature deaths could result.
Neither the Clean Air Act nor relevant case
law requires or permits such a result.
Indeed, the suggestion that EPA cannot and
should not rely upon the Pope, Dockery, and
Schwartz studies, unless and until interested
parties have had an opportunity to examine
and reanalyze the underlying raw data, is
extraordinary. Given the precautionary nature
of section 109 of the Act and the need to
allow an adequate margin of safety, see Lead
Industries, 647 F.2d at 1154, 1155, there are
limits on EPA's discretion to disregard even
studies that are clearly flawed, if they are
nonetheless "useful" in indicating the kind
and extent of health effects that may result
from the presence of a pollutant in the
ambient air. See sections 109(b)(l) and
108(a)(2)oftheAct.
A few commenters cited Kennecott v. EPA,
684 F.2d 1007 (D.C. Cir. 1982) and argued
that under sections 307(d)(8) and
307(d)(9)(D) of the Act, a failure by EPA to
obtain and include in the docket the data
underlying the Pope and Dockery studies
would constitute an "error" that is "so
serious and related to matters of such central
relevance to the rule that there is a substantial
likelihood that the rule would have been
significantly changed if such errorf] had not
been made."93 EPA disagrees. Peer reviewed
studies conducted by outside parties were not
at issue in Kennecott. Kennecott involved a
dispute over financial analyses that EPA itself
had previously conducted and used in earlier
rulemakings. The court determined that the
financial analyses at issue must have provided
at least part of the factual basis for EPA's
rule, and EPA referenced these analyses in
the preamble to the final rule without placing
them in the docket until one week before
promulgation. The factual circumstances in
Kennecott are substantially different than the
current situation and thus, Kennecott cannot
fairly be read to establish the applicable legal
standard with regard to EPA's reliance on
peer reviewed studies for use in setting the
NAAQS.
In this case, EPA—well before proposal—
has placed the information that it relied upon
in the docket. This information is in the form
of studies. These studies have been subject to
extensive scrutiny and peer review. To date
no specific allegation has been made that the
studies are clearly in error or that the data
underlying them are the subject of fraud,
scientific misconduct, or gross error going to
the basic validity of the studies.94 Instead,
various commenters have merely stated their
view that were the raw data behind these
studies available, they would be able to better
verify and assess the results reached in the
studies.
As one commenter noted, "In the absence
of data on which EPA's proposal is based,
[key scientific] issues remain shrouded in
uncertainty and skepticism. The disclosure of
the data would allow for robust scientific
analysis and discussion of these issues." A
similarly hypothetical concern is raised by
another commenter who stated that "seeing
the data would clarify substantial questions of
methodology" and "had the Harvard data
been available, a far broader evaluation of the
defects of the Harvard Studies would have
been possible with the same expenditure of
time and money." Yet, despite having spent
"in the neighborhood of a million dollars to
duplicate and reanalyze the Harvard data set"
this commenter was unable to allege any
particular defect in the methodology or results
of these studies and noted instead that "the
track record to date suggests that the claimed
associations to PM2.5 and health effects
would not have held up under such a broader
evaluation."
93 One commenter argued that EPA's failure to place
the "data" in the docket was not an "error" but a
"refusal to comply with the clear language of the law that
should be reviewed by the courts under section
307(d)(9)(C), rather than 307(d)(9)p)." As noted
previously, EPA does not agree with this interpretation of
section 307(d)(3) of the Act. Under applicable caselaw, the
term "data" may include information in many forms,
including studies that EPA has placed in the docket. See
Endangered Species Committee v. Babbitt, 852 F. Supp.
32, 37 (D.D.C., 1994) ("data can come in many forms:
it can be a scientific report, it can be graphs and
tabulations * * * it can be raw numbers").
94 A number of commenters did argue these studies do
not form a sufficient basis for EPA's decision to revise
the NAAQS and that attempts to replicate these studies
have not been universally successful. These same
commenters also listed a number of hypothetical questions
and issues that might be resolved through a review of the
underlying data and suggested that before EPA may
properly rely on these studies to revise the NAAQS, a
variety of confounders (such as smoking) should also be
ruled out by reviewing the data. As set forth more fully
in Unit II. of this preamble, neither EPA nor CAS AC
agrees that any of these factors precludes reliance on the
studies in question.
EPA is not required to await the results of
such an inquiry before proceeding to regulate
to protect human health and the environment.
The concerns raised by the commenters
regarding these studies remain hypothetical;
the comments themselves raise no allegations
of fraud, scientific misconduct or gross error
that calls into question the fundamental
validity of the studies. Given this fact, EPA
does not agree with the commenters that
reliance on these studies and/or a failure to
place the underlying data in the docket
constitutes an error, much less an error that
is "so serious and related to matters of such
central relevance that there is a substantial
likelihood that the rule would have been
significantly changed." EPA is entitled to
rely upon these studies and it has satisfied its
obligation to provide the "factual data" upon
which the proposed rule is based by placing
these studies in the docket.
In fact, the concerns raised by the
commenters ultimately boil down to a
disagreement with EPA over the level of
scientific certainty necessary to adopt the
NAAQS revisions. In setting standards under
the Clean Air Act, EPA is not required to
resolve all scientific issues to the complete
satisfaction of every interested party. As
noted by the D.C. Circuit in Lead Industries
Association v. EPA, 647 F.2d 1130, 1160
(D.C. Cir. 1980):
To be sure, the Administrator's conclusions were
not unchallenged; both LIA and the Administrator
are able to point to an impressive array of experts
supporting each of their respective positions.
However, disagreement among the experts is
inevitable when the issues involved are at the
"very frontiers of scientific knowledge," and such
disagreement does not preclude us from finding
that the Administrator's decisions are adequately
supported by the evidence in the record. It may be
that LIA expects this court to conclude that LIA's
experts are right, and the experts whose testimony
supports EPA are wrong. If so, LIA has seriously
misconceived our role * * * . It is not our function
to resolve disagreement among the experts or to
judge the merits of competing expert views * * *
. Cf. Hercules, Inc., v. EPA, 598 F.2d 91, 115 (D.C.
Cir. 1978) ("[cjhoice among scientific test data is
precisely the type of judgment that must be made
by EPA, not this court").
647 F.2d at 1160 (footnotes omitted).
The EPA's rationale for proposing to add
a fine particle standard was detailed in the
preamble to the proposed rule, most notably
at 61 FR 65654-65662, December 13, 1996.
This decision is based on the extensive
review of the science and policy issues
contained in the PM Criteria Document and
Staff Paper; the CASAC concluded, after
extensive review, that both of these
documents were appropriate for use in
decision making on standards. These
documents contain a full discussion of both
what is known about PM and the information
gaps and uncertainties. Considering the full
weight of the scientific evidence, including
the uncertainties, the CASAC recommended
that the Administrator adopt fine particle
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
39
standards and a number of panel members
based their support for a PM2.5 standard on
the following reasoning:
[TJhere is strong consistency and coherence of
information indicating that high concentrations of
urban air pollution adversely affect human health,
there are already NAAQS that deal with all of the
major components of that pollution except PMi.5,
and there are strong reasons to believe that PMi.s
is at least as important as PMio-2.5 in producing
adverse health effects.
Wolff, 1996.
Given the consistency and coherence of the
evidence that premature mortality and
sickness occur in large numbers of Americans
at concentrations permitted by the current
standards, it would be irresponsible to delay
action that would put more appropriate air
quality goals into place based on the most
recent scientific information. After a review
of the comments submitted, the Agency's
conclusion that it is appropriate to rely on the
existing studies remains unchanged.
D. 1990 Amendments
Contrary to the view expressed in some
public comments, the provisions of subpart 4
of Part D of Title I of the Act, enacted in
1990, do not preclude EPA from adopting
PM2.5 as an additional indicator for PM and
establishing standards for PM2.5. The
provisions of subpart 4 of Part D of Title I
of the Act simply do not limit EPA's clear
authority under section 109 of the Act to
revise the PM standards.
The basic contention is that because the
provisions of subpart 4 of Part D of Title I
of the Act refer to PMio, they prohibit EPA
from regulating any other type of PM, for
example, by revising the existing NAAQS for
PM by adopting an ambient air quality
standard for PM2 5. These provisions,
however, do not lead to such a conclusion.
Moreover, this view ignores provisions
indicating that Congress believed that EPA
could revise any existing NAAQS or adopt
a new NAAQS.
At the outset, it should be noted that
Congress expressly authorized EPA to revise
any ambient air quality standard and to adopt
a new NAAQS in section 109 of the Act.
That section, which requires EPA to review
and revise, as appropriate, each NAAQS
every five years, contains no language
expressly or implicitly prohibiting EPA from
revising a NAAQS or adopting a new
NAAQS. If Congress had intended to
preclude EPA from reviewing and revising a
NAAQS or adopting a new NAAQS, which
are part of EPA's fundamental functions,
Congress would have specifically done so.
Clearly, Congress knew how to preclude EPA
from exercising otherwise existing regulatory
authority and did so in other instances. See
section 202(b)(l)(C) of the Act (expressly
precluding EPA from modifying certain
motor vehicle standards prior to model year
2004); section 112(b)(2) of the Act
(preventing EPA from adding to the list of
hazardous air pollutants any air pollutants that
are listed under section 108(a) of the Act
unless they meet the specific exceptions of
section 112(b)(2) of the Act); section
249(e)(3), (f) and section 250(b) (limiting
EPA's authority regarding certain clean-fuel
vehicle programs). No such language was
included either in section 109 of the Act or
elsewhere in the Act and no such implication
may properly be based on the provisions of
subpart 4 of Part D of Title I of the Act.
Second, other provisions of the Act
expressly contemplate EPA's ability to
promulgate a new or revised NAAQS, and
provide no indication that such ability is
limited to standards other than those whose
implementation is the subject of subparts 2,
3 and 4 of Part D of Title I of the Act. For
example, section 110(a)(2)(H)(i) of the Act
provides that SIPs are to provide for revisions
"from time to time as may be necessary to
take account of revisions of such national
primary or secondary ambient air quality
standard * * * ." Section 107(d)(l)(A) of the
Act provides a process for designating areas
as attainment, nonattainment, or
unclassifiable "after promulgation of a new
or revised standard for any pollutant under
section 109 * * * ." Section 172(e) of the
Act addresses modifications of national
primary ambient air quality standards.
Finally, section 172(a)(l) of the Act expressly
contemplates that EPA may revise a standard
in effect at the time of enactment of the 1990
Clean Air Act Amendments. Section
172(a)(l)(A) of the Act provides EPA with
authority to classify nonattainment areas on
or after the designation of an area as
nonattainment with respect to "any revised
standard, including a revision of any standard
in effect on the date of the enactment of the
Clean Air Act Amendments of 1990."
Plainly, Congress had no intention of
prohibiting EPA from revising any of the
ambient standards in effect at the time of the
enactment of the 1990 amendments.
Third, the provisions of subpart 4 of Part
D of Title I of the Act do not support the
contention that they somehow preclude EPA
from exercising its authority to adopt a
revised PM NAAQS based on a metric other
than PMio. The fact that Congress laid out
an implementation program for the PM
standard existing at the time of the 1990
amendments in no way suggests that
Congress intended to preclude EPA from
exercising the authority it provided EPA to
revise the NAAQS when the health data on
which EPA bases such decisions warranted a
change in the standard.
The fact that Congress drafted subpart 4 of
Part D of Title I of the Act in 1990 to specify
the implementation regime for the PM
standard then in effect, a PMio standard, in
terms that explicitly refer to PMio in no way
suggests that Congress meant to preclude
EPA from adopting a PM standard based on
another metric if scientific information
supported such a change. Obviously, PMio
was the standard in existence in 1990 and
Congress drafted subpart 4 of Part D of Title
I of the Act, the purpose of which was to
delineate an implementation regime for that
standard, in terms of that standard. There is
simply no language in subpart 4 of Part D of
Title I of the Act that limits EPA's ability to
establish a different PM standard if such a
standard were warranted under section 109 of
the Act or indicates any implicit intent on the
part of Congress to limit EPA's authority
under section 109 of the Act in such a way.
Subpart 4 of Part D of Title I of the Act
simply does not speak to the question of
whether EPA may establish a PM standard
based on a different metric. In addition,
section 107(d)(4) of the Act, the only
provision outside of subpart 4 of Part D of
Title I of the Act invoked as a basis for the
view that the Act prohibits EPA from
adopting a PM2 5 standard, does not support
that view. That provision simply preserved
pre-existing designations for "total suspended
particulates," the PM metric utilized prior to
PMio, for certain purposes. It provides no
suggestion that Congress intended to prohibit
EPA from adopting a metric other than PMi0.
Indeed, if anything, it indicates that Congress
was fully aware that EPA had previously
changed the PM metric used in the PM
NAAQS and confirms the view that Congress
would have explicitly barred EPA from
changing the metric had it intended to do so.
Finally, for the reasons stated in this unit,
EPA's analysis of its ability to implement a
PM2.5 standard under the provisions of
subpart 1 of Part D of Title I does not support
the view that Congress prohibited EPA from
promulgating such a standard. Congress
clearly specified an approach to the
implementation of the PMio standard in the
provisions of subpart 4 of Part D of Title I
of the Act. The EPA believes that the clear
and express linkage of that approach to the
PMio standard indicates that a different PM
standard should be implemented under the
general principles of subpart 1 of Part D of
Title I of the Act. That Congress directed
specifically how EPA and the States should
implement the PMio standard does not carry
with it the implication that Congress intended
to prohibit EPA from exercising its otherwise
clear and express authority to adopt a PM
standard based on a different metric in order
to carry out one of its fundamental missions,
the establishment of ambient air quality
standards to protect public health with an
adequate margin of safety. It is entirely
reasonable and logical for Congress to, on the
one hand, specify an implementation regime
for the PM standard in effect at the time of
enactment of the 1990 amendments, but, on
the other hand, leave EPA free to exercise the
authority provided it by Congress in section
109 of the Act to adopt a new or revised
standard when EPA determined that such a
standard was needed to protect public health
with an adequate margin of safety. Congress
-------
40
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
explicitly required EPA to review and revise
as appropriate the NAAQS every five years.
If Congress did not intend for EPA to revise
the NAAQS when warranted, it would not
have required EPA to review and revise them.
If Congress had intended to prohibit EPA
from exercising such a fundamental authority
it would have clearly specified, as it did in
other instances, that EPA could not do so.
V. Revisions to 40 CFR Part 50, Appendix
K—Intrepretation of the PM NAAQS
Because the revocation of the existing
PMio standards will become effective at a
later date (as discussed in Unit VII. of this
preamble), EPA is retaining 40 CFR part 50,
Appendix K, although it is being published
today in revised format to conform with the
format of the other appendices in this part.
A new Appendix N to 40 CFR part 50
explains the computations necessary for
determining when the primary and secondary
PM2.5 and PMio standards being adopted
today are met. The discussion in this unit
sometimes refers to the contents of the new
Appendix N as revisions to Appendix K, so
as to highlight how the new Appendix N
differs from the current Appendix K.
Key elements of the new 40 CFR part 50,
Appendix N, particularly as they differ from
those of Appendix K, are outlined in this unit.
A. PM2.5 Computations and Data Handling
Conventions
As discussed in Unit II.E. of this preamble,
the form of the annual PM2.5 standard is a
spatially averaged annual mean averaged over
3 years, and the form of the 24-hour PM2.5
standard is a 98th percentile concentration
averaged over 3 years.
With regard to the annual PM2.5 standard,
the EPA proposed a form expressed as the
annual arithmetic mean, averaged over 3
years and spatially averaged over all
designated monitoring sites to represent
population exposures. As discussed in Unit
II.E.I. of this preamble, the form of the
annual PM2.5 standard has been clarified to
make explicit that implementing agencies
have the flexibility to base comparison of the
standard level with measured values from
either a single community-oriented site or an
average of measured values from such
monitors within the constraints enumerated in
40 CFR part 58. The new Appendix N of 40
CFR part 50 reflects this clarification. The
spatial average, if used, is to be carried out
using data from monitoring sites designated
in a State PM Monitoring Network
Description in accordance with the provisions
of 40 CFR part 58.
Also, the EPA proposed that, for spatial
averaging, the requirements for 3 years of
data for comparison with the standard be
fulfilled by the spatial averaging network as
a whole, not by individual monitors within
the network. The EPA received comments
regarding the application of the 75 percent
data completeness requirement to spatial
averaging. The commenters stated that the
inclusion or exclusion of a site not meeting
the data completeness requirements from a
spatial average, based on the level of the
single site average, would bias the spatial
average for that year. The EPA has responded
to the comment by demonstrating in Example
1 in 40 CFR part 50, Appendix N the
application of the data completeness criterion
that is consistent with a spatially averaged
network. Specifically, the application of the
data completeness requirement has been
altered in the example if a particular site has
quarters in a year that do not meet the
minimum data completeness requirement.
Instead of comparing a site's annual average
to the level of the standard to decide whether
or not to keep the site in the calculations, the
annual average for all the sites (the spatial
average) is compared to the level of the
standard. If the spatial average is above the
level of the standard, the site is kept in the
calculations. If it is below, the site is omitted
from the calculations.
The EPA also proposed that averaging over
calendar quarters be retained for the annual
average form of the standard. Although
several commenters stated that the step of
calculating quarterly averages to obtain the
annual average was unnecessary, the EPA
maintains that quarterly averages are
important to ensure representative sampling
in areas with extreme seasonal variation.
Regarding the 75 percent data
completeness requirement, the proposal stated
that a given year meets data completeness
requirements when at least 75 percent of the
scheduled sampling days for each quarter
have valid data, and high values measured in
incomplete quarters shall not be ignored but
shall be included if their value causes the
annual calculation to be above the level of the
standard. Some commenters felt that this
treatment was unfair in that measured data
below the standard in incomplete quarters are
not retained. In addition, the commenters felt
that this could create a bias where a single
sample could inflate an annual average to a
level above the standard. The EPA agrees and
has incorporated in 40 CFR part 50,
Appendix N the following provisions.
(1) A statement has been added that less
than complete data may be used in certain
cases subject to the approval of the
appropriate Regional Administrator in
accordance with EPA guidance for dealing
with less than complete data. This statement
was considered necessary for those situations
where measured data and air quality analyses
would indicate that the area met or did not
meet the standard although it did not exactly
meet the data completeness requirements.
(2) A provision has been added that a
minimal amount of data is needed before the
requirement to retain high values in an
incomplete quarter comes into effect for the
annual standards. Sites with at least 11
samples but less than 75 percent data
completeness in a quarter will have to include
high values if they result in calculated values
which are above the level of the standard.
This provision is based upon the change in
sampling frequency set forth in the revisions
to 40 CFR part 58 which effectively doubles
the minimum sampling frequency from 1-in-
6 day sampling to l-in-3 day sampling. The
data completeness requirement for the annual
form of the standard under the original 1-in-
6 day sampling schedule is equivalent to a
minimum of 37.5 percent under the new
sampling schedule of l-in-3 days. This is
equivalent to a minimum of 11 samples in
each quarter. Therefore, a minimum of 11
samples in a quarter should be sufficient for
an annual average above the level of the
standard to be used under the new sampling
schedule.
(3) In sharp contrast, this minimum
requirement was considered unnecessary for
the 24-hour form of the standard when the
98th percentile is above the level of the
standard. That is, for a site with a 98th
percentile above the level of the standard that
does not meet the 75 percent data
completeness requirement, the 98th percentile
would be equivalent to the maximum or
second maximum daily concentration in that
year. While adding more data samples up to
the minimum data completeness requirement
of 75 percent could help to ensure that the
second maximum value (rather than the
maximum value) corresponds to the 98th
percentile, this difference is not considered
significant enough to require some minimal
number of data samples when dealing with
the form of the 24-hour standard.
With regard to the 24-hour PM2.5 standard,
the proposed revision to 40 CFR part 50,
Appendix K defined the 98th percentile as the
daily value out of a year of monitoring data
below which 98 percent of all values in the
group fall. The calculation of the percentile
form has been revised to reflect general
comments that the form of the standard and
its calculation should be simplified. The EPA
maintains that the revised calculation is
consistent with the definition of the percentile
being that number below which a certain
percent of the data fall.
Regarding the expression of the annual
standard to the nearest 0.1 |ig/m3 and the 24-
hour standard to the nearest 1 |ig/m3, virtually
no commenters disagreed with the EPA's
proposed approach. The few that did,
however, took issue with the overall
stringency of the standards, not the rationale
discussed in the proposal. The EPA maintains
its position that instrument sensitivity and the
number of measured values used in
calculating the values to be compared to the
standard, as discussed at length in the
proposal, point to keeping the expressions of
the standards stated in this unit.
B. PMio Computations and Data Handling
Conventions
As discussed in Unit II.G. of this preamble,
the EPA proposed retaining the current
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
41
annual arithmetic mean, averaged over 3
years, as the form of the annual PMio
standard, and changing the form of the 24-
hour PMio standard to a 98th percentile value
form, averaged over 3 years. As discussed in
Unit II. G. of this preamble, the form of the
daily PMio standard has been revised to a
99th percentile instead of the 98th percentile,
and the related calculations have been revised
accordingly. The same revision described
above in Unit V.A. of this preamble to
simplify the formula used to calculate the
percentile form of the 24-hour PM2.5 standard
also applies to the PMio 99th percentile
calculation.
The revisions made to the annual and 24-
hour PM2.5 standards regarding the 75 percent
data completeness requirement also apply to
the annual and 24-hour PMio standards.
Appendix N of 40 CFR part 50 reflects this
change.
As with the PM2.5 standards, the EPA
maintains its position that instrument
sensitivity and the number of measured
values used in calculating the values to be
compared to the standard, as discussed in
detail in the proposal, point to keeping the
expressions of the standards to the nearest 1
|ig/m3 for the annual standard and to the
nearest 10 |ig/m3 for the 24-hour standard.
C. Changes That Apply to Both PM^.5 and
PM\o Computations
In the proposal, the EPA stated that
revisions to 40 CFR part 50, Appendix K
would not address the treatment of
exceptional events data, which are considered
part of the standards implementation process.
Since several commenters mentioned the
handling of these events in conjunction with
the proposed revisions to Appendix K, the
EPA has addressed this concern in Appendix
N of 40 CFR part 50, which states that
whether to exclude, retain, or make
adjustments to data affected by uncontrollable
or natural events is subject to the approval of
the appropriate Regional Administrator.
Comments were also received expressing
the desire of some areas to conduct seasonal
sampling, reducing the frequency of
monitoring during a period of expected low
concentrations to save resources. The
proposed revision to 40 CFR part 50,
Appendix K did not prohibit this course of
action, and referred matters of sampling
frequency to 40 CFR 58.13. For clarification,
40 CFR part 50, Appendix N adds that
exceptions to specified sampling frequencies,
such as a reduced frequency during a season
of expected low concentrations, shall be
subject to the approval of the appropriate
Regional Administrator.
VI. Reference Methods for the
Determination of Participate Matter as
PMio and PM2 5 in the Atmosphere
A. Revisions to 40 CFR Part 50, Appendix
J—Reference Method for PM\ o
Because the revocation of the existing
PMio standards will become effective at a
later date (as discussed in Unit VII. of this
preamble), EPA is retaining Appendix J in its
current form. A new Appendix M to 40 CFR
part 50 establishes the reference method for
measuring PMio in the ambient air for the
revised PMio standards. The discussion in
this unit sometimes refers to the contents of
the new Appendix M as revisions to
Appendix J, so as to highlight how the new
Appendix M differs from the current
Appendix J. As discussed below, the only
revision to the Reference Method for PMio
relates to the calculation of the volume of air
sampled.
During the course of this standards review,
EPA has received a number of comments
regarding the appropriateness of the current
practice of adjusting measured PMio
concentrations to reflect standard conditions
of temperature and pressure (25° C and 760
mm Hg, respectively), as required by 40 CFR
part 50, Appendix J. The practice was
originally adopted to provide a standard basis
for comparing all pollutants measured in
terms of mass per unit volume (e.g., |ig/m3).
As EPA has reviewed the ambient standards
for gaseous pollutants, however, technical
changes have been made to express them on
a pollutant volume/air volume basis (i.e.,
ppm) that is insensitive to differences in
altitude and temperature. Such an approach is
not applicable to particulate pollutants. The
question arises whether continuing the past
practice of making temperature and pressure
adjustments for PM is appropriate or
necessary.
Information in the Criteria Document on
the health and welfare effects of PM provides
no clear basis for making such adjustments.
Recent health effects studies have been
conducted in cool and warm climates, and in
cities at high altitude, e.g., Denver, as well
as near sea level, e.g., Philadelphia (U.S.
EPA, 1996a). These studies provide no
evidence that risk associated with PM
exposures is affected by variations in altitude.
Accordingly, any effect that would be
accounted for by temperature and pressure
adjustments would be below the detection
limits of epidemic logical studies. While
extremes of altitude might be expected to
increase the delivered dose of PM in those
not acclimatized to such locations, the
dosimetric studies summarized in the Criteria
Document provide no clear support for any
quantitative adjustment to standard
conditions. With respect to welfare effects,
visibility is directly related to the actual mass
of fine particles in the atmosphere.
Adjustment of PM concentrations collected at
higher altitudes to standard conditions would
therefore lead to an overstatement of the
effect of PM on visibility in such locations.
Similarly, there is no evidence in the Criteria
Document suggesting that effects on materials
damage and soiling are dependent on altitude.
Based on this assessment, EPA proposed to
delete the requirement to adjust PMio
concentrations to standard conditions of
temperature and pressure from 40 CFR part
50, Appendix J for the revised standards and
to make corresponding revisions in 40 CFR
50.3. Comments received on this issue were
divided. A number of commentors supported
EPA's proposal for the reasons set forth
above. A few States opposed the change
because the lack of adjustment for very cold
temperature in areas near sea level could
make the standard more stringent. Some
commentors were concerned that the
proposed change would relax protection
afforded for areas at high altitude. A few
commentors expressed concern that
"sojoumers" who visit high altitude area
would have higher ventilation rates and
receive reduced protection as compared to
local residents whose ventilation patterns
were more adapted to these conditions.
The EPA does not believe that the
localized comparisons regarding increased or
decreased stringency of standards relative to
the proposed change are an appropriate
rationale for keeping the current adjustment
for temperature and pressure. The issue is
whether the available scientific evidence on
the health and welfare effects of PM provides
a basis for continuing with the traditional
adjustments. The comments with respect to
sojoumers at altitude are relevant, but this
issue was considered in reaching the proposed
decision. Furthermore, commentors provided
neither laboratory nor epidemiologic evidence
that would support their theoretical concerns
regarding increased annual or 24-hour PM
effects at altitudes typical of mountainous
urban areas in the United States.
Based on its assessment of the available
evidence and public comments, EPA
concludes that a continuation of the practice
of adjusting PMio concentrations to standard
conditions of temperature and pressure is not
warranted or appropriate. Accordingly, this
requirement is not included in 40 CFR part
50, Appendix M and corresponding revisions
are made in 40 CFR 50.3. In addition, EPA
is also incorporating the proposed minor
modifications to 40 CFR part 50, Appendix
J in Appendix M.
B. 40 CFR Part 50, Appendix L - New
Reference Method for PA/2.5
1. Introduction. A new reference method
for the measurement of fine particles (as
PM2.5) in the ambient air has been developed
for the primary purpose of determining
attainment of the new PM2.5 standards. The
method is described in the new 40 CFR part
50, Appendix L, and joins the other reference
methods (or measurement principles)
-------
42
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
specified for other criteria pollutants in other
appendices to 40 CFR part 50.
In developing the proposed new reference
method for PM2.5, EPA staff consulted with
a number of individuals and groups in the
monitoring community, including instrument
manufacturers, academics, consultants, and
experts in State and local agencies. The
approach and key specifications were
submitted to the CASAC Technical
Subcommittee for Fine Particle Monitoring,
which held a public meeting to discuss the
proposed new reference method for PM2.5
and related monitoring issues on March 1,
1996. Comments on the proposed method
were provided orally and in writing by
interested parties. The Technical
Subcommittee indicated their overall
satisfaction with the method in a letter (Price,
1996) forwarded by CASAC to the
Administrator.
On December 13, 1996, EPA proposed the
new 40 CFR part 50, Appendix L at 61 FR
65676 for public comment. The proposal
described in detail the approach taken and the
design specifications and performance
requirements for the new PM2.5 sampler. On
January 14, 1997, EPA held a public hearing
on the proposed new 40 CFR part 50,
Appendix L and associated 40 CFR parts 53
and 58 requirements.
2. Basic reference method approach. In
addition to the primary purpose of the new
PM2.5 reference method (determining
attainment of the standards), EPA considered
a variety of possible secondary goals and
objectives that the PM2.5 reference method
might also fulfill. Subsequently, various
alternative PM2.5 measurement techniques
were evaluated. From this analysis, EPA
proposed to base its PM2.5 reference method
on a conventional type sampler that collects
24-hour integrated PM2.5 samples on a 47
mm Teflon filter that is subsequently
moisture and temperature conditioned and
analyzed gravimetrically. The sampler is a
low volume sampler that operates at a flow
rate of 1 cubic meter per hour, for a total
sample volume of 24 m3 for the specified 24-
hour sample collection period. The sampler is
easy to operate, operates over a wide range
of ambient conditions, produces a
measurement that is comparable to large sets
of previously collected PM data in existing
databases, and provides a physical sample
that can be further analyzed for chemical
composition.
3. Public comments and responses—a.
Sampler design. The EPA received many
general comments concerning the proposed
sampler design. Commenters suggested the
use of a different indicator, use of a different
size cut, inclusion of additional constituents
(e.g., acid aerosols, carbon, metals, and semi-
volatiles), and/or use of a multi-filter method.
Early in the development process, design
decisions were based on public input and the
advice of CASAC on these and other basic
design issues. Other factors affecting the
basic design of the method were the need for
historical continuity, high measurement
precision, and simplicity of operation, all in
response to current national monitoring
objectives and available resources. In
selecting the basic measurement approach,
substantial weight was given to maintaining
comparability to PM2.5 samplers, such as the
"dichotomous sampler," that were widely
used to obtain the data upon which the new
standards are based. Given this objective,
EPA concludes that the conventional PM
measurement approach is appropriate and will
provide PM2 5 measurements that are
comparable to the air quality data used in the
health studies that provide the basis for the
PM2.5 standards.
Although the sampler is conventional in
configuration, its design is much more
sophisticated than that of previous PM
samplers. This more sophisticated sampler,
together with improved manufacturing and
operational quality assurance, is necessary to
achieve the more stringent data quality
objectives established for PM2.5 monitoring
data. To meet precision requirements, the
critical mechanical components of the inlet,
particle size separator, downtube, and upper
portion of the filter holder are specified by
design. All other aspects of the sampler are
specified by performance-based
specifications.
Several commenters felt that the portions
of the sampler that were specified by design
would stifle further improvements and
innovations. Although the EPA specifies
methods by performance whenever possible,
for the PM2.5 reference method, development
of adequate performance specifications for
inlet aspiration and particle size
discrimination would have been a very
difficult, costly, lengthy, and problematic
process. Moreover, manufacturer testing of
proposed inlet and particle size discrimination
devices against such performance
specifications would require elaborate
specialized facilities and would be extremely
costly. For these reasons, the EPA believes
that specification of these critical components
by design is a prudent and very cost-effective
way to ensure good inter-manufacturer and
intra-manufacturer precision of the PM2.5
measurements. Therefore, these components
are specified by design, and other aspects of
the sampler are specified by performance, as
proposed. Innovations and improved samplers
or measurement methods are encouraged and
provided for as Class II and III equivalent
methods (see 40 CFR part 53).
b. Inlet and impactor design. Several
commenters addressed the inlet design, noting
that the inlet could allow entrance of
precipitation and possibly insects. In fact, the
inlet selected for the sampler has been used
effectively for many years to obtain many of
the PM2.5 measurements that formed the basis
of the epidemiological studies. While EPA
acknowledges that there have been some
reports of intrusion of precipitation, the
Agency believes the problem is relatively
minor. Nevertheless, a modification of the
inlet has been developed to further reduce the
possibility of precipitation (and possibly
small insects) reaching the sample filter to
damage the PM2 5 sample. Extensive wind
tunnel tests have shown no significant
compromise in the PM2.5 aspiration
performance of the modified inlet.
In addition, a new provision has been
added, in 40 CFR part 50, Appendix L,
section 7.3.8, to require that the sampling air
entrance of the inlet be at a height of 2 + 0.2
meters above the supporting surface to help
ensure homogeneous air samples when
collocated samplers of different types are
operated simultaneously.
Other commenters addressed the sharpness
of the size cut and how it is obtained, e.g.,
whether more than two stages should be used
and what size cut should be used for each
stage. These aspects were carefully
considered in selecting the sampler
configuration. The selection by EPA of the
previously used PMio inlet established the
size cut for the first stage, and the second
stage was designed to be simple, reliable, and
low in cost for user agencies. In EPA's
estimation, the advantages of this
configuration outweigh any modest advantage
that might have been gained by designing a
new inlet/separation configuration that would
further refine the cut points at each of two
(or more) stages.
A few commenters questioned whether the
inlet was wind speed dependent at high wind
speeds. The selected inlet has been shown to
perform well up to 24 km/hr with 10 |im
aerosols and is expected to perform well at
higher speeds with 2.5 jam aerosols. The EPA
again determined that the advantages of using
the selected inlet outweighed the possible
minor improvement in wind-speed
characteristics that might have been obtained
by a newly-designed inlet.
Some commenters felt that other types of
particle discrimination techniques such as
cyclones and virtual impactors, should be
allowed. Again, these alternatives were
evaluated previously and the specified inlet
and impactor were determined to best meet
the various objectives of the sampler.
However, EPA has provided for
considerations of other particle size selection
techniques or devices for approval if
incorporated into candidate equivalent
methods for PM2 5.
Several commenters addressed the
impactor design, noting that the impactor
should be changed to sharpen the size-cut
characteristic, to address concerns regarding
possible contamination and/or performance
loss due to impactor oil, and to improve ease
of access to service. To address the first
concern, the initial prototype impactor has
been modified slightly to sharpen its size-cut.
The current impactor is designed to lower
cost and to optimize cut sharpness, loading
capacity, manufacturing simplicity,
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
43
manufacturing quality control, serviceability,
and reliability. A report containing the
penetration efficiency of the impactor is
available in Docket No. A-95-54. With regard
to impactor oil concerns, the impactor oil
selected has a very low vapor pressure, and
testing has indicated no contamination of the
sample filters with impactor oil. The EPA
believes that the impactor design is as
accessible as possible, given the design
objectives. Some flexibility may be allowed
for manufacturers to develop improved
closure devices or other external
modifications. Proper maintenance will, of
course, be very important and will be stressed
in the associated operator instruction manuals
and in other training and guidance materials.
The EPA has been performing field and
laboratory tests that will provide detailed
guidance for all necessary preventive
maintenance. Proper installation procedures
for the oil and the impactor filter, as well as
all other maintenance requirements, will be
available in the quality assurance procedures
and guidance contained in the new section
2.12 of Appendix L to be added to EPA's
Quality Assurance Handbook for Air
Pollution Measurement Systems (EPA/600/R-
94/038b).
c. Anodized aluminum surface. All internal
surfaces exposed to sample air prior to the
filter are required to be anodized aluminum
as stated in 40 CFR part 50, Appendix L,
section 7.3.7. A few commenters expressed
concern that the anodized aluminum surfaces
in high volume PMio samplers have shown
substantial pitting, particularly in the venturi
flow control device. The anodized aluminum
surfaces are required in the PM2.5 sampler to
maintain comparability to previously used
samplers. The EPA believes that the much
lower flow rate in the PM2.5 sampler will
greatly reduce the pitting tendency, and the
active flow control in the PM2.5 sampler is
not dependent on the physical dimensions of
a critical orifice as it is in a venturi flow
control device.
d. Filter for PM^. 5 sample collection. The
proposed reference method called for the
sample to be collected on a 47 mm Teflon
filter. Many of the comments received on the
measurement method concerned the proposed
filter medium and its performance.
Commenters expressed concerns with the use
of Teflon filters and with the selection of a
single-filter method. Several commenters
recommended that alternative filter media be
allowed, in most cases to support speciation
and/or to allow the capture of all PM
components. Other comments noted potential
advantages of other media in operating
characteristics or chemistry requirements.
Operational concerns expressed about Teflon
filters included tearing, possible loss of
integrity, and high cost. Other concerns were
that Teflon is generally not conducive to
carbon analysis, and that Teflon filters may
not hold deposited PM. Many commenters
recommended use of a multi-filter sampler to
support chemical speciation in addition to
compliance determination.
To address some of these general concerns
about the performance of the specified filter
material, some minor refinements to the filter
specifications concerning the filter diameter
and the filter support ring have been made to
ensure proper performance of the filter in the
specified filter holder. Additional
clarifications have been made to the
maximum moisture pickup and the filter
weight stability requirements. Although
Teflon may preclude certain chemical
analyses (e.g., elemental and organic carbon),
the EPA believes that Teflon filter material
is the best overall choice to meet the
objectives of compliance monitoring and to
provide good measurement precision. Other
filter media are likely to provide reduced
gravimetric precision and preclude more
types of subsequent chemical analysis.
Additional or alternative samplers or filter
types can be considered as candidate
equivalent methods under 40 CFR part 53 and
can be used for non-compliance monitoring,
where necessary.
Compliance monitoring based on mass
concentration of PM2.5 is the primary
objective of the reference method. Multi-filter
capability would have substantially increased
the cost and complexity of the sampler.
However, multi-filter samplers can be
considered as candidate equivalent methods.
In addition, multi-filter samplers can be used
as special purpose monitors (SPMs) to
perform characterization studies, develop
control strategies, and conduct other special
studies as has been done previously for PMio.
In response to numerous comments
received on 40 CFR part 50, Appendix L and
on the provisions of 40 CFR part 58
regarding the need for chemical speciation,
the EPA is assigning a high priority to a
chemical speciation trends network through
section 105 of the Act grant allocation
program and will issue guidance describing
the monitoring methods and scenarios under
which speciation should be performed. The
program will incorporate additional PM2.5
samplers that allow for the simultaneous
collection of aerosols on multiple filter media.
The associated requirement for archiving
filters has been removed from 40 CFR part
50, Appendix L, section 10.17 and relocated
to 40 CFR part 58, Appendix A. This change
has been made because this is a supplemental
monitoring requirement and not an integral
part of the reference method for determining
compliance with the PM2.5 NAAQS.
Provisions of 40 CFR part 50, Appendix L
have been clarified to apply not only to a
single-sample sampler, but also to a
sequential-sample sampler, provided that all
specifications are met and no deviations,
modifications, or exceptions are made to the
inlet, downtube, impactor, or the upper
portion of the filter holder. Samplers that
have minor changes or modifications in these
components, have changes that alter the
aerosol's flow path, or contain other
significant deviations will be required to meet
the requirements of Class I equivalent
methods, in the amendments to 40 CFR part
53. Further, a provision has been added to
require that sequential sample filters stored in
a sequential sampler be adequately covered
and protected from contamination during
storage periods in the sampler.
A few commenters expressed concern
about who must carry out filter tests to
determine if they meet the filter
specifications. In response, the filter
specifications have been clarified to indicate
that filter manufacturers should generally
carry out most or all of the filter performance
tests in order to certify that their filters meet
the filter specifications for the PM2.5
reference method. In addition, EPA conducts
acceptance tests on filters procured for
NAMS/SLAMS networks prior to distribution
to State and local agencies.
Some commenters requested additional
information on the requirement that an ID
number be attached to each filter. Preliminary
information indicates that it is not practical
at this time for either filter manufacturers or
users to print an ID number directly on the
filter. However, EPA is continuing to pursue
this goal. In the meantime, alternative means,
such as attaching an appropriate ID number
to the filter's storage container, will be
necessary. Additional details and possible
alternative filter identification methods will
be provided in new section 2.12 of the
Quality Assurance Handbook for Air
Pollution Measurement Systems.
e. Filter handling/weighing/conditioning
requirements. Many commenters felt that the
filter handling requirements for collected
PM2 5 samples were too burdensome.
However, handling of the exposed filter
between retrieval from the sampler and
commencement of the conditioning period is
expected to be one of the most significant
sources of PM2.5 measurement variability.
Thus, EPA concludes that specific
requirements for this activity are necessary,
and this position was supported by several
commenters.
Some commenters felt that the samples
should be kept cold until analysis to prevent
volatile losses. In response to this concern,
the restriction on the maximum temperature
exposure for collected samples has been
reduced from 32 to 25° C, and a
recommendation has been added for sampler
operators to keep the samples as cool as
practical between retrieval from the sampler
and delivery to the conditioning environment.
Further, the length of time permitted between
retrieval of the filter and post-collection
weighing is increased from 10 to 30 days,
provided that the sample is maintained at 4°
C or less between retrieval and the start of
the conditioning period. The new section 2.12
of the Quality Assurance Handbook for Air
Pollution Measurement Systems will provide
guidance and techniques for keeping samples
-------
44
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
cool during this period and may suggest
devices to document maximum temperature
exposure of the sample.
Commenters also requested additional
specifications and guidance for field blanks.
The EPA will provide additional clarification
and detailed procedures and guidance
regarding field blanks in the new section 2.12
of the Quality Assurance Handbook for Air
Pollution Measurement Systems.
Other commenters felt that the filter
weighing requirements were too restrictive.
Because filter weighing is one of the most
significant sources of PM2.5 measurement
variability, specific requirements and
restrictions are deemed necessary. However,
in response to some of the concerns
expressed, the proposed requirement that both
pre- and post-weighings be carried out by the
same analyst has been reduced to a non-
mandatory recommendation. Detailed
recommendations and guidance on filter
weighing, based on information obtained in
current field tests, will be provided in the new
section 2.12 of the Quality Assurance
Handbook for Air Pollution Measurement
Systems.
Several commenters questioned the filter
conditioning requirements, with some
requesting a lower humidity range. Since
humidity can profoundly affect the weight of
the PM2.5 on the filter, EPA maintains that
filter conditioning requirements need to be
tight to control measurement variability and
to ensure satisfactory precision. But in
response to at least one of the concerns, the
filter conditioning humidity requirement has
been changed to allow conditioning at a
relative humidity within +5 RH percent of the
mean ambient humidity during sampling
(down to a minimum of 20 RH percent) for
samples collected at average ambient
humidities lower than 30 percent. The EPA
will provide further details on filter
conditioning controls in the new section 2.12
of the Quality Assurance Handbook for Air
Pollution Measurement Systems.
f Sampler performance requirements.
Several commenters addressed sampler
performance requirements, including sampler
flow control specifications, filter temperature
control, sampler performance under extreme
conditions, and data reporting. In response to
concerns that various sampler flow control
specifications are too tight, EPA contends that
good flow control is necessary to maintain
uniform sampling, to ensure correct particle
size discrimination, and to control
measurement variability. Sampler
manufacturers have been able to meet the
specified flow control requirements, and field
studies to date confirm that prototype
samplers are able to meet these flow control
requirements.
In response to comments about the ambient
temperature plus 3° C filter temperature
control requirement, EPA believes that fairly
tight control of the sample filter temperature
is necessary to minimize losses of semi-
volatile components over a wide temperature
range, and tight temperature control has been
strongly recommended by the CASAC.
Monitoring of the filter temperature
difference from ambient temperature is
necessary to verify that the sampler filter
temperature control is functioning properly.
Testing to date indicates that the proposed 3°
C (above ambient temperature) limit is
somewhat difficult to meet; however, a 5° C
limit can be reasonably met. Therefore, the
filter temperature control requirement has
been relaxed slightly from 3° C to not more
than 5° C above the concurrent ambient
temperature. Ambient and filter temperature
sensors will require periodic calibration or
verification of accuracy. In response to a
frequent comment, the method has been
clarified to indicate that exceedance of the
filter temperature difference limit would not
necessarily invalidate the sample.
In response to concerns about the
performance of the sampler under extreme
weather conditions (e.g., high or low
temperatures, low pressures, high winds, high
or low humidity, fog, dust storms), the EPA
has established sampler specifications that are
intended to cover reasonably normal
environmental conditions at about 95 percent
of expected monitoring sites. Qualification
test requirements in 40 CFR part 53 address
most, if not all, of these operational
requirements. Specification of the sampler
performance for sites with extreme
environmental conditions would substantially
raise the cost of the sampler for other users,
most of whom do not require the extra
capability. Users requiring operation of
samplers under extreme conditions are
encouraged to develop supplemental
specifications for modified samplers to cover
those specific conditions. Sampler
manufacturers have indicated a commitment
to respond to the need for modified samplers
for such extreme conditions.
Although concerns were expressed that the
amount of data required to be reported from
each sampler is excessive, EPA stresses that
only a portion of the data collected by the
sampler needs to be reported to AIRS. These
limited data reporting requirements (i.e.,
ambient and filter temperature, barometric
pressure, sample volume, variation in sample
run flow rate) are important to establish or
verify the reliability and confidence of the
PM2.5 measurements and to aid in utilization
of those data. The substantial amount of
additional data generated by the sampler are
of use to the site operator to provide
confirmation of a given sample's validity, and
to aid in troubleshooting should outlier
measurements appear in the monitoring data.
A variety of current electronic devices and
systems may be used to acquire and handle
the data, and these devices can easily
accommodate the amount of data required to
be reported, as well as the additional, optional
data. Printers, modem connections, and
alternative data output connections or devices
are not precluded.
4. Additional changes. Additional
clarifying changes have also been made
throughout 40 CFR part 50, Appendix L,
based on comments received or recently
obtained field test information. In 40 CFR
part 50, Appendix L, section 3.1, the lower
concentration of the method has been revised
from 1 to 2 |ig/m3, based on the results of
field blanks associated with available field
test data. In 40 CFR part 50, Appendix L,
section 3.3, the sample period specification
has been augmented to clarify that a
measured PM2.5 concentration for a sample
period less than 23 hours that is greater than
the NAAQS level(s) is to be considered a
valid measurement for comparison to the
NAAQS, even though not valid for other
purposes. Sections 4 (Accuracy) and 5
(Precision) have been revised to properly
reflect associated changes to the data quality
and method performance assessment
requirements set forth in 40 CFR part 58,
Appendix A.
A provision has been added in 40 CFR part
50, Appendix L, section 7.4.17 to require
sampler manufacturers to make available
computer software to input sampler output
data and translate the data into a standard
spreadsheet format (since no specific format
is specified for output of the sample data
acquired by the sampler).
The requirements for the sampler to display
current flow rate, temperature, filter
temperature, and barometric pressure readings
have been changed to require updating of
these readings at least every 30 seconds. This
change is based on operational experience of
prototype samplers in 40 CFR part 50,
Appendix L, section 7.4.5.1, and will make
it easier for the operators to perform status
checks and calibrations. In 40 CFR part 50,
Appendix L, section 7.4.8.1, the requirements
for the ambient temperature sensor have been
changed to specify an external sensor with a
passive sun shield, to provide better
uniformity in the ambient temperature
measurements among different types of
reference method samplers. The reference
method has also been clarified to indicate that
PM2.5 samples for which the sampler reported
an out-of-specification (FLAG) occurrence
during or after the sample period are not
necessarily invalid, and that such samples
should be reviewed by a quality assurance
officer (40 CFR part 50, Appendix L, section
10.12). Finally, a new reference has been
added in section 13 of the Act to provide
applicable standards for meteorological
measurements and measurement systems.
5. Decision on 40 CFR part 50, Appendix
L. After fully considering the public
comments on the proposed new reference
method for PM2 5, EPA has concluded that
the proposed design and performance
specifications for the reference sampler, with
the modifications discussed in this unit, will
achieve the design objectives set forth in the
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
45
proposal and outlined above. Therefore, EPA
is adopting the sampler and other method
requirements specified in 40 CFR part 50,
Appendix L as the reference method for
measuring PM2.5 in the ambient air.
Since proposal, a series of field tests have
been performed using prototype samplers
manufactured in accordance with the
proposed design and performance
specifications. The results of these field tests
confirm that the prototype samplers perform
in accordance with design expectations.
Operational experience gained through these
field tests did, however, identify the need for
minor modifications as discussed above in
this unit. In addition, EPA made other
modifications to the proposed design and
performance specification in response to
public comment as discussed above. As part
of this process, EPA performed laboratory
tests to ensure that the modifications achieved
the intended objective.
While the results of these field tests and
laboratory tests were largely confirmatory in
nature and did not indicate a need to alter the
basic design and performance specifications,
they did identify areas that needed further
refinement. Given that these tests were
performed, by necessity, during and after the
close of the public comment period and
because the results were not available for
placement in the docket until late in the
rulemaking process, EPA is announcing, in a
separate Federal Register notice being
signed today, a supplemental comment period
for the limited purpose of taking comments
on these field and laboratory test results.
VH. Effective Date of the Revised PM
Standards and Applicability of the Current
PM10 Standards
In summary, the primary and secondary
NAAQS for PM have been revised by
establishing annual and 24-hour PM2.5
standards; and by changing the form of the
existing 24-hour PMio standards. The existing
PMio annual standards have been retained.
Section 50.3 (reference conditions) of 40 CFR
part 50 has been revised to remove the
adjustment of measured PMio concentrations
to standard conditions of temperature and
pressure with respect to the revised PM
standards. (Although EPA is retaining the
current annual PMio standards, the revision of
40 CFR 50.3 potentially may affect the
effective stringency of the annual standards.)
A new Appendix M has been added to 40
CFR part 50 that reflects the revision of 40
CFR 50.3. A new Appendix N to 40 CFR part
50 has been added to reflect the forms of the
PM2.5 and revised PMio standards. Finally, a
new Appendix L to 40 CFR part 50 has been
added that specifies the reference method for
measuring PM2.5 in the ambient air.
The revised PM NAAQS, the revisions to
40 CFR 50.3, and the new Appendices M, N,
and L to 40 CFR part 50 will become
effective September 16, 1997. Inherent in the
establishment of this revised set of PM
standards and related provisions is the
revocation of the current set of PMio
standards and associated provisions. To
provide for an effective transition from the
existing PM standards to the revised PM
standards —in light of the need to establish
PM2.5 monitoring networks, designate areas,
and develop control strategies for PM2.5—the
Administrator has determined that the
effective date of the revocation of the current
set of PMio standards and associated
provisions should be delayed so that the
existing standards and associated provisions
will continue to apply for an interim period.
The duration of the interim period would
depend on whether the area in question has
attained the current PMio standards, as
described below in this unit.
First, section 172(e) of the Act provides
that, if the Administrator relaxes a national
primary ambient air quality standard, she
shall, within 12 months after the relaxation,
promulgate requirements applicable to all
areas that have not attained that standard as
of the date of the relaxation. Those
requirements shall provide for controls that
are not less stringent than the controls
applicable to areas designated nonattainment
before such relaxation. Although the set of
revised PM standards, viewed as a whole, is
more stringent than the set of current PM
standards, it appears that the shift from the
current PMio standards to the revised PMio
standards, viewed in and of itself, represents
a relaxation of the current PMio standards. As
a result, section 172(e) of the Act requires
EPA to issue a rule within 12 months to apply
implementation requirements no less stringent
than the currently applicable requirements for
those areas that have not yet attained the
current PMio standard(s) by today's
promulgation. However, the Act does not
specifically provide how to ensure that States
with current PMio problems should maintain
the necessary public health protection in the
interim between promulgation of a relaxed
standard and issuance of a rule under section
172(e) of the Act. For that reason, EPA
believes that it is both necessary and
appropriate to defer the effective date of the
revocation of the current PMio standards, for
areas that have not attained those standards,
until EPA issues the rule called for by section
172(e)oftheAct.
Second, since it will take many years for
States to identify PM problems under the
revised standards and to develop effective
means for addressing those problems, EPA
believes it is necessary for even those areas
that have already attained the current PMio
standards (and hence are not subject to the
terms of section 172(e) of the Act) to
continue their current PMio implementation
efforts for the purpose of protecting public
health in the transition to implementation of
the revised standards.
In order to deal with both of these
categories of areas—those that are not
attaining the current PMio standards and
those that are in attainment of the current
PMio standards—EPA is taking a two-
pronged approach towards deferral of the
effective date of the revocation of the current
PMio standards. For those areas that are not
attaining the current PMio standards at the
time of the promulgation of the revised PMio
standards, the current standards will continue
to apply until EPA has completed its
rulemaking under section 172(e) of the Act
to prevent backsliding in those areas. This
will assure that no backsliding can occur in
the interim period between the promulgation
of the revised standards and the completion
of the rulemaking under section 172(e) of the
Act. For those areas that are attaining the
current PMio standards at the time of
promulgation of the revised PMio standards,
the existing PMio standards will continue to
apply until the areas have an approved SIP
that includes any control measures that had
been adopted and implemented at the State
level to meet the current PMio NAAQS and
have an approved section 110 SIP for
purposes of implementing the revised PM
standards. If an area has already received
approval of a PMio SIP embodying all of the
measures that had been adopted and
implemented at the State level, no further Part
D submission or approval would be
necessary. If an area has already submitted
such measures, EPA would need to take
action to approve them. Finally, if an area has
not yet submitted such measures to EPA for
inclusion in the SIP, the area would need to
submit them and EPA would need to approve
them. This submission and approval would
serve to satisfy both the area's remaining
subpart D obligations and, in part, its new
obligations under section 110(a)(l) of the Act
regarding the implementation of the revised
PM NAAQS. EPA emphasizes that it is not
requiring an approval of a modeled
attainment demonstration for the current
PMio NAAQS, only an approval of the
control measures that had in fact been
adopted and implemented and that, therefore,
were responsible for the area's attainment of
the current PMio standards.
The existing definition of reference
conditions and 40 CFR part 50, Appendices
J and K will remain in force as long as the
current PMio standards apply to an area.
Additional policies and guidance for assuring
an effective transition will be set forth in
future EPA guidance, policies, and/or rules.
VHL Regulatory and Environmental
Impact Analyses
As discussed in Unit IV of this preamble,
the Clean Air Act and judicial decisions make
clear that the economic and technological
feasibility of attaining ambient standards are
not to be considered in setting NAAQS,
although such factors may be considered in
the development of State plans to implement
the standards. Accordingly, although, as
described below, a Regulatory Impact
Analysis (RIA) has been prepared, neither the
-------
46
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
RIA nor the associated contractor reports
have been considered in issuing this final
rule.
A. Executive Order 12866
Under Executive Order 12866, 58 FR
51735 (October 4, 1993), the Agency must
determine whether a regulatory action is
"significant" and, therefore, subject to Office
of Management and Budget (OMB) review
and other requirements of the Executive
Order. The order defines "significant
regulatory action" as any regulatory action
that is likely to result in a rule that may:
(1) Have an annual effect on the economy
of $100 million or more or adversely affect
in a material way the economy, a sector of
the economy, productivity, competition, jobs,
the environment, public health or safety, or
State, local, or tribal governments or
communities.
(2) Create a serious inconsistency or
otherwise interfere with an action taken or
planned by another Agency.
(3) Materially alter the budgetary impact of
entitlements, grants, user fees, or loan
programs or the rights and obligations of
recipients thereof.
(4) Raise novel legal or policy issues
arising out of legal mandates, the President's
priorities, or the principles set forth in the
Executive Order.
In view of its important policy
implications, this action has been judged to
be a "significant regulatory action" within
the meaning of the Executive Order. As a
result, under section 6 of the Executive Order,
EPA has prepared an RIA, entitled
"Regulatory Impact Analysis for Particulate
Matter and Ozone National Ambient Air
Quality Standards and Proposed Regional
Haze Rule (July 1997)." This RIA assesses
the costs, economic impacts, and benefits
associated with potential State
implementation strategies for attaining the
PM and O3 NAAQS and the proposed
Regional Haze Rule. Changes made in
response to OMB suggestions or
recommendations will be documented in the
public docket and made available for public
inspection at EPA's Air and Radiation Docket
Information Center (Docket No. A-95-58).
The RIA will be publicly available in hard
copy by contacting the U.S. Environmental
Protection Agency Library at the address
under ' 'Availability of Related Information''
and in electronic form as discussed above in
"Electronic Availability."
B. Regulatory Flexibility Analysis
The Regulatory Flexibility Act (RFA), 5
U.S.C. 601 et seq., provides that, whenever
an agency is required to publish a general
notice of rulemaking for a proposal, the
agency must prepare an initial regulatory
flexibility analysis for the proposal unless the
head of the agency certifies that the rule will
not, if promulgated, have a significant
economic impact on a substantial number of
small entities (section 605(b)). The EPA
certified the proposed NAAQS rule based on
its conclusion that the rule would not
establish requirements applicable to small
entities and therefore would not have a
significant economic impact on small entities
within the meaning of the RFA. See 61 FR
65638, 65668 (PM proposal) and 61 FR
65716, 65764 (ozone proposal), both
published December 13, 1996. Accordingly,
the Agency did not prepare an initial
regulatory flexibility analysis for the
proposal, but it did conduct a more general
analysis of the potential impact on small
entities of possible State strategies for
implementing any new or revised NAAQS.
At the heart of EPA's certification of the
proposed NAAQS rule was the Agency's
interpretation of the word "impact" as used
in the RFA. Is the "impact" to be analyzed
under the RFA a rule's impact on the small
entities that will be subject to the rule's
requirements, or the rule's impact on small
entities in general, whether or not they will
be subject to the rule? In the case of NAAQS
rules, the question arises because of the
congressionally designed mixture of Federal
and State responsibilities in setting and
implementing the NAAQS.
As EPA explained in the proposal, NAAQS
rules establish air quality standards that States
are primarily responsible for meeting. Under
section 110 and Part D of Title I of the Act,
every State develops a State Implementation
Plan (SIP) containing the control measures
that will achieve a newly promulgated
NAAQS. States have broad discretion in the
choice of control measures. As the U.S.
Supreme Court noted in Train v. NRDC, 421
U.S. 60(1975), 95 S.Ct. 1470:
[P]rimary [NAAQS] deal with the quality of
outdoor air and are fixed on a nationwide basis at
a level which the agency determines will protect
the public health. It is the attainment and
maintenance of these standards which section
110(a)(2)(A) requires that State plans provide. In
complying with this requirement, a State's plan
must include "emission limitations" which are
regulations of the composition of substances
emitted into the ambient air from such sources as
power plants, service stations and the like. They are
the specific rules to which operators of pollution
sources are subject and which, if enforced, should
result in ambient air which meets the national
standards.
The Agency is plainly charged by the Act with
the responsibility for setting the national ambient
air standards. Just as plainly, it is relegated to a
secondary role in the process of determining and
enforcing the specific, source-by-source emission
limitations which are necessary if the national
standards are to be met. Under 110(a)(2), the
Agency is required to approve a State plan which
provides for the timely attainment and maintenance
of the ambient air standards, and which also
satisfies that sections other general requirements.
The Act gives the agency no authority to question
the wisdom of a state's choices of emission
limitations if they are part of a plan which satisfies
the standards of 110(a)(2) and the Agency may
devise and promulgate a plan of its own only if
the State fails to submit an implementation plan
which satisfies those standards. Section 110(c).
421 U.S. 60 at 78-79 (emphasis in original).
In short, NAAQS rules themselves do not
establish any control requirements applicable
to small entities. State rules implementing the
NAAQS may establish such requirements and
the extent to which they do depends primarily
on each State's strategy for meeting the
NAAQS.95
To determine the proper interpretation of
impact under the RFA, EPA considered the
RFA's stated purpose, its requirements for
regulatory flexibility analyses, its legislative
history, the amendments made by the Small
Business Regulatory Enforcement Fairness
Act of 1996 (SBREFA) (Pub. L. 104-121),
and caselaw. The EPA concluded that all of
these traditional tools of statutory
construction point in one direction—that an
agency is required to assess the impact of a
rule on the small entities that will be subject
to the rule's requirements, because the
purpose of a regulatory flexibility analysis is
to consider ways of easing or even waiving
a rule's requirements as they will apply to
small entities, consistent with the statute
authorizing the rule. That purpose cannot be
served in the case of the rules like the
NAAQS that do not have requirements that
apply to small entities.
More specifically, EPA noted that its
interpretation of "impact" flows from the
express purpose of the RFA itself. As the
RFA's "Findings and Purposes" section
(Pub. L. 96-354, section 2) makes clear,
Congress enacted the RFA in 1980 out of
concern that agencies were writing one-size-
fits-all regulations that in fact did not fit the
size and resources of small entities. Congress
noted that it is generally easier for big
businesses to comply with regulations, and
that small businesses are therefore at a
competitive disadvantage in complying with
uniform rules. Congress also noted that small
entities' relative contribution to the problem
a rule is supposed to solve may not warrant
applying the same requirements to large and
small entities alike. In the RFA itself,
Congress therefore stated:
It is the purpose of this Act to establish as a
principle of regulatory issuance that agencies shall
endeavor, consistent with the objectives of the rule
and of applicable statutes, to fit regulatory and
informational requirements to the scale of the
businesses, organizations, and governmental
jurisdictions subject to regulation.
(Pub. L. 96-354, section 2(b))
The EPA further noted that the RFA
sections governing initial and final regulatory
95 It is worth noting that Federal rules that apply
nationally also play a role in reducing emissions governed
by NAAQS. For instance, EPA rules under Title II of the
Act require reductions in ozone-forming emissions from
on and off-road vehicles and the fuels that power them.
When EPA issues such rules, it conducts the analysis
required under the RFA. For example, EPA performed
regulatory flexibility analyses for the reformulated gasoline
rule issued under section 211(k) of the Act. See 59 FR
7716, February 16, 1994.
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
47
flexibility analyses reflect this statement of
purpose. Sections 603 and 604 of the RFA
require that initial and final regulatory
flexibility analyses identify the types and
estimate the numbers of small entities "to
which the proposed will apply" (sections
603(b)(3) and 604(a)(3) of the RFA).
Similarly, they require a description of the
"projected reporting, recordkeeping, and
other compliance requirements of the
proposal, including an estimate of the classes
of small entities which will be subject to the
requirement" (sections 603(b)(4) and
604(a)(4)). At the core of the analyses is the
requirement that agencies identify and
consider "significant regulatory alternatives"
that would "accomplish the stated objectives
of applicable statutes and which minimize
any significant economic impact of the
proposal on small entities" (sections 603(c)
and 604(a)(5)). Among the types of
alternatives agencies are to consider are the
establishment of different "compliance or
reporting requirements or timetables" for
small entities and the exemption of small
entities "from coverage of the rule, or any
part" of the rule (section 603(c)(l) and (4)
of the RFA). The RFA thus makes clear that
regulatory flexibility analyses are to focus on
how to minimize rule requirements on small
entities.
As EPA further explained, since regulatory
flexibility analyses are not required for a rule
that will not have a "significant economic
impact on a substantial number of small
entities", it makes sense to interpret
"impact" in light of the requirements for
such analyses. Regulatory flexibility analyses,
as described in this unit, are to consider how
a rule will apply to small entities and how
its requirements may be minimized with
respect to small entities. In this context,
"impact" is appropriately interpreted to
mean the impact of a rule on the small entities
subject to the rule's requirements.
The Agency cited two Federal court cases
in support of its interpretation. In Mid-Tex
Elec. Co-op v. FERC, 111, F.2d 327, 342
(B.C. Cir. 1985), petitioners claimed that the
RFA required an agency to analyze the
effects of a rule on small entities that were
not regulated by the rule but might be
indirectly impacted by it. Petitioners noted
that the Small Business Administration (SBA)
also interpreted the RFA to require analysis
of a rule's impact on small entities not
regulated by the rule, and argued that the
court should defer to the SBA's position in
light of its compliance monitoring role under
the RFA. After reviewing the RFA's
"Findings and Purposes" section, its
legislative history, and its requirements for
regulatory flexibility analyses, the Mid-Tex
court rejected petitioners' interpretation. As
the court explained:
The problem Congress stated it discerned was
the high cost to small entities of compliance with
uniform regulations, and the remedy Congress
fashioned—careful consideration of those costs in
regulatory flexibility analyses—is accordingly
limited to small entities subject to the proposed
regulation * * *. [W]e conclude that an agency may
properly certify that no regulatory flexibility
analysis is necessary when it determines that the
rule will not have a significant economic impact
on a substantial number of small entities that are
subject to the requirements of the rule.
Id. at 342. Notably, Congress let this
interpretation stand when it recently amended
the RFA in enacting SBREFA.
The EPA also cited a recent case affirming
the Mid-Tex court's interpretation. In United
Distribution Companies v. FERC, 88 F.3d
1105, 1170 (B.C. Cir. 1996), the court noted
that the Mid-Tex court:
* * * conducted an extensive analysis of RFA
provisions governing when a regulatory flexibility
analysis is required and concluded that no analysis
is necessary when an agency determines ' 'that the
rule will not have a significant economic impact
on a substantial number of small entities that are
subject to the requirements of the rule".
Id., citing and quoting Mid-Tex (emphasis
added by United Bistribution court). The
Agency went on to explain that given the
Federal/State partnership for attaining healthy
air, the proposed NAAQS, if adopted, would
not establish any requirements applicable to
small entities. Instead, any new or revised
standard would establish levels of air quality
that States would be primarily responsible for
achieving by adopting plans containing
specific control measures for that purpose.
The proposed NAAQS rule was thus not
susceptible to regulatory flexibility analysis
as prescribed by the amended RFA. Since it
would establish no requirements applicable to
small entities, it afforded no opportunity for
EPA to fashion for small entities less
burdensome compliance or reporting
requirements or timetables, or exemptions
from all or part of the rule. For these reasons,
EPA certified that the proposal' 'will not, if
promulgated, have a significant economic
impact on a substantial number of small
entities," within the meaning of the RFA.
Because EPA was not required to prepare an
initial regulatory flexibility analysis for the
rule, it was also not required to convene a
Small Business Advocacy Review Panel for
the rule under section 609(b) of the RFA, as
added by SBREFA.
Notwithstanding its certification of the
proposal, EPA recognized that the proposed
NAAQS, if adopted, would begin a process
of State implementation that could eventually
lead to small entities having to comply with
new or different control measures, depending
on the implementation plans developed by the
States. EPA also recognized that the Act does
not allow EPA to dictate or second-guess how
States should exercise their discretion in
regulating to attain any new or revised
NAAQS. Under those circumstances, EPA
concluded that the best way to take account
of small entity concerns regarding any new
or revised NAAQS was to work with small
entity representatives and States to provide
information and guidance on how States
could address small entity concerns when
they write their implementation plans.
In line with this approach, as part of RIA
it prepared for the proposed NAAQS, EPA
analyzed how hypothetical State plans for
implementing the proposal might affect small
entities. The analysis was necessarily
speculative and limited, since it depended on
projections about what States might do
several years in the future and did not take
into account any new strategies that might be
developed and recommended by the FACA
subcommittee formed to help devise potential
strategies for implementing a new or revised
NAAQS (see discussion of RIA and FACA
process in this document). Nevertheless, the
analysis provided as much information on
potential small entity impacts as was
reasonably available at the time of the
proposal.
The Agency also took steps to ensure that
small entities' voices were heard in the
NAAQS rulemaking itself. With Jere Glover,
Chief Counsel for Advocacy of the SBA,
EPA convened outreach meetings modeled on
the SBREFA panel process to solicit and
convey small entities' concerns with the
proposed NAAQS. Two meetings were held
as part of that process, on January 7 and
February 28, 1997, with a total attendance of
41 representatives of small businesses, small
governments, and small nonprofit
organizations. Both meetings were attended
by representatives of SBA and OMB, as well
as of EPA. The key concerns raised by small
entities at those meetings related to the
scientific foundation of the proposed NAAQS
and the potential cost of implementing it, the
same concerns raised by other industry
commenters on the proposal. The Agency
produced a report on the meetings to ensure
that small entity concerns were part of the
rulemaking record when EPA made its final
decision on the proposal.
In light of States' pivotal role in NAAQS
implementation, EPA also undertook a
number of additional activities to assist and
encourage the States to be sensitive to small
entity impacts as they implement any new or
revised NAAQS. With the SBA, EPA began
an interagency panel process to collect advice
and recommendations from small entity
representatives on how States could lessen
any impacts on small entities. The EPA plans
to issue materials in two phases to help States
develop their implementation plans. In view
of States' discretion in implementing the
NAAQS, these materials will mostly take the
form of guidance, which is not subject to the
RFA's requirement for initial regulatory
flexibility analysis. (Under section 603 of the
RFA, that requirement applies only to binding
rules that are required to undergo notice and
comment rulemaking procedures.) But
regardless of the form such materials take,
EPA is employing panel procedures to ensure
that small entities have an opportunity to raise
-------
48
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
any concerns prior to the materials being
issued in draft form.
To supplement the input the Agency
receives from the ongoing FACA process
(described previously in this document), EPA
also added more small entity representatives
to the Subcommittee on implementation of
any new or revised NAAQS. These
representatives have formed a small entity
caucus to develop and bring to the
Subcommittee a focused approach to small
entity issues. These new Subcommittee
members are also part of the group in the
aforementioned panel process. By means of
these various processes, EPA hopes to
promote the consideration of small entity
concerns and advice throughout the NAAQS
implementation process.
In response to the proposal, a number of
commenters questioned EPA's decision to
certify that the proposed NAAQS will not
have a significant impact on a substantial
number of small entities. Some commenters
disagreed with EPA's view that the proposed
NAAQS would not establish regulatory
requirements applicable to small entities.
These commenters argued that a number of
control requirements applicable to small
entities would automatically result from
promulgation of the proposed NAAQS, such
as new reasonable further progress, SIP and
Federal Implementation Plan (FIP)
requirements. Other commenters stated that it
is possible for EPA to assess the impacts of
the NAAQS revision on small entities and
that, to a limited extent, EPA has already
done so. Further, a number of commenters
argued that EPA has a legal obligation under
the RFA, as amended by SBREFA, to choose
a NAAQS alternative that minimizes the
impact on small entities. Some commenters
questioned EPA's interpretations of the Mid-
Tex and United Distribution cases. In
addition, other commenters stated that EPA's
position regarding the NAAQS and the RFA
is inconsistent with its past practice and the
legislative history of the RFA. Finally, a few
commenters noted that the panel process EPA
conducted for the proposed NAAQS did not
satisfy the requirements of SBREFA.
EPA disagrees that promulgation of the
NAAQS will automatically result in control
requirements applicable to small entities that
EPA can and must analyze under the RFA.
As noted previously in this unit, a NAAQS
rule only establishes a standard of air quality
that other provisions of the Act call on States
(or in case of State inaction, the Federal
government) to achieve by adopting
implementation plans containing specific
control measures for that purpose. Following
promulgation of a new or revised NAAQS,
section 110 of the Act requires States and
EPA to engage in a designation process to
determine what areas within each State's
borders are attaining or not attaining the
NAAQS. Under section 110 and Parts C and
D of Title I of the Act, States then conduct
a planning process to develop and adopt their
SIPS. Depending on an area's designation for
the particular NAAQS, these and other Title
I provisions of the Act require a State's SIP
to contain certain control programs in
addition to the control measures that the State
decides are also needed to attain and maintain
the NAAQS.
The fact that the Act requires SIPs to
contain certain control programs under certain
circumstances does not mean that EPA either
can or must conduct a regulatory flexibility
analysis of a rule establishing a NAAQS. Just
from the standpoint of feasibility, EPA cannot
know which areas will be subject to what
mandatory SIP programs until after the
designation process is completed. Beyond
that, any mandatory SIP programs are still
implemented by the States, and States have
considerable discretion in how they
implement them. For instance, the reasonable
further progress requirement under section
172 of the Act leaves States broad discretion
to determine the rate of progress and the
control measures to achieve that progress.96
As a result, EPA cannot be certain where and
how any mandatory programs will be
implemented with respect to small (or large)
entities. Much less can EPA know about how
States will exercise their discretion to develop
additional controls needed to attain and
maintain the NAAQS.
Even if EPA could know exactly how any
mandatory SIP programs would apply to
small entities, the purpose of the RFA is not
served by attempting a regulatory flexibility
analysis of State implementation of those
programs. As explained previously in this
unit, the RFA and the caselaw interpreting it
clearly establish that the purpose of the RFA
is to promote Federal agency efforts to tailor
a rule's requirements to the scale of the small
entities that will be subject to it. That purpose
cannot be served in the case of a NAAQS rule
since the rule does not establish requirements
applicable to small entities. In promulgating
a NAAQS, the only choice before EPA
concerns the level of the standard, not its
implementation. While mandatory SIP
programs may ultimately follow from
promulgation of the NAAQS, there is nothing
EPA can do in setting the NAAQS to tailor
those programs as they apply to small entities.
Whether and how the programs will apply in
particular nonattainment areas is beyond the
scope of the NAAQS rulemaking and, indeed,
beyond EPA's reach in any rulemaking to the
extent the applicability and terms of the
programs are prescribed by statute.97
MThe SIP requirements of subpart 4 of Part D of Title
I of the Act apply to SIPs for areas designated as not
attaining NAAQS for PMio. Those requirements will not
apply to SIPs to implement the PM2.5 NAAQS. Further,
to the extent SIPs for areas in nonattainment with the
applicable PMio NAAQS remain subject to subpart 4
requirements, there will be no incremental change in the
impact on sources regulated by the States' SIPs pursuant
to those requirements as a result of this promulgation.
97 If and when the Agency issues any rules addressing
State implementation of any statutorily required actions,
Moreover, any mandatory SIP programs are
supplemented by discretionary State controls
that EPA has no power to tailor under the
PsFA or the Act (see Train v. NRDC, quoted
previously in this unit).
The commenters' suggestions for
minimizing the potential impact of the
NAAQS rule on small entities run afoul of
both the PsFA and the Act. Some suggested
that EPA set a less stringent standard (or no
standard at all in the case of PM2.5) to reduce
the chance that small entities would become
subject to new or tighter SIP requirements.
Others suggested that EPA require States to
exempt small entities from new or tighter SIP
requirements. However, as explained
previously in this document, the PsFA neither
requires nor authorizes EPA to set a less
stringent NAAQS than the applicable Clean
Air Act provisions allow in order to reduce
potential small entity impacts. Indeed, the
RFA provides that any means of providing
regulatory flexibility to small entities be
consistent with the statute authorizing the
rule. Moreover, even if EPA set a less
stringent standard, States could still exercise
their discretion to obtain any needed emission
reductions from small entities. As the
Supreme Court in Train v. NRDC made clear,
EPA has no authority to forbid States from
obtaining reductions from any particular
category of stationary sources, including
small entities. See also, Virginia v. EPA, No.
108 F.3d 1397, 1408 (D.C. Cir. 1997),
quoting Union Electric v. EPA, 427 U.S. 246,
269 (1976) ("section 110 left to the states the
power to determine which sources would be
burdened by regulations and to what extent").
EPA's approval of SIPs for the new or
revised NAAQS also will not establish new
requirements, but will instead simply approve
requirements that a State is already imposing.
And again, EPA does not have authority to
disapprove a State's plan except to the extent
that the plan fails to demonstrate attainment
and maintenance of the NAAQS as required
by Title I of the Clean Air Act. In cases
where EPA promulgates a FIP, EPA might
establish control requirements applicable to
small entities, and in such a circumstance,
EPA would conduct the analyses required by
the RFA.
Some commenters argued that under the
RFA as amended by SBREFA, EPA now has
an obligation to choose the alternative that
minimizes the impact on small entities when
setting the NAAQS. As indicated previously
in this unit, EPA disagrees with the
commenters' argument for the reasons stated
in this document's discussion of the Agency's
authority to consider costs and other factors
not related to public health in setting and
revising primary NAAQS. In a nutshell, both
the text and legislative history of the RFA
make clear that the RFA does not override the
substantive provisions of the statute
EPA would analyze and address the impact of those rules
on small entities as appropriate under the RFA.
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
49
authorizing the rule, but only requires
agencies to identify and consider ways of
minimizing the economic impact on small
entities subject to the rule in a manner
consistent with the authorizing statute.
Some commenters disagreed with EPA's
interpretation of the Mid-Tex and United
Distribution cases. In particular, these
commenters noted that in those cases the
relevant regulatory agency, Federal Energy
Regulatory Commission (FERC), wholly
lacked jurisdiction to regulate the small
entities at issue. According to these
commenters, EPA does have the ability and
jurisdiction to regulate small entities in the
case of the NAAQS, and therefore EPA's
reliance on Mid-Tex and United Distribution
is misplaced.
The commenters' attempt to distinguish the
FERC cases from the NAAQS rulemaking
wholly overlooks the courts' reasoning, which
in fact fully supports EPA's certification of
the proposed NAAQS. As described
previously in this unit, the Mid-Tex court
exhaustively reviewed the relevant sections of
the RFA and its legislative history. Its
analysis revealed that Congress passed the
RFA out of concern with one-size-fits-all
regulations and fashioned a remedy limited to
regulations that apply to small entities. This
principle is fully applicable to the NAAQS,
which creates no rule requirements that apply
to small entities.
The fact that FERC had no regulatory
authority over the small entities indirectly
affected by its rules played no essential role
in the court's rationale. FERC could (and
apparently did in the Mid-Tex rulemaking)
estimate the potential indirect impact of its
rules on small entities. Presumably, FERC
could have also mitigated any indirect impact
by changing some aspect of the rule (or else
the small entities would have had no
incentive to sue the agency). The court
nevertheless found it unnecessary for FERC
to do either, based on its reading of the RFA
as limited to analysis of a rule's impact on
the small entities subject to the rule's
requirements. In reaching its decision, the
court noted that requiring agencies to
"consider every indirect effect that any
regulation might have on small businesses *
* * is a very broad and ambitious agenda, *
* * that Congress is unlikely to have
embarked on * * * without airing the
matter." Mid-Tex, 773 F.d. at 343.
The commenters also overstate EPA's
regulatory authority over small entities with
respect to the regulation of criteria pollutants.
Various provisions of the Clean Air Act
authorize EPA to regulate various types of
sources at the Federal level to accomplish
specified goals. However, EPA's authority to
more generally regulate sources, including
small entities, in the manner of SIPs is limited
to instances of State default of SIP
responsibilities. When that occurs, EPA may
issue a FIP containing specific control
measures, and to the extent a proposed FIP
would establish control measures applicable
to small entities, EPA would analyze the
small entity impact of those measures as
required by the RFA. In 1994, for example,
EPA prepared an initial regulatory flexibility
analysis when it proposed a FIP for Los
Angeles. See 59 FR 23264 (May 5, 1994).
As noted previously in this unit, Congress
let the Mid-Tex interpretation stand when it
recently amended the RFA in enacting
SBREFA. If it had disagreed with the court's
decision, it would have revised the relevant
statutory provisions or otherwise indicated its
disagreement when it enacted SBREFA.
Instead, Congress actually reinforced the
Mid-Tex court's interpretation of the RFA in
enacting section 212(a) of SBREFA. That
section requires that an agency issue a "small
entity compliance guide" for "each rule * *
* for which an agency is required to prepare
a final regulatory flexibility analysis under
section 604" of the RFA. The guide is "to
assist small entities in complying with the
rule" by "explaining] the actions a small
entity is required to take to comply" with the
rule (section 212(a) of SBREFA). Obviously,
it makes no sense to prepare a small entity
compliance guide for a rule that does not
apply to small entities. Thus SBREFA stands
as further confirmation that Congress
intended the RFA to address only rules that
establish requirements small entities must
meet. Since SBREFA's passage, the United
Distribution court has affirmed the Mid-Tex
court's interpretation.
Some commenters noted that EPA's
informal panel process did not comply with
the requirements of SBREFA. The EPA did
not convene a SBREFA panel because such
a panel is not required for rules like the
NAAQS that do not apply to small entities.
Under the RFA as amended by SBREFA,
since the Agency certified the proposal, it was
not required to convene a panel for it.
Nevertheless, EPA conducted the voluntary
panel process described previously in this
unit, as well as other voluntary small business
outreach efforts. The process could not
comply with the analytical requirements of
the RFA for the reasons given in this unit.
However, it could and did ensure that EPA
heard directly from small entities about the
NAAQS proposals.
A few commenters stated that EPA's view
of the NAAQS and the RFA is inconsistent
with EPA's past positions regarding the RFA
and NAAQS revisions. Some commenters
also cited the RIA for the proposed NAAQS
and noted that this analysis demonstrates
EPA's ability to estimate the impact of the
NAAQS on small entities, thereby
undercutting EPA's argument that it is not
able to perform a regulatory flexibility
analysis when setting the NAAQS.
Past Federal Register documents make
clear that the nature of the NAAQS makes a
regulatory flexibility analysis inapplicable to
NAAQS rulemakings. For instance, in 1984,
EPA stated that a "NAAQS for NOX by itself
has no direct impact on small entities.
However, it forces each State to design and
implement control strategies for areas not in
attainment." See 49 FR 6866, 6876 (February
23, 1984); see also, 50 FR 37484, 37499
(September 13, 1985); 50 FR 25532, 25542
(June 19, 1985) (NAAQS forNO2 do not
impact small entities directly). EPA stated
again in 1987 that the NAAQS "themselves
do not contain emission limits or other
pollution controls. Rather, such controls are
contained in state implementation plans." See
52 FR 24634, 24654 (July 1, 1987).
EPA has typically performed an analysis to
assess, to the extent practicable, the potential
impact of retaining or revising the NAAQS
on small entities, depending on possible State
strategies for implementing the NAAQS.
These analyses have provided as much insight
into the potential small entity impacts of
implementing revised NAAQS as could be
provided at the NAAQS rulemaking stage. In
some instances, these preliminary analyses
were described as ' 'regulatory flexibility
analysfes]" or as analyses "pursuant to this
[Regulatory Flexibility] Act." See, e.g., 52
FR 24634, 24654 (July 1, 1987); 50 FR
37484, 37499 (September 13, 1985).
However, these analyses were based on
hypothetical State control strategies, and EPA
made the point on various occasions that any
conclusions to be drawn from such analyses
were speculative, given that the NAAQS
themselves do not impose requirements on
small entities. Although these past analyses
reflected the Agency's best efforts to evaluate
potential impacts, they were not regulatory
flexibility analyses containing the necessary
elements required by the RFA. These
analyses, for example, did not describe the
reporting, recordkeeping, and other
compliance requirements of the proposed
NAAQS rules that would apply to small
entities, since the NAAQS rules did not apply
to small entities. Nor did they determine how
the proposed NAAQS rules could be eased or
waived for small entities. Such an analysis is
not possible in the case of the NAAQS. To
the extent EPA labeled these analyses
regulatory flexibility analyses in the past, that
label was inappropriate. EPA's current
practice is to describe such an analysis more
accurately as a general analysis of the
potential cost impacts on small entities. See,
e.g., 61 FR 65638, 65669, 65747 (December
13, 1996) (current O3 and PM NAAQS
proposals).98 EPA's analytical approach to
98 As commenters pointed out, the RIA for the proposed
PM NAAQS does state that "[t]he screening analysis *
* * provides enough information for an initial regulatory
flexibility analysis (RFA) if such an analysis were to be
done." That statement was mistaken and was not made
in the RIA for the proposed ozone NAAQS. While both
RIAs attempted to gauge the potential impact on small
entities of State implementation of the proposed NAAQS,
neither could or did identify any specific control or
information requirements contained in the NAAQS rule
that would apply to small entities. Indeed, both RIAs made
clear that the impact being analyzed was that of potential
Continued
-------
50
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
small entity impacts of the NAAQS has thus
remained consistent over time.
One commenter noted that the legislative
history of the RFA suggests that the RFA was
intended to apply to the NAAQS. As noted
previously in this unit, EPA's reading of both
the RFA and SBREFA, based on the language
of the statute as amended and its legislative
histories and applicable caselaw, is that the
RFA requirements at issue do not apply to the
NAAQS. The legislative history cited by the
commenter does not change this conclusion.
In fact, the statement by Senator Culver on
which the commenter relies does not indicate
that the NAAQS should be subject to
regulatory flexibility analyses. Rather,
Senator Culver uses the NAAQS as an
example of the type of standard that agencies
would not change as a result of the RFA.
According to Senator Culver, section 606 of
the RFA "succinctly states that this bill does
not alter the substantive standard contained in
underlying statutes which defines the
agency's mandate." 126 Cong. Rec. S 21455
(August 6, 1980) daily ed. After citing section
109 of the Act, Senator Culver goes on to
describe EPA's bubble policy (which
addresses the limits on emissions from a
particular facility) as the type of flexible
regulation that agencies should consider, once
EPA has set a NAAQS. "The important point
for purposes of this discussion is that the
'bubble concept,' a type of flexible
regulation, in no manner altered the basic
statutory substantive standard of the EPA *
* *. No regulatory flexibility analysis alters
the substantive standard otherwise applicable
by law to agency action." Id. Thus, contrary
to the suggestion of the commenter, Senator
Culver's statement actually confirms that the
time to consider regulatory flexibility is when
regulations applicable to sources are being
established, not when a NAAQS itself is
being set.
Under section 604 of the RFA, whenever
an agency promulgates a final rule under
section 553 of the Administrative Procedure
Act, after being required by that section or
any other law to publish a general notice of
proposed rulemaking (NPRM), the agency is
required to prepare a final regulatory
flexibility analysis. RFA section 605(b)
provides, however, that section 603 (re initial
regulatory flexibility analyses) and section
604 do not apply if the agency certifies that
the rule will not have a significant economic
impact on a substantial number of small
entities and publishes such certification at the
time of publication of the NPRM or at the
time of the final rule.
As noted above, EPA certified this final
rule at the time of the NPRM. After
considering the public comments on the
State measures to attain the NAAQS, and that such an
analysis was inherently speculative and uncertain. Thus,
the RIAs actually confirm EPA's statement in the
preambles for the proposed NAAQS that conducting a
complete regulatory flexibility analysis is not feasible for
rules setting or revising a NAAQS.
certification, EPA continues to believe that
this final rule will not have a significant
economic impact on a substantial number of
small entities for the reasons explained above
and that it therefore appropriately certified the
rule. Further, as required by the Clean Air
Act, EPA is promulgating this final rule under
section 307(d) of the Clean Air Act. For all
the foregoing reasons, EPA has not prepared
a final regulatory flexibility analysis for the
rule. The Agency has nonetheless analyzed in
the final RIA for the rule the potential impact
on small entities of hypothetical State plans
for implementing the NAAQS. The Agency
also plans to issue guidance to the States on
reducing the potential impact on small entities
of implementing the NAAQS.
C. Impact on Reporting Requirements
There are no reporting requirements
directly associated with the finalization of
ambient air quality standards under section
109 of the Act (42 U.S.C. 7400). There are,
however, reporting requirements associated
with related sections of the Act, particularly
sections 107, 110, 160, and 317 (42 U.S.C.
7407, 7410,7460, and 7617).
In EPA's final revisions to the air quality
surveillance requirements (40 CFR part 58)
for PM, the associated RIA addresses the
Paperwork Reduction Act requirements
through an Information Collection Request.
D. Unfunded Mandates Reform Act
Title II of the Unfunded Mandates Reform
Act of 1995 (UMRA), Pub. L. 104^1,
establishes requirements for Federal agencies
to assess the effects of certain regulatory
actions on State, local, and tribal governments
and the private sector. Under section 202 of
UMRA, EPA generally must prepare a
written statement, including a cost-benefit
analysis, for proposed and final rules with
"Federal mandates" that may result in
expenditures by State, local, and tribal
governments, in the aggregate, or by the
private sector, of $100 million or more in any
1 year. This requirement does not apply if
EPA is prohibited by law from considering
section 202 of UMRA estimates and analyses
in adopting the rule in question. Before
promulgating a final rule for which a written
statement is needed, section 205 of UMRA
generally requires EPA to identify and
consider a reasonable number of regulatory
alternatives and adopt the least costly, most
cost-effective, or least burdensome alternative
that achieves the objectives of the rule. These
requirements do not apply when they are
inconsistent with applicable law. Moreover,
section 205 of UMRA allows EPA to adopt
an alternative other than the least costly, most
cost-effective, or least burdensome alternative
if the Administrator publishes with the final
rule an explanation of why that alternative
was not adopted. Before EPA establishes any
regulatory requirements that may significantly
or uniquely affect small governments,
including tribal governments, it must have
developed under section 203 of UMRA a
small government agency plan. The plan must
provide for notifying potentially affected
small governments, enabling officials of
affected small governments to have
meaningful and timely input in the
development of EPA regulatory proposals
with significant Federal intergovernmental
mandates, and informing, educating, and
advising small governments on compliance
with the regulatory requirements. Section 204
of UMRA requires each agency to develop
"an effective process to permit elected
officers of state, local and tribal governments
* * * to provide meaningful and timely
input" in the development of regulatory
proposals containing a significant Federal
intergovernmental mandate."
The EPA has determined that the
provisions of sections 202 and 205 of UMRA
do not apply to this decision. "Unless
otherwise prohibited by law," EPA is to
prepare a written statement under section 202
of UMRA that is to contain assessments and
estimates of the costs and benefits of a rule
containing a Federal mandate. Congress
clarified that' 'unless otherwise prohibited by
law" referred to whether an agency was
prohibited from considering the information
in the rulemaking process, not to whether an
agency was prohibited from collecting the
information. The Conference Report on
UMRA states, "This section [202] does not
require the preparation of any estimate or
analysis if the agency is prohibited by law
from considering the estimate or analysis in
adopting the rule." 141 Cong. Rec. H3063
(daily ed. March 13, 1995). Because the
Clean Air Act prohibits EPA, when setting
the NAAQS, from considering the types of
estimates and assessments described in
section 202 of UMRA, UMRA does not
require EPA to prepare a written statement
under section 202.10° The requirements in
section 205 of UMRA do not apply because
those requirements only apply to rules "for
99 As noted in unit VIII.B., a NAAQS rule only
establishes a standard of air quality that other provisions
of the Act call on States (or in the case of State inaction,
the Federal government) to achieve by adopting
implementation plans containing specific control measures
for the purpose. Thus, it is questionable whether the
NAAQS itself imposes an enforceable duty and thus
whether it is a significant Federal mandate within the
meaning of UMRA. EPA need not and does not reach this
issue in this document. For the reasons given in this unit,
even if the NAAQS were determined to be a significant
Federal mandate, EPA does not have any obligations under
sections 202 and 205 of UMRA, and EPA has met any
obligations it would have under section 204 of UMRA.
100In addition to the estimates and assessments
described in section 202 of UMRA, written statements are
also to include an identification of the Federal law under
which the rule is promulgated (section 202(a)(l) of
UMRA) and a description of outreach efforts under section
204 of UMRA (section 202(a)(5) of UMRA). Although
these requirements do not apply here because a written
statement is not required under section 202 of UMRA, this
preamble identifies the Federal law under which this rule
is being promulgated and a written statement describing
EPA's outreach efforts with State, local, and tribal
governments will be placed in the docket.
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
51
which a written statement is required under
section 202 * * *."
The EPA has determined that the
provisions of section 203 of UMRA do not
apply to this decision. Section 203 of UMRA
only requires the development of a small
government agency plan for requirements
with which small governments might have to
comply. Since setting the NAAQS does not
establish requirements with which small
governments might have to comply, section
203 of UMRA does not apply. The EPA
acknowledges, however, that any
corresponding revisions to associated SIP
requirements and air quality surveillance
requirements, 40 CFR parts 51 and 58,
respectively, might result in such
effects.Accordingly, EPA did address
unfunded mandates when it proposed
revisions to 40 CFR part 58, and will do so,
as appropriate, when it proposes any revision
to 40 CFR part 51.
With regard to the outreach described in
section 204 of UMRA, EPA did follow a
process for providing elected officials with an
opportunity for meaningful and timely input
into the proposed NAAQS revisions, although
EPA did not describe this process in the
proposal. The EPA conducted a series of pre-
proposal outreach meetings with State and
local officials and their representatives that
permitted these officials to provide
meaningful and timely input on issues related
to the NAAQS and the monitoring issues
associated with them. Beginning in January,
1996, EPA briefed State and local air
pollution control officials at national meetings
with State and Territorial Air Pollution
Program Administrators (STAPPA) /
Association of Local Air Pollution Control
Officials (ALAPCO) in Washington, DC,
North Carolina, Chicago, and Nevada. The
EPA also held briefings for the Washington,
DC representatives of several State and local
organizations, including National Conference
of State Legislators, U.S. Conference of
Mayors, National Governors Association,
National League of Cities, and STAPPA/
ALAPCO. EPA also held separate briefings
and discussions with State and local officials
at meetings set up by the National Governors
Association, the U.S. Conference of Mayors
and the Council of State Governments. The
EPA also conducted in-depth briefings at
each EPA regional office and regional staff
also had several meetings and discussions
with their State counterparts about the
standards. The efforts described in this
paragraph of this preamble, which provided
elected officials with opportunity for
meaningful and timely input into the
proposed NAAQS revisions, met any
requirements imposed by section 204 of
UMRA. The docket will contain a written
statement describing these outreach efforts,
including a summary of the comments and
concerns presented by State, local, and tribal
governments and a summary of EPA's
evaluation of those comments and concerns.
Several commenters disagreed with EPA
that sections 202, 203, and 205 of UMRA do
not apply to this decision. These commenters
argued that EPA is not prohibited from
considering costs in setting NAAQS under
the Clean Air Act and applicable judicial
decisions. Some commenters also expressed
the view that there is no conflict between
UMRA and the Clean Air Act with regard to
the NAAQS. These commenters argued that
UMRA and the NAAQS can be harmonized
by reading UMRA as an information
gathering statute and that EPA should
therefore perform the analyses required by
UMRA, regardless of whether costs may be
considered. Finally, at least one commenter
argued that in past NAAQS reviews, EPA did
not dispute its UMRA obligations.
As discussed more fully in Unit IV. of this
preamble, EPA is prohibited from considering
cost in setting the NAAQS. Given that fact
(as noted in Unit IV. of this preamble),
sections 202 and 205 of UMRA do not
apply.101 As the Conference Report clarifies,
UMRA itself states that the section 202
estimates and analyses are not required in
cases such as the NAAQS, where an agency
is prohibited by law from considering section
202 estimates and analyses. Reading UMRA
in the manner suggested by the commenters
would effectively read this provision out of
UMRA; UMRA contains an exception for
rules like the NAAQS, it must be given
effect.
With regard to EPA's position regarding
UMRA in previous NAAQS review exercises,
EPA simply made plain in those situations
that because it did not plan on revising the
NAAQS, it determined, without further
review, that sections 202, 203, and 205 of
UMRA did not apply. EPA thus stated that:
Because the Administrator has decided not to
revise the existing primary NAAQS for SCh, this
action will not impose any new expenditures on
governments or on the private sector, or establish
any new regulatory requirements affecting small
governments. Accordingly, EPA has determined
that the provisions of sections 202, 203 and 205
do not apply to this final decision.
61 FR 25566, 25577, May 22, 1996; see also
61 FR 52852, 52856, October 8, 1996 (Same
statement for NO2 NAAQS). As this
statement makes clear, EPA only determined
that sections 202, 203, and 205 of UMRA did
not apply to the NAAQS when EPA fails to
101 One commenter argued that in reviewing the SO2
NAAQS, EPA determined that it need not revise the S02
NAAQS, but could instead pursue an alternative regulatory
program under other authority. This commenter argued that
EPA has similar flexibility in reviewing the PM and Ozone
NAAQS, and thus UMRA requires EPA to identify the
least burdensome alternative (such as retaining the current
NAAQS) as part of that process. As discussed more fully
in Unit IV. of this preamble, EPA does not agree that it
has flexibility to choose such an alternative; nor does EPA
agree with the commenter's characterization of the action
it took in deciding not to revise the SO2 NAAQS. In fact,
in deciding not to revise the SO2 NAAQS, EPA
determined, for reasons independent of section 303 of the
Clean Air Act that a NAAQS revision was not warranted.
See 61 FR 25566, 25575 (May 22, 1996).
revise the standard. Having made that
determination, EPA had no reason to catalog
additional bases for finding UMRA
inapplicable. Nothing in that statement was
intended to preclude EPA, or precludes EPA,
from concluding for other reasons (such as
those discussed in this unit) that UMRA also
does not apply when EPA in fact revises an
applicable NAAQS.
E. Environmental Justice
Executive Order 12848 (58 FR 7629,
February 11, 1994) requires that each Federal
agency make achieving environmental justice
part of its mission by identifying and
addressing, as appropriate, disproportionately
high and adverse human health or
environmental effects of its programs,
policies, and activities on minorities and low-
income populations. These requirements have
been addressed to the extent practicable in the
RIA cited in this unit.
F. Submission to Congress and the
Comptroller General
Under 5 U.S.C. 801(a)(l)(A), as added by
the Small Business Regulatory Enforcement
Fairness Act of 1996 (SBREFA), EPA
submitted a report containing this rule and
other required information to the U.S. Senate,
the U.S. House of Representatives, and the
Comptroller General of the United States
prior to publication of the rule in this issue
of the Federal Register. This rule is a
"major rule" for purposes of SBREFA.
IX. Response to Petition for Administrator
Browner's Rescusal
On March 13, 1997, the Washington Legal
Foundation (WLF), filed a petition with EPA
asking that I, Carol Browner, disqualify
myself in rulemaking regarding the NAAQS
for PM and ozone. The petition claims that
my public statements indicate a "clear and
convincing showing" that I had "already
decided to revise the NAAQS for PM and
ozone" and that I therefore "could not give
meaningful consideration" to comments
adverse to the proposed rule. On May 12,
1997, EPA's General Counsel, Jonathan Z.
Cannon, sent a letter to WLF regarding the
petition. This letter and the WLF petition
were then placed in the dockets for the
proposed ozone and PM standards pending
"consideration and final response in
connection with the Agency's final actions."
Contrary to WLF's assertions, I have
maintained an open mind throughout these
proceedings, and have based today's
decisions on the rulemaking record—
including consideration of comments opposed
to the proposal. The law does not require the
Administrator of EPA to disqualify herself
merely for expressing views on a proposed
regulation; in fact, it is part of my
responsibility to engage in the public debate
on the proposals. Moreover, the assertions in
WLF's petition do not accurately represent
my views. The petition takes quotes out of
-------
52
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
context and repeatedly misinterprets my
statements. For example, WLF quotes a
statement that I made at the Children's
Environmental Health Network Research
Conference as an indication that I had
"prejudged the issue." However, my
statement that "I will not be swayed" did not
refer to adopting the NAAQS as proposed.
Instead, as is clear from reviewing the entire
speech, I was addressing my broader concern
about children's health and the range of EPA
standards affecting children's health. I also
appeared at several congressional hearings
and testified before members of Congress,
some of whom were strongly opposed to the
proposals. At those hearings, I explained the
basis for the proposals and put forward the
reasons why I concluded the proposals were
appropriate, given the information before me
at the time. At the same time, I made clear
that I took very seriously my obligation to
keep an open mind, and to consider fully and
fairly all significant comments that the
Agency received. For these reasons and
others, as set forth in Mr. Cannon's May 12,
1997 response to WLF, which I adopt in full,
I have decided not to recuse myself from any
aspect of considering revisions to the
NAAQS for ozone and PM. Accordingly, I
am hereby denying WLF's petition.
X. References
(1) Abbey, D.E.; Mills, P.K.; Petersen, F.F.;
Beeson, W.L. (1991) Long-term ambient
concentrations of total suspended particulates and
oxidants as related to incidence of chronic disease
in California Seventh-Day Adventists.
Environmental Health Perspectives 94:43-50.
(2) Abt Associates (1996a) Proposed
Methodology for PM Risk Analyses in Selected
Cities (Draft). Prepared by Abt Associates, Inc.,
Hampden Square, Suite 500, 4800 Montgomery
Lane, Bethesda, MD 20814-5341. February 12,
1996.
(3) Abt Associates (1996b) A Particulate Matter
Risk Analysis for Philadelphia and Los Angeles.
Prepared by Abt Associates for US EPA, OAQPS,
Hampden Square, Suite 500, 4800 Montgomery
Lane, Bethesda, MD 20814-5341. July 3, 1996
(Updated November 1996).
(4) Abt Associates (1997a) Revision of Mortality
Incidence Estimates Based on Pope et al. (1995)
in the Abt Particulate Matter Risk Assessment
Report. Prepared by Abt Associates, Inc., Hampden
Square, Suite 500, 4800 Montgomery Lane,
Bethesda, MD 29814-5341. June 5, 1997.
(5) Abt Associates (1997b) Revision of Mortality
Incidence Estimates Based on Pope et al. (1995)
in the Abt Particulate Matter Risk Assessment
Report. Prepared by Abt Associates, Inc., Hampden
Square, Suite 500, 4800 Montgomery Lane,
Bethesda, MD 29814-5341. June 6, 1997.
(6) American Industrial Health Council. (1997)
Comments on EPA's Proposed Rule for National
Ambient Air Quality Standards for Particulate
Matter. Docket No. A-95-54, IV-D-2340. March
3, 1997.
(7) American Petroleum Institute. (1997)
Comments of the American Petroleum Institute on
the Proposed National Ambient Air Quality
Standards for Particulate Matter. Docket No. A—
95-54, IV-D-2247. March 12, 1997.
(8) Borja-Aburto, V.H.; Loomis, D.P.;
Bangdiwala, S.I.; Shy, C.M.; Rascon-Pacheco, R.A.
(1997) Ozone, suspended parti culates, and daily
mortality in Mexico City. American Journal of
Epidemiology 145:258-268.
(9) Braun-Fahrlander, C.; Ackermann-Liebrich,
U.; Schwartz, J.; Gnehm, H.P.; Rutishauser, M.;
Wanner, H.U. (1992) Air pollution and respiratory
symptoms in preschool children. American Review
of Respiratory Disease 145:42-47.
(10) Brown, K. (1997) Predictability of personal
exposure to airborne particulate matter from
ambient concentrations in four communities. In:
Comments of the American Petroleum Institute on
the proposed national ambient air quality standards
for particulate matter (Appendix F). Docket No. A—
95-54, IV-D-2247.
(11) California Environmental Protection
Agency. (1997) Comments on the proposed
national ambient air quality standards. Docket No.
A-95-54, IV-D-2251. March 11, 1997.
(12) Chestnut, L.G.; Rowe, R.D. (1990)
Preservation values for visibility in the national
parks. Washington, DC: U.S. Environmental
Protection Agency.
(13) Chestnut, L.G.; Dennis, R.L.; Latimer, D.A.
(1994) Economic benefits of improvements in
visibility: acid rain provisions of the 1990 Clean
Air Act Amendments. Proceedings of the aerosols
and atmospheric optics: radiative balance and
visual air quality. Air & Waste Management
International Specialty Conference, pp. 791-802.
(14) Cooper, J.A.; Tawney, C.W. (1996)
Comments on Office of Air Quality Planning and
Standards recommended PMi.s concentrations
equivalent to current PMio standards. In:
Comments of the American Iron and Steel Institute.
Docket No. A-95-54, IV-D-2242. March 12,
1997.
(15) Costa, D.L; Dreher, K.L. (1997)
Bioavailable transition metals in particulate matter
mediate cardiopulmonary injury in healthy and
compromised animal models. Environmental
Health Perspectives (In press)
(16) Damberg, R.; Polkowsky, B. (1996)
Methodology for Estimating Visibility
Improvements Due to Reductions in Fine Particle
Concentrations. Memorandum to the docket for the
PM National Ambient Air Quality Standards
review. September 1996.
(17) Davis, J.M.; Sacks, J.; Saltzman, N.; Smith,
R.L.; Styer, P. (1996) Airborne particulate matter
and daily mortality in Birmingham, Alabama.
Technical Report #55, National Institute of
Statistical Sciences, P.O. Box 14162, Research
Triangle Park, NC 27709.
(18) Delfmo RJ, Murphy-Moulton AM, Burnett
RT, Brook JR, Becklake MR. (1997) Effects of air
pollution on emergency room visits for respiratory
illnesses in Montreal, Quebec. American Journal of
Respiratory and Critical Care Medicine 155:568-
576
(19) Dickey, J.H. (1997) Letter to President Bill
Clinton. Docket No. A-95-54, IV-D-2301. March
12, 1997.
(20) Dockery, D.W.; Cunningham, J.;
Damokosh, A.I.; Neas, L.M.; Spengler, J.D.;
Koutrakis, P.; Ware, J.H.; Raizenne, M.; Speizer,
F.E. (1996) Health effects of acid aerosols on North
American children: respiratory symptoms.
Environmental Health Perspectives 104:500-505.
(21) Dockery, D.W.; Pope, C.A., III; Wu, W.;
Spengler, J.D.; Ware, J.H.; Fay, M.E.; Ferris, E.G.,
Jr.; Speizer, F.E. (1993) An association between air
pollution and mortality in six U.S. cities. New
England Journal of Medicine 329:1753-1759.
(22) Dockery, D.W.; Schwartz, J.; Spengler, J.D.
(1992) Air pollution and daily mortality:
associations with particulates and acid aerosols.
Environmental Research 59:362-373.
(23) Dockery, D.W.; Speizer, F.E.; Stram, D.O.;
Ware, J.H.; Spengler, J.D.; Ferris, E.G., Jr. (1989)
Effects of inhalable particles on respiratory health
of children. American Review of Respiratory
Disease 139:587-594
(24) Fairley, D. (1990) The relationship of daily
mortality to suspended particulates in Santa Clara
County, 1980-86. Environmental Health
Perspectives 89:159-168.
(25) Fitz-Simons, T.; Mintz, D.; Wayland, M.
(1996) Proposed methodology for predicting PM2.5
from PMio values to assess the impact of
alternative forms and levels of the PM National
Ambient Air Quality Standards. Document
transmitted to members of the Clean Air Act
Scientific Advisory Committee on June 26, 1996.
(26) Freas, W.P. (1997) Memorandum to the
files: Ratios of Particulate Matter Annual
Arithmetic Mean to Annual Geometric Mean
Concentrations. Docket No. A-95-54, IV-B-2.
May 29, 1997.
(27) Friedlander, S.K. (1982) Letter from
Sheldon K. Friedlander, Chair, Clean Air Science
Advisory Committee to Administrator Anne. M.
Gorsuch. Clean Air Scientific Advisory Committee
Review and Closure of the Office of Air Quality
Planning and Standards Staff Paper for Particulate
Matter. January 29, 1982.
(28) Gamble, J.F.; Lewis, R.J. (1996) Health and
respirable particulate (PMio) air pollution: a causal
or statistical association? Environmental Health
Perspectives 104:838-850.
(29) Grand Canyon Visibility Transport
Commission (1996) Recommendations for
improving Western vistas. Report of the Grand
Canyon Visibility Transport Commission to the
U.S. Environmental Protection Agency. June 1996.
(30) Godleski, J.J.; Sioutas, C.; Katler, M.;
Catalano, P.; Koutrakis, P. (1996) Death from
inhalation of concentrated ambient air particles in
animal models of pulmonary disease. Proceedings
of the Second Annual Colloquium on Particulate
Air Pollution and Human Health. Lee, J.; Phalen,
R.;eds. 4:136-143.
(31) Gordian, M.E.; Ozkaynak, H,; Xue, J.;
Morris, S.S.; Spengler, J.D. (1996) Particulate air
pollution and respiratory disease in Anchorage,
Alaska. Environmental Health Perspectives
104:290-297.
(32) Hefflin, B.J.; Jalaludin, B.; McClure, E.;
Cobb, N.; Johnson, C.A.; Jecha, L; Etzel, R.A.
(1994) Surveillance for dust storms and respiratory
diseases in Washington State, 1991. Archives of
Environmental Health 49:170-174.
(33) Health Effects Institute. (1997) Particulate
Air Pollution and Daily Mortality: Analyses of the
Effects of Weather and Multiple Air Pollutants
(The Phase I.B Report of the Particle Epidemiology
Evaluation Project). Health Effects Institute,
Cambridge, MA.
(34) Health Effects Institute. (1995) Particulate
Air Pollution and Daily Mortality: Replication and
Valication of Selected Studies (The Phase LA
Report of the Particle Epidemiology Evaluation
Project). Health Effects Institute, Cambridge, MA.
(35) Hill, A.B. (1965) The environment and
disease: associations or causation? Proceedings of
the Royal Society of Medicine 58:295-300.
(36) Ito, K.; Kinney, P.L.; Thurston, G.D. (1995)
Variations in PMio concentrations within two
metropolitan areas and their implications for health
effects analyses. Inhalation Toxicology 7:735-745.
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
53
(37) Katsouyanni, K.; Touloumi, G.; Spix, C.;
Schwartz, J.; Balducci, F.; Medina, S.; Rossi, G.;
Wojtyniak, B.; Sunyer, J.; Bacharova, L.; Schouten,
IP.; Ponka, A.; Anderson, H.R. (1997) Short-term
effects of ambient sulphur dioxide and particulate
matter on mortality in 12 European cities: results
from the APHEA project. British Medical Journal
(In press)
(38) Killingsworth, C.R.; Alessandrini, F.;
Krishna Murthy, G.G.; Catalano, P.J.; Paulauskis,
ID.; Godleski, II (1997) Inflammation,
chemokine expression, and death in monocrotaline-
treated rats following fuel oil fly ash inhalation.
Inhalation Toxicology (In press)
(39) Koman, P.O. (1996) Memorandum to the
files regarding PMi.s Air Quality Values from PM
Health Studies. September 30, 1996.
(40) Koman, P.O. (1997) Supplemental
Information Regarding PMi.5 Air Quality Values
from Key Short-term Fine Particle Exposure
Studies. lune 30, 1997.
(41) Li, Y.; Roth, H.D. (1995) Daily mortality
analysis by using different regression models in
Philadelphia County, 1973-1990. Inhalation
Toxicology 7:45-58.
(42) Lipfert, F.W., Wyzga, R.E. (1997) Air
pollution and mortality: the implications of
uncertainties in regression modeling and exposure
measurement. Journal of Air & Waste Management
47:517-523.
(43) Lipfert, F.W.; Urch, R.B. (1997) Personal
exposures of asthmatics to air pollution in Toronto
and implications for epidemiological studies.
Submitted as attachment to comments on the
proposed National Ambient Air Quality Standard
for particulate matter by R.E. Wyzga, Docket No.
A-95-54, IV-D-2672. March 11, 1997.
(44) Lipfert, F.W. (1995) Estimating air
pollution-mortality risks from cross-sectional
studies: prospective vs. ecologic study designs. In:
Particulate matter: health and regulatory issues:
proceedings of an international specialty
conference; April; Pittsburgh, PA. Pittsburgh, PA:
Air & Waste Management Association; p. 78-102.
(Air & Waste Management Association Publication
VIP-49)
(45) Lipfert, F.W.; Wyzga, R.E. (1995) Air
pollution and mortality: issues and uncertainties.
Journal of Air & Waste Management Association
45:949-966.
(46) Lippmann, M.; Shy, C.; Stolwijk, I;
Speizer, F. (1996) Letter to Administrator Carol M.
Browner. Supplement to the Closure Letter from
the Clean Air Scientific Advisory Committee.
March 20, 1996.
(47) Lippman, M. (1986) Letter from Morton
Lippmann, Chair, Clean Air Act Scientific
Advisory Committee, to Administrator Lee M.
Thomas.
(48) Lipsett, M.; Hurley, S.; Ostro, B. (1997) Air
Pollution and emergency room visits for asthma in
Santa Clara County, California. Environmental
Health Perspectives 105(2):216-222
(49) Lyon, I.L.; Mori, M.; Gao, R. (1995) Is
there a causal association between excess mortality
and exposure to PMio air pollution? Additional
analyses by location, year, season and cause of
death. Inhalation Toxicology 7:603-614.
(50) Martin, A.E.; Bradley, W.H. (1960)
Mortality, fog and atmospheric pollution: an
investigation during the winter of 1958-1959.
Monthly Bulletin of the Minister of Health, Public
Health Laboratory Service (GB) 19:56-73.
(51) Ministry of Health - London- Her Majesty's
Stationary Office (1954) Reports on Public Health
and Medical Subjects No. 95. Mortality and
Morbidity During the London Fog of December
1952.
(52) Moolgavkar, S.H.; Luebeck, E.G.;
Anderson, E.L. (1997) Air pollution and hospital
admissions for respiratory causes in Minneapolis-
St. Paul and Birmingham. Epidemiology 8:364—
370.
(53) Moolgavkar, S.H.; Luebeck, E.G. (1996)
Particulate air pollution and mortality: a critical
review of the evidence. Epidemiology 7:420-428.
(54) Moolgavkar, S.H.; Luebeck, E.G.; Hall,
T.A.; Anderson, E.L. (1995a) Air pollution and
daily mortality in Philadelphia. Epidemiology
6:476-484.
(55) Moolgavkar, S.H.; Luebeck, E.G.; Hall,
T.A.; Anderson, E.L. (1995b) Particulate air
pollution, sulfur dioxide, and daily mortality: a
reanalysis of the Steubenville data. Inhalation
Toxicology 7:35^14.
(56) Neas L.M.; Dockery D.W.; Koutrakis P.;
Tollerud, D.I; Speizer, F.E. (1995) The association
of ambient air pollution with twice daily peak
expiratory flow rate measurements in children.
American Journal of Epidemiology 141:111-122.
(57) National Mining Association. (1997)
Comments of the National Mining Association on
the proposed revisions to the national ambient air
quality standards for particulate matter. Docket No.
A-95-54, IV-D-2158. March 12, 1997.
(58) Ozkaynak, H.; Spengler, ID. (1996) The
role of outdoor particulate matter in assessing total
human exposure. In: Particles in Our Air:
Concentrations and Health Effects. Harvard
University Press. Wilson, R; Spengler, ID.; eds.
(59) Peters, A.; Wichmann, H-E.; Tuch, T.;
Heinrich, I; Heyder, I (1997) Respiratory effects
are associated with the number of ultrafine
particles. American Journal of Respiratory Critical
Care Medicine 155 :(In press).
(60) Pacific Gas and Electric Company (1997)
Comments on the proposed PM2.5 fine particulate
standard. Docket No. A-95-54, IV-D-2183. March
11, 1997.
(61) Pope, C.A., III, Kalkstein, IS. (1996)
Synoptic weather modeling and estimates of the
exposure-response relationship between daily
mortality and particulate air pollution.
Environmental Health Perspectives 104:414-420.
(62) Pope, C.A., III; Thun, M.I; Namboori,
M.M.; Dockery, D.W.; Evans, IS.; Speizer, F.E.;
Heath, D.W., Ir. (1995) Particulate air pollution as
a predictor of mortality in a prospective study of
U.S. adults. American Journal of Respiratory and
Critical Care Medicine 151:669-674.
(63) Pope, C.A., III; Schwartz, I; Ransom, M.R.
(1992) Daily mortality and PMio pollution in Utah
Valley. Archives Environmental Health 47:211-
217.
(64) Pope, C.A., III; Dockery, D.W. (1992)
Acute health effects of PMio pollution on
symptomatic and asymptomatic children. American
Review of Respiratory Disease 145:1123-1128.
(65) Raizenne, M.; Neas, L.M.; Damokosh, A.I.;
Dockery, D.W.; Spengler, ID.; Koutrakis, P.;
Ware, I.H.; Speizer, F.E. (1996) Health effects of
acid aerosols on North American children:
pulmonary function. Environmental Health
Perspectives 104:506-514.
(66) Roth, H.D.; Li, Y. (1997) Analysis of the
association between air pollutants with mortality
and hospital admissions in Birmingham, Alabama:
1986-1990. (Submitted for publication)
(67) Roth, H.D. et al. (1997) Analysis of the
association between total particulate matter and
daily mortality in the Czech Republic: 1986-1994.
Submitted as an attachment to comments of R.E.
Wyzga, Docket No. A-95-54, IV-D-2672. March
11, 1997.
(68) Sacks, I; Karr, A.F.; Smith, R.L.; Davis,
J.M. (1997) Comment on Scientific Input to
Decision-Making on Airborne Particulate
Standards. Docket No. A-95-54, IV-D-14,298.
March 12, 1997.
(69) Samet, I.M.; Zeger, S.L.; Kelsall, I.E.; Xu,
I (1996a) Air pollution and mortality in
Philadelphia 1973-1988: Report to the Health
Effects Institute on Phase I.B of the Particle
Epidemiology Evaluation Project. Health Effects
Institute, Cambridge, MA.
(70) Samet, I.M.; Zeger, S.L.; Kelsall, I.E.; Xu,
I; Kalkstein, L.S. (1996b) Weather, air pollution,
and mortality in Philadelphia, 1973-1980: Report
to the Health Effects Institute on Phase I.B of the
Particle Epidemiology Evaluation Project. Health
Effects Institute, Cambridge, MA.
(71) Samet, I.M.; Zeger, S.L.; Berthane, K.
(1995) The association of mortality and particulate
air pollution. In: Particulate Air Pollution and Daily
Mortality: Replication and Valication of Selected
Studies (The Phase LA Report of the Particle
Epidemiology Evaluation Project) pp. 3-104.
Health Effects Institute, Cambridge, MA.
(72) Scarlett, IF.; Abbott, K.I; Peacock, I.L.;
Strachan, D.P.; Anderson, H.R. (1996) Acute
effects of summer air pollution on respiratory
function in primary school children in southern
England. Thorax 51:1109-1114.
(73) Schulze, W.D.; Brookshire, D.S.; Walther,
E.G.; MacFarland, K. K.; Thayer, M.A.;
Whitworth, R.L.; Ben-David, S.; Malm, W.;
Molenar, Ir. (1983) The economic benefits of
preserving visibility in the national parklands of the
southwest. Natural Resources Journal 23:149-173.
(74) Schwartz, I; Dockery, D.W.; Neas, L.M.
(1996) Is daily mortality associated specifically
with fine particles? Journal of Air & Waste
Management Association 46:927-939.
(75) Schwartz, I; Dockery, D.W.; Neas, L.M.;
Wypij, D.; Ware, I.H.; Spengler, ID.; Koutrakis,
P.; Speizer, F.E.; Ferris, Ir., E.G. (1994) Acute
effects of summer air pollution on respiratory
symptom reporting in children. American Journal
of Respiratory and Critical Care Medicine
150:1234-1242.
(76) Schwartz, I (1994) Air pollution and
hospital admissions for the elderly in Birmingham,
Alabama. American Journal of Epidemiology
139:589-598.
(77) Schwartz, I (1993) Air pollution and daily
mortality in Birmingham, Alabama. American
Journal of Epidemiology 137:1136-1147.
(78) Schwartz, I; Dockery, D.W. (1992a)
Increased mortality in Philadelphia associated with
daily air pollution concentrations. American Review
of Respiratory Disease 145:600-604.
(79) Schwartz, I; Dockery, D.W. (1992b)
Particulate air pollution and daily mortality in
Steubenville, Ohio. American Journal of
Epidemiology 135:12-19
(80) Sisler, I; Malm, W.; Molenar, I; Gebhardt,
K. (1996) Spatial and Seasonal Patterns and Long
Term Variability of the Chemical Composition of
the Haze in the U.S.: An Analysis of Data from
the IMPROVE Network. Fort Collins, CO:
Cooperative Institute for Research in the
Atmosphere, Colorado State University. luly 1996.
(81) Styer, P.; McMillan, N.; Gao, F.; Davis, I;
Sacks, I (1995) Effect of outdoor airborne
particulate matter on daily death counts.
Environmental Health Perspectives 103:490^197.
(82) Swiss EKL. (1996) Report No. 270. Swiss
Federal Commission of Air Hygiene (EKL).
Docket No. A-95-54, IV-I-59.
-------
54
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
(83) Thurston, G.D. (1997) Letter to President
William J. Clinton. Docket No. A-95-54, IV-F-
97. January 10, 1997.
(84) Thurston, G.D.; Ito, K.; Hayes, C.G.; Bates,
D.V.; Lippmann, M. (1994) Respiratory hospital
admissions and summertime haze air pollution in
Toronto, Ontario: consideration of the role of acid
aerosols. Environmental Research 55:271-290.
(85) Utility Air Regulatory Group. (1997)
Comments on the proposed air quality standards.
Docket No. A-95-54, IV-D-2250. March 14, 1997
(86) United Kingdom Department of
Environment. (1997) The United Kingdom
National Air Quality Strategy. United Kingdom of
the Environment. Scottish Office. Docket No. A—
95-54, IV-I-58. March 1997.
(87) U.S. Department of Health, Education and
Welfare (1964) Smoking and health: report of the
Advisory Committee to the Surgeon General of the
Public Health Service. Washington, DC: Public
Health Service; p. 60.
(88) U.S. Environmental Protection Agency
(1996a) Air Quality Criteria for Particulate Matter.
Research Triangle Park, NC: National Center for
Environmental Assessment. Office of Research and
Development. April 12, 1996.
(89) U.S. Environmental Protection Agency
(1996b) Review of the National Ambient Air
Quality Standards for Particulate Matter: Policy
Assessment of Scientific and Technical
Information—Office of Air Quality Planning and
Standards Staff Paper. Office of Air Quality
Planning and Standards. Office of Air and
Radiation. July 1996.
(90) U.S. Environmental Protection Agency
(1996c) Transcript of the Clean Air Scientific
Advisory Committee's Review of the Particulate
Matter Staff Paper Meetings held on May 6-7,
1996 in Chapel Hill, NC.
(91) U.S. Environmental Protection Agency
(1993) Office of Air Quality Planning and
Standards Effects of the 1990 Clean Air Act
Amendments on Visibility in Class I Areas: An
EPA Report to Congress. Research Triangle Park,
NC.
(92) Valdberg, P.A. (1997) Causality has not
been demonstrated between outdoor levels of
particulate matter (PM) and daily mortality and
morbidity. In: Comments of the Engine
Manufacturers Association on the proposed
revisions to NAAQS for particulate matter and
ozone. Docket No. A-95-54, IV-D-2328. March
11, 1997.
(93) Ware, J.H.; Ferris, E.G., Jr.; Dockery, D.W.;
Spengler, J.D.; Strain, D.O.; Speizer, F.E. (1986)
Effects of ambient sulfur oxides and suspended
particles on respiratory health of children.
American Review of Respiratory Disease 133:834—
842.
(94) World Health Organization. (1997) Update
and Revision of the WHO Air Quality Guidelines
for Europe. European Centre for Environment and
Health. (In press). Docket No. A-95-54, IV-I-60.
(95) Wolff, G.T. (1996a) Letter from George T.
Wolff, Chair, Clean Air Scientific Advisory
Committee, to Administrator Carol M. Browner.
Closure letter on draft Air Quality Criteria for
Particulate Matter. March 15, 1996.
(96) Wolff, G.T. (1996b) Letter from George T.
Wolff, Chair, Clean Air Scientific Advisory
Committee, to Administrator Carol M. Browner.
Clsoure letter on draft OAQPS Staff Paper (Review
of the National Ambient Air Quality Standards for
Particulate Matter: policy Assessment of Scientific
and Technical Information). June 13, 1996.
(97) Woodruff, T.J.; Grille, J.; Schoendorf, K.C.;
1997. The relationship between selected causes of
postneonatal infant mortality and particulate air
pollution in the United States. Environmental
Health Perspectives 105:(In press)
(98) Wordley, J.; Walters, S.; Ayres, J.R. (1997)
Short-term variations in hospital admissions and
mortality and particulate air pollution.
Occupational and Environmental Medicine 54:108-
116.
(99) Wyzga, R.E.; Lipfert, F.W. (1995)
Temperature-pollution interactions with daily
mortality in Philadelphia. In: Particulate matter:
health and regulatory issues: proceedings of an
international specialty conference; April;
Pittsburgh, PA. Air & Waste Management
Association, Pittsburgh, PA (Air & Waste
Management Association Publication VIP-49)
List of Subjects in 40 CFR Part 50
Environmental protection, Air pollution
control, Carbon monoxide, Lead, Nitrogen
dioxide, Ozone, Particulate matter, Sulfur
oxides.
Dated: July 16, 1997.
Carol M. Browner,
Administrator.
Therefore, 40 CFR Chapter I is amended
as follows:
PART 50—NATIONAL PRIMARY AND
SECONDARY AMBIENT AIR QUALITY
STANDARDS
1. The authority citation for part 50
continues to read as follows:
Authority: Sees. 109 and 301(a), Clean Air Act,
as amended (42 U.S.C. 7409, 7601(a)).
2. Section 50.3 is revised to read as
follows:
§50.3 Reference conditions.
All measurements of air quality that are
expressed as mass per unit volume (e.g.,
micrograms per cubic meter) other than for
the particulate matter (PMi0 and PM2.5)
standards contained in § 50.7 shall be
corrected to a reference temperature of 25 °C
and a reference pressure of 760 millimeters
of mercury (1,013.2 millibars). Measurements
of PMio and PM2.5 for purposes of
comparison to the standards contained in
§ 50.7 shall be reported based on actual
ambient air volume measured at the actual
ambient temperature and pressure at the
monitoring site during the measurement
period.
3. Section 50.6 is amended by revising the
section heading and adding paragraph (d) to
read as follows:
§50.6 National primary and secondary
ambient air quality standards for PM,,,.
*****
(d) The PMio standards set forth in this
section will no longer apply to an area not
attaining these standards as of September 16,
1997, once EPA takes final action to
promulgate a rule pursuant to section 172(e)
of the Clean Air Act, as amended (42 U.S.C.
7472(e)) applicable to the area. The PM10
standards set forth in this section will no
longer apply to an area attaining these
standards as of September 16, 1997, once
EPA approves a State Implementation Plan
(SIP) applicable to the area containing all
PMio control measures adopted and
implemented by the state prior to September
16, 1997, and a section 110 SIP implementing
the PM standards published on July 18, 1997.
SIP approvals are codified in 40 CFR part 52.
4. Section 50.7 is added to read as follows:
§50.7 National primary and secondary
ambient air quality standards for particulate
matter.
(a) The national primary and secondary
ambient air quality standards for particulate
matter are:
(1) 15.0 micrograms per cubic meter (|j,g/
m3) annual arithmetic mean concentration,
and 65 l-ig/m3 24-hour average concentration
measured in the ambient air as PM2.5
(particles with an aerodynamic diameter less
than or equal to a nominal 2.5 micrometers)
by either:
(i) A reference method based on Appendix
L of this part and designated in accordance
with part 53 of this chapter; or
(ii) An equivalent method designated in
accordance with part 53 of this chapter.
(2) 50 micrograms per cubic meter (|j,g/m3)
annual arithmetic mean concentration, and
150 |ig/m3 24-hour average concentration
measured in the ambient air as PMio
(particles with an aerodynamic diameter less
than or equal to a nominal 10 micrometers)
by either:
(i) A reference method based on Appendix
M of this part and designated in accordance
with part 53 of this chapter; or
(ii) An equivalent method designated in
accordance with part 53 of this chapter.
(b) The annual primary and secondary
PM2.5 standards are met when the annual
arithmetic mean concentration, as determined
in accordance with Appendix N of this part,
is less than or equal to 15.0 micrograms per
cubic meter.
(c) The 24-hour primary and secondary
PM2.5 standards are met when the 98th
percentile 24-hour concentration, as
determined in accordance with Appendix N
of this part, is less than or equal to 65
micrograms per cubic meter.
(d) The annual primary and secondary
PMio standards are met when the annual
arithmetic mean concentration, as determined
in accordance with Appendix N of this part,
is less than or equal to 50 micrograms per
cubic meter.
(e) The 24-hour primary and secondary
PMio standards are met when the 99th
percentile 24-hour concentration, as
determined in accordance with Appendix N
of this part, is less than or equal to 150
micrograms per cubic meter.
5. Appendix K is revised (for conformity
with the format of the other appendices in this
part) to read as follows:
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
55
Appendix K to Part 50—Interpretation of
the National Ambient Air Quality
Standards for Particulate Matter
1.0 General.
(a) This appendix explains the computations
necessary for analyzing particulate matter data to
determine attainment of the 24-hour and annual
standards specified in 40 CFR 50.6. For the
primary and secondary standards, particulate matter
is measured in the ambient air as PMio (particles
with an aerodynamic diameter less than or equal
to a nominal 10 micrometers) by a reference
method based on appendix J of this part and
designated in accordance with part 53 of this
chapter, or by an equivalent method designated in
accordance with part 53 of this chapter. The
required frequency of measurements is specified in
part 58 of this chapter.
(b) The terms used in this appendix are defined
as follows:
Average refers to an arithmetic mean. All
particulate matter standards are expressed in terms
of expected annual values: Expected number of
exceedances per year for the 24-hour standards and
expected annual arithmetic mean for the annual
standards.
Daily value for PMio refers to the 24-hour
average concentration of PMio calculated or
measured from midnight to midnight (local time).
Exceedance means a daily value that is above the
level of the 24-hour standard after rounding to the
nearest 10 |lg/m3 (i.e., values ending in 5 or greater
are to be rounded up).
Expected annual value is the number approached
when the annual values from an increasing number
of years are averaged, in the absence of long-term
trends in emissions or meteorological conditions.
Year refers to a calendar year.
(c) Although the discussion in this appendix
focuses on monitored data, the same principles
apply to modeling data, subject to EPA modeling
guidelines.
2.0 Attainment Determinations.
2.1 24-Hour Primary and Secondary
Standards.
(a) Under 40 CFR 50.6(a) the 24-hour primary
and secondary standards are attained when the
expected number of exceedances per year at each
monitoring site is less than or equal to one. In the
simplest case, the number of expected exceedances
at a site is determined by recording the number of
exceedances in each calendar year and then
averaging them over the past 3 calendar years.
Situations in which 3 years of data are not available
and possible adjustments for unusual events or
trends are discussed in sections 2.3 and 2.4 of this
appendix. Further, when data for a year are
incomplete, it is necessary to compute an estimated
number of exceedances for that year by adjusting
the observed number of exceedances. This
procedure, performed by calendar quarter, is
described in section 3.0 of this appendix. The
expected number of exceedances is then estimated
by averaging the individual annual estimates for the
past 3 years.
(b) The comparison with the allowable expected
exceedance rate of one per year is made in terms
of a number rounded to the nearest tenth (fractional
values equal to or greater than 0.05 are to be
rounded up; e.g., an exceedance rate of 1.05 would
be rounded to 1.1, which is the lowest rate for
nonattainment).
2.2 Annual Primary and Secondary Standards.
Under 40 CFR 50.6(b), the annual primary and
secondary standards are attained when the expected
annual arithmetic mean PMio concentration is less
than or equal to the level of the standard. In the
simplest case, the expected annual arithmetic mean
is determined by averaging the annual arithmetic
mean PMio concentrations for the past 3 calendar
years. Because of the potential for incomplete data
and the possible seasonality in PMio
concentrations, the annual mean shall be calculated
by averaging the four quarterly means of PMio
concentrations within the calendar year. The
equations for calculating the annual arithmetic
mean are given in section 4.0 of this appendix.
Situations in which 3 years of data are not available
and possible adjustments for unusual events or
trends are discussed in sections 2.3 and 2.4 of this
appendix. The expected annual arithmetic mean is
rounded to the nearest 1 |lg/m3 before comparison
with the annual standards (fractional values equal
to or greater than 0.5 are to be rounded up).
2.3 Data Requirements.
(a) 40 CFR 58.13 specifies the required
minimum frequency of sampling for PMio. For the
purposes of making comparisons with the
particulate matter standards, all data produced by
National Air Monitoring Stations (NAMS), State
and Local Air Monitoring Stations (SLAMS) and
other sites submitted to EPA in accordance with the
Part 58 requirements must be used, and a minimum
of 75 percent of the scheduled PMio samples per
quarter are required.
(b) To demonstrate attainment of either the
annual or 24-hour standards at a monitoring site,
the monitor must provide sufficient data to perform
the required calculations of sections 3.0 and 4.0 of
this appendix. The amount of data required varies
with the sampling frequency, data capture rate and
the number of years of record. In all cases, 3 years
of representative monitoring data that meet the 75
percent criterion of the previous paragraph should
be utilized, if available, and would suffice. More
than 3 years may be considered, if all additional
representative years of data meeting the 75 percent
criterion are utilized. Data not meeting these
criteria may also suffice to show attainment;
however, such exceptions will have to be approved
by the appropriate Regional Administrator in
accordance with EPA guidance.
(c) There are less stringent data requirements for
showing that a monitor has failed an attainment test
and thus has recorded a violation of the particulate
matter standards. Although it is generally necessary
to meet the minimum 75 percent data capture
requirement per quarter to use the computational
equations described in sections 3.0 and 4.0 of this
appendix, this criterion does not apply when less
data is sufficient to unambiguously establish
nonattainment. The following examples illustrate
how nonattainment can be demonstrated when a
site fails to meet the completeness criteria.
Nonattainment of the 24-hour primary standards
can be established by the observed annual number
of exceedances (e.g., four observed exceedances in
a single year), or by the estimated number of
exceedances derived from the observed number of
exceedances and the required number of scheduled
samples (e.g., two observed exceedances with every
other day sampling). Nonattainment of the annual
standards can be demonstrated on the basis of
quarterly mean concentrations developed from
observed data combined with one-half the
minimum detectable concentration substituted for
missing values. In both cases, expected annual
values must exceed the levels allowed by the
standards.
2.4 Adjustment for Exceptional Events and
Trends.
(a) An exceptional event is an uncontrollable
event caused by natural sources of particulate
matter or an event that is not expected to recur at
a given location. Inclusion of such a value in the
computation of exceedances or averages could
result in inappropriate estimates of their respective
expected annual values. To reduce the effect of
unusual events, more than 3 years of representative
data may be used. Alternatively, other techniques,
such as the use of statistical models or the use of
historical data could be considered so that the event
may be discounted or weighted according to the
likelihood that it will recur. The use of such
techniques is subject to the approval of the
appropriate Regional Administrator in accordance
with EPA guidance.
(b) In cases where long-term trends in emissions
and air quality are evident, mathematical techniques
should be applied to account for the trends to
ensure that the expected annual values are not
inappropriately biased by unrepresentative data. In
the simplest case, if 3 years of data are available
under stable emission conditions, this data should
be used. In the event of a trend or shift in emission
patterns, either the most recent representative
year(s) could be used or statistical techniques or
models could be used in conjunction with previous
years of data to adjust for trends. The use of less
than 3 years of data, and any adjustments are
subject to the approval of the appropriate Regional
Administrator in accordance with EPA guidance.
3.0 Computational Equations for the 24-hour
Standards.
3.1 Estimating Exceedances for a Year.
(a) If PMio sampling is scheduled less frequently
than every day, or if some scheduled samples are
missed, a PMio value will not be available for each
day of the year. To account for the possible effect
of incomplete data, an adjustment must be made
to the data collected at each monitoring location to
estimate the number of exceedances in a calendar
year. In this adjustment, the assumption is made
that the fraction of missing values that would have
exceeded the standard level is identical to the
fraction of measured values above this level. This
computation is to be made for all sites that are
scheduled to monitor throughout the entire year and
meet the minimum data requirements of section 2.3
of this appendix. Because of possible seasonal
imbalance, this adjustment shall be applied on a
quarterly basis. The estimate of the expected
number of exceedances for the quarter is equal to
the observed number of exceedances plus an
increment associated with the missing data. The
following equation must be used for these
computations:
Equation 1
eq=vq + [(vq/nq) x (Nq -nq}\ = vq x Nq/nq
where:
eq=the estimated number of exceedances for
calendar quarter q;
vq=the observed number of exceedances for
calendar quarter q;
Nq=the number of days in calendar quarter q;
nq=the number of days in calendar quarter q with
PMio data; and
q=the index for calendar quarter, q=l, 2, 3 or 4.
(b) The estimated number of exceedances for a
calendar quarter must be rounded to the nearest
hundredth (fractional values equal to or greater than
0.005 must be rounded up).
-------
56
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
(c) The estimated number of exceedances for the
year, e, is the sum of the estimates for each
calendar quarter.
Equation 2
e = Leq
q=\
(d) The estimated number of exceedances for a
single year must be rounded to one decimal place
(fractional values equal to or greater than 0.05 are
to be rounded up). The expected number of
exceedances is then estimated by averaging the
individual annual estimates for the most recent 3
or more representative years of data. The expected
number of exceedances must be rounded to one
decimal place (fractional values equal to or greater
than 0.05 are to be rounded up).
(e) The adjustment for incomplete data will not
be necessary for monitoring or modeling data
which constitutes a complete record, i.e., 365 days
per year.
(f) To reduce the potential for overestimating the
number of expected exceedances, the correction for
missing data will not be required for a calendar
quarter in which the first observed exceedance has
occurred if:
(1) There was only one exceedance in the
calendar quarter;
(2) Everyday sampling is subsequently initiated
and maintained for 4 calendar quarters in
accordance with 40 CFR 58.13; and
(3) Data capture of 75 percent is achieved during
the required period of everyday sampling. In
addition, if the first exceedance is observed in a
calendar quarter in which the monitor is already
sampling every day, no adjustment for missing data
will be made to the first exceedance if a 75 percent
data capture rate was achieved in the quarter in
which it was observed.
Example 1
a. During a particular calendar quarter, 39 out
of a possible 92 samples were recorded, with one
observed exceedance of the 24-hour standard.
Using Equation 1, the estimated number of
exceedances for the quarter is:
eq=lx92/39=2.359 or 2.36.
b. If the estimated exceedances for the other 3
calendar quarters in the year were 2.30, 0.0 and 0.0,
then, using Equation 2, the estimated number of
exceedances for the year is 2.36+2.30+0.0+0.0
which equals 4.66 or 4.7. If no exceedances were
observed for the 2 previous years, then the expected
number of exceedances is estimated by: (I/
3)x(4.7+0+0)=1.57 or 1.6. Since 1.6 exceeds the
allowable number of expected exceedances, this
monitoring site would fail the attainment test.
Example 2
In this example, everyday sampling was initiated
following the first observed exceedance as required
by 40 CFR 58.13. Accordingly, the first observed
exceedance would not be adjusted for incomplete
sampling. During the next three quarters, 1.2
exceedances were estimated. In this case, the
estimated exceedances for the year would be
1.0+1.2+0.0+0.0 which equals 2.2. If, as before, no
exceedances were observed for the two previous
years, then the estimated exceedances for the 3—
year period would then be (l/3)x(2.2+0.0+0.0)=0.7,
and the monitoring site would not fail the
attainment test.
3.2 Adjustments for Non-Scheduled Sampling
Days.
(a) If a systematic sampling schedule is used and
sampling is performed on days in addition to the
days specified by the systematic sampling schedule,
e.g., during episodes of high pollution, then an
adjustment must be made in the eqution for the
estimation of exceedances. Such an adjustment is
needed to eliminate the bias in the estimate of the
quarterly and annual number of exceedances that
would occur if the chance of an exceedance is
different for scheduled than for non-scheduled
days, as would be the case with episode sampling.
(b) The required adjustment treats the systematic
sampling schedule as a stratified sampling plan. If
the period from one scheduled sample until the day
preceding the next scheduled sample is defined as
a sampling stratum, then there is one stratum for
each scheduled sampling day. An average number
of observed exceedances is computed for each of
these sampling strata. With nonscheduled sampling
days, the estimated number of exceedances is
defined as:
Equation 3
where:
he
quarter;
Nq=the number of days in the quarter;
mq=the number of strata with samples during the
quarter;
Vj=the number of observed exceedances in stratum
j; and
kj=the number of actual samples in stratum j.
(c) Note that if only one sample value is
recorded in each stratum, then Equation 3 reduces
to Equation 1.
Example 3
A monitoring site samples according to a
systematic sampling schedule of one sample every
6 days, for a total of 15 scheduled samples in a
quarter out of a total of 92 possible samples.
During one 6-day period, potential episode levels
of PMio were suspected, so 5 additional samples
were taken. One of the regular scheduled samples
was missed, so a total of 19 samples in 14 sampling
strata were measured. The one 6-day sampling
stratum with 6 samples recorded 2 exceedances.
The remainder of the quarter with one sample per
stratum recorded zero exceedances. Using Equation
3, the estimated number of exceedances for the
quarter is:
eq=(92/14)x(2/6+0+. . .+0)=2.19.
4.0 Computational Equations for Annual
Standards.
4.1 Calculation of the Annual Arithmetic Mean.
(a) An annual arithmetic mean value for PMio is
determined by averaging the quarterly means for
the 4 calendar quarters of the year. The following
equation is to be used for calculation of the mean
for a calendar quarter:
Equation 4
nq= the number of samples in the quarter, and
Xi= the ith concentration value recorded in the
quarter.
(b) The quarterly mean, expressed in |lg/m3,
must be rounded to the nearest tenth (fractional
values of 0.05 should be rounded up).
(c) The annual mean is calculated by using the
following equation:
Equation 5
where:
xq= the quarterly mean concentration for quarter q,
q=l, 2, 3, or 4,
X =
where:
x=the annual mean; and
xq=the mean for calendar quarter q.
(d) The average of quarterly means must be
rounded to the nearest tenth (fractional values of
0.05 should be rounded up).
(e) The use of quarterly averages to compute the
annual average will not be necessary for monitoring
or modeling data which results in a complete
record, i.e., 365 days per year.
(f) The expected annual mean is estimated as the
average of three or more annual means. This multi-
year estimate, expressed in |lg/m3, shall be rounded
to the nearest integer for comparison with the
annual standard (fractional values of 0.5 should be
rounded up).
Example 4
Using Equation 4, the quarterly means are
calculated for each calendar quarter. If the quarterly
means are 52.4, 75.3, 82.1, and 63.2 |lg/m3, then
the annual mean is:
x = (l/4)x(52.4+75.3+82.1+63.2)= 68.25 or 68.3.
4.2 Adjustments for Non-scheduled Sampling
Days, (a) An adjustment in the calculation of the
annual mean is needed if sampling is performed on
days in addition to the days specified by the
systematic sampling schedule. For the same reasons
given in the discussion of estimated exceedances,
under section 3.2 of this appendix, the quarterly
averages would be calculated by using the
following equation:
Equation 6
m,
where:
xq=the quarterly mean concentration for quarter q,
q=l, 2, 3, or 4;
Xij=the ith concentration value recorded in stratum
j;
kj=the number of actual samples in stratum j; and
mq=the number of strata with data in the quarter.
(b) If one sample value is recorded in each
stratum, Equation 6 reduces to a simple arithmetic
average of the observed values as described by
Equation 4.
Example 5
a. During one calendar quarter, 9 observations
were recorded. These samples were distributed
among 7 sampling strata, with 3 observations in
one stratum. The concentrations of the 3
observations in the single stratum were 202, 242,
and 180 |lg/m3. The remaining 6 observed
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
57
concentrations were 55, 68, 73, 92, 120, and 155
|-lg/m3. Applying the weighting factors specified in
Equation 6, the quarterly mean is:
Xq = (l/7)x [(l/3)x(202+242+180)
+155+68+73+92+120+155=110.1
b. Although 24-hour measurements are rounded
to the nearest 10 |lg/m3 for determinations of
exceedances of the 24—hour standard, note that
these values are rounded to the nearest 1 |lg/m3 for
the calculation of means.
6. Appendix L is added to read as follows:
Appendix L to Part 50—Reference Method
For the Determination of Fine Particulate
Matter as PM2 5 in the Atmosphere
1.0 Applicability.
1.1 This method provides for the measurement
of the mass concentration of fine particulate matter
having an aerodynamic diameter less than or equal
to a nominal 2.5 micrometers (PMi.s) in ambient
air over a 24-hour period for purposes of
determining whether the primary and secondary
national ambient air quality standards for fine
particulate matter specified in § 50.6 of this part are
met. The measurement process is considered to be
nondestructive, and the PMi.s sample obtained can
be subjected to subsequent physical or chemical
analyses. Quality assessment procedures are
provided in part 58, Appendix A of this chapter,
and quality assurance guidance are provided in
references 1, 2, and 3 in section 13.0 of this
appendix.
1.2 This method will be considered a reference
method for purposes of part 58 of this chapter only
if:
(a) The associated sampler meets the
requirements specified in this appendix and the
applicable requirements in part 53 of this chapter,
and
(b) The method and associated sampler have
been designated as a reference method in
accordance with part 53 of this chapter.
1.3 PMi.s samplers that meet nearly all
specifications set forth in this method but have
minor deviations and/or modifications of the
reference method sampler will be designated as
"Class I" equivalent methods for PMi.s in
accordance with part 53 of this chapter.
2.0 Principle.
2.1 An electrically powered air sampler draws
ambient air at a constant volumetric flow rate into
a specially shaped inlet and through an inertial
particle size separator (impactor) where the
suspended particulate matter in the PMi.5 size
range is separated for collection on a
polytetrafluoroethylene (PTFE) filter over the
specified sampling period. The air sampler and
other aspects of this reference method are specified
either explicitly in this appendix or generally with
reference to other applicable regulations or quality
assurance guidance.
2.2 Each filter is weighed (after moisture and
temperature conditioning) before and after sample
collection to determine the net gain due to collected
PMi.s. The total volume of air sampled is
determined by the sampler from the measured flow
rate at actual ambient temperature and pressure and
the sampling time. The mass concentration of
PMi.s in the ambient air is computed as the total
mass of collected particles in the PM2.5 size range
divided by the actual volume of air sampled, and
is expressed in micrograms per cubic meter of air
(|lg/m3).
3.0 PM2.5 Measurement Range.
3.1 Lower concentration limit. The lower
detection limit of the mass concentration
measurement range is estimated to be
approximately 2 |lg/am3, based on noted mass
changes in field blanks in conjunction with the 24
m3 nominal total air sample volume specified for
the 24-hour sample.
3.2 Upper concentration limit. The upper limit
of the mass concentration range is determined by
the filter mass loading beyond which the sampler
can no longer maintain the operating flow rate
within specified limits due to increased pressure
drop across the loaded filter. This upper limit
cannot be specified precisely because it is a
complex function of the ambient particle size
distribution and type, humidity, the individual filter
used, the capacity of the sampler flow rate control
system, and perhaps other factors. Nevertheless, all
samplers are estimated to be capable of measuring
24-hour PM2.5 mass concentrations of at least 200
|lg/m3 while maintaining the operating flow rate
within the specified limits.
3.3 Sample period. The required sample period
for PM2.5 concentration measurements by this
method shall be 1,380 to 1500 minutes (23 to 25
hours). However, when a sample period is less than
1,380 minutes, the measured concentration (as
determined by the collected PM2.5 mass divided by
the actual sampled air volume), multiplied by the
actual number of minutes in the sample period and
divided by 1,440, may be used as if it were a valid
concentration measurement for the specific purpose
of determining a violation of the NAAQS. This
value assumes that the PM2.5 concentration is zero
for the remaining portion of the sample period and
therefore represents the minimum concentration
that could have been measured for the full 24-hour
sample period. Accordingly, if the value thus
calculated is high enough to be an exceedance, such
an exceedance would be a valid exceedance for the
sample period. When reported to AIRS, this data
value should receive a special code to identify it
as not to be commingled with normal concentration
measurements or used for other purposes.
4.0 Accuracy.
4.1 Because the size and volatility of the
particles making up ambient particulate matter vary
over a wide range and the mass concentration of
particles varies with particle size, it is difficult to
define the accuracy of PM2.5 measurements in an
absolute sense. The accuracy of PM2.5
measurements is therefore defined in a relative
sense, referenced to measurements provided by this
reference method. Accordingly, accuracy shall be
defined as the degree of agreement between a
subject field PM2.5 sampler and a collocated PM2.5
reference method audit sampler operating
simultaneously at the monitoring site location of
the subject sampler and includes both random
(precision) and systematic (bias) errors. The
requirements for this field sampler audit procedure
are set forth in part 58, Appendix A of this chapter.
4.2 Measurement system bias. Results of
collocated measurements where the duplicate
sampler is a reference method sampler are used to
assess a portion of the measurement system bias
according to the schedule and procedure specified
in part 58, Appendix A of this chapter.
4.3 Audits with reference method samplers to
determine system accuracy and bias. According to
the schedule and procedure specified in part 58,
Appendix A of this chapter, a reference method
sampler is required to be located at each of selected
PM2.5 SLAMS sites as a duplicate sampler. The
results from the primary sampler and the duplicate
reference method sampler are used to calculate
accuracy of the primary sampler on a quarterly
basis, bias of the primary sampler on an annual
basis, and bias of a single reporting organization
on an annual basis. Reference 2 in section 13.0 of
this appendix provides additional information and
guidance on these reference method audits.
4.4 Flow rate accuracy and bias. Part 58,
Appendix A of this chapter requires that the flow
rate accuracy and bias of individual PM2.5 samplers
used in SLAMS monitoring networks be assessed
periodically via audits of each sampler's
operational flow rate. In addition, part 58,
Appendix A of this chapter requires that flow rate
bias for each reference and equivalent method
operated by each reporting organization be assessed
quarterly and annually. Reference 2 in section 13.0
of this appendix provides additional information
and guidance on flow rate accuracy audits and
calculations for accuracy and bias.
5.0 Precision. A data quality objective of 10
percent coefficient of variation or better has been
established for the operational precision of PM2.5
monitoring data.
5.1 Tests to establish initial operational precision
for each reference method sampler are specified as
a part of the requirements for designation as a
reference method under § 53.58 of this chapter.
5.2 Measurement System Precision. Collocated
sampler results, where the duplicate sampler is not
a reference method sampler but is a sampler of the
same designated method as the primary sampler,
are used to assess measurement system precision
according to the schedule and procedure specified
in part 58, Appendix A of this chapter. Part 58,
Appendix A of this chapter requires that these
collocated sampler measurements be used to
calculate quarterly and annual precision estimates
for each primary sampler and for each designated
method employed by each reporting organization.
Reference 2 in section 13.0 of this appendix
provides additional information and guidance on
this requirement.
6.0 Filter for PM2.5 Sample Collection. Any filter
manufacturer or vendor who sells or offers to sell
filters specifically identified for use with this PM2.5
reference method shall certify that the required
number of filters from each lot of filters offered
for sale as such have been tested as specified in
this section 6.0 and meet all of the following design
and performance specifications.
6.1 Size. Circular, 46.2 mm diameter ±0.25 mm.
6.2 Medium. Polytetrafluoroethylene (PTFE
Teflon), with integral support ring.
6.3 Support ring. Polymethylpentene (PMP) or
equivalent inert material, 0.38 ±0.04 mm thick,
outer diameter 46.2 mm ±0.25 mm, and width of
3.68 mm ( ±0.00, -0.51 mm).
6.4 Pore size. 2 |lm as measured by ASTM F
316-94.
6.5 Filter thickness. 30 to 50 |0m.
6.6 Maximum pressure drop (clean filter). 30 cm
FbO column @ 16.67 L/min clean air flow.
6.7 Maximum moisture pickup. Not more than 10
|lg weight increase after 24-hour exposure to air of
40 percent relative humidity, relative to weight
after 24-hour exposure to air of 35 percent relative
humidity.
6.8 Collection efficiency. Greater than 99.7
percent, as measured by the OOP test (ASTM D
2986-91) with 0.3 |lm particles at the sampler's
operating face velocity.
6.9 Filter weight stability. Filter weight loss shall
be less than 20 |lg, as measured in each of the
following two tests specified in sections 6.9.1 and
6.9.2 of this appendix. The following conditions
apply to both of these tests: Filter weight loss shall
be the average difference between the initial and
the final filter weights of a random sample of test
-------
58
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
filters selected from each lot prior to sale. The
number of filters tested shall be not less than 0.1
percent of the filters of each manufacturing lot, or
10 filters, whichever is greater. The filters shall be
weighed under laboratory conditions and shall have
had no air sample passed through them, i.e., filter
blanks. Each test procedure must include initial
conditioning and weighing, the test, and final
conditioning and weighing. Conditioning and
weighing shall be in accordance with sections 8.0
through 8.2 of this appendix and general guidance
provided in reference 2 of section 13.0 of this
appendix.
6.9.1 Test for loose, surface particle
contamination. After the initial weighing, install
each test filter, in turn, in a filter cassette (Figures
L-27, L-28, and L-29 of this appendix) and drop
the cassette from a height of 25 cm to a flat hard
surface, such as a particle-free wood bench. Repeat
two times, for a total of three drop tests for each
test filter. Remove the test filter from the cassette
and weigh the filter. The average change in weight
must be less than 20 |lg.
6.9.2 Test for temperature stability. After
weighing each filter, place the test filters in a
drying oven set at 40 °C ±2 °C for not less than
48 hours. Remove, condition, and reweigh each test
filter. The average change in weight must be less
than 20 |lg.
6.10 Alkalinity. Less than 25 microequivalents/
gram of filter, as measured by the guidance given
in reference 2 in section 13.0 of this appendix.
6.11 Supplemental requirements. Although not
required for determination of PM2.5 mass
concentration under this reference method,
additional specifications for the filter must be
developed by users who intend to subject PMi.s
filter samples to subsequent chemical analysis.
These supplemental specifications include
background chemical contamination of the filter
and any other filter parameters that may be required
by the method of chemical analysis. All such
supplemental filter specifications must be
compatible with and secondary to the primary filter
specifications given in this section 6.0 of this
appendix.
7.OPA/2.5 Sampler.
1.1 Configuration. The sampler shall consist of
a sample air inlet, downtube, particle size separator
(impactor), filter holder assembly, air pump and
flow rate control system, flow rate measurement
device, ambient and filter temperature monitoring
system, barometric pressure measurement system,
timer, outdoor environmental enclosure, and
suitable mechanical, electrical, or electronic control
capability to meet or exceed the design and
functional performance as specified in this section
7.0 of this appendix. The performance
specifications require that the sampler:
(a) Provide automatic control of sample
volumetric flow rate and other operational
parameters.
(b) Monitor these operational parameters as well
as ambient temperature and pressure.
(c) Provide this information to the sampler
operator at the end of each sample period in digital
form, as specified in Table L-l of section 7.4.19
of this appendix.
7.2 Nature of specifications. The PM2.5 sampler
is specified by a combination of design and
performance requirements. The sample inlet,
downtube, particle size discriminator, filter
cassette, and the internal configuration of the filter
holder assembly are specified explicitly by design
figures and associated mechanical dimensions,
tolerances, materials, surface finishes, assembly
instructions, and other necessary specifications. All
other aspects of the sampler are specified by
required operational function and performance, and
the design of these other aspects (including the
design of the lower portion of the filter holder
assembly) is optional, subject to acceptable
operational performance. Test procedures to
demonstrate compliance with both the design and
performance requirements are set forth in subpart
E of part 53 of this chapter.
7.3 Design specifications. Except as indicated in
this section 7.3 of this appendix, these components
must be manufactured or reproduced exactly as
specified, in an ISO 9001-registered facility, with
registration initially approved and subsequently
maintained during the period of manufacture. See
§ 53.1 (t) of this chapter for the definition of an
ISO-registered facility. Minor modifications or
variances to one or more components that clearly
would not affect the aerodynamic performance of
the inlet, downtube, impactor, or filter cassette will
be considered for specific approval. Any such
proposed modifications shall be described and
submitted to the EPA for specific individual
acceptability either as part of a reference or
equivalent method application under part 53 of this
chapter or in writing in advance of such an
intended application under part 53 of this chapter.
7.3.1 Sample inlet assembly. The sample inlet
assembly, consisting of the inlet, downtube, and
impactor shall be configured and assembled as
indicated in Figure L-l of this appendix and shall
meet all associated requirements. A portion of this
assembly shall also be subject to the maximum
overall sampler leak rate specification under section
7.4.6 of this appendix.
7.3.2 Inlet. The sample inlet shall be fabricated
as indicated in Figures L-2 through L-l 8 of this
appendix and shall meet all associated
requirements.
7.3.3 Downtube. The downtube shall be
fabricated as indicated in Figure L-19 of this
appendix and shall meet all associated
requirements.
7.3.4 Impactor.
7.3.4.1 The impactor (particle size separator)
shall be fabricated as indicated in Figures L-20
through L-24 of this appendix and shall meet all
associated requirements. Following the
manufacture and finishing of each upper impactor
housing (Figure L-21 of this appendix), the
dimension of the impaction jet must be verified by
the manufacturer using Class ZZ go/no-go plug
gauges that are traceable to NIST.
7.3.4.2 Impactor filter specifications:
(a) Size. Circular, 35 to 37 mm diameter.
(b) Medium. Borosilicate glass fiber, without
binder.
(c) Pore size. 1 to 1.5 micrometer, as measured
byASTMF316-80.
(d) Thickness. 300 to 500 micrometers.
7.3.4.3 Impactor oil specifications:
(a) Composition.
Tetramethyltetraphenyltrisiloxane, single-
compound diffusion oil.
(b) Vapor pressure. Maximum 2 x 10~8 mm Hg
at 25 °C.
(c) Viscosity. 36 to 40 centistokes at 25 °C.
(d) Density. 1.06 to 1.07 g/cm3 at 25 °C.
(e) Quantity. 1 mL ±0.1 mL.
7.3.5 Filter holder assembly. The sampler shall
have a sample filter holder assembly to adapt and
seal to the down tube and to hold and seal the
specified filter, under section 6.0 of this appendix,
in the sample air stream in a horizontal position
below the downtube such that the sample air passes
downward through the filter at a uniform face
velocity. The upper portion of this assembly shall
be fabricated as indicated in Figures L-25 and L—
26 of this appendix and shall accept and seal with
the filter cassette, which shall be fabricated as
indicated in Figures L-27 through L-29 of this
appendix.
(a) The lower portion of the filter holder
assembly shall be of a design and construction that:
(1) Mates with the upper portion of the assembly
to complete the filter holder assembly,
(2) Completes both the external air seal and the
internal filter cassette seal such that all seals are
reliable over repeated filter changings, and
(3) Facilitates repeated changing of the filter
cassette by the sampler operator.
(b) Leak-test performance requirements for the
filter holder assembly are included in section 7.4.6
of this appendix.
(c) If additional or multiple filters are stored in
the sampler as part of an automatic sequential
sample capability, all such filters, unless they are
currently and directly installed in a sampling
channel or sampling configuration (either active or
inactive), shall be covered or (preferably) sealed in
such a way as to:
(1) Preclude significant exposure of the filter to
possible contamination or accumulation of dust,
insects, or other material that may be present in the
ambient air, sampler, or sampler ventilation air
during storage periods either before or after
sampling; and
(2) To minimize loss of volatile or semi-volatile
PM sample components during storage of the filter
following the sample period.
7.3.6 Flow rate measurement adapter. A flow
rate measurement adapter as specified in Figure L—
30 of this appendix shall be furnished with each
sampler.
7.3.7 Surface finish. All internal surfaces
exposed to sample air prior to the filter shall be
treated electrolytically in a sulfuric acid bath to
produce a clear, uniform anodized surface finish of
not less than 1000 mg/ft2 (1.08 mg/cm2) in
accordance with military standard specification
(mil. spec.) 8625F, Type II, Class 1 in reference
4 of section 13.0 of this appendix. This anodic
surface coating shall not be dyed or pigmented.
Following anodization, the surfaces shall be sealed
by immersion in boiling deionized water for not
less than 15 minutes. Section 53.51(d)(2) of this
chapter should also be consulted.
7.3.8 Sampling height. The sampler shall be
equipped with legs, a stand, or other means to
maintain the sampler in a stable, upright position
and such that the center of the sample air entrance
to the inlet, during sample collection, is maintained
in a horizontal plane and is 2.0 ±0.2 meters above
the floor or other horizontal supporting surface.
Suitable bolt holes, brackets, tie-downs, or other
means should be provided to facilitate mechanically
securing the sample to the supporting surface to
prevent toppling of the sampler due to wind.
7.4 Performance specifications.
7.4.1 Sample flow rate. Proper operation of the
impactor requires that specific air velocities be
maintained through the device. Therefore, the
design sample air flow rate through the inlet shall
be 16.67 L/min (1.000 m3/hour) measured as actual
volumetric flow rate at the temperature and
pressure of the sample air entering the inlet.
7.4.2 Sample airflow rate control system. The
sampler shall have a sample air flow rate control
system which shall be capable of providing a
sample air volumetric flow rate within the specified
range, under section 7.4.1 of this appendix, for the
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
59
specified filter, under section 6.0 of this appendix,
at any atmospheric conditions specified, under
section 7.4.7 of this appendix, at a filter pressure
drop equal to that of a clean filter plus up to 75
cm water column (55 mm Hg), and over the
specified range of supply line voltage, under
section 7.4.15.1 of this appendix. This flow control
system shall allow for operator adjustment of the
operational flow rate of the sampler over a range
of at least ±15 percent of the flow rate specified
in section 7.4.1 of this appendix.
7.4.3 Sample flow rate regulation. The sample
flow rate shall be regulated such that for the
specified filter, under section 6.0 of this appendix,
at any atmospheric conditions specified, under
section 7.4.7 of this appendix, at a filter pressure
drop equal to that of a clean filter plus up to 75
cm water column (55 mm Hg), and over the
specified range of supply line voltage, under
section 7.4.15.1 of this appendix, the flow rate is
regulated as follows:
7.4.3.1 The volumetric flow rate, measured or
averaged over intervals of not more than 5 minutes
over a 24-hour period, shall not vary more than ±5
percent from the specified 16.67 L/min flow rate
over the entire sample period.
7.4.3.2 The coefficient of variation (sample
standard deviation divided by the mean) of the flow
rate, measured over a 24-hour period, shall not be
greater than 2 percent.
7.4.3.3 The amplitude of short-term flow rate
pulsations, such as may originate from some types
of vacuum pumps, shall be attenuated such that
they do not cause significant flow measurement
error or affect the collection of particles on the
particle collection filter.
7.4.4 Flow rate cut off. The sampler's sample air
flow rate control system shall terminate sample
collection and stop all sample flow for the
remainder of the sample period in the event that
the sample flow rate deviates by more than 10
percent from the sampler design flow rate specified
in section 7.4.1 of this appendix for more than 60
seconds. However, this sampler cut-off provision
shall not apply during periods when the sampler is
inoperative due to a temporary power interruption,
and the elapsed time of the inoperative period shall
not be included in the total sample time measured
and reported by the sampler, under section 7.4.13
of this appendix.
7.4.5 Flow rate measurement.
7.4.5.1 The sampler shall provide a means to
measure and indicate the instantaneous sample air
flow rate, which shall be measured as volumetric
flow rate at the temperature and pressure of the
sample air entering the inlet, with an accuracy of
±2 percent. The measured flow rate shall be
available for display to the sampler operator at any
time in either sampling or standby modes, and the
measurement shall be updated at least every 30
seconds. The sampler shall also provide a simple
means by which the sampler operator can manually
start the sample flow temporarily during non-
sampling modes of operation, for the purpose of
checking the sample flow rate or the flow rate
measurement system.
7.4.5.2 During each sample period, the sampler's
flow rate measurement system shall automatically
monitor the sample volumetric flow rate, obtaining
flow rate measurements at intervals of not greater
than 30 seconds.
(a) Using these interval flow rate measurements,
the sampler shall determine or calculate the
following flow-related parameters, scaled in the
specified engineering units:
(1) The instantaneous or interval-average flow
rate, in L/min.
(2) The value of the average sample flow rate
for the sample period, in L/min.
(3) The value of the coefficient of variation
(sample standard deviation divided by the average)
of the sample flow rate for the sample period, in
percent.
(4) The occurrence of any time interval during
the sample period in which the measured sample
flow rate exceeds a range of ±5 percent of the
average flow rate for the sample period for more
than 5 minutes, in which case a warning flag
indicator shall be set.
(5) The value of the integrated total sample
volume for the sample period, in m3.
(b) Determination or calculation of these values
shall properly exclude periods when the sampler is
inoperative due to temporary interruption of
electrical power, under section 7.4.13 of this
appendix, or flow rate cut off, under section 7.4.4
of this appendix.
(c) These parameters shall be accessible to the
sampler operator as specified in Table L—1 of
section 7.4.19 of this appendix. In addition, it is
strongly encouraged that the flow rate for each 5-
minute interval during the sample period be
available to the operator following the end of the
sample period.
7.4.6 Leak test capability.
7.4.6.1 External leakage. The sampler shall
include an external air leak-test capability
consisting of components, accessory hardware,
operator interface controls, a written procedure in
the associated Operation/Instruction Manual, under
section 7.4.18 of this appendix, and all other
necessary functional capability to permit and
facilitate the sampler operator to conveniently carry
out a leakiest of the sampler at a field monitoring
site without additional equipment. The sampler
components to be subjected to this leak test include
all components and their interconnections in which
external air leakage would or could cause an error
in the sampler's measurement of the total volume
of sample air that passes through the sample filter.
(a) The suggested technique for the operator to
use for this leak test is as follows:
(1) Remove the sampler inlet and installs the
flow rate measurement adapter supplied with the
sampler, under section 7.3.6 of this appendix.
(2) Close the valve on the flow rate measurement
adapter and use the sampler air pump to draw a
partial vacuum in the sampler, including (at least)
the impactor, filter holder assembly (filter in place),
flow measurement device, and interconnections
between these devices, of at least 55 mm Hg (75
cm water column), measured at a location
downstream of the filter holder assembly.
(3) Plug the flow system downstream of these
components to isolate the components under
vacuum from the pump, such as with a built-in
valve.
(4) Stop the pump.
(5) Measure the trapped vacuum in the sampler
with a built-in pressure measuring device.
(6) (i) Measure the vacuum in the sampler with
the built-in pressure measuring device again at a
later time at least 10 minutes after the first pressure
measurement.
(ii) Caution: Following completion of the test,
the adaptor valve should be opened slowly to limit
the flow rate of air into the sampler. Excessive air
flow rate may blow oil out of the impactor.
(7) Upon completion of the test, open the adaptor
valve, remove the adaptor and plugs, and restore
the sampler to the normal operating configuration.
(b) The associated leak test procedure shall
require that for successful passage of this test, the
difference between the two pressure measurements
shall not be greater than the number of mm of Hg
specified for the sampler by the manufacturer,
based on the actual internal volume of the sampler,
that indicates a leak of less than 80 mL/min.
(c) Variations of the suggested technique or an
alternative external leak test technique may be
required for samplers whose design or
configuration would make the suggested technique
impossible or impractical. The specific proposed
external leak test procedure, or particularly an
alternative leak test technique, proposed for a
particular candidate sampler may be described and
submitted to the EPA for specific individual
acceptability either as part of a reference or
equivalent method application under part 53 of this
chapter or in writing in advance of such an
intended application under part 53 of this chapter.
7.4.6.2 Internal, filter bypass leakage. The
sampler shall include an internal, filter bypass leak-
check capability consisting of components,
accessory hardware, operator interface controls, a
written procedure in the Operation/Instruction
Manual, and all other necessary functional
capability to permit and facilitate the sampler
operator to conveniently carry out a test for internal
filter bypass leakage in the sampler at a field
monitoring site without additional equipment. The
purpose of the test is to determine that any portion
of the sample flow rate that leaks past the sample
filter without passing through the filter is
insignificant relative to the design flow rate for the
sampler.
(a) The suggested technique for the operator to
use for this leak test is as follows:
(1) Carry out an external leak test as provided
under section 7.4.6.1 of this appendix which
indicates successful passage of the prescribed
external leak test.
(2) Install a flow-impervious membrane material
in the filter cassette, either with or without a filter,
as appropriate, which effectively prevents air flow
through the filter.
(3) Use the sampler air pump to draw a partial
vacuum in the sampler, downstream of the filter
holder assembly, of at least 55 mm Hg (75 cm
water column).
(4) Plug the flow system downstream of the filter
holder to isolate the components under vacuum
from the pump, such as with a built-in valve.
(5) Stop the pump.
(6) Measure the trapped vacuum in the sampler
with a built-in pressure measuring device.
(7) Measure the vacuum in the sampler with the
built-in pressure measuring device again at a later
time at least 10 minutes after the first pressure
measurement.
(8) Remove the flow plug and membrane and
restore the sampler to the normal operating
configuration.
(b) The associated leak test procedure shall
require that for successful passage of this test, the
difference between the two pressure measurements
shall not be greater than the number of mm of Hg
specified for the sampler by the manufacturer,
based on the actual internal volume of the portion
of the sampler under vacuum, that indicates a leak
of less than 80 mL/min.
(c) Variations of the suggested technique or an
alternative internal, filter bypass leak test technique
may be required for samplers whose design or
configuration would make the suggested technique
impossible or impractical. The specific proposed
internal leak test procedure, or particularly an
alternative internal leak test technique proposed for
a particular candidate sampler may be described
-------
60
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
and submitted to the EPA for specific individual
acceptability either as part of a reference or
equivalent method application under part 53 of this
chapter or in writing in advance of such intended
application under part 53 of this chapter.
7.3.5 Filter holder assembly. The sampler shall
have a sample filter holder assembly to adapt and
seal to the down tube and to hold and seal the
specified filter, under section 6.0 of this appendix,
in the sample air stream in a horizontal position
below the downtube such that the sample air passes
downward through the filter at a uniform face
velocity. The upper portion of this assembly shall
be fabricated as indicated in Figures L-25 and L—
26 of this appendix and shall accept and seal with
the filter cassette, which shall be fabricated as
indicated in Figures L-27 through L-29 of this
appendix.
(a) The lower portion of the filter holder
assembly shall be of a design and construction that:
(1) Mates with the upper portion of the assembly
to complete the filter holder assembly,
(2) Completes both the external air seal and the
internal filter cassette seal such that all seals are
reliable over repeated filter changings, and
(3) Facilitates repeated changing of the filter
cassette by the sampler operator.
(b) Leak-test performance requirements for the
filter holder assembly are included in section 7.4.6
of this appendix.
(c) If additional or multiple filters are stored in
the sampler as part of an automatic sequential
sample capability, all such filters, unless they are
currently and directly installed in a sampling
channel or sampling configuration (either active or
inactive), shall be covered or (preferably) sealed in
such a way as to:
(1) Preclude significant exposure of the filter to
possible contamination or accumulation of dust,
insects, or other material that may be present in the
ambient air, sampler, or sampler ventilation air
during storage periods either before or after
sampling; and
(2) To minimize loss of volatile or semi-volatile
PM sample components during storage of the filter
following the sample period.
7.3.6 Flow rate measurement adapter. A flow
rate measurement adapter as specified in Figure L—
30 of this appendix shall be furnished with each
sampler.
7.3.7 Surface finish. All internal surfaces
exposed to sample air prior to the filter shall be
treated electrolytically in a sulfuric acid bath to
produce a clear, uniform anodized surface finish of
not less than 1000 mg/ft2 (1.08 mg/cm2) in
accordance with military standard specification
(mil. spec.) 8625F, Type II, Class 1 in reference
4 of section 13.0 of this appendix. This anodic
surface coating shall not be dyed or pigmented.
Following anodization, the surfaces shall be sealed
by immersion in boiling deionized water for not
less than 15 minutes. Section 53.51(d)(2) of this
chapter should also be consulted.
7.3.8 Sampling height. The sampler shall be
equipped with legs, a stand, or other means to
maintain the sampler in a stable, upright position
and such that the center of the sample air entrance
to the inlet, during sample collection, is maintained
in a horizontal plane and is 2.0 ±0.2 meters above
the floor or other horizontal supporting surface.
Suitable bolt holes, brackets, tie-downs, or other
means should be provided to facilitate mechanically
securing the sample to the supporting surface to
prevent toppling of the sampler due to wind.
7.4 Performance specifications.
7.4.1 Sample flow rate. Proper operation of the
impactor requires that specific air velocities be
maintained through the device. Therefore, the
design sample air flow rate through the inlet shall
be 16.67 L/min (1.000 m3/hour) measured as actual
volumetric flow rate at the temperature and
pressure of the sample air entering the inlet.
7.4.2 Sample air flow rate control system. The
sampler shall have a sample air flow rate control
system which shall be capable of providing a
sample air volumetric flow rate within the specified
range, under section 7.4.1 of this appendix, for the
specified filter, under section 6.0 of this appendix,
at any atmospheric conditions specified, under
section 7.4.7 of this appendix, at a filter pressure
drop equal to that of a clean filter plus up to 75
cm water column (55 mm Hg), and over the
specified range of supply line voltage, under
section 7.4.15.1 of this appendix. This flow control
system shall allow for operator adjustment of the
operational flow rate of the sampler over a range
of at least ±15 percent of the flow rate specified
in section 7.4.1 of this appendix.
7.4.3 Sample flow rate regulation. The sample
flow rate shall be regulated such that for the
specified filter, under section 6.0 of this appendix,
at any atmospheric conditions specified, under
section 7.4.7 of this appendix, at a filter pressure
drop equal to that of a clean filter plus up to 75
cm water column (55 mm Hg), and over the
specified range of supply line voltage, under
section 7.4.15.1 of this appendix, the flow rate is
regulated as follows:
7.4.3.1 The volumetric flow rate, measured or
averaged over intervals of not more than 5 minutes
over a 24-hour period, shall not vary more than ±5
percent from the specified 16.67 L/min flow rate
over the entire sample period.
7.4.3.2 The coefficient of variation (sample
standard deviation divided by the mean) of the flow
rate, measured over a 24-hour period, shall not be
greater than 2 percent.
7.4.3.3 The amplitude of short-term flow rate
pulsations, such as may originate from some types
of vacuum pumps, shall be attenuated such that
they do not cause significant flow measurement
error or affect the collection of particles on the
particle collection filter.
7.4.4 Flow rate cut off. The sampler's sample air
flow rate control system shall terminate sample
collection and stop all sample flow for the
remainder of the sample period in the event that
the sample flow rate deviates by more than 10
percent from the sampler design flow rate specified
in section 7.4.1 of this appendix for more than 60
seconds. However, this sampler cut-off provision
shall not apply during periods when the sampler is
inoperative due to a temporary power interruption,
and the elapsed time of the inoperative period shall
not be included in the total sample time measured
and reported by the sampler, under section 7.4.13
of this appendix.
7.4.5 Flow rate measurement.
7.4.5.1 The sampler shall provide a means to
measure and indicate the instantaneous sample air
flow rate, which shall be measured as volumetric
flow rate at the temperature and pressure of the
sample air entering the inlet, with an accuracy of
±2 percent. The measured flow rate shall be
available for display to the sampler operator at any
time in either sampling or standby modes, and the
measurement shall be updated at least every 30
seconds. The sampler shall also provide a simple
means by which the sampler operator can manually
start the sample flow temporarily during non-
sampling modes of operation, for the purpose of
checking the sample flow rate or the flow rate
measurement system.
7.4.5.2 During each sample period, the sampler's
flow rate measurement system shall automatically
monitor the sample volumetric flow rate, obtaining
flow rate measurements at intervals of not greater
than 30 seconds.
(a) Using these interval flow rate measurements,
the sampler shall determine or calculate the
following flow-related parameters, scaled in the
specified engineering units:
(1) The instantaneous or interval-average flow
rate, in L/min.
(2) The value of the average sample flow rate
for the sample period, in L/min.
(3) The value of the coefficient of variation
(sample standard deviation divided by the average)
of the sample flow rate for the sample period, in
percent.
(4) The occurrence of any time interval during
the sample period in which the measured sample
flow rate exceeds a range of ±5 percent of the
average flow rate for the sample period for more
than 5 minutes, in which case a warning flag
indicator shall be set.
(5) The value of the integrated total sample
volume for the sample period, in m3.
(b) Determination or calculation of these values
shall properly exclude periods when the sampler is
inoperative due to temporary interruption of
electrical power, under section 7.4.13 of this
appendix, or flow rate cut off, under section 7.4.4
of this appendix.
(c) These parameters shall be accessible to the
sampler operator as specified in Table L—1 of
section 7.4.19 of this appendix. In addition, it is
strongly encouraged that the flow rate for each 5-
minute interval during the sample period be
available to the operator following the end of the
sample period.
7.4.6 Leak test capability.
7.4.6.1 External leakage. The sampler shall
include an external air leak-test capability
consisting of components, accessory hardware,
operator interface controls, a written procedure in
the associated Operation/Instruction Manual, under
section 7.4.18 of this appendix, and all other
necessary functional capability to permit and
facilitate the sampler operator to conveniently carry
out a leak test of the sampler at a field monitoring
site without additional equipment. The sampler
components to be subjected to this leak test include
all components and their interconnections in which
external air leakage would or could cause an error
in the sampler's measurement of the total volume
of sample air that passes through the sample filter.
(a) The suggested technique for the operator to
use for this leak test is as follows:
(1) Remove the sampler inlet and installs the
flow rate measurement adapter supplied with the
sampler, under section 7.3.6 of this appendix.
(2) Close the valve on the flow rate measurement
adapter and use the sampler air pump to draw a
partial vacuum in the sampler, including (at least)
the impactor, filter holder assembly (filter in place),
flow measurement device, and interconnections
between these devices, of at least 55 mm Hg (75
cm water column), measured at a location
downstream of the filter holder assembly.
(3) Plug the flow system downstream of these
components to isolate the components under
vacuum from the pump, such as with a built-in
valve.
(4) Stop the pump.
(5) Measure the trapped vacuum in the sampler
with a built-in pressure measuring device.
(6) (i) Measure the vacuum in the sampler with
the built-in pressure measuring device again at a
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
61
later time at least 10 minutes after the first pressure
measurement.
(ii) Caution: Following completion of the test,
the adaptor valve should be opened slowly to limit
the flow rate of air into the sampler. Excessive air
flow rate may blow oil out of the impactor.
(7) Upon completion of the test, open the adaptor
valve, remove the adaptor and plugs, and restore
the sampler to the normal operating configuration.
(b) The associated leak test procedure shall
require that for successful passage of this test, the
difference between the two pressure measurements
shall not be greater than the number of mm of Hg
specified for the sampler by the manufacturer,
based on the actual internal volume of the sampler,
that indicates a leak of less than 80 mL/min.
(c) Variations of the suggested technique or an
alternative external leak test technique may be
required for samplers whose design or
configuration would make the suggested technique
impossible or impractical. The specific proposed
external leak test procedure, or particularly an
alternative leak test technique, proposed for a
particular candidate sampler may be described and
submitted to the EPA for specific individual
acceptability either as part of a reference or
equivalent method application under part 53 of this
chapter or in writing in advance of such an
intended application under part 53 of this chapter.
7.4.6.2 Internal, filter bypass leakage. The
sampler shall include an internal, filter bypass leak-
check capability consisting of components,
accessory hardware, operator interface controls, a
written procedure in the Operation/Instruction
Manual, and all other necessary functional
capability to permit and facilitate the sampler
operator to conveniently carry out a test for internal
filter bypass leakage in the sampler at a field
monitoring site without additional equipment. The
purpose of the test is to determine that any portion
of the sample flow rate that leaks past the sample
filter without passing through the filter is
insignificant relative to the design flow rate for the
sampler.
(a) The suggested technique for the operator to
use for this leak test is as follows:
(1) Carry out an external leak test as provided
under section 7.4.6.1 of this appendix which
indicates successful passage of the prescribed
external leak test.
(2) Install a flow-impervious membrane material
in the filter cassette, either with or without a filter,
as appropriate, which effectively prevents air flow
through the filter.
(3) Use the sampler air pump to draw a partial
vacuum in the sampler, downstream of the filter
holder assembly, of at least 55 mm Hg (75 cm
water column).
(4) Plug the flow system downstream of the filter
holder to isolate the components under vacuum
from the pump, such as with a built-in valve.
(5) Stop the pump.
(6) Measure the trapped vacuum in the sampler
with a built-in pressure measuring device.
(7) Measure the vacuum in the sampler with the
built-in pressure measuring device again at a later
time at least 10 minutes after the first pressure
measurement.
(8) Remove the flow plug and membrane and
restore the sampler to the normal operating
configuration.
(b) The associated leak test procedure shall
require that for successful passage of this test, the
difference between the two pressure measurements
shall not be greater than the number of mm of Hg
specified for the sampler by the manufacturer,
based on the actual internal volume of the portion
of the sampler under vacuum, that indicates a leak
of less than 80 mL/min.
(c) Variations of the suggested technique or an
alternative internal, filter bypass leak test technique
may be required for samplers whose design or
configuration would make the suggested technique
impossible or impractical. The specific proposed
internal leak test procedure, or particularly an
alternative internal leak test technique proposed for
a particular candidate sampler may be described
and submitted to the EPA for specific individual
acceptability either as part of a reference or
equivalent method application under part 53 of this
chapter or in writing in advance of such intended
application under part 53 of this chapter.
7.4.7 Range of operational conditions. The
sampler is required to operate properly and meet
all requirements specified in this appendix over the
following operational ranges.
7'.4.7'.1 Ambient temperature. -30 to +45 °C
(Note: Although for practical reasons, the
temperature range over which samplers are required
to be tested under part 53 of this chapter is -20
to +40 °C, the sampler shall be designed to operate
properly over this wider temperature range.).
7.4.7.2 Ambient relative humidity. 0 to 100
percent.
7.4.7.3 Barometric pressure range. 600 to 800
mm Hg.
7'.4.8 Ambient temperature sensor. The sampler
shall have capability to measure the temperature of
the ambient air surrounding the sampler over the
range of-30 to +45 °C, with a resolution of 0.1
°C and accuracy of ±2.0 °C, referenced as
described in reference 3 in section 13.0 of this
appendix, with and without maximum solar
insolation.
7.4.8.1 The ambient temperature sensor shall be
mounted external to the sampler enclosure and shall
have a passive, naturally ventilated sun shield. The
sensor shall be located such that the entire sun
shield is at least 5 cm above the horizontal plane
of the sampler case or enclosure (disregarding the
inlet and downtube) and external to the vertical
plane of the nearest side or protuberance of the
sampler case or enclosure. The maximum
temperature measurement error of the ambient
temperature measurement system shall be less than
1.6 °C at 1 m/s wind speed and 1000 W/m2 solar
radiation intensity.
7.4.8.2 The ambient temperature sensor shall be
of such a design and mounted in such a way as
to facilitate its convenient dismounting and
immersion in a liquid for calibration and
comparison to the filter temperature sensor, under
section 7.4.11 of this appendix.
7.4.8.3 This ambient temperature measurement
shall be updated at least every 30 seconds during
both sampling and standby (non-sampling) modes
of operation. A visual indication of the current
(most recent) value of the ambient temperature
measurement, updated at least every 30 seconds,
shall be available to the sampler operator during
both sampling and standby (non-sampling) modes
of operation, as specified in Table L-l of section
7.4.19 of this appendix.
7.4.8.4 This ambient temperature measurement
shall be used for the purpose of monitoring filter
temperature deviation from ambient temperature, as
required by section 7.4.11 of this appendix, and
may be used for purposes of effecting filter
temperature control, under section 7.4.10 of this
appendix, or computation of volumetric flow rate,
under sections 7.4.1 to 7.4.5 of this appendix, if
appropriate.
7.4.8.5 Following the end of each sample period,
the sampler shall report the maximum, minimum,
and average temperature for the sample period, as
specified in Table L-l of section 7.4.19 of this
appendix.
7.4.9 Ambient barometric sensor. The sampler
shall have capability to measure the barometric
pressure of the air surrounding the sampler over a
range of 600 to 800 mm Hg referenced as described
in reference 3 in section 13.0 of this appendix; also
see part 53, subpart E of this chapter. This
barometric pressure measurement shall have a
resolution of 5 mm Hg and an accuracy of ±10 mm
Hg and shall be updated at least every 30 seconds.
A visual indication of the value of the current (most
recent) barometric pressure measurement, updated
at least every 30 seconds, shall be available to the
sampler operator during both sampling and standby
(non-sampling) modes of operation, as specified in
Table L-l of section 7.4.19 of this appendix. This
barometric pressure measurement may be used for
purposes of computation of volumetric flow rate,
under sections 7.4.1 to 7.4.5 of this appendix, if
appropriate. Following the end of a sample period,
the sampler shall report the maximum, minimum,
and mean barometric pressures for the sample
period, as specified in Table L-l of section 7.4.19
of this appendix.
7.4. W Filter temperature control (sampling and
post-sampling). The sampler shall provide a means
to limit the temperature rise of the sample filter (all
sample filters for sequential samplers), from
insolation and other sources, to no more 5 °C above
the temperature of the ambient air surrounding the
sampler, during both sampling and post-sampling
periods of operation. The post-sampling period is
the non-sampling period between the end of the
active sampling period and the time of retrieval of
the sample filter by the sampler operator.
7.4.11 Filter temperature sensor(s).
7.4.11.1 The sampler shall have the capability to
monitor the temperature of the sample filter (all
sample filters for sequential samplers) over the
range of -30 to +45 °C during both sampling and
non-sampling periods. While the exact location of
this temperature sensor is not explicitly specified,
the filter temperature measurement system must
demonstrate agreement, within 1 °C, with a test
temperature sensor located within 1 cm of the
center of the filter downstream of the filter during
both sampling and non-sampling modes, as
specified in the filter temperature measurement test
described in part 53, subpart E of this chapter. This
filter temperature measurement shall have a
resolution of 0.1 °C and accuracy of ±1.0 °C,
referenced as described in reference 3 in section
13.0 of this appendix. This temperature sensor shall
be of such a design and mounted in such a way
as to facilitate its reasonably convenient
dismounting and immersion in a liquid for
calibration and comparison to the ambient
temperature sensor under section 7.4.8 of this
appendix.
7.4.11.2 The filter temperature measurement
shall be updated at least every 30 seconds during
both sampling and standby (non-sampling) modes
of operation. A visual indication of the current
(most recent) value of the filter temperature
measurement, updated at least every 30 seconds,
shall be available to the sampler operator during
both sampling and standby (non-sampling) modes
of operation, as specified in Table L-l of section
7.4.19 of this appendix.
7.4.11.3 For sequential samplers, the temperature
of each filter shall be measured individually unless
it can be shown, as specified in the filter
-------
62
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
temperature measurement test described in § 53.57
of this chapter, that the temperature of each filter
can be represented by fewer temperature sensors.
7.4.11.4 The sampler shall also provide a
warning flag indicator following any occurrence in
which the filter temperature (any filter temperature
for sequential samplers) exceeds the ambient
temperature by more than 5 °C for more than 30
consecutive minutes during either the sampling or
post-sampling periods of operation, as specified in
Table L-l of section 7.4.19 of this appendix, under
section 10.12 of this appendix, regarding sample
validity when a warning flag occurs. It is further
recommended (not required) that the sampler be
capable of recording the maximum differential
between the measured filter temperature and the
ambient temperature and its time and date of
occurrence during both sampling and post-sampling
(non-sampling) modes of operation and providing
for those data to be accessible to the sampler
operator following the end of the sample period,
as suggested in Table L-l of section 7.4.19 of this
appendix.
7.4.12 Clock/timer system.
(a) The sampler shall have a programmable real-
time clock timing/control system that:
(1) Is capable of maintaining local time and date,
including year, month, day-of-month, hour, minute,
and second to an accuracy of ±1.0 minute per
month.
(2) Provides a visual indication of the current
system time, including year, month, day-of-month,
hour, and minute, updated at least each minute, for
operator verification.
(3) Provides appropriate operator controls for
setting the correct local time and date.
(4) Is capable of starting the sample collection
period and sample air flow at a specific, operator-
settable time and date, and stopping the sample air
flow and terminating the sampler collection period
24 hours (1440 minutes) later, or at a specific,
operator-settable time and date.
(b) These start and stop times shall be readily
sellable by Ihe sampler operator to wilhin ±1.0
minule. The system shall provide a visual
indication of Ihe currenl slarl and slop lime sellings,
readable to ±1.0 minule, for verification by Ihe
operator, and Ihe slarl and slop limes shall also be
available via Ihe dala oulpul port, as specified in
Table L-l of section 7.4.19 of Ihis appendix. Upon
execution of a programmed sample period start, Ihe
sampler shall automatically resel all sample period
information and warning flag indications pertaining
to a previous sample period. Refer also to section
7.4.15.4 of Ihis appendix regarding retention of
currenl dale and lime and programmed slart and
slop limes during a temporary eleclrical power
interruption.
7.4.13 Sample time determination. The sampler
shall be capable of determining Ihe elapsed sample
collection lime for each PMi.s sample, accurale to
wilhin ±1.0 minule, measured as Ihe lime belween
Ihe slarl of Ihe sampling period, under seclion
7.4.12 of Ihis appendix and Ihe termination of Ihe
sample period, under seclion 7.4.12 of Ihis
appendix or seclion 7.4.4 of Ihis appendix. This
elapsed sample lime shall nol include periods when
Ihe sampler is inoperalive due to a temporary
interruption of eleclrical power, under seclion
7.4.15.4 oflhis appendix. In Ihe evenllhallhe
elapsed sample lime determined for Ihe sample
period is nol wilhin Ihe range specified for Ihe
required sample period in seclion 3.3 oflhis
appendix, Ihe sampler shall sel a warning flag
indicalor. The dale and lime of Ihe slart of Ihe
sample period, Ihe value of Ihe elapsed sample lime
for Ihe sample period, and Ihe flag indicalor slalus
shall be available lo Ihe sampler operator following
Ihe end of Ihe sample period, as specified in Table
L-l of seclion 7.4.19 oflhis appendix.
7.4.14 Outdoor environmental enclosure. The
sampler shall have an ouldoor enclosure (or
enclosures) suitable lo prolecl Ihe filler and olher
non-wealherproof componenls of Ihe sampler from
precipitation, wind, dusl, exlremes of lemperalure
and humidily; lo help mainlain lemperalure conlrol
of Ihe filler (or fillers, for sequential samplers); and
lo provide reasonable security for sampler
componenls and sellings.
7.4.15 Electrical power supply.
7.4.15.1 The sampler shall be operable and
function as specified herein when operated on an
eleclrical power supply vollage of 105 lo 125 volls
AC (RMS) al a frequency of 59 lo 61 Hz. Optional
operation as specified al additional power supply
voltages and/or frequencies shall nol be precluded
by Ihis requiremenl.
7.4.15.2 The design and conslruclion of Ihe
sampler shall comply wilh all applicable National
Eleclrical Code and Underwriters Laboratories
eleclrical safely requiremenls.
7.4.15.3 The design of all eleclrical and
eleclronic conlrols shall be such as lo provide
reasonable resistance lo interference or malfunction
from ordinary or typical levels of slray
eleclromagnelic fields (EMF) as may be found al
various monitoring sites and from typical levels of
eleclrical Iransienls or eleclronic noise as may often
or occasionally be presenl on various eleclrical
power lines.
7.4.15.4 In Ihe evenl of temporary loss of
eleclrical supply power lo Ihe sampler, Ihe sampler
shall nol be required lo sample or provide olher
specified functions during such loss of power,
excepl lhal Ihe internal clock/timer system shall
mainlain ils local lime and dale selling wilhin ±1
minule per week, and Ihe sampler shall relain all
olher lime and programmable sellings and all dala
required lo be available lo Ihe sampler operator
following each sample period for al leasl 7 days
wilhoul eleclrical supply power. When eleclrical
power is absenl al Ihe operalor-sel lime for starting
a sample period or is interrupted during a sample
period, Ihe sampler shall automatically slart or
resume sampling when eleclrical power is reslored,
if such restoration of power occurs before Ihe
operalor-sel slop lime for Ihe sample period.
7.4.15.5 The sampler shall have Ihe capability lo
record and relain a record of Ihe year, monlh, day-
of-monlh, hour, and minule of Ihe slart of each
power interruption of more lhan 1 minule duration,
up lo 10 such power interruptions per sample
period. (More lhan 10 such power interruptions
shall invalidate Ihe sample, excepl where an
exceedance is measured, under seclion 3.3 oflhis
appendix.) The sampler shall provide for Ihese
power interruption dala lo be available lo Ihe
sampler operator following Ihe end of Ihe sample
period, as specified in Table L-l of seclion 7.4.19
oflhis appendix.
7.4.16 Control devices and operator interface.
The sampler shall have mechanical, eleclrical, or
eleclronic conlrols, conlrol devices, eleclrical or
eleclronic circuils as necessary lo provide Ihe
liming, flow rale measuremenl and conlrol,
lemperalure conlrol, dala slorage and compulation,
operator interface, and olher functions specified.
Operator-accessible conlrols, dala displays, and
inlerface devices shall be designed lo be simple,
slraighlforward, reliable, and easy lo learn, read,
and operate under field condilions. The sampler
shall have provision for operator inpul and slorage
of up lo 64 characlers of numeric (or alphanumeric)
dala for purposes of site, sampler, and sample
identification. This information shall be available lo
Ihe sampler operator for verification and change
and for oulpul via Ihe dala oulpul port along wilh
olher dala following Ihe end of a sample period,
as specified in Table L-l of seclion 7.4.19 oflhis
appendix. All dala required lo be available lo Ihe
operator following a sample collection period or
obtained during standby mode in a posl-sampling
period shall be relained by Ihe sampler until resel,
eilher manually by Ihe operator or automatically by
Ihe sampler upon initiation of a new sample
collection period.
7.4.17 Data output port requirement. The
sampler shall have a slandard RS-232C dala oulpul
connection Ihrough which digilal dala may be
exported lo an external dala slorage or Iransmission
device. All informalion which is required lo be
available al Ihe end of each sample period shall be
accessible Ihrough Ihis dala oulpul connection. The
informalion lhal shall be accessible Ihough Ihis
oulpul port is summarized in Table L-l of seclion
7.4.19 oflhis appendix. Since no specific formal
for Ihe oulpul dala is provided, Ihe sampler
manufaclurer or vendor shall make available lo
sampler purchasers appropriate computer software
capable of receiving exported sampler dala and
correclly Iranslaling Ihe dala into a slandard
spreadsheel formal and optionally any olher
formals as may be useful lo sampler users. This
requiremenl shall nol preclude Ihe sampler from
offering olher types of oulpul connections in
addition lo Ihe required RS-232C port.
7.4.18 Operation/instruction manual. The
sampler shall include an associated comprehensive
operation or inslruclion manual, as required by part
53 oflhis chapter, which includes detailed
operating inslruclions on Ihe selup, operation,
calibration, and maintenance of Ihe sampler. This
manual shall provide complete and detailed
descriptions of Ihe operational and calibration
procedures prescribed for field use of Ihe sampler
and all inslrumenls utilized as part oflhis reference
melhod. The manual shall include adequate
warning of potential safely hazards lhal may resull
from normal use or malfunction of Ihe melhod and
a description of necessary safely precautions. The
manual shall also include a clear descriplion of all
procedures pertaining lo installation, operation,
periodic and corrective mainlenance, and
Iroubleshooling, and shall include parts
identification diagrams.
7.4.19 Data reporting requirements. The various
informalion lhal Ihe sampler is required lo provide
and how il is lo be provided is summarized in Ihe
following Table L-l.
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
63
TABLE L-1 .—SUMMARY OF INFORMATION To BE PROVIDED BY THE SAMPLER
Information to be pro-
vided
Flow rate, 30-second
maximum interval.
Flow rate, average for
the sample period.
Flow rate, CV, for
sample period.
Flow rate, 5-min. aver-
age out of spec.
(FLAG6).
Sample volume, total
Temperature, ambient,
30-second interval.
Temperature, ambient,
min., max., average
for the sample pe-
riod.
Baro pressure, ambi-
ent, 30-second inter-
val.
Baro pressure, ambi-
ent, min., max., av-
erage for the sample
period.
Filter temperature, 30-
second interval.
Filter temperature dif-
ferential, 30-second
interval, out of spec.
(FLAG6).
Filter temperature,
maximum differential
from ambient, date,
time of occurrence.
Date and time
Sample start and stop
time settings.
Sample period start
time.
Elapsed sample time
Elapsed sample time,
out of spec. (FLAG6).
Power interruptions >1
min., start time of
first 10.
User-entered informa-
tion, such as sam-
pler and site identi-
fication.
Appendix
L section
reference
7.4.5.1 ....
7.4.5.2 ....
7.4.5.2 ....
7.4.5.2 ....
7.4.5.2 ....
7.4.8
748
749
7.4.9
7.4.11
7.4.11
7.4.11
7412
7.4.12
7412
7.4.13
7.4.13
7.4.15.5
7.4.16
Availability
Anytime1
/
*
/
/
/
*
/
/
/
*
/
End of pe-
riod2
/
/
/
/
/
/
/
/
/
/
/
/
/
Visual dis-
play3
/
*
/
/
/
/
/
/
/
/
/
/
/
/
/
/
Data out-
put4
/
/•
/•
/•
/•
/•
/•
/
/•
/•
/•
/
/•
Format
Digital reading5
XXX
XX.X
XXX
On/Off
XXX
XX.X
XXX
XXX
XXX
XX.X
On/Off
X.X, YY/MM/DD
HH:mm.
YY/MM/DD HH:mm ....
YY/MM/DD HH:mm ....
YYYY/MM/DD HH:mm
HH:mm
On/Off
1HH:mm, 2HH:mm,
etc ....
As entered
Units
L/min
L/min
%
m3
°C
°C
mm Hg
mm Hg
°C
°C, Yr./Mon./Day Hrs.
min
Yr./Mon./Day Hrs. min
Yr./Mon./Day Hrs. min
Yr./Mon./Day Hrs. min
Hrs. min
Hrs. min
/ Provision of this information is required.
Provision of this information is optional. If information related to the entire sample period is optionally provided prior to the end of the sample
period, the value provided should be the value calculated for the portion of the sampler period completed up to the time the information is pro-
vided.
• Indicates that this information is also required to be provided to the AIRS data bank; see §§58.26 and 58.35 of this chapter.
1 Information is required to be available to the operator at any time the sampler is operating, whether sampling or not.
2 Information relates to the entire sampler period and must be provided following the end of the sample period until reset manually by the oper-
ator or automatically by the sampler upon the start of a new sample period.
3 Information shall be available to the operator visually.
4 Information is to be available as digital data at the sampler's data output port specified in section 7.4.16 of this appendix following the end of
the sample period until reset manually by the operator or automatically by the sampler upon the start of a new sample period.
5 Digital readings, both visual and data output, shall have not less than the number of significant digits and resolution specified.
6 Flag warnings may be displayed to the operator by a single-flag indicator or each flag may be displayed individually. Only a set (on) flag
warning must be indicated; an off (unset) flag may be indicated by the absence of a flag warning. Sampler users should refer to section 10.12 of
this appendix regarding the validity of samples for which the sampler provided an associated flag warning.
8.0 Filter Weighing. See reference 2 in section 13.0
of this appendix, for additional, more detailed
guidance.
8.1 Analytical balance. The analytical balance
used to weigh filters must be suitable for weighing
the type and size of filters specified, under section
6.0 of this appendix, and have a readability of ±1
|lg. The balance shall be calibrated as specified by
the manufacturer at installation and recalibrated
immediately prior to each weighing session. See
-------
64
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
reference 2 in section 13.0 of this appendix for
additional guidance.
8.2 Filter conditioning. All sample filters used
shall be conditioned immediately before both the
pre- and post-sampling weighings as specified
below. See reference 2 in section 13.0 of this
appendix for additional guidance.
8.2.1 Mean temperature. 20 - 23 °C.
8.2.2 Temperature control. ±2 °C over 24 hours.
8.2.3 Mean humidity. Generally, 30-40 percent
relative humidity; however, where it can be shown
that the mean ambient relative humidity during
sampling is less than 30 percent, conditioning is
permissible at a mean relative humidity within ±5
relative humidity percent of the mean ambient
relative humidity during sampling, but not less than
20 percent.
8.2.4 Humidity control. ±5 relative humidity
percent over 24 hours.
8.2.5 Conditioning time. Not less than 24 hours.
8.3 Weighing procedure.
8.3.1 New filters should be placed in the
conditioning environment immediately upon arrival
and stored there until the pre-sampling weighing.
See reference 2 in section 13.0 of this appendix for
additional guidance.
8.3.2 The analytical balance shall be located in
the same controlled environment in which the
filters are conditioned. The filters shall be weighed
immediately following the conditioning period
without intermediate or transient exposure to other
conditions or environments.
8.3.3 Filters must be conditioned at the same
conditions (humidity within ±5 relative humidity
percent) before both the pre- and post-sampling
weighings.
8.3.4 Both the pre- and post-sampling weighings
should be carried out on the same analytical
balance, using an effective technique to neutralize
static charges on the filter, under reference 2 in
section 13.0 of this appendix. If possible, both
weighings should be carried out by the same
analyst.
8.3.5 The pre-sampling (tare) weighing shall be
within 30 days of the sampling period.
8.3.6 The post-sampling conditioning and
weighing shall be completed within 240 hours (10
days) after the end of the sample period, unless the
filter sample is maintained at 4 °C or less during
the entire time between retrieval from the sampler
and the start of the conditioning, in which case the
period shall not exceed 30 days. Reference 2 in
section 13.0 of this appendix has additional
guidance on transport of cooled filters.
8.3.7 Filter blanks.
8.3.7.1 New field blank filters shall be weighed
along with the pre-sampling (tare) weighing of each
lot of PM2.5 filters. These blank filters shall be
transported to the sampling site, installed in the
sampler, retrieved from the sampler without
sampling, and reweighed as a quality control check.
8.3.7.2 New laboratory blank filters shall be
weighed along with the pre-sampling (tare)
weighing of each set of PMi.s filters. These
laboratory blank filters should remain in the
laboratory in protective containers during the field
sampling and should be reweighed as a quality
control check.
8.3.8 Additional guidance for proper filter
weighing and related quality assurance activities is
provided in reference 2 in section 13.0 of this
appendix.
9.0 Calibration. Reference 2 in section 13.0 of this
appendix contains additional guidance.
9.1 General requirements.
9.1.1 Multipoint calibration and single-point
verification of the sampler's flow rate measurement
device must be performed periodically to establish
and maintain traceability of subsequent flow
measurements to a flow rate standard.
9.1.2 An authoritative flow rate standard shall be
used for calibrating or verifying the sampler's flow
rate measurement device with an accuracy of ±2
percent. The flow rate standard shall be a separate,
stand-alone device designed to connect to the flow
rate measurement adapter, Figure L-30 of this
appendix. This flow rate standard must have its
own certification and be traceable to a National
Institute of Standards and Technology (NIST)
primary standard for volume or flow rate. If
adjustments to the sampler's flow rate measurement
system calibration are to be made in conjunction
with an audit of the sampler's flow measurement
system, such adjustments shall be made following
the audit. Reference 2 in section 13.0 of this
appendix contains additional guidance.
9.1.3 The sampler's flow rate measurement
device shall be re-calibrated after electromechanical
maintenance or transport of the sampler.
9.2 Flow rate calibration/verification procedure.
9.2.1 PM2.5 samplers may employ various types
of flow control and flow measurement devices. The
specific procedure used for calibration or
verification of the flow rate measurement device
will vary depending on the type of flow rate
controller and flow rate measurement employed.
Calibration shall be in terms of actual ambient
volumetric flow rates (Qa), measured at the
sampler's inlet downtube. The generic procedure
given here serves to illustrate the general steps
involved in the calibration of a PM2.5 sampler. The
sampler operation/instruction manual required
under section 7.4.18 of this appendix and the
Quality Assurance Handbook in reference 2 in
section 13.0 of this appendix provide more specific
and detailed guidance for calibration.
9.2.2 The flow rate standard used for flow rate
calibration shall have its own certification and be
traceable to a NIST primary standard for volume
or flow rate. A calibration relationship for the flow
rate standard, e.g., an equation, curve, or family of
curves relating actual flow rate (Qa) to the flow rate
indicator reading, shall be established that is
accurate to within 2 percent over the expected
range of ambient temperatures and pressures at
which the flow rate standard may be used. The flow
rate standard must be re-calibrated or re-verified at
least annually.
9.2.3 The sampler flow rate measurement device
shall be calibrated or verified by removing the
sampler inlet and connecting the flow rate standard
to the sampler's downtube in accordance with the
operation/instruction manual, such that the flow
rate standard accurately measures the sampler's
flow rate. The sampler operator shall first carry out
a sampler leak check and confirm that the sampler
passes the leak test and then verify that no leaks
exist between the flow rate standard and the
sampler.
9.2.4 The calibration relationship between the
flow rate (in actual L/min) indicated by the flow
rate standard and by the sampler's flow rate
measurement device shall be established or verified
in accordance with the sampler operation/
instruction manual. Temperature and pressure
corrections to the flow rate indicated by the flow
rate standard may be required for certain types of
flow rate standards. Calibration of the sampler's
flow rate measurement device shall consist of at
least three separate flow rate measurements
(multipoint calibration) evenly spaced within the
range of-10 percent to +10 percent of the
sampler's operational flow rate, section 7.4.1 of this
appendix. Verification of the sampler's flow rate
shall consist of one flow rate measurement at the
sampler's operational flow rate. The sampler
operation/instruction manual and reference 2 in
section 13.0 of this appendix provide additional
guidance.
9.2.5 If during a flow rate verification the
reading of the sampler's flow rate indicator or
measurement device differs by ±2 percent or more
from the flow rate measured by the flow rate
standard, a new multipoint calibration shall be
performed and the flow rate verification must then
be repeated.
9.2.6 Following the calibration or verification,
the flow rate standard shall be removed from the
sampler and the sampler inlet shall be reinstalled.
Then the sampler's normal operating flow rate (in
L/min) shall be determined with a clean filter in
place. If the flow rate indicated by the sampler
differs by ±2 percent or more from the required
sampler flow rate, the sampler flow rate must be
adjusted to the required flow rate, under section
7.4.1 of this appendix.
9.3 Periodic calibration or verification of the
calibration of the sampler's ambient temperature,
filter temperature, and barometric pressure
measurement systems is also required. Reference 3
of section 13.0 of this appendix contains additional
guidance.
10.0 PA/2.5 Measurement Procedure The detailed
procedure for obtaining valid PMi.5 measurements
with each specific sampler designated as part of a
reference method for PM2.5 under part 53 of this
chapter shall be provided in the sampler-specific
operation or instruction manual required by section
7.4.18 of this appendix. Supplemental guidance is
provided in section 2.12 of the Quality Assurance
Handbook listed in reference 2 in section 13.0 of
this appendix. The generic procedure given here
serves to illustrate the general steps involved in the
PMi.s sample collection and measurement, using a
PM2.5 reference method sampler.
10.1 The sampler shall be set up, calibrated, and
operated in accordance with the specific, detailed
guidance provided in the specific sampler's
operation or instruction manual and in accordance
with a specific quality assurance program
developed and established by the user, based on
applicable supplementary guidance provided in
reference 2 in section 13.0 of this appendix.
10.2 Each new sample filter shall be inspected
for correct type and size and for pinholes, particles,
and other imperfections. Unacceptable filters
should be discarded. A unique identification
number shall be assigned to each filter, and an
information record shall be established for each
filter. If the filter identification number is not or
cannot be marked directly on the filter, alternative
means, such as a number-identified storage
container, must be established to maintain positive
filter identification.
10.3 Each filter shall be conditioned in the
conditioning environment in accordance with the
requirements specified in section 8.2 of this
appendix.
10.4 Following conditioning, each filter shall be
weighed in accordance with the requirements
specified in section 8.0 of this appendix and the
presampling weight recorded with the filter
identification number.
10.5 A numbered and preweighed filter shall be
installed in the sampler following the instructions
provided in the sampler operation or instruction
manual.
10.6 The sampler shall be checked and prepared
for sample collection in accordance with
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
65
instructions provided in the sampler operation or
instruction manual and with the specific quality
assurance program established for the sampler by
the user.
10.7 The sampler's timer shall be set to start the
sample collection at the beginning of the desired
sample period and stop the sample collection 24
hours later.
10.8 Information related to the sample collection
(site location or identification number, sample date,
filter identification number, and sampler model and
serial number) shall be recorded and, if appropriate,
entered into the sampler.
10.9 The sampler shall be allowed to collect the
PMi.s sample during the set 24-hour time period.
10.10 Within 96 hours of the end of the sample
collection period, the filter, while still contained in
the filter cassette, shall be carefully removed from
the sampler, following the procedure provided in
the sampler operation or instruction manual and the
quality assurance program, and placed in a
protective container. This protective container shall
be made of metal and contain no loose material that
could be transferred to the filter. The protective
container shall hold the filter cassette securely such
that the cover shall not come in contact with the
filter's surfaces. Reference 2 in section 13.0 of this
appendix contains additional information.
10.11 The total sample volume in actual m3 for
the sampling period and the elapsed sample time
shall be obtained from the sampler and recorded
in accordance with the instructions provided in the
sampler operation or instruction manual. All
sampler warning flag indications and other
information required by the local quality assurance
program shall also be recorded.
10.12 All factors related to the validity or
representativeness of the sample, such as sampler
tampering or malfunctions, unusual meteorological
conditions, construction activity, fires or dust
storms, etc. shall be recorded as required by the
local quality assurance program. The occurrence of
a flag warning during a sample period shall not
necessarily indicate an invalid sample but rather
shall indicate the need for specific review of the
QC data by a quality assurance officer to determine
sample validity.
10.13 After retrieval from the sampler, the
exposed filter containing the PM2.5 sample should
be transported to the filter conditioning
environment as soon as possible ideally to arrive
at the conditioning environment within 24 hours for
conditioning and subsequent weighing. During the
period between filter retrieval from the sampler and
the start of the conditioning, the filter shall be
maintained as cool as practical and continuously
protected from exposure to temperatures over 25
°C. See section 8.3.6 of this appendix regarding
time limits for completing the post-sampling
weighing. See reference 2 in section 13.0 of this
appendix for additional guidance on transporting
filter samplers to the conditioning and weighing
laboratory.
10.14. The exposed filter containing the PMi.s
sample shall be re-conditioned in the conditioning
environment in accordance with the requirements
specified in section 8.2 of this appendix.
10.15. The filter shall be reweighed immediately
after conditioning in accordance with the
requirements specified in section 8.0 of this
appendix, and the postsampling weight shall be
recorded with the filter identification number.
10.16 The PMi.s concentration shall be
calculated as specified in section 12.0 of this
appendix.
11.0 Sampler Maintenance
The sampler shall be maintained as described by
the sampler's manufacturer in the sampler-specific
operation or instruction manual required under
section 7.4.18 of this appendix and in accordance
with the specific quality assurance program
developed and established by the user based on
applicable supplementary guidance provided in
reference 2 in section 13.0 of this appendix.
12.0 Calculations
12.1 (a) The PMi.s concentration is calculated
as:
PM2.5=(Wf-Wi)/Va
where:
PMi.s = mass concentration of PMi.s, |lg/m3;
Wf, Wi = final and initial weights, respectively,
of the filter used to collect the PMi.s particle
sample, |lg;
Va = total air volume sampled in actual volume
units, as provided by the sampler, m3.
(b) Note: Total sample time must be between
1,380 and 1,500 minutes (23 and 25 hrs) for a fully
valid PMi.s sample; however, see also section 3.3
of this appendix.
13.0 References.
1. Quality Assurance Handbook for Air Pollution
Measurement Systems, Volume I, Principles. EPA/
600/R-94/038a, April 1994. Available from CERI,
ORD Publications, U.S. Environmental Protection
Agency, 26 West Martin Luther King Drive,
Cincinnati, Ohio 45268.
2. Copies of secton 2.12 of the Quality
Assurance Handbook for Air Pollution
Measurement Systems, Volume II, Ambient Air
Specific Methods, EPA/600/R-94/038b, are
available from Department E (MD-77B), U.S. EPA,
Research Triangle Park, NC 27711.
3. Quality Assurance Handbook for Air Pollution
Measurement Systems, Volume IV: Meteorological
Measurements, (Revised Edition) EPA/600/R-94/
03 8d, March, 1995. Available from CERI, ORD
Publications, U.S. Environmental Protection
Agency, 26 West Martin Luther King Drive,
Cincinnati, Ohio 45268.
4. Military standard specification (mil. spec.)
8625F, Type II, Class 1 as listed in Department of
Defense Index of Specifications and Standards
(DODISS), available from DODSSP-Customer
Service, Standardization Documents Order Desk,
700 Robbins Avenue, Building 4D, Philadelphia,
PA 1911-5094.
14.0 Figures L—l through L—30 to Appendix L.
-------
66
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
FIGURE L-l. PM2.5 SAMPLER, ASSEMBLY
ATTACH WATER COLLECTOR HARDWARE
(FOR EXAMPLE: 1/4" NPT GLASS JAR
BRASS, LONG NIPPLE, 1/4" MNPT X 2" LONG
BRASS, BUSHING, 1/4" FNPT X 3/8" MNPT
BRASS, PLUG, 1/4" MNPT)
DOTTED LINE INDICATES ~\
TOP OF SAMPLER CASE \
1 +/-1
TOLERANCES
2PLCS
+/-0.010
3PLOS
+/- 0.005
FRAC.
+/- 1/64
ANGLE
+/- 15'
ALL DIMENSIONS ARE INCHES
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
67
CQ
§
W
o
S
o
2
E
O
O
LU
CO
o:
9
a
z
a:
LU
Q. CO
I- CO "-
o ^
-------
68
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
O
CO
&
a
OH
CL,
D
CQ
S
w
CO
co
1
CJ
I
O
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
69
§
o
ffl
s
I
o
S
o
N
Q
O
y
o
Q.
LU
LU
Q
O
X
gt
Q CO
< 5
Q co
(O Q.
-------
70
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
w
0.
O
LU
12
W s
si
SSI
m ^ P- yj
V 8ld
O — ;;;•;=•
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
71
co
in
9
o
^
LU
O
*f
co
LU
O
o
tq
2
oo
<
O
I
O
tq
2
a
Q
O
to
LU
I
o
z
LU
o:
<
co
O
CO
a -j.
o +
-------
72
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
W
H-l
UH
PJ
Q
Q
§
Pi
U
2
E
LU
s
LLJ
O
lls
HI
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
73
O
s
u
O
[I
CO
(Ł3
-------
74
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
I
o
z
CO
z
o
CO
z
L1J
I
Q.
UJ
• Q
XZ
Pi
w
u
<
O.
CiO
§
&
U
_L
s
CO
UJ
o
o
p
w o
O
I
..
w
UJ
1-
CC
CD
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
75
w
w
IOT
W W
O
u
w
u.
U-l
Q
I
O
o
o
W
oi
O
-------
76
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
o
w
§
o
I
O
E
Illlli
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
77
o
o
i
CD
I
u
p-1
GO
P-)
J
o
Pi
u
§
o
-------
78
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
PQ
J
N
N
O
CJ
<
o.
s
o
CO Q
« UJ
il
V)
UJ
o
z
<
O
>
Q_
O
ce
8
a:
LU
CO
UJ
Z g
O cog
CC UJ ±.
O O 5
2 ui z
^o-i
^0:3,
co x
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
79
w
H-l
N
i
1
oi
O
**
HJ
o
to
HI
o
i
LU
P
HI
a:
CD
HI
§
OL
h
O W§
2 LU .=
Sal
s
O
O
CD
W
V) •*. <Ł {Ł -r
W < UJ < to
HI 111 I- LU =
a. m
:j
o LL
-------
80
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
>-
m
8
w
5
a:
t
o
CM
O
O
I
o
§
o
P
O
u~
_L
O
•*
csi
x
o
X
o
CM
O
O
I
HI
a
z
UJ
w
a
o
3
(Nj
O
X
z
u
CM
CD
8 b
CO
UJ
§
UJ
a.
c\i
X
1
CO
UJ
i- co o i
o -*.-..
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
81
2
O
LJJ
m O
OT Z
W
w
w
o
a
g
2
o
8
o
O
3
Q.
W
JJ
Z
1O
8
o o
p
o
D
O
HH
U.
O
a:
o
i.S
Q.
O
CO
O
ig
Ł0
g>
oŁ
HW
O _i
<<
Q ^C
33
in Q;
o m
3 SI 5
WR a
;§s^
< <*Q
1111
5 O u.
-------
82
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
H-l
(X
tq
s
1
u
I-H
s
o
O
05 =
LU
o -
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
83
R
PU,
<
Q
<
R
u
1
o
o
P
CO
» o
W
IL1
-------
84
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
85
03
s
CO
LU
I
o
«
z
LU
O
O
OH
2
O
O
§
o
(N
ua
Di
D
a
EZ
8
o o
8
3
u. O
€1
Z
o
-------
86
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
w
OH
o-
d
5o
B
o
H
|
O
§
I
u-i
(N
D
a
I Q
C3Z
3 .
i +
U)
in
o
; — j? _ s o::
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
87
z
o
I
O
LU
Cfl
W
PL,
PL,
_f
W
CJ
0.
I-H
|
U
§
i/->
04
I
o
H^
LlH
L._J
CO
UJ
o
o
p
o
-------
88
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
§
H
CJ
uu
C/D
a!
O
Pi
O
r
o
z
z
o
OT
UJ
O
8
o:^
in s
§1
§
p
o
1
CJ
I
O
r
7
O (
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
89
a
Ł
o
o
g
^
o
ffi
I
o
I/-}
-------
90
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
CO
w
W
Q
I
I
o
E-
JJ
<
cr:
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
91
si
Q
I
aj
vo
04
a
E
-------
92
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
W
O
O
IT)
(N
CM T-
§§
O O
O
CO
10
LU
Q
ill
I
O
i
I-H
O
PJ
CLn
0.
a:
(0
O
w
z
LU
5
Q
_1
w
LU
O
i
a:
in
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
93
U
00
o
HH
u.
o:
LU
O
y)
HI
I
z
o
M
Z
LU
O
o
o o
o
p
8 <=>
-------
94
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
o
HH
U
PJ
OO
(*
u
Ł
o
W
oo
00
ON
<2 I
% n
Ł ^
UJ CO
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
95
s
V)
<
W
W
H
-------
96
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
7. Appendix M is added to read as follows:
Appendix M to Part 50—Reference
Method for the Determination of
Particulate Matter as PMio in the
Atmosphere
1.0 Applicability.
1.1 This method provides for the measurement
of the mass concentration of participate matter with
an aerodynamic diameter less than or equal to a
nominal 10 micrometers (PMio) in ambient air
over a 24-hour period for purposes of determining
attainment and maintenance of the primary and
secondary national ambient air quality standards for
particulate matter specified in § 50.6 of this chapter.
The measurement process is nondestructive, and the
PMio sample can be subjected to subsequent
physical or chemical analyses. Quality assurance
procedures and guidance are provided in part 58,
Appendices A and B of this chapter and in
references 1 and 2 of section 12.0 of this appendix.
2.0 Principle.
2.1 An air sampler draws ambient air at a
constant flow rate into a specially shaped inlet
where the suspended particulate matter is inertially
separated into one or more size fractions within the
PMio size range. Each size fraction in the PMio
size range is then collected on a separate filter over
the specified sampling period. The particle size
discrimination characteristics (sampling
effectiveness and 50 percent cutpoint) of the
sampler inlet are prescribed as performance
specifications in part 53 of this chapter.
2.2 Each filter is weighed (after moisture
equilibration) before and after use to determine the
net weight (mass) gain due to collected PMio. The
total volume of air sampled, measured at the actual
ambient temperature and pressure, is determined
from the measured flow rate and the sampling time.
The mass concentration of PMio in the ambient air
is computed as the total mass of collected particles
in the PMio size range divided by the volume of
air sampled, and is expressed in micrograms per
actual cubic meter (|lg/m3).
2.3 A method based on this principle will be
considered a reference method only if the
associated sampler meets the requirements
specified in this appendix and the requirements in
part 53 of this chapter, and the method has been
designated as a reference method in accordance
with part 53 of this chapter.
3.0 Range.
3.1 The lower limit of the mass concentration
range is determined by the repeatability of filter
tare weights, assuming the nominal air sample
volume for the sampler. For samplers having an
automatic filter-changing mechanism, there may be
no upper limit. For samplers that do not have an
automatic filter-changing mechanism, the upper
limit is determined by the filter mass loading
beyond which the sampler no longer maintains the
operating flow rate within specified limits due to
increased pressure drop across the loaded filter.
This upper limit cannot be specified precisely
because it is a complex function of the ambient
particle size distribution and type, humidity, filter
type, and perhaps other factors. Nevertheless, all
samplers should be capable of measuring 24-hour
PMio mass concentrations of at least 300 |lg/m3
while maintaining the operating flow rate within
the specified limits.
4.0 Precision.
4.1 The precision of PMio samplers must be 5
|lg/m3 for PMio concentrations below 80 |lg/m3
and 7 percent for PMio concentrations above 80
|lg/m3, as required by part 53 of this chapter, which
prescribes a test procedure that determines the
variation in the PMio concentration measurements
of identical samplers under typical sampling
conditions. Continual assessment of precision via
collocated samplers is required by part 58 of this
chapter for PMio samplers used in certain
monitoring networks.
5.0 Accuracy.
5.1 Because the size of the particles making up
ambient particulate matter varies over a wide range
and the concentration of particles varies with
particle size, it is difficult to define the absolute
accuracy of PMio samplers. Part 53 of this chapter
provides a specification for the sampling
effectiveness of PMio samplers. This specification
requires that the expected mass concentration
calculated for a candidate PMio sampler, when
sampling a specified particle size distribution, be
within ±10 percent of that calculated for an ideal
sampler whose sampling effectiveness is explicitly
specified. Also, the particle size for 50 percent
sampling effectiveness is required to be 10±0.5
micrometers. Other specifications related to
accuracy apply to flow measurement and
calibration, filter media, analytical (weighing)
procedures, and artifact. The flow rate accuracy of
PMio samplers used in certain monitoring networks
is required by part 58 of this chapter to be assessed
periodically via flow rate audits.
6.0 Potential Sources of Error.
6.1 Volatile Particles. Volatile particles
collected on filters are often lost during shipment
and/or storage of the filters prior to the post-
sampling weighing3. Although shipment or storage
of loaded filters is sometimes unavoidable, filters
should be reweighed as soon as practical to
minimize these losses.
6.2 Artifacts. Positive errors in PMio
concentration measurements may result from
retention of gaseous species on filters 4-5. Such
errors include the retention of sulfur dioxide and
nitric acid. Retention of sulfur dioxide on filters,
followed by oxidation to sulfate, is referred to as
artifact sulfate formation, a phenomenon which
increases with increasing filter alkalinity 6. Little or
no artifact sulfate formation should occur using
filters that meet the alkalinity specification in
section 7.2.4 of this appendix, Artifact nitrate
formation, resulting primarily from retention of
nitric acid, occurs to varying degrees on many filter
types, including glass fiber, cellulose ester, and
many quartz fiber filters 5-7-8-9-10. Loss of true
atmospheric particulate nitrate during or following
sampling may also occur due to dissociation or
chemical reaction. This phenomenon has been
observed on Teflon® filters 8 and inferred for quartz
fiber filters n-12. The magnitude of nitrate artifact
errors in PMio mass concentration measurements
will vary with location and ambient temperature;
however, for most sampling locations, these errors
are expected to be small.
6.3 Humidity. The effects of ambient humidity
on the sample are unavoidable. The filter
equilibration procedure in section 9.0 of this
appendix is designed to minimize the effects of
moisture on the filter medium.
6.4 Filter Handling. Careful handling of filters
between presampling and postsampling weighings
is necessary to avoid errors due to damaged filters
or loss of collected particles from the filters. Use
of a filter cartridge or cassette may reduce the
magnitude of these errors. Filters must also meet
the integrity specification in section 7.2.3 of this
appendix.
6.5 Flow Rate Variation. Variations in the
sampler's operating flow rate may alter the particle
size discrimination characteristics of the sampler
inlet. The magnitude of this error will depend on
the sensitivity of the inlet to variations in flow rate
and on the particle distribution in the atmosphere
during the sampling period. The use of a flow
control device, under section 7.1.3 of this appendix,
is required to minimize this error.
6.6 Air Volume Determination. Errors in the air
volume determination may result from errors in the
flow rate and/or sampling time measurements. The
flow control device serves to minimize errors in the
flow rate determination, and an elapsed time meter,
under section 7.1.5 of this appendix, is required to
minimize the error in the sampling time
measurement.
7.0 Apparatus.
7.1 PMio Sampler.
7.1.1 The sampler shall be designed to:
(a) Draw the air sample into the sampler inlet
and through the particle collection filter at a
uniform face velocity.
(b) Hold and seal the filter in a horizontal
position so that sample air is drawn downward
through the filter.
(c) Allow the filter to be installed and removed
conveniently.
(d) Protect the filter and sampler from
precipitation and prevent insects and other debris
from being sampled.
(e) Minimize air leaks that would cause error in
the measurement of the air volume passing through
the filter.
(f) Discharge exhaust air at a sufficient distance
from the sampler inlet to minimize the sampling
of exhaust air.
(g) Minimize the collection of dust from the
supporting surface.
7.1.2 The sampler shall have a sample air inlet
system that, when operated within a specified flow
rate range, provides particle size discrimination
characteristics meeting all of the applicable
performance specifications prescribed in part 53 of
this chapter. The sampler inlet shall show no
significant wind direction dependence. The latter
requirement can generally be satisfied by an inlet
shape that is circularly symmetrical about a vertical
axis.
7.1.3 The sampler shall have a flow control
device capable of maintaining the sampler's
operating flow rate within the flow rate limits
specified for the sampler inlet over normal
variations in line voltage and filter pressure drop.
7.1.4 The sampler shall provide a means to
measure the total flow rate during the sampling
period. A continuous flow recorder is
recommended but not required. The flow
measurement device shall be accurate to ±2
percent.
7.1.5 A timing/control device capable of
starting and stopping the sampler shall be used to
obtain a sample collection period of 24 ±1 hr
(1,440 ±60 min). An elapsed time meter, accurate
to within ±15 minutes, shall be used to measure
sampling time. This meter is optional for samplers
with continuous flow recorders if the sampling time
measurement obtained by means of the recorder
meets the ±15 minute accuracy specification.
7.1.6 The sampler shall have an associated
operation or instruction manual as required by part
53 of this chapter which includes detailed
instructions on the calibration, operation, and
maintenance of the sampler.
7.2 Filters.
7.2.1 Filter Medium. No commercially
available filter medium is ideal in all respects for
all samplers. The user's goals in sampling
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
97
determine the relative importance of various filter
characteristics, e.g., cost, ease of handling, physical
and chemical characteristics, etc., and,
consequently, determine the choice among
acceptable filters. Furthermore, certain types of
filters may not be suitable for use with some
samplers, particularly under heavy loading
conditions (high mass concentrations), because of
high or rapid increase in the filter flow resistance
that would exceed the capability of the sampler's
flow control device. However, samplers equipped
with automatic filter-changing mechanisms may
allow use of these types of filters. The
specifications given below are minimum
requirements to ensure acceptability of the filter
medium for measurement of PMio mass
concentrations. Other filter evaluation criteria
should be considered to meet individual sampling
and analysis objectives.
7.2.2 Collection Efficiency. >99 percent, as
measured by the OOP test (ASTM-2986) with 0.3
|lm particles at the sampler's operating face
velocity.
7.2.3 Integrity. ±5 |lg/m3 (assuming sampler's
nominal 24-hour air sample volume). Integrity is
measured as the PMio concentration equivalent
corresponding to the average difference between
the initial and the final weights of a random sample
of test filters that are weighed and handled under
actual or simulated sampling conditions, but have
no air sample passed through them, i.e., filter
blanks. As a minimum, the test procedure must
include initial equilibration and weighing,
installation on an inoperative sampler, removal
from the sampler, and final equilibration and
weighing.
7.2.4 Alkalinity. <25 microequivalents/gram of
filter, as measured by the procedure given in
reference 13 of section 12.0 of this appendix
following at least two months storage in a clean
environment (free from contamination by acidic
gases) at room temperature and humidity.
7.3 Flow Rate Transfer Standard. The flow
rate transfer standard must be suitable for the
sampler's operating flow rate and must be
calibrated against a primary flow or volume
standard that is traceable to the National Institute
of Standard and Technology (NIST). The flow rate
transfer standard must be capable of measuring the
sampler's operating flow rate with an accuracy of
±2 percent.
7.4 Filter Conditioning Environment.
7.4.1 Temperature range. 15 to 30 C.
7.4.2 Temperature control. ±3 C.
7.4.3 Humidity range. 20% to 45% RH.
7.4.4 Humidity control. ±5% RH.
7.5 Analytical Balance. The analytical balance
must be suitable for weighing the type and size of
filters required by the sampler. The range and
sensitivity required will depend on the filter tare
weights and mass loadings. Typically, an analytical
balance with a sensitivity of 0.1 mg is required for
high volume samplers (flow rates >0.5 m3/min).
Lower volume samplers (flow rates <0.5 m3/min)
will require a more sensitive balance.
8.0 Calibration.
8.1 General Requirements.
8.1.1 Calibration of the sampler's flow
measurement device is required to establish
traceability of subsequent flow measurements to a
primary standard. A flow rate transfer standard
calibrated against a primary flow or volume
standard shall be used to calibrate or verify the
accuracy of the sampler's flow measurement
device.
8.1.2 Particle size discrimination by inertial
separation requires that specific air velocities be
maintained in the sampler's air inlet system.
Therefore, the flow rate through the sampler's inlet
must be maintained throughout the sampling period
within the design flow rate range specified by the
manufacturer. Design flow rates are specified as
actual volumetric flow rates, measured at existing
conditions of temperature and pressure (Qa).
8.2 Flow Rate Calibration Procedure.
8.2.1 PMio samplers employ various types of
flow control and flow measurement devices. The
specific procedure used for flow rate calibration or
verification will vary depending on the type of flow
controller and flow rate indicator employed.
Calibration is in terms of actual volumetric flow
rates (Qa) to meet the requirements of section 8.1
of this appendix. The general procedure given here
serves to illustrate the steps involved in the
calibration. Consult the sampler manufacturer's
instruction manual and reference 2 of section 12.0
of this appendix for specific guidance on
calibration. Reference 14 of section 12.0 of this
appendix provides additional information on
various other measures of flow rate and their
interrelationships.
8.2.2 Calibrate the flow rate transfer standard
against a primary flow or volume standard
traceable to NIST. Establish a calibration
relationship, e.g., an equation or family of curves,
such that traceability to the primary standard is
accurate to within 2 percent over the expected
range of ambient conditions, i.e., temperatures and
pressures, under which the transfer standard will be
used. Recalibrate the transfer standard periodically.
8.2.3 Following the sampler manufacturer's
instruction manual, remove the sampler inlet and
connect the flow rate transfer standard to the
sampler such that the transfer standard accurately
measures the sampler's flow rate. Make sure there
are no leaks between the transfer standard and the
sampler.
8.2.4 Choose a minimum of three flow rates
(actual m3/min), spaced over the acceptable flow
rate range specified for the inlet, under section
7.1.2 of the appendix, that can be obtained by
suitable adjustment of the sampler flow rate. In
accordance with the sampler manufacturer's
instruction manual, obtain or verify the calibration
relationship between the flow rate (actual m3/min)
as indicated by the transfer standard and the
sampler's flow indicator response. Record the
ambient temperature and barometric pressure.
Temperature and pressure corrections to subsequent
flow indicator readings may be required for certain
types of flow measurement devices. When such
corrections are necessary, correction on an
individual or daily basis is preferable. However,
seasonal average temperature and average
barometric pressure for the sampling site may be
incorporated into the sampler calibration to avoid
daily corrections. Consult the sampler
manufacturer's instruction manual and reference 2
in section 12.0 of this appendix for additional
guidance.
8.2.5 Following calibration, verify that the
sampler is operating at its design flow rate (actual
m3/min) with a clean filter in place.
8.2.6 Replace the sampler inlet.
9.0 Procedure.
9.1 The sampler shall be operated in
accordance with the specific guidance provided in
the sampler manufacturer's instruction manual and
in reference 2 in section 12.0 of this appendix. The
general procedure given here assumes that the
sampler's flow rate calibration is based on flow
rates at ambient conditions (Qa) and serves to
illustrate the steps involved in the operation of a
PMio sampler.
9.2 Inspect each filter for pinholes, particles,
and other imperfections. Establish a filter
information record and assign an identification
number to each filter.
9.3 Equilibrate each filter in the conditioning
environment (see 7.4) for at least 24 hours.
9.4 Following equilibration, weigh each filter
and record the presampling weight with the filter
identification number.
9.5 Install a preweighed filter in the sampler
following the instructions provided in the sampler
manufacturer's instruction manual.
9.6 (a) Turn on the sampler and allow it to
establish run-temperature conditions. Record the
flow indicator reading and, if needed, the ambient
temperature and barometric pressure. Determine the
sampler flow rate (actual m3/min) in accordance
with the instructions provided in the sampler
manufacturer's instruction manual.
(b) Note: No onsite temperature or pressure
measurements are necessary if the sampler's flow
indicator does not require temperature or pressure
corrections or if seasonal average temperature and
average barometric pressure for the sampling site
are incorporated into the sampler calibration, under
section 8.2.4 of this appendix. If individual or daily
temperature and pressure corrections are required,
ambient temperature and barometric pressure can
be obtained by on-site measurements or from a
nearby weather station. Barometric pressure
readings obtained from airports must be station
pressure, not corrected to sea level, and may need
to be corrected for differences in elevation between
the sampling site and the airport.
9.7 If the flow rate is outside the acceptable
range specified by the manufacturer, check for
leaks, and if necessary, adjust the flow rate to the
specified setpoint. Stop the sampler.
9.8 Set the timer to start and stop the sampler
at appropriate times. Set the elapsed time meter to
zero or record the initial meter reading.
9.9 Record the sample information (site
location or identification number, sample date,
filter identification number, and sampler model and
serial number).
9.10 Sample for 24±1 hours.
9.11 Determine and record the average flow
rate (Qa) in actual m3/min for the sampling period
in accordance with the instructions provided in the
sampler manufacturer's instruction manual. Record
the elapsed time meter final reading and, if needed,
the average ambient temperature and barometric
pressure for the sampling period, in note following
section 9.6 of this appendix.
9.12 Carefully remove the filter from the
sampler, following the sampler manufacturer's
instruction manual. Touch only the outer edges of
the filter.
9.13 Place the filter in a protective holder or
container, e.g., petri dish, glassine envelope, or
manila folder.
9.14 Record any factors such as meteorological
conditions, construction activity, fires or dust
storms, etc., that might be pertinent to the
measurement on the filter information record.
9.15 Transport the exposed sample filter to the
filter conditioning environment as soon as possible
for equilibration and subsequent weighing.
9.16 Equilibrate the exposed filter in the
conditioning environment for at least 24 hours
under the same temperature and humidity
conditions used for presampling filter equilibration
(see section 9.3 of this appendix).
9.17 Immediately after equilibration, reweigh
the filter and record the postsampling weight with
the filter identification number.
-------
98
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
10.0 Sampler Maintenance.
10.1 The PMio sampler shall be maintained in
strict accordance with the maintenance procedures
specified in the sampler manufacturer's instruction
manual.
11.0 Calculations.
11.1 Calculate the total volume of air sampled
as:
V = Qat
where:
V = total air sampled, at ambient temperature and
pressure,m3;
Qa = average sample flow rate at ambient
temperature and pressure, m3/min; and
t = sampling time, min.
11.2 (a) Calculate the PMio concentration as:
PMio = (Wf-Wi)xl06/V
where:
PMio = mass concentration of PMio, l-lg/m3;
Wf, Wi = final and initial weights of filter
collecting PMio particles, g; and
10s = conversion of g to |lg.
(b) Note: If more than one size fraction in the
PMio size range is collected by the sampler, the
sum of the net weight gain by each collection filter
p(Wf—Wi)] is used to calculate the PMio mass
concentration.
12.0 References.
1. Quality Assurance Handbook for Air Pollution
Measurement Systems, Volume I, Principles. EPA-
600/9-76-005, March 1976. Available from CERI,
ORD Publications, U.S. Environmental Protection
Agency, 26 West St. Clair Street, Cincinnati, OH
45268.
2. Quality Assurance Handbook for Air Pollution
Measurement Systems, Volume II, Ambient Air
Specific Methods. EPA-600/4-77-027a, May
1977. Available from CERI, ORD Publications,
U.S. Environmental Protection Agency, 26 West St.
Clair Street, Cincinnati, OH 45268.
3. Clement, R.E., and F.W. Karasek. Sample
Composition Changes in Sampling and Analysis of
Organic Compounds in Aerosols. Int. J. Environ.
Analyt. Chem., 7:109, 1979.
4. Lee, R.E., Jr., and J. Wagman. A Sampling
Anomaly in the Determination of Atmospheric
Sulfate Concentration. Amer. Ind. Hyg. Assoc. J.,
27:266, 1966.
5. Appel, B.R., S.M. Wall, Y. Tokiwa, and M.
Haik. Interference Effects in Sampling Particulate
Nitrate in Ambient Air. Atmos. Environ., 13:319,
1979.
6. Coutant, R.W. Effect of Environmental
Variables on Collection of Atmospheric Sulfate.
Environ. Sci. Technol., 11:873, 1977.
7. Spicer, C.W., and P. Schumacher. Interference
in Sampling Atmospheric Particulate Nitrate.
Atmos. Environ., 11:873, 1977.
8. Appel, B.R., Y. Tokiwa, and M. Haik.
Sampling of Nitrates in Ambient Air. Atmos.
Environ., 15:283, 1981.
9. Spicer, C.W., and P.M. Schumacher.
Particulate Nitrate: Laboratory and Field Studies of
Major Sampling Interferences. Atmos. Environ.,
13:543, 1979.
10. Appel, B.R. Letter to Larry Purdue, U.S.
EPA, Environmental Monitoring and Support
Laboratory. March 18, 1982, Docket No. A-82-37,
II-I-l.
11. Pierson, W.R., W.W. Brachaczek, T.J.
Korniski, T.J. Truex, and J.W. Butler. Artifact
Formation of Sulfate, Nitrate, and Hydrogen Ion on
Backup Filters: Allegheny Mountain Experiment. J.
Air Pollut. Control Assoc., 30:30, 1980.
12. Dunwoody, C.L. Rapid Nitrate Loss From
PMio Filters. J. Air Pollut. Control Assoc., 36:817,
1986.
13. Harrell, R.M. Measuring the Alkalinity of
Hi-Vol Air Filters. EMSL/RTP-SOP-QAD-534,
October 1985. Available from the U.S.
Environmental Protection Agency, EMSL/QAD,
Research Triangle Park, NC 27711.
14. Smith, F., P.S. Wohlschlegel, R.S.C. Rogers,
and D.J. Mulligan. Investigation of Flow Rate
Calibration Procedures Associated With the High
Volume Method for Determination of Suspended
Particulates. EPA-600/4-78-047, U.S.
Environmental Protection Agency, Research
Triangle Park, NC 27711, 1978.
8. Appendix N is added to read as follows:
Appendix N to Part 50—Interpretation of
the National Ambient Air Quality
Standards for Particulate Matter
1.0 General.
(a) This appendix explains the data handling
conventions and computations necessary for
determining when the annual and 24-hour primary
and secondary national ambient air quality
standards for PM specified in § 50.7 of this chapter
are met. Particulate matter is measured in the
ambient air as PMio and PMi.s (particles with an
aerodynamic diameter less than or equal to a
nominal 10 and 2.5 micrometers, respectively) by
a reference method based on Appendix M of this
part for PMio and on Appendix L of this part for
PM2.5, as applicable, and designated in accordance
with part 53 of this chapter, or by an equivalent
method designated in accordance with part 53 of
this chapter. Data handling and computation
procedures to be used in making comparisons
between reported PMio and PMi.s concentrations
and the levels of the PM standards are specified
in the following sections.
(b) Data resulting from uncontrollable or natural
events, for example structural fires or high winds,
may require special consideration. In some cases,
it may be appropriate to exclude these data because
they could result in inappropriate values to compare
with the levels of the PM standards. In other cases,
it may be more appropriate to retain the data for
comparison with the level of the PM standards and
then allow the EPA to formulate the appropriate
regulatory response. Whether to exclude, retain, or
make adjustments to the data affected by
uncontrollable or natural events is subject to the
approval of the appropriate Regional Administrator.
(c) The terms used in this appendix are defined
as follows:
Average and mean refer to an arithmetic mean.
Daily value for PM refers to the 24-hour average
concentration of PM calculated or measured from
midnight to midnight (local time) for PMio or
PM2.5.
Designated monitors are those monitoring sites
designated in a State PM Monitoring Network
Description for spatial averaging in areas opting for
spatial averaging in accordance with part 58 of this
chapter.
98th percentile (used for PM2.s) means the daily
value out of a year of monitoring data below which
98 percent of all values in the group fall.
99thpercentile (used for PMio) means the daily
value out of a year of monitoring data below which
99 percent of all values in the group fall.
Year refers to a calendar year.
(d) Sections 2.1 and 2.5 of this appendix contain
data handling instructions for the option of using
a spatially averaged network of monitors for the
annual standard. If spatial averaging is not
considered for an area, then the spatial average is
equivalent to the annual average of a single site and
is treated accordingly in subsequent calculations.
For example, paragraph (a)(3) of section 2.1 of this
appendix could be eliminated since the spatial
average would be equivalent to the annual average.
2.0 Comparisons with thePM^s Standards.
2.1 Annual PM2.5 Standard.
(a) The annual PM2.5 standard is met when the
3-year average of the spatially averaged annual
means is less than or equal to 15.0 |lg/m3. The 3-
year average of the spatially averaged annual
means is determined by averaging quarterly means
at each monitor to obtain the annual mean PMi.s
concentrations at each monitor, then averaging
across all designated monitors, and finally
averaging for 3 consecutive years. The steps can
be summarized as follows:
(1) Average 24-hour measurements to obtain
quarterly means at each monitor.
(2) Average quarterly means to obtain annual
means at each monitor.
(3) Average across designated monitoring sites
to obtain an annual spatial mean for an area (this
can be one site in which case the spatial mean is
equal to the annual mean).
(4) Average 3 years of annual spatial means to
obtain a 3-year average of spatially averaged
annual means.
(b) In the case of spatial averaging, 3 years of
spatial averages are required to demonstrate that the
standard has been met. Designated sites with less
than 3 years of data shall be included in spatial
averages for those years that data completeness
requirements are met. For the annual PMi.s
standard, a year meets data completeness
requirements when at least 75 percent of the
scheduled sampling days for each quarter have
valid data. However, years with high concentrations
and more than a minimal amount of data (at least
11 samples in each quarter) shall not be ignored
just because they are comprised of quarters with
less than complete data. Thus, in computing annual
spatially averaged means, years containing quarters
with at least 11 samples but less than 75 percent
data completeness shall be included in the
computation if the resulting spatially averaged
annual mean concentration (rounded according to
the conventions of section 2.3 of this appendix) is
greater than the level of the standard.
(c) Situations may arise in which there are
compelling reasons to retain years containing
quarters which do not meet the data completeness
requirement of 75 percent or the minimum number
of 11 samples. The use of less than complete data
is subject to the approval of the appropriate
Regional Administrator.
(d) The equations for calculating the 3-year
average annual mean of the PMi.s standard are
given in section 2.5 of this appendix.
2.2 24-Hour PM2.5 Standard.
(a) The 24-hour PMi.s standard is met when the
3-year average of the 98th percentile values at each
monitoring site is less than or equal to 65 |lg/m3.
This comparison shall be based on 3 consecutive,
complete years of air quality data. A year meets
data completeness requirements when at least 75
percent of the scheduled sampling days for each
quarter have valid data. However, years with high
concentrations shall not be ignored just because
they are comprised of quarters with less than
complete data. Thus, in computing the 3-year
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
99
average 98th percentile value, years containing
quarters with less than 75 percent data
completeness shall be included in the computation
if the annual 98th percentile value (rounded
according to the conventions of section 2.3 of this
appendix) is greater than the level of the standard.
(b) Situations may arise in which there are
compelling reasons to retain years containing
quarters which do not meet the data completeness
requirement. The use of less than complete data is
subject to the approval of the appropriate Regional
Administrator.
(c) The equations for calculating the 3-year
average of the annual 98* percentile values is
given in section 2.6 of this appendix.
2.3 Rounding Conventions. For the purposes of
comparing calculated values to the applicable level
of the standard, it is necessary to round the final
results of the calculations described in sections 2.5
and 2.6 of this appendix. For the annual PMi.s
standard, the 3-year average of the spatially
averaged annual means shall be rounded to the
nearest 0.1 |lg/m3 (decimals 0.05 and greater are
rounded up to the next 0.1, and any decimal lower
than 0.05 is rounded down to the nearest 0.1). For
the 24-hour PMi.5 standard, the 3-year average of
the annual 98th percentile values shall be rounded
to the nearest 1 |lg/m3 (decimals 0.5 and greater
are rounded up to nearest whole number, and any
decimal lower than 0.5 is rounded down to the
nearest whole number).
2.4 Monitoring Considerations.
(a) Section 58.13 of this chapter specifies the
required minimum frequency of sampling for
PMi.s. Exceptions to the specified sampling
frequencies, such as a reduced frequency during a
season of expected low concentrations, are subject
to the approval of the appropriate Regional
Administrator. Section 58.14 of 40 CFR part 58
and section 2.8 of Appendix D of 40 CFR part 58,
specify which monitors are eligible for making
comparisons with the PM standards. In determining
a spatial mean using two or more monitoring sites
operating in a given year, the annual mean for an
individual site may be included in the spatial mean
if and only if the mean for that site meets the
criterion specified in §2.8 of Appendix D of 40
CFR part 58. In the event data from an otherwise
eligible site is excluded from being averaged with
data from other sites on the basis of this criterion,
then the 3-year mean from that site shall be
compared directly to the annual standard.
(b) For the annual PMi.5 standard, when
designated monitors are located at the same site and
are reporting PMi.s values for the same time
periods, and when spatial averaging has been
chosen, their concentrations shall be averaged
before an area-wide spatial average is calculated.
Such monitors will then be considered as one
monitor.
2.5 Equations for the Annual PM2.5 Standard.
(a) An annual mean value for PM2.5 is
determined by first averaging the daily values of
a calendar quarter:
Equation 1
1 "l
— 1 V
X = > X
q.y.s ^-i i.q.y.i
nq 1=1
q.y.s
"q 1=1
where:
xq,y,s = the mean for quarter q of year y for site
s;
nq = the number of monitored values in the quarter;
and
Xi.q.y.s = the i* value in quarter q for year y for
site s.
(b) The following equation is then to be used for
calculation of the annual mean:
Equation 2
where:
xy>s = the annual mean concentration for year y (y
= 1, 2, or 3) and for site s; and
Xq.y.s = the mean for quarter q of year y for site
(c) (1) The spatially averaged annual mean for
year y is computed by first calculating the annual
mean for each site designated to be included in a
spatial average, xy>s, and then computing the
average of these values across sites:
Equation 3
s=l
where:
Xy = the spatially averaged mean for year y;
xy>s = the annual mean for year y and site s; and
ns = the number of sites designated to be averaged.
(2) In the event that an area designated for
spatial averaging has two or more sites at the same
location producing data for the same time periods,
the sites are averaged together before using
Equation 3 by:
Equation 4
Table 1.—Results from Equations 1 and 2
1
_
Xs* —
where:
xy,s* = the annual mean for year y for the sites at
the same location (which will now be
considered one site);
nc = the number of sites at the same location
designated to be included in the spatial
average; and
xy>s = the annual mean for year y and site s.
(d) The 3-year average of the spatially averaged
annual means is calculated by using the following
equation:
Equation 5
X = —
X,.
where:
x = the 3-year average of the spatially averaged
annual means; and
xy = the spatially averaged annual mean for year
y.
Example 1 — Area Designated for Spatial Averaging
That Meets the Primary Annual PM?.. 5 Standard.
a. In an area designated for spatial averaging,
four designated monitors recorded data in at least
1 year of a particular 3-year period. Using
Equations 1 and 2, the annual means for PM2.5 at
each site are calculated for each year. The
following table can be created from the results.
Data completeness percentages for the quarter with
the fewest number of samples are also shown.
Year 1
Year 2
YearS
3-year mean
Annual mean (|ig/m3)
% data completeness
Annual mean (|ig/m3)
% data completeness
Annual mean (|ig/m3)
% data completeness
Site #1
127
80
12.6
90
125
90
Site #2
0
17.5
63
185
80
Site #3
0
15.2
38
14 1
85
Site #4
0
0
169
50
Spatial mean
127
15.05
1550
14.42
b. The data from these sites are averaged in the
order described in section 2.1 of this appendix.
Note that the annual mean from site #3 in year 2
and the annual mean from site #4 in year 3 do not
meet the 75 percent data completeness criteria.
Assuming the 38 percent data completeness
represents a quarter with fewer than 11 samples,
site #3 in year 2 does not meet the minimum data
completeness requirement of 11 samples in each
quarter. The site is therefore excluded from the
calculation of the spatial mean for year 2. However,
since the spatial mean for year 3 is above the level
of the standard and the minimum data requirement
of 11 samples in each quarter has been met, the
annual mean from site #4 in year 3 is included in
the calculation of the spatial mean for year 3 and
in the calculation of the 3-year average. The 3-year
average is rounded to 14.4 |lg/m3, indicating that
this area meets the annual PMi.s standard.
-------
100
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
Example 2—Area With Two Monitors at the Same
Location That Meets the Primary Annual PM^.5
Standard.
a. In an area designated for spatial averaging, six
designated monitors, with two monitors at the same
location (#5 and #6), recorded data in a particular
3-year period. Using Equations 1 and 2, the annual
means for PMi.s are calculated for each year. The
following table can be created from the results.
Table 2.—Results From Equations 1 and 2
Annual mean (|ig/m3)
Year 1
Year 2
YearS
3-Year mean
Site #1
129
145
14.4
Site #2
99
133
12.4
Site #3
126
122
11.5
Site #4
11 1
109
9.7
Site #5
145
16 1
12.3
Site #6
146
160
12.1
Average of
#5 and #6
1455
1605
12.20
Spatial
mean
1221
1339
12.04
12.55
b. The annual means for sites #5 and #6 are
averaged together using Equation 4 before the
spatial average is calculated using Equation 3 since
they are in the same location. The 3-year mean is
rounded to 12.6 |lg/m3, indicating that this area
meets the annual PM2.5 standard.
Example 3—Area With a Single Monitor That
Meets the Primary Annual PM2.s Standard.
a. Given data from a single monitor in an area,
the calculations are as follows. Using Equations 1
and 2, the annual means for PM2.5 are calculated
for each year. If the annual means are 10.28, 17.38,
and 12.25 |lg/m3, then the 3-year mean is:
x = (I/3)x(10.28 + 17.38 + 12.25) = 13.303
b. This value is rounded to 13.3, indicating that
this area meets the annual PMi.5 standard.
2.6 Equations for the 24-Hour PMi.s Standard.
(a) When the data for a particular site and year
meet the data completeness requirements in section
2.2 of this appendix, calculation of the 98th
percentile is accomplished by the following steps.
All the daily values from a particular site and year
comprise a series of values (xi, X2, X3, ..., xn), that
can be sorted into a series where each number is
equal to or larger than the preceding number (x[ij,
X[2], xpj, ..., X[n]). In this case, xpj is the smallest
number and X[n] is the largest value. The 98*
percentile is found from the sorted series of daily
values which is ordered from the lowest to the
highest number. Compute (0.98) X (n) as the
number "i.d", where "i" is the integer part of the
result and "d" is the decimal part of the result.
The 98th percentile value for year y, Po.98, y, is
given by Equation 6:
Equation 6
where:
Po.98,y = 98th percentile for year y;
X[i+ij = the (i+l)* number in the ordered series of
numbers; and
i = the integer part of the product of 0.98 and n.
(b) The 3-year average 98th percentile is then
calculated by averaging the annual 98th percentiles:
Equation 7
(c) The 3-year average 98th percentile is rounded
according to the conventions in section 2.3 of this
appendix before a comparison with the standard is
made.
Example 4—Ambient Monitoring Site WithEvery-
Day Sampling That Meets the Primary 24-Hour
PM2.s Standard.
a. In each year of a particular 3 year period,
varying numbers of daily PM2.5 values (e.g., 281,
304, and 296) out of a possible 365 values were
recorded at a particular site with the following
ranked values (in |lg/m3):
Table 3.—Ordered Monitoring Data For 3 Years
Year!
j rank
275
276
277
Xj value
57.9
59.0
62.2
Year 2
j rank
296
297
298
Xj value
54.3
57.1
63.0
YearS
j rank
290
291
292
Xj value
66.0
68.4
69.8
b. Using Equation 6, the 98* percentile values
for each year are calculated as follows:
0.98 x 281 = 275.38 =>/ + ! = 276 => P098?1 = X[276]= 59.0/^g/m
0.98 x 304 = 297.92 => / +1 = 298 => P,
0.98,2
-------
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
101
0.98 x 296 = 290.07 =>/ + ! = 291 => P0983 =X[29l] =68.4/ug/
c. 1. Using Equation 7, the 3-year average 98th
percentile is calculated as follows:
P —
•'0.98 ~~
59.0 + 63.0 + 68.4
= 63.46/ig / m3, which rounds to 63/ug I m3
2. Therefore, this site meets the 24-hour PMi.s
standard.
3.0 Comparisons with the PMio Standards.
3.1 AnnualPM\o Standard.
(a) The annual PMio standard is met when the
3-year average of the annual mean PMio
concentrations at each monitoring site is less than
or equal to 50 |lg/m3. The 3-year average of the
annual means is determined by averaging quarterly
means to obtain annual mean PMio concentrations
for 3 consecutive, complete years at each
monitoring site. The steps can be summarized as
follows:
(1) Average 24-hour measurements to obtain a
quarterly mean.
(2) Average quarterly means to obtain an annual
mean.
(3) Average annual means to obtain a 3-year
mean.
(b) For the annual PMio standard, a year meets
data completeness requirements when at least 75
percent of the scheduled sampling days for each
quarter have valid data. However, years with high
concentrations and more than a minimal amount of
data (at least 11 samples in each quarter) shall not
be ignored just because they are comprised of
quarters with less than complete data. Thus, in
computing the 3-year average annual mean
concentration, years containing quarters with at
least 11 samples but less than 75 percent data
completeness shall be included in the computation
if the annual mean concentration (rounded
according to the conventions of section 2.3 of this
appendix) is greater than the level of the standard.
(c) Situations may arise in which there are
compelling reasons to retain years containing
quarters which do not meet the data completeness
requirement of 75 percent or the minimum number
of 11 samples. The use of less than complete data
is subject to the approval of the appropriate
Regional Administrator.
(d) The equations for calculating the 3-year
average annual mean of the PMio standard are
given in section 3.5 of this appendix.
3.2 24-Hour PMio Standard.
(a) The 24-hour PMio standard is met when the
3-year average of the annual 99th percentile values
at each monitoring site is less than or equal to 150
|lg/m3. This comparison shall be based on 3
consecutive, complete years of air quality data. A
year meets data completeness requirements when at
least 75 percent of the scheduled sampling days for
each quarter have valid data. However, years with
high concentrations shall not be ignored just
because they are comprised of quarters with less
than complete data. Thus, in computing the 3-year
average of the annual 99th percentile values, years
containing quarters with less than 75 percent data
completeness shall be included in the computation
if the annual 99th percentile value (rounded
according to the conventions of section 2.3 of this
appendix) is greater than the level of the standard.
(b) Situations may arise in which there are
compelling reasons to retain years containing
quarters which do not meet the data completeness
requirement. The use of less than complete data is
subject to the approval of the appropriate Regional
Administrator.
(c) The equation for calculating the 3-year
average of the annual 99th percentile values is
given in section 2.6 of this appendix.
3.3 Rounding Conventions. For the annual PMio
standard, the 3-year average of the annual PMio
means shall be rounded to the nearest 1 |lg/m3
(decimals 0.5 and greater are rounded up to the
next whole number, and any decimal less than 0.5
is rounded down to the nearest whole number). For
the 24-hour PMio standard, the 3-year average of
the annual 99th percentile values of PMio shall be
rounded to the nearest 10 |lg/m3 (155 |lg/m3 and
greater would be rounded to 160 |lg/m3 and 154
|lg/m3 and less would be rounded to 150 |lg/m3).
3.4 Monitoring Considerations. Section 58.13 of
this chapter specifies the required minimum
frequency of sampling for PMio. Exceptions to the
specified sampling frequencies, such as a reduced
frequency during a season of expected low
concentrations, are subject to the approval of the
appropriate Regional Administrator. For making
comparisons with the PMio NAAQS, all sites
meeting applicable requirements in part 58 of this
chapter would be used.
3.5 Equations for the Annual PMio Standard.
(a) An annual arithmetic mean value for PMio
is determined by first averaging the 24-hour values
of a calendar quarter using the following equation:
Equation 8
T = > T
q,y „ ^-i t,q,y
Hq i=\
where:
xq,y = the mean for quarter q of year y;
nq = the number of monitored values in the quarter;
and
Xi.q.y = the i* value in quarter q for year y.
(b) The following equation is then to be used for
calculation of the annual mean:
Equation 9
T = — > T
y A i-i q,y
4 9=1
where:
xy = the annual mean concentration for year y,
(y=l, 2, or 3); and
xq_y = the mean for a quarter q of year y.
(c) The 3-year average of the annual means is
calculated by using the following equation:
Equation 10
where:
x = the 3-year average of the annual means; and
xy = the annual mean for calendar year y.
Example 5—Ambient Monitoring Site That Does
Not Meet the Annual PMio Standard.
a. Given data from a PMio monitor and using
Equations 8 and 9, the annual means for PMio are
calculated for each year. If the annual means are
52.42, 82.17, and 63.23 |lg/m3, then the 3-year
average annual mean is:
5? = (1/3) x (52.42 + 82.17 + 63.23) = 65.94, which is rounded to 66
b. Therefore, this site does not meet the annual
PMio standard.
3.6 Equation for the 24-Hour PMio Standard.
(a) When the data for a particular site and year
meet the data completeness requirements in section
3.2 of this appendix, calculation of the 99th
percentile is accomplished by the following steps.
All the daily values from a particular site and year
comprise a series of values (xi, X2, X3, ..., xn) that
can be sorted into a series where each number is
equal to or larger than the preceding number (xpj,
xp], X[3j, ..., X[nj). In this case, X[i] is the smallest
number and x[n] is the largest value. The 99th
percentile is found from the sorted series of daily
values which is ordered from the lowest to the
highest number. Compute (0.99) x (n) as the
number "i.d", where "i" is the integer part of the
result and "d" is the decimal part of the result.
The 99th percentile value for year y, Po.99,y, is
given by Equation 11:
Equation 11
-MX99,y = X[i+l]
where:
Po.99,y = the 99th percentile for year y;
-------
102
Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication
xp+i] = the (i+1)* number in the ordered series of
numbers; and
i = the integer part of the product of 0.99 and n.
(b) The 3-year average 99th percentile value is
then calculated by averaging the annual 99*
percentiles:
Equation 12
().99,y
Example 6—Ambient Monitoring Site With
Sampling Every Sixth Day That Meets the Primary
24-Hour PMw Standard.
a. In each year of a particular 3 year period,
p _ y=1 varying numbers of PMio daily values (e.g., 110,
099 3 98, and 100) out of a possible 121 daily values
(c) The 3-year average 99* percentile is rounded were recorded at a particular site with the following
according to the conventions in section 3.3 of this ranked values (in |lg/m3):
appendix before a comparison with the standard is
made.
Table 4.—Ordered Monitoring Data For 3 Years
YeaM
j rank
108
109
110
Xj value
120
128
130
Year 2
j rank
96
97
98
Xj value
143
148
150
YearS
j rank
98
99
100
Xj value
140
144
147
b. Using Equation 11, the 99th percentile values
for each year are calculated as follows:
0.99 x 110 = 108.9 =>/ + ! =109:
P0 99 j = X[W9] = 128/ig / m *
0.99 x 98 = 97.02 => / +1 = 98 => P0 99 2 = X[9S] = 150/ug I m
0.99 x 100 = 99 => i +1 = 100 => P0 99 3 = X[wo] = 147'/j.g I m
c. 1. Using Equation 12, the 3-year average 99th
percentile is calculated as follows:
128 + 50 + 147
= 141.7 /ig / m3 rounds to 140/ig / m3
2. Therefore, this site meets the 24-hour PMio
standard.
[FR Doc. 97-18577 Filed 7-17-97; 8:45 am]
BILLING CODE 6560-50-F
-------
EPA-454/B-95-003b
USER'S GUIDE FOR THE
INDUSTRIAL SOURCE COMPLEX (ISC3) DISPERSION MODELS
VOLUME II - DESCRIPTION OF MODEL ALGORITHMS
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air Quality Planning and Standards
Emissions, Monitoring, and Analysis Division
Research Triangle Park, North Carolina 27711
September 1995
-------
DISCLAIMER
The information in this document has been reviewed in its
entirety by the U.S. Environmental Protection Agency (EPA), and
approved for publication as an EPA document. Mention of trade
names, products, or services does not convey, and should not be
interpreted as conveying official EPA endorsement, or
recommendation.
11
-------
PREFACE
This User's Guide provides documentation for the
Industrial Source Complex (ISC3) models, referred to hereafter
as the Short Term (ISCST3) and Long Term (ISCLT3) models. This
volume describes the dispersion algorithms utilized in the
ISCST3 and ISCLT3 models, including the new area source and dry
deposition algorithms, both of which are a part of Supplement C
to the Guideline on Air Quality Models (Revised).
This volume also includes a technical description for the
following algorithms that are not included in Supplement C:
pit retention (ISCST3 and ISCLT3), wet deposition (ISCST3
only), and COMPLEXl (ISCST3 only). The pit retention and wet
deposition algorithms have not undergone extensive evaluation
at this time, and their use is optional. COMPLEXl is
incorporated to provide a means for conducting screening
estimates in complex terrain. EPA guidance on complex terrain
screening procedures is provided in Section 5.2.1 of the
Guideline on Air Quality Models (Revised).
Volume I of the ISC3 User's Guide provides user
instructions for the ISC3 models.
111
-------
ACKNOWLEDGEMENTS
The User's Guide for the ISC3 Models has been prepared by
Pacific Environmental Services, Inc., Research Triangle Park,
North Carolina. This effort has been funded by the
Environmental Protection Agency (EPA) under Contract No. 68-
D30032, with Desmond T. Bailey as Work Assignment Manager
(WAM). The technical description for the dry deposition
algorithm was developed from material prepared by Sigma
Research Corporation and funded by EPA under Contract No. 68-
D90067, with Jawad S. Touma as WAM.
IV
-------
CONTENTS
PREFACE ill
ACKNOWLEDGEMENTS iv
FIGURES vii
TABLES viii
SYMBOLS ix
1.0 THE ISC SHORT-TERM DISPERSION MODEL EQUATIONS 1-1
1.1 POINT SOURCE EMISSIONS 1-2
I.I.I The Gaussian Equation 1-2
1.1.2 Downwind and Crosswind Distances 1-3
1.1.3 Wind Speed Profile 1-4
1.1.4 Plume Rise Formulas 1-5
1.1.5 The Dispersion Parameters 1-14
1.1.6 The Vertical Term 1-31
1.1.7 The Decay Term 1-42
1.2 NON-POINT SOURCE EMISSIONS 1-43
1.2.1 General 1-43
1.2.2 The Short-Term Volume Source Model . . . 1-43
1.2.3 The Short-Term Area Source Model .... 1-46
1.2.4 The Short-Term Open Pit Source Model . . 1-50
1.3 THE ISC SHORT-TERM DRY DEPOSITION MODEL .... 1-54
1.3.1 General 1-54
1.3.2 Deposition Velocities 1-55
1.3.3 Point and Volume Source Emissions .... 1-60
1.3.4 Area and Open Pit Source Emissions . . . 1-61
1.4 THE ISC SHORT-TERM WET DEPOSITION MODEL .... 1-61
1.5 ISC COMPLEX TERRAIN SCREENING ALGORITHMS .... 1-63
1.5.1 The Gaussian Sector Average Equation . . 1-63
1.5.2 Downwind, Crosswind and Radial Distances 1-65
1.5.3 Wind Speed Profile 1-65
1.5.4 Plume Rise Formulas 1-65
1.5.5 The Dispersion Parameters 1-66
1.5.6 The Vertical Term 1-66
1.5.7 The Decay Term 1-68
1.5.8 The Plume Attenuation Correction Factor . 1-68
1.5.9 Wet Deposition in Complex Terrain . . . 1-69
1.6 ISC TREATMENT OF INTERMEDIATE TERRAIN 1-69
2.0 THE ISC LONG-TERM DISPERSION MODEL EQUATIONS 2-1
2.1 POINT SOURCE EMISSIONS 2-1
2.1.1 The Gaussian Sector Average Equation . . .2-1
2.1.2 Downwind and Crosswind Distances 2-3
2.1.3 Wind Speed Profile 2-3
2.1.4 Plume Rise Formulas 2-3
2.1.5 The Dispersion Parameters 2-4
2.1.6 The Vertical Term 2-5
2.1.7 The Decay Term 2-6
v
-------
2.1.8 The Smoothing Function 2-6
2.2 NON-POINT SOURCE EMISSIONS 2-7
2.2.1 General 2-7
2.2.2 The Long-Term Volume Source Model 2-7
2.2.3 The Long-Term Area Source Model 2-7
2.2.4 The Long-Term Open Pit Source Model . . . 2-11
2.3 THE ISC LONG-TERM DRY DEPOSITION MODEL 2-11
2.3.1 General 2-11
2.3.2 Point and Volume Source Emissions .... 2-11
2.3.3 Area and Open Pit Source Emissions ... 2-12
3.0 REFERENCES 3-1
INDEX INDEX-1
VI
-------
FIGURES
Figure Page
1-1 LINEAR DECAY FACTOR, A AS A FUNCTION OF EFFECTIVE
STACK HEIGHT, He. A SQUAT BUILDING IS ASSUMED FOR
SIMPLICITY 1-71
1-2 ILLUSTRATION OF TWO TIERED BUILDING WITH DIFFERENT
TIERS DOMINATING DIFFERENT WIND DIRECTIONS .... 1-72
1-3 THE METHOD OF MULTIPLE PLUME IMAGES USED TO SIMULATE
PLUME REFLECTION IN THE ISC MODEL 1-73
1-4 SCHEMATIC ILLUSTRATION OF MIXING HEIGHT INTERPOLATION
PROCEDURES 1-74
1-5 ILLUSTRATION OF PLUME BEHAVIOR IN COMPLEX TERRAIN
ASSUMED BY THE ISC MODEL 1-75
1-6 ILLUSTRATION OF THE DEPLETION FACTOR FQ AND THE
CORRESPONDING PROFILE CORRECTION FACTOR P(x,z). . 1-76
1-7 VERTICAL PROFILE OF CONCENTRATION BEFORE AND AFTER
APPLYING FQ AND P(x,z) SHOWN IN FIGURE 1-6 . . . . 1-77
1-8 EXACT AND APPROXIMATE REPRESENTATION OF LINE SOURCE BY
MULTIPLE VOLUME SOURCES 1-78
1-9 REPRESENTATION OF AN IRREGULARLY SHAPED AREA SOURCE
BY 4 RECTANGULAR AREA SOURCES 1-79
1-10 EFFECTIVE AREA AND ALONGWIND LENGTH FOR AN OPEN PIT
SOURCE 1-80
1-11 WET SCAVENGING RATE COEFFICIENT AS A FUNCTION OF PARTICLE
SIZE (JINDAL & HEINOLD, 1991) 1-81
VI1
-------
TABLES
Table Page
1-1 PARAMETERS USED TO CALCULATE PASQUILL-GIFFORD •• . . 1-16
1-2 PARAMETERS USED TO CALCULATE PASQUILL-GIFFORD •• . . 1-17
1-3 BRIGGS FORMULAS USED TO CALCULATE McELROY-POOLER •• 1-19
1-4 BRIGGS FORMULAS USED TO CALCULATE McELROY-POOLER •• 1-19
1-5 COEFFICIENTS USED TO CALCULATE LATERAL VIRTUAL
DISTANCES FOR PASQUILL-GIFFORD DISPERSION RATES . . 1-21
1-6 SUMMARY OF SUGGESTED PROCEDURES FOR ESTIMATING
INITIAL LATERAL DIMENSIONS • »0 AND INITIAL VERTICAL
DIMENSIONS '10 FOR VOLUME AND LINE SOURCES 1-46
VI11
-------
SYMBOLS
Symbol Definition
A Linear decay term for vertical dispersion in
Schulman-Scire downwash (dimensionless)
Ae Effective area for open pit emissions (dimensionless)
D Exponential decay term for Gaussian plume equation
(dimensionless)
DB Brownian diffusivity (cm/s)
Dr Relative pit depth (dimensionless)
de Effective pit depth (m)
dp Particle diameter for particulate emissions (urn)
ds Stack inside diameter (m)
Fb Buoyancy flux parameter (m4/s3)
Fd Dry deposition flux (g/m2)
Fm Momentum flux parameter (m4/s2)
FQ Plume depletion factor for dry deposition
(dimensionless)
FT Terrain adjustment factor (dimensionless)
Fw Wet deposition flux (g/m2)
f Frequency of occurrence of a wind speed and stability
category combination (dimensionless)
g Acceleration due to gravity (9.80616 m/s2)
hb Building height (m)
he Plume (or effective stack) height (m)
hs Physical stack height (m)
hter Height of terrain above stack base (m)
hs' Release height modified for stack-tip downwash (m)
IX
-------
hw Crosswind projected width of building adjacent to a
stack (m)
k von Karman constant (= 0.4)
L Monin-Obukhov length (m)
Ly Initial plume length for Schulman-Scire downwash
sources with enhanced lateral plume spread (m)
Lb Lesser of the building height and crosswind projected
building width (m)
0 Alongwind length of open pit source (m)
P(x,y) Profile adjustment factor (dimensionless)
p Wind speed power law profile exponent (dimensionless)
QA Area Source pollutant emission rate (g/s)
Qe Effective emission rate for effective area source for
an open pit source (g/s)
Q± Adjusted emission rate for particle size category for
open pit emissions (g/s)
Qs Pollutant emission rate (g/s)
Q.. Total amount of pollutant emitted during time period •
(g)
R Precipitation rate (mm/hr)
R0 Initial plume radius for Schulman-Scire downwash
sources (m)
R(z,zd) Atmospheric resistance to vertical transport (s/cm)
r Radial distance range in a polar receptor network (m)
ra Atmospheric resistance (s/cm)
rd Deposition layer resistance (s/cm)
^i
Stability parameter = 9
Smoothing term for smoothing across adjacent sectors in
the Long Term model (dimensionless)
x
-------
SCF Splip correction factor (dimensionless)
SG Schmidt number = */DB (dimensionless)
Of- 9
Stokes number = (V /g) (u../*} (dimensionless)
&
Ta Ambient temperature (K)
Ts Stack gas exit temperature (K)
uref Wind speed measured at reference anemometer height
(m/s)
us Wind speed adjusted to release height (m/s)
u. Surface friction velocity (m/s)
V Vertical term of the Gaussian plume equation
(dimensionless)
Vd Vertical term with dry deposition of the Gaussian plume
equation (dimensionless)
vd Particle deposition velocity (cm/s)
vg Gravitational settling velocity for particles (cm/s)
vs Stack gas exit velocity (m/s)
X X-coordinate in a Cartesian grid receptor network (m)
x0 Length of side of square area source (m)
Y Y-coordinate in a Cartesian grid receptor network (m)
•• Direction in a polar receptor network (degrees)
x Downwind distance from source to receptor (m)
xy Lateral virtual point source distance (m)
xz Vertical virtual point source distance (m)
xf Downwind distance to final plume rise (m)
x* Downwind distance at which turbulence dominates
entrainment (m)
y Crosswind distance from source to receptor (m)
z Receptor/terrain height above mean sea level (m)
XI
-------
zd Dry deposition reference height (m)
zr Receptor height above ground level (i.e. flagpole) (m)
zref Reference height for wind speed power law (m)
zs Stack base elevation above mean sea level (m)
z± Mixing height (m)
z0 Surface roughness height (m)
•• Entrainment coefficient used in buoyant rise for
Schulman-Scire downwash sources = 0.6
•T Jet entrainment coefficient used in gradual momentum
AIR
plume rise calculations •• — •
3
Us
Plume rise (m)
Potential temperature gradient with height (K/m)
Escape fraction of particle size category for open pit
emissions (dimensionless)
Precipitation scavenging ratio (s"1)
Precipitation rate coefficient (s-mm/hr)"1
pi = 3.14159
Decay coefficient = 0.693/T1/2 (s"1)
Stability adjustment factor (dimensionless)
Fraction of mass in a particular settling velocity
category for particulates (dimensionless)
Particle density (g/cm
3^
Density of air (g/cm
3^
•• Horizontal (lateral) dispersion parameter (m)
••0 Initial horizontal dispersion parameter for virtual
point source (m)
••e Effective lateral dispersion parameter including
effects of buoyancy-induced dispersion (m)
XII
-------
•• Vertical dispersion parameter (m)
• *0 Initial vertical dispersion parameter for virtual point
source (m)
• *e Effective vertical dispersion parameter including
effects of buoyancy-induced dispersion (m)
Viscosity of air - 0.15 cm2/s
u Absolute viscosity of air - 1.81 x 10"4 g/cm/s
Concentration (ug/m3)
•3 Concentration with dry deposition effects (ug/m3)
XI11
-------
1.0 THE ISC SHORT-TERM DISPERSION MODEL EQUATIONS
The Industrial Source Complex (ISC) Short Term model
provides options to model emissions from a wide range of
sources that might be present at a typical industrial source
complex. The basis of the model is the straight-line,
steady-state Gaussian plume equation, which is used with some
modifications to model simple point source emissions from
stacks, emissions from stacks that experience the effects of
aerodynamic downwash due to nearby buildings, isolated vents,
multiple vents, storage piles, conveyor belts, and the like.
Emission sources are categorized into four basic types of
sources, i.e., point sources, volume sources, area sources, and
open pit sources. The volume source option and the area source
option may also be used to simulate line sources. The
algorithms used to model each of these source types are
described in detail in the following sections. The point
source algorithms are described in Section 1.1. The volume,
area and open pit source model algorithms are described in
Section 1.2. Section 1.3 gives the optional algorithms for
calculating dry deposition for point, volume, area and open pit
sources, and Section 1.4 describes the optional algorithms for
calculating wet deposition. Sections 1.1 through 1.4 describe
calculations for simple terrain (defined as terrain elevations
below the release height). The modifications to these
calculations to account for complex terrain are described in
Section 1.5, and the treatment of intermediate terrain is
discussed in Section 1.6.
The ISC Short Term model accepts hourly meteorological
data records to define the conditions for plume rise,
transport, diffusion, and deposition. The model estimates the
concentration or deposition value for each source and receptor
combination for each hour of input meteorology, and calculates
user-selected short-term averages. For deposition values,
either the dry deposition flux, the wet deposition flux, or the
total deposition flux may be estimated. The total deposition
1-1
-------
flux is simply the sum of the dry and wet deposition fluxes at
a particular receptor location. The user also has the option
of selecting averages for the entire period of input
meteorology.
1.1 POINT SOURCE EMISSIONS
The ISC Short Term model uses a steady-state Gaussian
plume equation to model emissions from point sources, such as
stacks and isolated vents. This section describes the Gaussian
point source model, including the basic Gaussian equation, the
plume rise formulas, and the formulas used for determining
dispersion parameters.
I.I.I The Gaussian Equation
The ISC short term model for stacks uses the steady-state
Gaussian plume equation for a continuous elevated source. For
each source and each hour, the origin of the source's
coordinate system is placed at the ground surface at the base
of the stack. The x axis is positive in the downwind
direction, the y axis is crosswind (normal) to the x axis and
the z axis extends vertically. The fixed receptor locations
are converted to each source's coordinate system for each
hourly concentration calculation. The calculation of the
downwind and crosswind distances is described in Section 1.1.2.
The hourly concentrations calculated for each source at each
receptor are summed to obtain the total concentration produced
at each receptor by the combined source emissions.
For a steady-state Gaussian plume, the hourly
concentration at downwind distance x (meters) and crosswind
distance y (meters) is given by:
QKVD
9 • 41 • • • •
^s y z
exp
(1-1)
where:
1-2
-------
Q = pollutant emission rate (mass per unit time)
K = a scaling coefficient to convert calculated
concentrations to desired units (default value of
1 x 10s for Q in g/s and concentration in ug/m3)
V = vertical term (See Section 1.1.6)
D = decay term (See Section 1.1.7)
standard deviation of lateral and vertical
concei
1.1.5;
y' z
concentration distribution (m) (See Section
us = mean wind speed (m/s) at release height (See
Section 1.1.3)
Equation (1-1) includes a Vertical Term (V), a Decay Term
(D), and dispersion parameters (•• and ••) as discussed below.
It should be noted that the Vertical Term includes the effects
of source elevation, receptor elevation, plume rise, limited
mixing in the vertical, and the gravitational settling and dry
deposition of particulates (with diameters greater than about
0.1 microns).
1.1.2 Downwind and Crosswind Distances
The ISC model uses either a polar or a Cartesian receptor
network as specified by the user. The model allows for the use
of both types of receptors and for multiple networks in a
single run. All receptor points are converted to Cartesian
(X,Y) coordinates prior to performing the dispersion
calculations. In the polar coordinate system, the radial
coordinate of the point (r, •} is measured from the
user-specified origin and the angular coordinate **is measured
clockwise from the north. In the Cartesian coordinate system,
the X axis is positive to the east of the user-specified origin
and the Y axis is positive to the north. For either type of
receptor network, the user must define the location of each
source with respect to the origin of the grid using Cartesian
coordinates. In the polar coordinate system, assuming the
1-3
-------
origin is at X = X0, Y = Y0, the X and Y coordinates of a
receptor at the point (r, •} are given by:
x( R) ••rsin»~»X0 (1-2)
Y( R) ••rcos*~*Y0 (1-3)
If the X and Y coordinates of the source are X(S) and Y(S), the
downwind distance x to the receptor, along the direction of
plume travel, is given by:
• • (XfrR) "X(S) ) sin(WD) "(Y(R) "Y(S) ) COS (WD (1-4)
where WD is the direction from which the wind is blowing. The
downwind distance is used in calculating the distance-dependent
plume rise (see Section 1.1.4) and the dispersion parameters
(see Section 1.1.5) . If any receptor is located within 1 meter
of a point source or within 1 meter of the effective radius of
a volume source, a warning message is printed and no
concentrations are calculated for the source-receptor
combination. The crosswind distance y to the receptor from the
plume centerline is given by:
f •• (X(R) "X(S) )cos(WD) "(Y(R) "Y(S) ) sin(WD) (1-5)
The crosswind distance is used in Equation (1-1) .
1.1.3 Wind Speed Profile
The wind power law is used to adjust the observed wind
speed, uref, from a reference measurement height, zref, to the
stack or release height, hs. The stack height wind speed, us,
is used in the Gaussian plume equation (Equation 1-1) , and in
the plume rise formulas described in Section 1.1.4. The power
law equation is of the form:
(1-6)
Zref
1-4
-------
where p is the wind profile exponent. Values of p may be
provided by the user as a function of stability category and
wind speed class. Default values are as follows:
Stability Category
A
B
C
D
E
F
Rural Exponent
0.07
0.07
0.10
0.15
0.35
0.55
Urban Exponent
0.15
0.15
0.20
0.25
0.30
0.30
The stack height wind speed, us, is not allowed to be less
than 1.0 m/s.
1.1.4 Plume Rise Formulas
The plume height is used in the calculation of the
Vertical Term described in Section 1.1.6. The Briggs plume
rise equations are discussed below. The description follows
Appendix B of the Addendum to the MPTER User's Guide (Chico and
Catalano, 1986) for plumes unaffected by building wakes. The
distance dependent momentum plume rise equations, as described
in (Bowers, et al., 1979), are used to determine if the plume
is affected by the wake region for building downwash
calculations. These plume rise calculations for wake
determination are made assuming no stack-tip downwash for both
the Huber-Snyder and the Schulman-Scire methods. When the
model executes the building downwash methods of Schulman and
Scire, the reduced plume rise suggestions of Schulman and Scire
(1980) are used, as described in Section 1.1.4.11.
1-5
-------
1.1.4.1 Stack-tip Downwash.
In order to consider stack-tip downwash, modification of
the physical stack height is performed following Briggs (1974,
p. 4) . The modified physical stack height hs' is found from:
h ' • *h
Vs
•1.5
for v,, < 1. 5u
(1-7)
or
hs ' • »hs for vs > 1.5
where hs is physical stack height (m) , vs is stack gas exit
velocity (m/s) , and ds is inside stack top diameter (m) . This
hs' is used throughout the remainder of the plume height
computation. If stack tip downwash is not considered, hs' = hs
in the following equations.
1.1.4.2 Buoyancy and Momentum Fluxes.
For most plume rise situations, the value of the Briggs
buoyancy flux parameter, Fb (m4/s3) , is needed. The following
equation is equivalent to Equation (12), (Briggs, 1975, p. 63):
A '*\
Pb"3vA(-J
where • T = Ts - Ta, Ts is stack gas temperature (K) , and Ta is
ambient air temperature (K).
For determining plume rise due to the momentum of the
plume, the momentum flux parameter, Fm (m4/s2) , is calculated
based on the following formula:
4TS
(1-9)
1-6
-------
1.1.4.3 Unstable or Neutral - Crossover Between Momentum
and Buoyancy.
For cases with stack gas temperature greater than or equal
to ambient temperature, it must be determined whether the plume
rise is dominated by momentum or buoyancy. The crossover
temperature difference, (*T)C, is determined by setting Briggs '
(1969, p. 59) Equation 5.2 equal to the combination of Briggs'
(1971, p. 1031) Equations 6 and 7, and solving for • T, as
follows :
for Fb < 55,
1/3
( • T)c "0.0297TS ^— (1-10)
and for Fb > 55,
2/3
( • T)c "0.00575TS ^— (1-11)
ds1/3
If the difference between stack gas and ambient temperature,
• T, exceeds or equals (*T)C, plume rise is assumed to be
buoyancy dominated, otherwise plume rise is assumed to be
momentum dominated.
1.1.4.4 Unstable or Neutral - Buoyancy Rise.
For situations where *T exceeds (*T)C as determined above,
buoyancy is assumed to dominate. The distance to final rise,
xf, is determined from the equivalent of Equation (7), (Briggs,
1971, p. 1031) , and the distance to final rise is assumed to be
3.5x*, where x* is the distance at which atmospheric turbulence
begins to dominate entrainment. The value of xf is calculated
as follows:
for Fb < 55:
•49Fb5/8 (1-12)
1-7
-------
and for Fb > 55:
"119Fb2/5 (1-13)
The final effective plume height, he (m) , is determined
from the equivalent of the combination of Equations (6) and (7]
(Briggs, 1971, p. 1031):
for Fb < 55:
h ••h' "21.425—— (1-14)
c s
u_
and for Fb > 55:
p3/5
he "h/ • .38.71—^— (1-15)
Us
1.1.4.5 Unstable or Neutral - Momentum Rise.
For situations where the stack gas temperature is less
than or equal to the ambient air temperature, the assumption is
made that the plume rise is dominated by momentum. If •T is
less than (*T)C from Equation (1-10) or (1-11), the assumption
is also made that the plume rise is dominated by momentum. The
plume height is calculated from Equation (5.2) (Briggs, 1969,
p. 59) :
vs
he "h/ "3ds — (1-16)
Briggs (1969, p. 59) suggests that this equation is most
applicable when vs/us is greater than 4.
1.1.4.6 Stability Parameter.
For stable situations, the stability parameter, s, is
calculated from the Equation (Briggs, 1971, p. 1031):
1-8
-------
B'idz
s "g - (1-17)
Ta
As a default approximation, for stability class E (or 5) d* /dz
is taken as 0.020 K/m, and for class F (or 6), d'idz is taken
as 0.035 K/m.
1.1.4.7 Stable - Crossover Between Momentum and Buoyancy.
For cases with stack gas temperature greater than or equal
to ambient temperature, it must be determined whether the plume
rise is dominated by momentum or buoyancy. The crossover
temperature difference, (*T)C , is determined by setting
Briggs' (1975, p. 96) Equation 59 equal to Briggs ' (1969, p.
59) Equation 4.28, and solving for • T, as follows:
ss
T) "0.019582 Tv \s (1-18)
If the difference between stack gas and ambient temperature,
• T, exceeds or equals (*T)C, plume rise is assumed to be
buoyancy dominated, otherwise plume rise is assumed to be
momentum dominated.
1.1.4.8 Stable - Buoyancy Rise.
For situations where *T exceeds (*T)C as determined above,
buoyancy is assumed to dominate. The distance to final rise,
xf, is determined by the equivalent of a combination of
Equations (48) and (59) in Briggs, (1975), p. 96:
xf "2.0715 — (1-19)
The plume height, he, is determined by the equivalent of
Equation (59) (Briggs, 1975, p. 96) :
he "h/ "2.61 —-I (1-20)
1-9
-------
1.1.4.9 Stable - Momentum Rise.
Where the stack gas temperature is less than or equal to
the ambient air temperature, the assumption is made that the
plume rise is dominated by momentum. If »T is less than (*T)C
as determined by Equation (1-18), the assumption is also made
that the plume rise is dominated by momentum. The plume height
is calculated from Equation 4.28 of Briggs ((1969), p. 59):
\ 1/3
F
h "h "1.5
e
U0
(1-21)
The equation for unstable-neutral momentum rise (1-16) is also
evaluated. The lower result of these two equations is used as
the resulting plume height, since stable plume rise should not
exceed unstable-neutral plume rise.
1.1.4.10 All Conditions - Distance Less Than Distance to
Final Rise.
Where gradual rise is to be estimated for unstable,
neutral, or stable conditions, if the distance downwind from
source to receptor, x, is less than the distance to final rise,
the equivalent of Equation 2 of Briggs ((1972), p. 1030) is
used to determine plume height:
60
Us
(1-22)
This height will be used only for buoyancy dominated
conditions; should it exceed the final rise for the appropriate
condition, the final rise is substituted instead.
For momentum dominated conditions, the following equations
(Bowers, et al, 1979) are used to calculate a distance
dependent momentum plume rise:
a) unstable conditions:
1-10
-------
\ 1/3
(1-23)
where x is the downwind distance (meters), with a maximum value
defined by xmax as follows:
4ds(vs --3us
VSUS
for
•49F
5/8
4 /„ 3
••119F
2/5
for 0 < Fb < 55m4/s
for Fb > 55m4/s3
(1-24)
b) stable conditions:
3F.,
sin (xys/us
1/3
(1-25)
where x is the downwind distance (meters), with a maximum value
defined by xmax as follows:
x •»0 . 5
max
(1-26)
The jet entrainment coefficient, •*, is given by,
(1-27)
As with the buoyant gradual rise, if the distance-dependent
momentum rise exceeds the final rise for the appropriate
condition, then the final rise is substituted instead.
1.1.4.10.1 Calculating the plume height for wake effects
determination.
The building downwash algorithms in the ISC models always
require the calculation of a distance dependent momentum plume
rise. When building downwash is being simulated, the equations
1-11
-------
described above are used to calculate a distance dependent
momentum plume rise at a distance of two building heights
downwind from the leeward edge of the building. However,
stack-tip downwash is not used when performing this calculation
(i.e. hs' = hs) . This wake plume height is compared to the
wake height based on the good engineering practice (GEP)
formula to determine whether the building wake effects apply to
the plume for that hour.
The procedures used to account for the effects of building
downwash are discussed more fully in Section 1.1.5.3. The
plume rise calculations used with the Schulman-Scire algorithm
are discussed in Section 1.1.4.11.
1.1.4.11 Plume Rise When Schulman and Scire Building
Downwash is Selected.
The Schulman-Scire downwash algorithms are used by the ISC
models when the stack height is less than the building height
plus one half of the lesser of the building height or width.
When these criteria are met, the ISC models estimate plume rise
during building downwash conditions following the suggestion of
Scire and Schulman (1980) . The plume rise during building
downwash conditions is reduced due to the initial dilution of
the plume with ambient air.
The plume rise is estimated as follows. The initial
dimensions of the downwashed plume are approximated by a line
source of length Ly and depth 2R0 where:
Ro"^A'z- x"3LB (1-28)
Ly ••^•(•••••) x"3LB, ••>•• (l-29a)
x.-3LB,
B,
1-12
-------
LB equals the minimum of hb and hw, where hb is the building
height and hw the projected (crosswind) building width. A is a
linear decay factor and is discussed in more detail in Section
1.1.5.3.2. If there is no enhancement of •• or if the enhanced
•• is less than the enhanced ••, the initial plume will be
represented by a circle of radius R0. The \/2 factor converts
the Gaussian •• to an equivalent uniform circular distribution
and y2» • converts •• to an equivalent uniform rectangular
distribution. Both •• and •• are evaluated at x = 3LB, and are
taken as the larger of the building enhanced sigmas and the
sigmas obtained from the curves (see Section 1.1.5.3). The
value of •• used in the calculation of Ly also includes the
linear decay term, A.
The rise of a downwashed finite line source was solved in
the BLP model (Scire and Schulman, 1980). The neutral
distance-dependent rise (Z) is given by:
, , 3LV 3R ^ , f 6R L 3R2
Z3 .. X o
• • 2 • • 2
The stable distance-dependent rise is calculated by:
3Ly 3Ro) ^2 6RL
-7 A A I
-------
Fm = momentum flux term (Equation 1-9) (m4/s2)
x = downwind distance (m)
us = wind speed at release height (m/s)
vs = stack exit velocity (m/s)
ds = stack diameter (m)
•• = entrainment coefficient (=0.6)
•• = jet entrainment coefficient ** — *
3
us
= stability parameter * *9
a
The larger of momentum and buoyant rise, determined separately
by alternately setting Fb or Fm = 0 and solving for Z, is
selected for plume height calculations for Schulman-Scire
downwash. In the ISC models, Z is determined by solving the
cubic equation using Newton's method.
1.1.5 The Dispersion Parameters
1.1.5.1 Point Source Dispersion Parameters.
Equations that approximately fit the Pasquill-Gifford
curves (Turner, 1970) are used to calculate •• and •• (in
meters) for the rural mode. The equations used to calculate ••
are of the form:
•• ••465.11628(x)tan(TH) (1-32)
where:
TH "0.017453293 [C"dln(x) ] (1-33)
1-14
-------
In Equations (1-32) and (1-33) the downwind distance x is in
kilometers, and the coefficients c and d are listed in Table
1-1. The equation used to calculate •• is of the form:
•• • »axb (1-34)
where the downwind distance x is in kilometers and •• is in
meters. The coefficients a and b are given in Table 1-2.
Tables 1-3 and 1-4 show the equations used to determine ••
and •• for the urban option. These expressions were determined
by Briggs as reported by Gifford (1976) and represent a best
fit to urban vertical diffusion data reported by McElroy and
Pooler (1968). While the Briggs functions are assumed to be
valid for downwind distances less than 100m, the user is
cautioned that concentrations at receptors less than 100m from
a source may be suspect.
1-15
-------
TABLE 1-1
PARAMETERS USED TO CALCULATE PASQUILL-GIFFORD ••
•• = 465.11628 (x)tan(TH)
TH = 0.017453293 [c - d ln(x)]
Pasquill
Stability
Category
A
B
C
D
E
F
24.1670
18.3330
12.5000
8.3330
6.2500
4.1667
2.5334
1.8096
1.0857
0.72382
0.54287
0.36191
where •• is in meters and x is in kilometers
1-16
-------
TABLE 1-2
PARAMETERS USED TO CALCULATE PASQUILL-GIFFORD ••
••(meters) = axb
Pasquill
Stability
Category
A*
B*
C*
D
0
0
0
0
0
0
0
0
0
1
3
10
X
<
.10
.16
.21
.26
.31
.41
.51
>
<
.21
>
<
.31
.01
.01
.01
(km)
c.10
- 0.
- 0.
- 0.
- 0.
- 0.
- 0.
- 3.
3.11
c.20
- 0.
0.40
All
c.30
- 1.
- 3.
- 10
- 30
15
20
25
30
40
50
11
40
00
00
.00
.00
>30.00
122
158
170
179
217
258
346
453
90
98
109
61
34
32
32
33
36
44
a
.800
.080
.220
.520
.410
.890
.750
.850
* *
.673
.483
.300
.141
.459
.093
.093
.504
.650
.053
(x in
0
1
1
1
1
1
1
2
0
0
1
0
0
0
0
0
0
0
km)
b
.94470
.05420
.09320
.12620
.26440
.40940
.72830
.11660
* *
.93198
.98332
.09710
.91465
.86974
.81066
.64403
.60486
.56589
.51179
If the calculated value of •• exceed 5000 m, •• is set to
5000 m.
•• is equal to 5000 m.
1-17
-------
TABLE 1-2
(CONTINUED)
PARAMETERS USED TO CALCULATE PASQUILL-GIFFORD ••
••(meters) = axb
Pasquill
Stability
Category
E
0
0
1
2
4
10
20
F
0
0
1
2
3
7
15
30
x (km)
< .
.10 -
.31 -
.01 -
.01 -
.01 -
.01 -
.01 -
>40
< .
.21 -
.71 -
.01 -
.01 -
.01 -
.01 -
.01 -
.01 -
>60
10
- 0.
- 1.
- 2 .
- 4.
- 10
- 20
- 40
.00
20
- 0.
- 1.
- 2.
- 3.
- 7.
- 15
- 30
- 60
.00
(x in km)
a
30
00
00
00
.00
.00
.00
70
00
00
00
00
.00
.00
.00
24
23
21
21
22
24
26
35
47
15
14
13
13
14
16
17
22
27
34
.260
.331
.628
.628
.534
.703
.970
.420
.618
.209
.457
.953
.953
.823
.187
.836
.651
.074
.219
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
b
.83660
.81956
.75660
.63077
.57154
.50527
.46713
.37615
.29592
.81558
.78407
.68465
.63227
.54503
.46490
.41507
.32681
.27436
.21716
1-18
-------
TABLE 1-3
BRIGGS FORMULAS USED TO CALCULATE McELROY-POOLER ••
Pasquill
Stability
Category
A
B
C
D
E
F
• • (meters) *
0
0
0
0
0
0
.32
.32
.22
.16
.11
.11
x
x
X
X
X
X
(1
(1
(1
(1
(1
(1
.0 H
.0 H
.0 H
.0 H
.0 H
.0 H
h 0
h 0
h 0
h 0
h 0
h 0
.0004
.0004
.0004
.0004
.0004
.0004
x)
x)
x)
x)
x)
x)
-1/2
-1/2
-1/2
-1/2
-1/2
-1/2
Where x is in meters
TABLE 1-4
BRIGGS FORMULAS USED TO CALCULATE McELROY-POOLER ••
Pasquill
Stability
Category
• • (meters)*
A
B
C
D
E
F
0.24 x (1.0 + 0.001 x)1/2
0.24 x (1.0 + 0.001 x)1/2
0.20 x
0.14 x (1.0 + 0.0003 x)"1/2
0.08 x (1.0 + 0.0015 x)"1/2
0.08 x (1.0 + 0.0015 x)"1/2
Where x is in meters.
1-19
-------
1.1.5.2 Lateral and Vertical Virtual Distances.
The equations in Tables 1-1 through 1-4 define the
dispersion parameters for an ideal point source. However,
volume sources have initial lateral and vertical dimensions.
Also, as discussed below, building wake effects can enhance the
initial growth of stack plumes. In these cases, lateral (xy)
and vertical (xz) virtual distances are added by the ISC models
to the actual downwind distance x for the •• and ••
calculations. The lateral virtual distance in kilometers for
the rural mode is given by:
(1-35)
where the stability-dependent coefficients p and q are given in
Table 1-5 and **0 is the standard deviation in meters of the
lateral concentration distribution at the source. Similarly,
the vertical virtual distance in kilometers for the rural mode
is given by:
(•• V/b
x, •• — (1-36)
where the coefficients a and b are obtained form Table 1-2 and
• *0 is the standard deviation in meters of the vertical
concentration distribution at the source. It is important to
note that the ISC model programs check to ensure that the xz
used to calculate • • at (x + xz) in the rural mode is the xz
calculated using the coefficients a and b that correspond to
the distance category specified by the quantity (x + xz) .
To determine virtual distances for the urban mode, the
functions displayed in Tables 1-3 and 1-4 are solved for x.
The solutions are quadratic formulas for the lateral virtual
distances; and for vertical virtual distances the solutions are
cubic equations for stability classes A and B, a linear
equation for stability class C, and quadratic equations for
1-20
-------
stability classes D, E, and F. The cubic equations are solved
by iteration using Newton's method.
TABLE 1-5
COEFFICIENTS USED TO CALCULATE LATERAL VIRTUAL DISTANCES
FOR PASQUILL-GIFFORD DISPERSION RATES
Pasquill
Stability
Category
A
B
C
D
E
F
1.1.5.3
Buildinq
P
209.14
154.46
103.26
68.26
51.06
33.92
Procedures Used to Account for
Wakes on Effluent Dispersion.
q
0.890
0.902
0.917
0.919
0.921
0.919
the Effects of
The procedures used by the ISC models to account for the
effects of the aerodynamic wakes and eddies produced by plant
buildings and structures on plume dispersion originally
followed the suggestions of Huber (1977) and Snyder (1976) .
Their suggestions are principally based on the results of
wind-tunnel experiments using a model building with a crosswind
dimension double that of the building height. The atmospheric
turbulence simulated in the wind-tunnel experiments was
intermediate between the turbulence intensity associated with
the slightly unstable Pasquill C category and the turbulence
intensity associated with the neutral D category. Thus, the
data reported by Huber and Snyder reflect a specific stability,
building shape and building orientation with respect to the
mean wind direction. It follows that the ISC wake-effects
1-21
-------
evaluation procedures may not be strictly applicable to all
situations. The ISC models also provide for the revised
treatment of building wake effects for certain sources, which
uses modified plume rise algorithms, following the suggestions
of Schulman and Hanna (1986). This treatment is largely based
on the work of Scire and Schulman (1980). When the stack
height is less than the building height plus half the lesser of
the building height or width, the methods of Schulman and Scire
are followed. Otherwise, the methods of Huber and Snyder are
followed. In the ISC models, direction-specific building
dimensions may be used with either the Huber-Snyder or
Schulman-Scire downwash algorithms.
The wake-effects evaluation procedures may be applied by
the user to any stack on or adjacent to a building. For
regulatory application, a building is considered sufficiently
close to a stack to cause wake effects when the distance
between the stack and the nearest part of the building is less
than or equal to five times the lesser of the height or the
projected width of the building. For downwash analyses with
direction-specific building dimensions, wake effects are
assumed to occur if the stack is within a rectangle composed of
two lines perpendicular to the wind direction, one at 5Lb
downwind of the building and the other at 2Lb upwind of the
building, and by two lines parallel to the wind direction, each
at 0.5Lb away from each side of the building, as shown below:
1-22
-------
Wind direction )))))))))))))>
* * 1/2 Lb
+)), ))))))))))))* ))
-
* *Building* *
* * * *
-))))), *
* * * *
-))- )))))))))))* ))
* * 1/2 Lb
-))))))))))))))))))))))))))))))))))))))))))- ))
*<))2Lb))>* *<)))))))))5Lb)))))))))>*
Lb is the lesser of the height and projected width of the
building for the particular direction sector. For additional
guidance on determining whether a more complex building
configuration is likely to cause wake effects, the reader is
referred to the Guideline for Determination of Good Engineering
Practice Stack Height (Technical Support Document for the Stack
Height Regulations) - Revised (EPA, 1985). In the following
sections, the Huber and Snyder building downwash method is
described followed by a description of the Schulman and Scire
building downwash method.
1.1.5.3.1 Huber and Snyder building downwash procedures.
The first step in the wake-effects evaluation procedures
used by the ISC model programs is to calculate the gradual
plume rise due to momentum alone at a distance of two building
heights using Equation (1-23) or Equation (1-25). If the plume
height, he, given by the sum of the stack height (with no
stack-tip downwash adjustment) and the momentum rise is greater
than either 2.5 building heights (2.5 hb) or the sum of the
building height and 1.5 times the building width (hb + 1.5 hw) ,
the plume is assumed to be unaffected by the building wake.
Otherwise the plume is assumed to be affected by the building
wake.
1-23
-------
The ISC model programs account for the effects of building
wakes by modifying both •• and •• for plumes with plume height
to building height ratios less than or equal to 1.2 and by
modifying only •• for plumes from stacks with plume height to
building height ratios greater than 1.2 (but less than 2.5).
The plume height used in the plume height to stack height
ratios is the same plume height used to determine if the plume
is affected by the building wake. The ISC models define
buildings as squat (hw > hb) or tall (hw < hb) . The ISC models
include a general procedure for modifying •• and •• at
distances greater than or equal to 3hb for squat buildings or
3hw for tall buildings. The air flow in the building cavity
region is both highly turbulent and generally recirculating.
The ISC models are not appropriate for estimating
concentrations within such regions. The ISC assumption that
this recirculating cavity region extends to a downwind distance
of 3hb for a squat building or 3hw for a tall building is most
appropriate for a building whose width is not much greater than
its height. The ISC user is cautioned that, for other types of
buildings, receptors located at downwind distances of 3hb
(squat buildings) or 3hw (tall buildings) may be within the
recirculating region.
The modified •• equation for a squat building is given by:
••' ••0.7hb • »0 .067 (x« Shb) for 3hb < x
or (1-37)
••••{x»»xz} for x>10ht
where the building height hb is in meters. For a tall
building, Huber (1977) suggests that the width scale hw replace
1-24
-------
hb in Equation (1-37) . The modified •• equation for a tall
building is then given by:
•0.7hw •'0.067 (x« 3hw) for 3hw10h
v
where hw is in meters. It is important to note that ••' is not
permitted to be less than the point source value given in
Tables 1-2 or 1-4, a condition that may occur.
The vertical virtual distance, xz, is added to the actual
downwind distance x at downwind distances beyond 10hb for squat
buildings or beyond 10hw for tall buildings, in order to
account for the enhanced initial plume growth caused by the
building wake. The virtual distance is calculated from
solutions to the equations for rural or urban sigmas provided
earlier.
As an example for the rural options, Equations (1-34) and
(1-37) can be combined to derive the vertical virtual distance
xz for a squat building. First, it follows from Equation
(1-37) that the enhanced •• is equal to 1.2hb at a downwind
distance of 10hb in meters or 0.01hb in kilometers. Thus, xz
for a squat building is obtained from Equation (1-34) as
follows:
•• {0.01hb} "1.2hb "a(0.01hb "xz)b (1-39)
(1-40)
1-25
-------
where the stability-dependent constants a and b are given in
Table 1-2. Similarly, the vertical virtual distance for tall
buildings is given by:
(1-41)
For the urban option, xz is calculated from solutions to the
equations in Table 1-4 for •• = 1.2hb or •• = 1.2 hw for tall or
squat buildings, respectively.
For a squat building with a building width to building
height ratio (hw/hb) less than or equal to 5, the modified ••
equation is given by:
••' ••0.35hw ••0.067 (x»3hb) for 3hb < 5
or (1-42)
••••{x»»x } for x>10ht
The lateral virtual distance is then calculated for this value
of ••.
For a building that is much wider than it is tall (hw/hb
greater than 5), the presently available data are insufficient
to provide general equations for ••. For a stack located
toward the center of such a building (i.e., away form either
end), only the height scale is considered to be significant.
1-26
-------
The modified •• equation for a very squat building is then
given by:
••' "0.35hb "0.067 (x«Shb) for 3hb < 3
or (1-43)
• • • •
v
;{x»»x } for x>10ht
For hw/hb greater than 5, and a stack located laterally
within about 2.5 hb of the end of the building, lateral plume
spread is affected by the flow around the end of the building.
With end effects, the enhancement in the initial lateral spread
is assumed not to exceed that given by Equation (1-42) with hw
replaced by 5 hb. The modified •• equation is given by:
••' • »1.75hb • »0.067 (x» Shb) for 3hb < 5
or (1-44)
••••{x»»x } for x>10t
The upper and lower bounds of the concentrations that can
be expected to occur near a building are determined
respectively using Equations (1-43) and (1-44). The user must
specify whether Equation (1-43) or Equation (1-44) is to be
used in the model calculations. In the absence of user
instructions, the ISC models use Equation (1-43) if the
building width to building height ratio hw/hb exceeds 5.
Although Equation (1-43) provides the highest
concentration estimates for squat buildings with building width
to building height ratios (hw/hb) greater than 5, the equation
is applicable only to a stack located near the center of the
building when the wind direction is perpendicular to the long
side of the building (i.e., when the air flow over the portion
1-27
-------
of the building containing the source is two dimensional).
Thus, Equation (1-44) generally is more appropriate then
Equation (1-43). It is believed that Equations (1-43) and
(1-44) provide reasonable limits on the extent of the lateral
enhancement of dispersion and that these equations are adequate
until additional data are available to evaluate the flow near
very wide buildings.
The modified •• equation for a tall building is given by:
••' ••0.35hw ••0.067 (x»3hw) for
or (1-45)
••••{x»»x } for x>10t
The ISC models print a message and do not calculate
concentrations for any source-receptor combination where the
source-receptor separation is less than 1 meter, and also for
distances less than 3 hb for a squat building or 3 hw for a
tall building under building wake effects. It should be noted
that, for certain combinations of stability and building height
and/or width, the vertical and/or lateral plume dimensions
indicated for a point source by the dispersion curves at a
downwind distance of ten building heights or widths can exceed
the values given by Equation (1-37) or (1-38) and by Equation
(1-42) or (1-43). Consequently, the ISC models do not permit
the virtual distances xy and xz to be less than zero.
1.1.5.3.2 Schulman and Scire refined building downwash
procedures.
The procedures for treating building wake effects include
the use of the Schulman and Scire downwash method. The
building wake procedures only use the Schulman and Scire method
when the physical stack height is less than hb + 0.5 LB, where
hb is the building height and LB is the lesser of the building
1-28
-------
height or width. In regulatory applications, the maximum
projected width is used. The features of the Schulman and
Scire method are: (1) reduced plume rise due to initial plume
dilution, (2) enhanced vertical plume spread as a linear
function of the effective plume height, and (3) specification
of building dimensions as a function of wind direction. The
reduced plume rise equations were previously described in
Section 1.1.4.11.
When the Schulman and Scire method is used, the ISC
dispersion models specify a linear decay factor, to be included
in the **'s calculated using Equations (1-37) and (1-38), as
follows:
••" • 'A»»' (1-46)
where ••' is from either Equation (1-37) or (1-38) and A is the
linear decay factor determined as follows:
A ••! if he < hb
nb ' *ne
A" ••! if hb hb • »2LB
where the plume height, he, is the height due to gradual
momentum rise at 2 hb used to check for wake effects. The
effect of the linear decay factor is illustrated in Figure 1-1.
For Schulman-Scire downwash cases, the linear decay term is
also used in calculating the vertical virtual distances with
Equations (1-40) to (1-41).
When the Schulman and Scire building downwash method is
used the ISC models require direction specific building heights
and projected widths for the downwash calculations. The ISC
models also accept direction specific building dimensions for
Huber-Snyder downwash cases. The user inputs the building
height and projected widths of the building tier associated
1-29
-------
with the greatest height of wake effects for each ten degrees
of wind direction. These building heights and projected widths
are the same as are used for GEP stack height calculations.
The user is referred to EPA (1986) for calculating the
appropriate building heights and projected widths for each
direction. Figure 1-2 shows an example of a two tiered
building with different tiers controlling the height that is
appropriate for use for different wind directions. For an east
or west wind the lower tier defines the appropriate height and
width, while for a north or south wind the upper tier defines
the appropriate values for height and width.
1.1.5.4 Procedures Used to Account for Buoyancy-Induced
Dispersion.
The method of Pasquill (1976) is used to account for the
initial dispersion of plumes caused by turbulent motion of the
plume and turbulent entrainment of ambient air. With this
method, the effective vertical dispersion • »e is calculated as
follows:
2 • fa
'" 3.5
1/2
(1-48)
where •• is the vertical dispersion due to ambient turbulence
and • fa is the plume rise due to momentum and/or buoyancy. The
lateral plume spread is parameterized using a similar
expression:
ye
2
3.5
l/:
(1-49)
where •• is the lateral dispersion due to ambient turbulence.
It should be noted that • fa is the distance-dependent plume
rise if the receptor is located between the source and the
distance to final rise, and final plume rise if the receptor is
located beyond the distance to final rise. Thus, if the user
1-30
-------
elects to use final plume rise at all receptors the
distance-dependent plume rise is used in the calculation of
buoyancy-induced dispersion and the final plume rise is used in
the concentration equations. It should also be noted that
buoyancy-induced dispersion is not used when the Schulman-Scire
downwash option is in effect.
1.1.6 The Vertical Term
The Vertical Term (V), which is included in Equation
(1-1), accounts for the vertical distribution of the Gaussian
plume. It includes the effects of source elevation, receptor
elevation, plume rise (Section 1.1.4), limited mixing in the
vertical, and the gravitational settling and dry deposition of
particulates. In addition to the plume height, receptor height
and mixing height, the computation of the Vertical Term
requires the vertical dispersion parameter (••) described in
Section 1.1.5.
1.1.6.1 The Vertical Term Without Dry Deposition.
In general, the effects on ambient concentrations of
gravitational settling and dry deposition can be neglected for
gaseous pollutants and small particulates (less than about 0.1
1-31
-------
microns in diameter). The Vertical Term without deposition
effects is then given by:
v •»exp
e.5
•exp
6.5
• •
z
exp
exp
where:
•exp
H3
—
hs +
Zr -
Zr +
Zr -
z_ +
• h
(2iZi
(2iZi
(2iZi
(2iz,
- h,
- h,
+ h,
+ h,
exp
H4
(1-50)
z_ =
receptor height above ground (flagpole)
mixing height (m)
(m)
The infinite series term in Equation (1-50) accounts for
the effects of the restriction on vertical plume growth at the
top of the mixing layer. As shown by Figure 1-3, the method of
image sources is used to account for multiple reflections of
the plume from the ground surface and at the top of the mixed
layer. It should be noted that, if the effective stack height,
he, exceeds the mixing height, zi; the plume is assumed to
fully penetrate the elevated inversion and the ground-level
concentration is set equal to zero.
1-32
-------
Equation (1-50) assumes that the mixing height in rural
and urban areas is known for all stability categories. As
explained below, the meteorological preprocessor program uses
mixing heights derived from twice-daily mixing heights
calculated using the Holzworth (1972) procedures. The ISC
models currently assume unlimited vertical mixing under stable
conditions, and therefore delete the infinite series term in
Equation (1-50) for the E and F stability categories.
The Vertical Term defined by Equation (1-50) changes the
form of the vertical concentration distribution from Gaussian
to rectangular (i.e., a uniform concentration within the
surface mixing layer) at long downwind distances.
Consequently, in order to reduce computational time without a
loss of accuracy, Equation (1-50) is changed to the form:
(1-51)
at downwind distances where the •*/zi ratio is greater than or
equal to 1.6.
The meteorological preprocessor program, RAMMET, used by
the ISC Short Term model uses an interpolation scheme to assign
hourly rural and urban mixing heights on the basis of the early
morning and afternoon mixing heights calculated using the
Holzworth (1972) procedures. The procedures used to
interpolate hourly mixing heights in urban and rural areas are
illustrated in Figure 1-4, where:
Hm{max} = maximum mixing height on a given day
Hm{min} = minimum mixing height on a given day
MN = midnight
SR = sunrise
SS = sunset
The interpolation procedures are functions of the stability
category for the hour before sunrise. If the hour before
sunrise is neutral, the mixing heights that apply are indicated
1-33
-------
by the dashed lines labeled neutral in Figure 1-4. If the hour
before sunrise is stable, the mixing heights that apply are
indicated by the dashed lines labeled stable. It should be
pointed out that there is a discontinuity in the rural mixing
height at sunrise if the preceding hour is stable. As
explained above, because of uncertainties about the
applicability of Holzworth mixing heights during periods of E
and F stability, the ISC models ignore the interpolated mixing
heights for E and F stability, and treat such cases as having
unlimited vertical mixing.
1.1.6.2 The Vertical Term in Elevated Simple Terrain.
The ISC models make the following assumption about plume
behavior in elevated simple terrain (i.e., terrain that exceeds
the stack base elevation but is below the release height):
• The plume axis remains at the plume stabilization
height above mean sea level as it passes over elevated
or depressed terrain.
• The mixing height is terrain following.
• The wind speed is a function of height above the
surface (see Equation (1-6) ) .
Thus, a modified plume stabilization height he' is
substituted for the effective stack height he in the Vertical
Term given by Equation (1-50). For example, the effective
plume stabilization height at the point x, y is given by:
he' "he **Zs **Z (x,y) (1-52)
where:
zs = height above mean sea level of the base of the
stack (m)
z|(xy) = height above mean sea level of terrain at the
receptor location (x,y) (m)
1-34
-------
It should also be noted that, as recommended by EPA, the ISC
models "truncate" terrain at stack height as follows: if the
terrain height z - zs exceeds the source release height, hs,
the elevation of the receptor is automatically "chopped off" at
the physical release height. The user is cautioned that
concentrations at these complex terrain receptors are subject
to considerable uncertainty. Figure 1-5 illustrates the
terrain-adjustment procedures used by the ISC models for simple
elevated terrain. The vertical term used with the complex
terrain algorithms in ISC is described in Section 1.5.6.
1.1.6.3 The Vertical Term With Dry Deposition.
Particulates are brought to the surface through the
combined processes of turbulent diffusion and gravitational
settling. Once near the surface, they may be removed from the
atmosphere and deposited on the surface. This removal is
modeled in terms of a deposition velocity (vd), which is
described in Section 1.3.1, by assuming that the deposition
flux of material to the surface is equal to the product vd»3,
where • 3 is the airborne concentration just above the surface.
As the plume of airborne particulates is transported downwind,
such deposition near the surface reduces the concentration of
particulates in the plume, and thereby alters the vertical
distribution of the remaining particulates. Furthermore, the
larger particles will also move steadily nearer the surface at
a rate equal to their gravitational settling velocity (vg) . As
a result, the plume centerline height is reduced, and the
vertical concentration distribution is no longer Gaussian.
A corrected source-depletion model developed by Horst
(1983) is used to obtain a "vertical term" that incorporates
both the gravitational settling of the plume and the removal of
plume mass at the surface. These effects are incorporated as
modifications to the Gaussian plume equation. First,
1-35
-------
gravitational settling is assumed to result in a "tilted
plume", so that the effective plume height (he) in Equation
(1-50) is replaced by
v
hed "he **hv "he Vg (1-53)
Us
where hv = (x/us)vg is the adjustment of the plume height due to
gravitational settling. Then, a new vertical term (Vd) that
includes the effects of dry deposition is defined as:
vd(x,z,hed) ••V(x,z,hed) FQ(x) P(x,z) (1-54)
V(x,z,hed) is the vertical term in the absence of any
deposition--it is just Equation (1-50), with the tilted plume
approximation. FQ(x) is the fraction of material that remains
in the plume at the downwind distance x (i.e., the mass that
has not yet been deposited on the surface). This factor may be
thought of as a source depletion factor, a ratio of the
"current" mass emission rate to the original mass emission
rate. P(x,z) is a vertical profile adjustment factor, which
modifies the reflected Gaussian distribution of Equation
(1-50), so that the effects of dry deposition on near-surface
concentrations can be simulated.
For large travel-times, hed in Equation (1-53) can become
less than zero. However, the tilted plume approximation is not
a valid approach in this region. Therefore, a minimum value of
zero is imposed on hed. In effect, this limits the settling of
the plume centerline, although the deposition velocity
continues to account for gravitational settling near the
surface. The effect of gravitational settling beyond the plume
touchdown point (where hed =0) is to modify the vertical
structure of the plume, which is accounted for by modifying the
vertical dispersion parameter (••).
1-36
-------
The process of adjusting the vertical profile to reflect
loss of plume mass near the surface is illustrated in Figures
1-6 and 1-7. At a distance far enough downwind that the plume
size in the vertical has grown larger than the height of the
plume, significant corrections to the concentration profile may
be needed to represent the removal of material from the plume
due to deposition. Figure 1-6 displays a depletion factor FQ,
and the corresponding profile correction factor P(z) for a
distance at which •• is 1.5 times the plume height. The
depletion factor is constant with height, whereas the profile
correction shows that most of the material is lost from the
lower portion of the plume. Figure 1-7 compares the vertical
profile of concentration both with and without deposition and
the corresponding depletion of material from the plume. The
depleted plume profile is computed using Equation (1-54).
Both FQ(x) and P(x,z) depend on the size and density of
the particles being modeled, because this effects the total
deposition velocity (See Section 1.3.2). Therefore, for a
given source of particulates, ISC allows multiple particle-size
categories to be defined, with the maximum number of particle
size categories controlled by a parameter statement in the
model code (see Volume I). The user must provide the mass-mean
particle diameter (microns), the particle density (g/cm3) , and
the mass fraction (• } for each category being modeled. If we
denote the value of FQ(x) and P(x,z) for the nth particle-size
category by FQn(x) and Pn(x,z) and substitute these in Equation
(1-54), we see that a different value for the vertical term is
obtained for each particle-size category, denoted as Vdn.
Therefore, the total vertical term is given by the sum of the
terms for each particle-size category, weighted by the
respective mass-fractions:
n-i
1-37
-------
FQ(x) is a function of the total deposition velocity (vd) ,
V(x,zd,hed) , and P(x,zd) :
Fn(x) »»EXP
•_vd v(x", zd,hed) P(X", zd) dx"
(1-56)
where zd is a height near the surface at which the deposition
flux is calculated. The deposition reference height is
calculated as the maximum of 1.0 meters and 20z0. This
equation reflects the fact that the material removed from the
plume by deposition is just the integral of the deposition flux
over the distance that the plume has traveled. In ISC, this
integral is evaluated numerically. For sources modeled in
elevated or complex terrain, the user can input a terrain grid
to the model, which is used to determine the terrain elevation
at various distances along the plume path during the evaluation
of the integral. If a terrain grid is not input by the user,
then the model will linearly interpolate between the source
elevation and the receptor elevation.
The profile correction factor P(x,z) is given by
•P(x,zd)
"EXP['vg R(z,
(l-57a)
v
vgR(z ~, zd)l)dz
g v /J/
where R(z,zd) is an atmospheric resistance to vertical
transport that is derived from Briggs' formulas for ••
(Gifford, 1976) . When the product vgR(z,zd) is of order 0.1 or
less, the exponential function is approximated (for small
argument) to simplify P(x,z):
1-38
-------
P( x, z) - P(x,zd) [l "(vd --Vg) R(z,zd)]
/ \ V (x, z", o) _./ .. \ , ..
• *(vd • ^g). V- R(z' zd)dz
(l-57b)
This simplification is important, since the integral in
Equation (l-57a) is evaluated numerically, whereas that in
Equation (l-57b) is computed using analytical approximations
The resistance R(z,zd) is obtained for the following
functional forms of •• defined by Briggs:
1-39
-------
Case 1 :
Rural: stability A, B
Urban: stability C
• • • *ax
R(z,zd).-jl _L m(z/zd)
• • au
Case 2:
Rural: stability C, D
Urban: stability D, E, F
• • • *ax/( 1 • »bx)1/2
R<"">->[?.i
b
ln(z/zd)
\
(1-58)
Case 3 :
Rural: stability E, F
• • • *ax/(l • »bx)
au
In
2b
\ z.jj
d) **
• • 17 \ • •
3b"
2a:
Case 4:
Urban: stability A, B
•ax(l • »bx)
1/2
au
^1 • *bx(z) • •!
V1 **bx(zd) •'!
^1 • *bx(z) • •!
V1 **bx(zd) "1.
For this last form, the x(z) and x(zd) must be solved for z and
zd (respectively) by finding the root of the implicit relation
\
— z • »a x \Jl • *bx
(1-59)
1-40
-------
The corresponding functions for P(x,zd) for the special case of
Equation (1-57) are given by:
1-41
-------
Case 1:
Rural: stability A, B
Urban: stability C
•ax
p>i(x'zd)
ua
-z
/zd)
Case 2:
Rural: stability C, D
Urban: stability D, E, F
• *ax/(l • »bx)1/2
p>i(x'zd) "!
vd **vg
ua
^
In
(/2'z/zd)
Case 3:
Rural: stability E, F
• • • *ax/(l • »bx)
vd '
•
v§
ua
3b2
2a2
• •
2
/ 2
h
A
2
•
2
* * Z
•
)
— In
/zd)
(1-60)
Case 4:
Urban: stability A, B
ax(l • »bx)
1/
ua
^
zl
/zd)
In | 1 "k zH/8 "J — k
1-42
-------
For the last form,k ••
a \
— , and
2
1-43
-------
(l •• .0006 ''f •• < 300m
zl
••1**0.6724»» •• > 300m
and (1-61)
•52 '"Si *zi * 1000m
'52
'Y/1000 •'! •'! > 1000m
The added complexity of this last form arises because a simple
analytical solution to Equation (1-57) could not be obtained
for the urban class A and B. The integral in P(x,zd) for •• =
ax(l + bx)1/2 listed above matches a numerical solution to
within about 2% for zd = 1 m.
When vertical mixing is limited by zi; the profile
correction factor P(x,zd) involves an integral from 0 to zi;
rather than from 0 to infinity. Furthermore, V contains terms
that simulate reflection from z = zi as well as z = 0 so that
the profile correction factor, P(x,zd), becomes a function of
mixing height, i.e, P(x,zd,zi). In the well-mixed limit,
P(x,zd,zi) has the same form as P(x,zd) in Equation (1-60) but
•• is replaced by a constant times z±:
/zd) - ln(Zl/zd)
(1-62)
Therefore a limit is placed on each term involving • • in
Equation (1-60) so that each term does not exceed the
1-44
-------
corresponding term in z±. Similarly, since the leading order
term in P(x,zd) for •• = ax(l + bx)1/2 corresponds to the
lnk/2 **/z\ term in Equation (1-62), •• is capped at zi/\]2 for
this P(x,zd) as well. Note that these caps to •• in Equation
(1-60) are broadly consistent with the condition on the use of
the well-mixed limit on V in Equation (1-51) which uses a ratio
*'/zi = 1.6. In Equation (1-62), the corresponding ratios are
••/z± = 1.4, 1.6, and 1.9.
In many applications, the removal of material from the
plume may be extremely small, so that FQ(x) and P(x,z) are
virtually unity. When this happens, the vertical term is
virtually unchanged (Vd = V, see Equation (1-54)) . The
deposition flux can then be approximated as vd» • rather than
vd»d. The plume depletion calculations are optional, so that
the added expense of computing FQ(x) and P(x,z) can be avoided.
Not considering the effects of dry depletion results in
conservative estimates of both concentration and deposition,
since material deposited on the surface is not removed from the
plume.
1.1.7 The Decay Term (D)
The Decay Term in Equation (1-1) is a simple method of
accounting for pollutant removal by physical or chemical
processes. It is of the form:
for • •> 0
(1-63)
or
• •! for • •• »0
where:
1-45
-------
• • = the decay coefficient (s"1) (a value of zero means
decay is not considered)
x = downwind distance (m)
For example, if T1/2 is the pollutant half life in seconds, the
user can obtain ••from the relationship:
0.693
(1-64)
Tl/2
The default value for • •is zero. That is, decay is not
considered in the model calculations unless ••is specified.
However, a decay half life of 4 hours (• • = 0.0000481 s"1) is
automatically assigned for S02 when modeled in the urban mode.
1.2 NON-POINT SOURCE EMISSIONS
1.2.1 General
The ISC models include algorithms to model volume, area
and open-pit sources, in addition to point sources. These non-
point source options of the ISC models are used to simulate the
effects of emissions from a wide variety of industrial sources.
In general, the ISC volume source model is used to simulate the
effects of emissions from sources such as building roof
monitors and line sources (for example, conveyor belts and rail
lines). The ISC area source model is used to simulate the
effects of fugitive emissions from sources such as storage
piles and slag dumps. The ISC open pit source model is used to
simulate fugitive emissions from below-grade open pits, such as
surface coal mines or stone quarries.
1.2.2 The Short-Term Volume Source Model
The ISC models use a virtual point source algorithm to
model the effects of volume sources, which means that an
imaginary or virtual point source is located at a certain
distance upwind of the volume source (called the virtual
distance) to account for the initial size of the volume source
1-46
-------
plume. Therefore, Equation (1-1) is also used to calculate
concentrations produced by volume source emissions.
There are two types of volume sources: surface-based
sources, which may also be modeled as area sources, and
elevated sources. An example of a surface-based source is a
surface rail line. The effective emission height he for a
surface-based source is usually set equal to zero. An example
of an elevated source is an elevated rail line with an
effective emission height he set equal to the height of the
rail line. If the volume source is elevated, the user assigns
the effective emission height he, i.e., there is no plume rise
associated with volume sources. The user also assigns initial
lateral (**0) and vertical (**0) dimensions for the volume
source. Lateral (xy) and vertical (xz) virtual distances are
added to the actual downwind distance x for the •• and ••
calculations. The virtual distances are calculated from
solutions to the sigma equations as is done for point sources
with building downwash.
The volume source model is used to simulate the effects of
emissions from sources such as building roof monitors and for
line sources (for example, conveyor belts and rail lines). The
north-south and east-west dimensions of each volume source used
in the model must be the same. Table 1-6 summarizes the
general procedures suggested for estimating initial lateral
(••0) and vertical (••0) dimensions for single volume sources
and for multiple volume sources used to represent a line
source. In the case of a long and narrow line source such as a
rail line, it may not be practical to divide the source into N
volume sources, where N is given by the length of the line
source divided by its width. The user can obtain an
approximate representation of the line source by placing a
smaller number of volume sources at equal intervals along the
line source, as shown in Figure 1-8. In general, the spacing
between individual volume sources should not be greater than
1-47
-------
twice the width of the line source. However, a larger spacing
can be used if the ratio of the minimum source-receptor
separation and the spacing between individual volume sources is
greater than about 3. In these cases, concentrations
calculated using fewer than N volume sources to represent the
line source converge to the concentrations calculated using N
volume sources to represent the line source as long as
sufficient volume sources are used to preserve the horizontal
geometry of the line source.
Figure 1-8 illustrates representations of a curved line
source by multiple volume sources. Emissions from a line
source or narrow volume source represented by multiple volume
sources are divided equally among the individual sources unless
there is a known spatial variation in emissions. Setting the
initial lateral dimension »»0 equal to W/2.15 in Figure 1-8(a)
or 2W/2.15 in Figure 1-8(b) results in overlapping Gaussian
distributions for the individual sources. If the wind
direction is normal to a straight line source that is
represented by multiple volume sources, the initial crosswind
concentration distribution is uniform except at the edges of
the line source. The doubling of **0 by the user in the
approximate line-source representation in Figure 1-8(b) is
offset by the fact that the emission rates for the individual
volume sources are also doubled by the user.
1-48
-------
TABLE 1-6
SUMMARY OF SUGGESTED PROCEDURES FOR ESTIMATING
INITIAL LATERAL DIMENSIONS • »0 AND
INITIAL VERTICAL DIMENSIONS • »0 FOR VOLUME AND LINE SOURCES
Procedure for Obtaining
Type of Source Initial Dimension
(a) Initial Lateral Dimensions (**0)
Single Volume Source **0 = length of side divided
by 4.3
Line Source Represented by • *0 = length of side divided
Adjacent Volume Sources (see by 2.15
Figure 1-8(a))
Line Source Represented by **0 = center to center
Separated Volume Sources (see distance divided by
Figure 1-8(b)) 2.15
(b) Initial Vertical Dimensions (••0)
Surface-Based Source (he ~ 0) • *0 = vertical dimension of
source divided by 2.15
Elevated Source (he > 0) on or * *0 = building height
Adjacent to a Building divided by 2.15
Elevated Source (he > 0) not • *0 = vertical dimension of
on or Adjacent to a Building source divided by 4.3
1.2.3 The Short-Term Area Source Model
The ISC Short Term area source model is based on a
numerical integration over the area in the upwind and crosswind
directions of the Gaussian point source plume formula given in
Equation (1-1). Individual area sources may be represented as
rectangles with aspect ratios (length/width) of up to 10 to 1.
In addition, the rectangles may be rotated relative to a north-
south and east-west orientation. As shown by Figure 1-9, the
effects of an irregularly shaped area can be simulated by
dividing the area source into multiple areas. Note that the
size and shape of the individual area sources in Figure 1-9
varies; the only requirement is that each area source must be a
1-49
-------
rectangle. As a result, an irregular area source can be
represented by a smaller number of area sources than if each
area had to be a square shape. Because of the flexibility in
specifying elongated area sources with the Short Term model, up
to an aspect ratio of about 10 to 1, the ISCST area source
algorithm may also be useful for modeling certain types of line
sources.
The ground-level concentration at a receptor located
downwind of all or a portion of the source area is given by a
double integral in the upwind (x) and crosswind (y) directions
as:
QAK
VD
y z
/
. V
( vV
. f) C 1 \
1 • • 1
V y/ .
\
r\\r
uy
i
dx
(1-65)
where:
QA = area source emission rate (mass per unit area per
unit time)
K = units scaling coefficient (Equation (1-1))
V = vertical term (see Section 1.1.6)
D = decay term as a function of x (see Section 1.1.7)
The Vertical Term is given by Equation (1-50) or Equation
(1-54) with the effective emission height, he, being the
physical release height assigned by the user. In general, he
should be set equal to the physical height of the source of
emissions above local terrain height. For example, the
emission height he of a slag dump is the physical height of the
slag dump.
Since the ISCST algorithm estimates the integral over the
area upwind of the receptor location, receptors may be located
within the area itself, downwind of the area, or adjacent to
the area. However, since •• goes to 0 as the downwind distance
goes to 0 (see Section 1.1.5.1), the plume function is infinite
1-50
-------
for a downwind receptor distance of 0. To avoid this
singularity in evaluating the plume function, the model
arbitrarily sets the plume function to 0 when the receptor
distance is less than 1 meter. As a result, the area source
algorithm will not provide reliable results for receptors
located within or adjacent to very small areas, with dimensions
on the order of a few meters across. In these cases, the
receptor should be placed at least 1 meter outside of the area.
In Equation (1-65), the integral in the lateral (i.e.,
crosswind or y) direction is solved analytically as follows:
exp
• •
(v I
— (1-66)
where erfc is the complementary error function.
In Equation (1-65), the integral in the longitudinal
(i.e., upwind or x) direction is approximated using numerical
methods based on Press, et al (1986). Specifically, the ISCST
model estimates the value of the integral, I, as a weighted
average of previous estimates, using a scaled down
extrapolation as follows:
VD _ / y} . T (I2N* *%)
1 " erfcM- dx--I2N-^ 1 {1_67)
•••••• I •• I 3
x y z V y/
where the integral term refers to the integral of the plume
function in the upwind direction, and IN and I2N refer to
successive estimates of the integral using a trapezoidal
approximation with N intervals and 2N intervals. The number of
intervals is doubled on successive trapezoidal estimates of the
integral. The ISCST model also performs a Romberg integration
by treating the sequence Ik as a polynomial in k. The Romberg
integration technique is described in detail in Section 4.3 of
Press, et al (1986). The ISCST model uses a set of three
criteria to determine whether the process of integrating in the
upwind direction has "converged." The calculation process will
1-51
-------
be considered to have converged, and the most recent estimate
of the integral used, if any of the following conditions is
true:
1) if the number of "halving intervals" (N) in the
trapezoidal approximation of the integral has reached
10, where the number of individual elements in the
approximation is given by 1 + 2N~1 = 513 for N of 10;
2) if the extrapolated estimate of the real integral
(Romberg approximation) has converged to within a
tolerance of 0.0001 (i.e., 0.01 percent), and at
least 4 halving intervals have been completed; or
3) if the extrapolated estimate of the real integral is
less than l.OE-10, and at least 4 halving intervals
have been completed.
The first condition essentially puts a time limit on the
integration process, the second condition checks for the
accuracy of the estimate of the integral, and the third
condition places a lower threshold limit on the value of the
integral. The result of these numerical methods is an estimate
of the full integral that is essentially equivalent to, but
much more efficient than, the method of estimating the integral
as a series of line sources, such as the method used by the PAL
2.0 model (Petersen and Rumsey, 1987).
1-52
-------
1.2.4 The Short-Term Open Pit Source Model
The ISC open pit source model is used to estimate impacts
for particulate emissions originating from a below-grade open
pit, such as a surface coal mine or a stone quarry. The ISC
models allow the open pit source to be characterized by a
rectangular shape with an aspect ratio (length/width) of up to
10 to 1. The rectangular pit may also be rotated relative to a
north-south and east-west orientation. Since the open pit
model does not apply to receptors located within the boundary
of the pit, the concentration at those receptors will be set to
zero by the ISC models.
The model accounts for partial retention of emissions
within the pit by calculating an escape fraction for each
particle size category. The variations in escape fractions
across particle sizes result in a modified distribution of mass
escaping from the pit. Fluid modeling has shown that within-
pit emissions have a tendency to escape from the upwind side of
the pit. The open pit algorithm simulates the escaping pit
emissions by using an effective rectangular area source using
the ISC area source algorithm described in Section 1.2.3. The
shape, size and location of the effective area source varies
with the wind direction and the relative depth of the pit.
Because the shape and location of the effective area source
varies with wind direction, a single open pit source should not
be subdivided into multiple pit sources.
The escape fraction for each particle size catagory, •*,
is calculated as follows:
(1 --vg /(«Ur))
where:
vg = is the gravitational settling velocity (m/s),
Ur = is the approach wind speed at 10m (m/s),
1-53
(1-68)
-------
•• = is the proportionality constant in the relationship
between flux from the pit and the product of Ur and
concentration in the pit (Thompson, 1994) .
The gravitational settling velocity, vg, is computed as
described in Section 1.3.2 for each particle size category.
Thompson (1994) used laboratory measurements of pollutant
residence times in a variety of pit shapes typical of actual
mines and determined that a single value of ••= 0.029 worked
well for all pits studied.
The adjusted emission rate (QJ for each particle size
category is then computed as:
Q; •••? "f ' Q (1-69)
where Q is the total emission rate (for all particles) within
the pit, •i is the original mass fraction for the given size
category, and •• is the escape fraction calculated from Equation
(1-68). The adjusted total emission rate (for all particles
escaping the pit), Qa, is the sum of the Q± for all particle
categories calculated from Equation 1-69. The mass fractions
(of particles escaping the pit), • •ai, for each category is:
'm "Qi / Qa (1-70)
Because of particle settling within the pit, the distribution
of mass escaping the pit is different than that emitted within
the pit. The adjusted total particulate emission rate, Qa, and
the adjusted mass fractions, • ai, reflect this change, and it
is these adjusted values that are used for modeling the open
pit emissions.
The following describes the specification of the location,
dimensions and adjusted emissions for the effective area source
1-54
-------
used for modeling open pit emissions. Consider an arbitrary
rectangular-shaped pit with an arbitrary wind direction as
shown in Figure 1-10. The steps that the model uses for
determining the effective area source are as follows:
1. Determine the upwind sides of the pit based on the
wind direction.
2. Compute the along wind length of the pit ((>) based on
the wind direction and the pit geometry . 0 varies
between the lengths of the two sides of the
rectangular pit as follows:
H. • »L-(1 • "/90) ••W-(»/90) (1-71)
where L is the long axis and W is the short axis of
the pit, and **is the wind direction relative to the
long axis (L) of the pit (therefore ••varies between
0° and 90°). Note that with this formulation and a
square pit, the value of 0 will remain constant with
wind direction at 0 = L = W. The along wind
dimension, 0, is the scaling factor used to normalize
the depth of the pit.
3. The user specifies the average height of emissions
from the floor of the pit (H) and the pit volume (V).
The effective pit depth (de) and the relative pit
depth (Dr) are then calculated as follows:
de "V/di-W) (1-72)
Dr •• (de«H)/« (1-73)
4. Based on observations and measurements in a wind
tunnel study (Perry, et al., 1994), it is clear that
the emissions within the pit are not uniformly
released from the pit opening. Rather, the emissions
show a tendency to be emitted primarily from an
upwind sub-area of the pit opening. Therefore an
effective area source (with Ae being the fractional
size relative to the entire pit opening) is used to
simulate the pit emissions. Ae represents a single
area source whose dimensions and location depend on
the effective depth of the pit and the wind
direction. Based on wind tunnel results, if Dr>0.2,
then the effective area is about 8% of the total
opening of the mine (i.e. Ae=0.08). If Dr<0.2, then
the fractional area increases as follows:
1-55
-------
De •• (1.0'i.7Dr1/3)1/2 (1-74)
When Dr = 0, which means that the height of emissions
above the floor equals the effective depth of the
pit, the effective area is equal to the total area of
the mine opening (i.e. Ae=1.0).
Having determined the effective area from which the model
will simulate the pit emissions, the specific dimensions of
this effective rectangular area are calculated as a function of
••such that (see Figure 1-10):
^(l-eos 26) T7 (1-75)
AW • »A -W
and
AL "A^08 20)-L (1-76)
Note that in equations 1-75 and 1-76, W is defined as the short
dimension of the pit and L is the long dimension; AW is the
dimension of the effective area aligned with the short side of
the pit and AL is the dimension of the effective area aligned
with the long side of the pit (see Figure 1-10). The
dimensions AW and AL are used by the model to define the shape
of the effective area for input to the area source algorithm
described in Section 1.2.3.
The emission rate, Qe, for the effective area is such that
Qe "On Me (1-77)
where Qa is the emission rate per unit area (from the pit after
adjustment for escape fraction) if the emissions were uniformly
released from the actual pit opening (with an area of L-W) .
That is, if the effective area is one-third of the total area,
1-56
-------
then the emission rate (per unit area) used for the effective
area is three times that from the full area.
Because of the high level of turbulence in the mine, the
pollutant is initially mixed prior to exiting the pit.
Therefore some initial vertical dispersion is included to
represent this in the effective area source. Using the
effective pit depth, de, as the representative dimension over
which the pollutant is vertically mixed in the pit, the initial
vertical dispersion value, ••0, is equal to de/4.3. Note that
4.3-**0 represents about 90% of a Gaussian plume (in the
vertical) , so that the mixing in the pit is assumed to
approximately equal the mixing in a plume .
Therefore, for the effective area source representing the
pit emissions, the initial dispersion is included with ambient
dispersion as:
U))1/2 (1-78)
For receptors close to the pit, the initial dispersion value
can be particularly important.
Once the model has determined the characteristics of the
effective area used to model pit emissions for a particular
hour, the area source algorithm described in Section 1.2.3 is
used to calculate the concentration or deposition flux values
at the receptors being modeled.
1.3 THE ISC SHORT-TERM DRY DEPOSITION MODEL
1.3.1 General
This section describes the ISC Short Term dry deposition
model, which is used to calculate the amount of material
1-57
-------
deposited (i.e., the deposition flux, Fd) at the surface from a
particle plume through dry deposition processes.
The Short Term dry deposition model is based on a dry
deposition algorithm (Pleim et al., 1984) contained in the Acid
Deposition and Oxidant Model (ADOM). This algorithm was
selected as a result of an independent model evaluation study
(EPA, 1994).
The deposition flux, Fd, is calculated as the product of
the concentration, »d, and a deposition velocity, vd, computed
at a reference height zd:
Fd '"j • vd (1-79)
The concentration value, *d, used in Equation (1-79) is
calculated according to Equation (1-1) with deposition effects
accounted for in the vertical term as described in Section
1.1.6.3. The calculation of deposition velocities is described
below.
1.3.2 Deposition Velocities
A resistance method is used to calculate the deposition
velocity, vd. The general approach used in the resistance
methods for estimating vd is to include explicit
parameterizations of the effects of Brownian motion, inertial
impaction, and gravitational settling. The deposition velocity
is written as the inverse of a sum of resistances to pollutant
transfer through various layers, plus gravitational settling
terms (Slinn and Slinn, 1980; Pleim et al., 1984):
1
r • • rj • • r rjV
•'-a •*-
-------
vg = the gravitational settling velocity (cm/s]
ra = the aerodynamic resistance (s/cm), and,
rd = the deposition layer resistance (s/cm).
Note that for large settling velocities, the deposition
velocity approaches the settling velocity (vd -> vg) , whereas,
for small settling velocities, vd tends to be dominated by the
ra and rd resistance terms.
In addition to the mass mean diameters (microns), particle
densities (gm/cm3) , and the mass fractions for each particle
size category being modeled, the dry deposition model also
requires surface roughness length (cm), friction velocity
(m/s), and Monin-Obukhov length (m). The surface roughness
length is specified by the user, and the meteorological
preprocessor (PCRAMMET or MPRM) calculates the friction
velocity and Monin-Obukhov length for input to the model.
The lowest few meters of the atmosphere can be divided
into two layers: a fully turbulent region where vertical fluxes
are nearly constant, and the thin quasi-laminar sublayer. The
resistance to transport through the turbulent, constant flux
layer is the aerodynamic resistance. It is usually assumed
that the eddy diffusivity for mass transfer within this layer
is similar to that for heat. The atmospheric resistance
formulation is based on Byun and Dennis (1995):
stable (L > 0) :
k u..
(1-81)
unstable (L < 0]
k u..
In
(z0/ L
(1-82)
1-59
-------
where, u, = the surface friction velocity (cm/s),
k = the von Karman constant (0.4),
z = the height above ground (m),
L = the Monin-Obukhov length (m),
zd = deposition reference height (m), and
z0 = the surface roughness length (m).
The coefficients used in the atmospheric resistance formulation
are those suggested by Dyer (1974). A minimum value for L of
1.0m is used for rural locations. Recommended minimum values
for urban areas are provided in the user's guides for the
meteorological preprocessor programs PCRAMMET and MPRM.
The approach used by Pleim et al. (1984) to parameterize
the deposition layer resistance terms is modified to include
Slinn's (1982) estimate for the inertial impaction term. The
resulting deposition layer resistance is:
where, Sc = the Schmidt number (Sc = */DB)
(dimensionless),
the viscosity of air (= 0.15 cm2/s),
DB = the Brownian diffusivity (cm2/s) of the
pollutant in air,
St = the Stokes number [St = (vg/g) (u,2 /•}]
(dimensionless),
g = the acceleration due to gravity (981 cm/s2) ,
The gravitational settling velocity, vg (cm/s), is
calculated as:
1-60
-------
where, •• = the particle density (g/cm3),
*AIR = tne air density (- 1.2 x 10"3 g/cm3),
dp = the particle diameter (urn),
Vi = the absolute viscosity of air (- 1.81 x 10"4
g/cm/s),
c2 = air units conversion constant (1 x 10"8
cm2/um2) , and
SCF = the slip correction factor, which is computed
as:
SCF ••!.•• ^-^ '- (1-85)
10'4 dp
and, x2, ai; a2, a3 are constants with values of 6.5 x 10"6,
1.257, 0.4, and 0.55 x 10"4, respectively.
The Brownian diffusivity of the pollutant (in cm/s) is
computed from the following relationship:
DD • »8 .09 x 10'
iO
T_S,
(1-86)
where Ta is the air temperature (°K).
The first term of Eqn. (1-83), involving the Schmidt
number, parameterizes the effects of Brownian motion. This
term controls the deposition rate for small particles. The
second term, involving the Stokes number, is a measure of the
importance of inertial impaction, which tends to dominate for
1-61
-------
intermediate-sized particles in the 2-20 urn diameter size
range.
The deposition algorithm also allows a small adjustment to
the deposition rates to account for possible phoretic effects.
Some examples of phoretic effects (Hicks, 1982) are:
THERMOPHORESIS: Particles close to a hot surface experience a
force directed away from the surface because, on the
average, the air molecules impacting on the side of the
particle facing the surface are hotter and more energetic.
DIFFUSIOPHORESIS: Close to an evaporating surface, a particle
is more likely to be impacted by water molecules on the
side of the particle facing the surface. Since the water
molecules have a lower molecular weight than the average
air molecule, there is a net force toward the surface,
which results in a small enhancement of the deposition
velocity of the particle.
A second effect is that the impaction of new water vapor
molecules at an evaporating surface displaces a certain
volume of air. For example, 18 g of water vapor
evaporating from 1 m2 will displace 22.4 liters of air at
standard temperature and pressure (STP) conditions (Hicks,
1982). This effect is called Stefan flow. The Stefan
flow effect tends to reduce deposition fluxes from an
evaporating surface. Conversely, deposition fluxes to a
surface experiencing condensation will be enhanced.
ELECTROPHORESIS: Attractive electrical forces have the
potential to assist the transport of small particles
through the quasi-laminar deposition layer, and thus could
increase the deposition velocity in situations with high
local field strengths. However, Hicks (1982) suggests
this effect is likely to be small in most natural
circumstances.
Phoretic and Stefan flow effects are generally small.
However, for particles in the range of 0.1 - 1.0 urn diameter,
which have low deposition velocities, these effects may not
always be negligible. Therefore, the ability to specify a
phoretic term to the deposition velocity is added (i.e., vd' =
vd + vd(phor), where vd' is the modified deposition velocity and
vd(Phor) is tne phoretic term) .
1-62
-------
Although the magnitude and sign of vd(phor) will vary, a
small, constant value of + 0.01 cm/s is used in the present
implementation of the model to represent combined phoretic
effects.
1-63
-------
1.3.3 Point and Volume Source Emissions
As stated in Equation (1-59) , deposition is modeled as the
product of the near- surf ace concentration (from Equation (1-1) )
times the deposition velocity (from Equation (1-80) ) .
Therefore, the vertical term given in Equation (1-54) that is
used to obtain the concentration at height z, subject to
particle settling and deposition, can be evaluated at height zd
for one particle size, and multiplied by a deposition velocity
for that particle size to obtain a corresponding "vertical
term" for deposition. Since more than one particle size
category is typically used, the deposition for the nth size
category must also include the mass fraction for the category:
F , • • • 5 ' V j
d n d n dn
(1-87)
. _ / vi"
exp
p • • • ••• 11
y z ^s
where K, •• Vd, and D were defined previously (Equations (1-1),
(1-54), and (1-63)) . The parameter Q.. is the total amount of
material emitted during the time period •• for which the
deposition calculation is made. For example, Q.. is the total
amount of material emitted during a 1-hour period if an hourly
deposition is calculated. To simplify the user input, and to
keep the maximum compatibility between input files for
concentration and deposition runs, the model takes emission
inputs in grams per second (g/s), and converts to grams per
hour for deposition calculations. For time periods longer than
an hour, the program sums the deposition calculated for each
hour to obtain the total deposition flux for the period. In
the case of a volume source, the user must specify the
effective emission height he and the initial source dimensions
•*0 and ••0. It should be noted that for computational
1-64
-------
NPD
purposes, the model calculates the quantity, . .*nvdnVdn, as
n't
the "vertical term."
1.3.4 Area and Open Pit Source Emissions
For area and open pit source emissions, Equation (1-65) is
changed to the form:
Fdn •••dn' Vdn
QAxK*nVdn
2 *' x
VdnD
y z
/
I y
•e s( yl2l
i * * i
V y/ .
\
r\\r
uy
/
(1-88)
dx
where K, D, Vd, and vd are defined in Equations (1-1) , (1-54),
(1-65) , and (1-80) . The parameter QA.. is the total mass per
unit area emitted over the time period ••for which deposition
is calculated. The area source integral is estimated as
described in Section 1.2.3.
1.4 THE ISC SHORT-TERM WET DEPOSITION MODEL
A scavenging ratio approach is used to model the
deposition of gases and particles through wet removal. In this
approach, the flux of material to the surface through wet
deposition (Fw) is the product of a scavenging ratio times the
concentration, integrated in the vertical:
Fw(x'y)** * *(x,*y,z)dz (1-89)
• •
o
where the scavenging ratio (• } has units of s"1. The
concentration value is calculated using Equation (1-1). Since
the precipitation is assumed to initiate above the plume
height, a wet deposition flux is calculated even if the plume
height exceeds the mixing height. Across the plume, the total
1-65
-------
flux to the surface must equal the mass lost from the plume so
that
• «D
••— Q (x) •• Fw(x, y) dy ••• «Q (x) /u (1-90)
dx
Solving this equation for Q(x), the source depletion
relationship is obtained as follows:
Q(x) --Qn e>Ax/u --Qn e'At (1-91)
where t = x/u is the plume travel time in seconds. As with dry
deposition (Section 1.3), the ratio Q(x)/Q0 is computed as a
wet depletion factor, which is applied to the flux term in
Equation (1-89). The wet depletion calculation is also
optional. Not considering the effects of wet depletion will
result in conservative estimates of both concentration and
deposition, since material deposited on the surface is not
removed from the plume.
The scavenging ratio is computed from a scavenging
coefficient and a precipitation rate (Scire et al., 1990):
R (1-92)
where the coefficient ••has units (s-mm/hr)"1, and the
precipitation rate R has units (mm/hr). The scavenging
coefficient depends on the characteristics of the pollutant
(e.g., solubility and reactivity for gases, size distribution
for particles) as well as the nature of the precipitation
(e.g., liquid or frozen). Jindal and Heinold (1991) have
analyzed particle scavenging data reported by Radke et al.
(1980), and found that the linear relationship of Equation
(1-90) provides a better fit to the data than the non-linear
assumption ••= »Rb. Furthermore, they report best-fit values
for • • as a function of particle size. These values of the
scavenging rate coefficient are displayed in Figure 1-11.
1-66
-------
Although the largest particle size included in the study is 10
urn, the authors suggest that •• should reach a plateau beyond 10
urn, as shown in Figure 1-11. The scavenging rate coefficients
for frozen precipitation are expected to be reduced to about
1/3 of the values in Figure 1-11 based on data for sulfate and
nitrate (Scire et al., 1990). The scavenging rate coefficients
are input to the model by the user.
The wet deposition algorithm requires precipitation type
(liquid or solid) and precipitation rate, which is prepared for
input to the model through the meteorological preprocessor
programs (PCRAMMET or MPRM).
1.5 ISC COMPLEX TERRAIN SCREENING ALGORITHMS
The Short Term model uses a steady-state, sector-averaged
Gaussian plume equation for applications in complex terrain
(i.e., terrain above stack or release height). Terrain below
release height is referred to as simple terrain; receptors
located in simple terrain are modeled with the point source
model described in Section 1.1. The sector average approach
used in complex terrain implies that the lateral (crosswind)
distribution of concentrations is uniform across a 22.5 degree
sector. The complex terrain screening algorithms apply only to
point source and volume source emissions; area source and open
pit emission sources are excluded. The complex terrain point
source model, which is based on the COMPLEXl model, is
described below. The description parallels the discussion for
the simple terrain algorithm in Section 1.1, and includes the
basic Gaussian sector-average equation, the plume rise
formulas, and the formulas used for determining dispersion
parameters.
1-67
-------
1.5.1 The Gaussian Sector Average Equation
The Short Term complex terrain screening algorithm for
stacks uses the steady-state, sector-averaged Gaussian plume
equation for a continuous elevated source. As with the simple
terrain algorithm described in Section 1.1, the origin of the
source's coordinate system is placed at the ground surface at
the base of the stack for each source and each hour. The x
axis is positive in the downwind direction, the y axis is
crosswind (normal) to the x axis and the z axis extends
vertically. The fixed receptor locations are converted to each
source's coordinate system for each hourly concentration
calculation. Since the concentrations are uniform across a
22.5 degree sector, the complex terrain algorithms use the
radial distance between source and receptor instead of downwind
distance. The calculation of the downwind, crosswind and
radial distances is described in Section 1.5.2. The hourly
concentrations calculated for each source at each receptor are
summed to obtain the total concentration produced at each
receptor by the combined source emissions.
For a Gaussian, sector-averaged plume, the hourly
concentration at downwind distance x (meters) and crosswind
distance y (meters) is given by:
, >R"
-------
y = lateral distance from the plume axis to the
receptor (m)
xy = lateral virtual distance for volume sources (see
Equation (1-35)), equals zero for point sources
(m)
us = mean wind speed (m/sec) at stack height
•• = standard deviation of the vertical concentration
distribution (m)
V = the Vertical Term (see Section 1.1.6)
D = the Decay Term (see Section 1.1.7)
CORR = the attenuation correction factor for receptors
above the plume centerline height (see Section
1.5.8)
Equation (1-93) includes a Vertical Term, a Decay Term,
and a vertical dispersion term (••) . The Vertical Term
includes the effects of source elevation, receptor elevation,
plume rise, limited vertical mixing, gravitational settling and
dry deposition.
1.5.2 Downwind, Crosswind and Radial Distances
The calculation of downwind and crosswind distances is
described in Section 1.1.2. Since the complex terrain
algorithms in ISC are based on a sector average, the radial
distance is used in calculating the plume rise (see Section
1.5.4) and dispersion parameters (see Section 1.5.5). The
radial distance is calculated as R = [x2 + y2]1/2, where x is the
downwind distance and y is the crosswind distance described in
Section 1.1.2.
1.5.3 Wind Speed Profile
See the discussion given in Section 1.1.3.
1.5.4 Plume Rise Formulas
The complex terrain algorithm in ISC uses the Briggs plume
rise equations described in Section 1.1.4. For distances less
1-69
-------
than the distance to final rise, the complex terrain algorithm
uses the distance-dependent plume height (based on the radial
distance) as described in Section 1.1.4.10. Since the complex
terrain algorithm does not incorporate the effects of building
downwash, the Schulman-Scire plume rise described in Section
1.1.4.11 is not used for complex terrain modeling. The plume
height is used in the calculation of the Vertical Term
described in Section 1.5.6.
1.5.5 The Dispersion Parameters
The dispersion parameters used in the complex terrain
algorithms of ISC are the same as the point source dispersion
parameters for the simple terrain algorithms described in
Section 1.1.5.1, except that the radial distance is used
instead of the downwind distance. Since the lateral
distribution of the plume in complex terrain is determined by
the sector average approach, the complex terrain algorithm does
not use the lateral dispersion parameter, ••. The procedure to
account for buoyancy-induced dispersion in the complex terrain
algorithm only affects the vertical dispersion term (see
Equation 1-48). Since the complex terrain algorithm does not
incorporate the effects of building downwash, the enhanced
dispersion parameters and virtual distances do not apply.
1.5.6 The Vertical Term
The Vertical Term used in the complex terrain algorithm in
ISC is the same as described in Section 1.1.6 for the simple
terrain algorithm, except that the plume height and dispersion
parameter input to the vertical term are based on the radial
distance, as described above, and that the adjustment of plume
height for terrain above stack base is different, as described
in Section 1.5.6.1.
1-70
-------
1.5.6.1 The Vertical Term in Complex Terrain.
The ISC complex terrain algorithm makes the following
assumption about plume behavior in complex terrain:
• The plume axis remains at the plume stabilization
height above mean sea level as it passes over complex
terrain for stable conditions (categories E and F), and
uses a "half-height" correction factor for unstable and
neutral conditions (categories A - D).
• The plume centerline height is never less than 10 m
above the ground level in complex terrain.
• The mixing height is terrain following, i.e, the mixing
height above ground at the receptor location is assumed
to be the same as the height above ground at the source
location.
• The wind speed is a function of height above the
surface (see Equation (1-6) ) .
Thus, a modified plume stabilization height he' is
substituted for the effective stack height he in the Vertical
Term given by Equation (1-50). The effective plume
stabilization height at the point x,y is given by:
he' "he •• (1«FT) Ht (1-94)
where:
he = plume height at point x,y without terrain
adjustment, as described in Section 1.5.4 (m)
Ht = z|(xy) - zs = terrain height of the receptor
location above the base of the stack (m)
z|(xy) = height above mean sea level of terrain at the
receptor location (x,y) (m)
zs = height above mean sea level of the base of the
stack (m)
FT = terrain adjustment factor, which is 0.5 for
stability categories A - D and 0.0 for stability
categories E and F.
1-71
-------
The effect of the terrain adjustment factor is that the plume
height relative to stack base is deflected upwards by an amount
equal to half of the terrain height as it passes over complex
terrain during unstable and neutral conditions. The plume
height is not deflected by the terrain under stable conditions.
1.5.6.2 The Vertical Term for Particle Deposition
The Vertical Term for particle deposition used in the
complex terrain algorithm in ISC is the same as described in
Section 1.1.6 for the simple terrain algorithm, except that the
plume height and dispersion parameter input to the vertical
term are based on the radial distance, as described above, and
that the adjustment of plume height for terrain above stack
base is different, as described in Section 1.5.6.2.
1.5.7 The Decay Term
See the discussion given in Section 1.1.7.
1.5.8 The Plume Attenuation Correction Factor
Deflection of the plume by complex terrain features during
stable conditions is simulated by applying an attenuation
correction factor to the concentration with height in the
sector of concern. This is represented by the variable CORR in
Equation (1-93). The attenuation correction factor has a value
of unity for receptors located at and below the elevation of
the plume centerline in free air prior to encountering terrain
effects, and decreases linearly with increasing height of the
receptor above plume level to a value of zero for receptors
located at least 400 m above the undisturded plume centerline
height. This relationship is shown in the following equation:
1-72
-------
CORR ••!.0
unstable/neutral
•1.0
'0.0
Hr < Om
Hr > 400m
(1-95)
(400" Hr)/400 • Hr < 400m
where:
CORR
attenuation correction factor, which is between 0
and 1
height of receptor above undisturbed plume height,
including height of receptor above local ground
(i.e., flagpole height)
1.5.9 Wet Deposition in Complex Terrain
See the discussion given in Section 1.4.
1.6 ISC TREATMENT OF INTERMEDIATE TERRAIN
In the ISC Short Term model, intermediate terrain is
defined as terrain that exceeds the height of the release, but
is below the plume centerline height. The plume centerline
height used to define whether a given receptor is on
intermediate terrain is the distance-dependent plume height
calculated for the complex terrain algorithm, before the
terrain adjustment (Section 1.5.6.2) is applied.
If the plume height is equal to or exceeds the terrain
height, then that receptor is defined as complex terrain for
that hour and that source, and the concentration is based on
the complex terrain screening algorithm only. If the terrain
1-73
-------
height is below the plume height but exceeds the physical
release height, then that receptor is defined as intermediate
terrain for that hour and source. For intermediate terrain
receptors, concentrations from both the simple terrain
algorithm and the complex terrain algorithm are obtained and
the higher of the two concentrations is used for that hour and
that source. If the terrain height is less than or equal to
the physical release height, then that receptor is defined as
simple terrain, and the concentration is based on the simple
terrain algorithm only.
For deposition calculations, the intermediate terrain
analysis is first applied to the concentrations at a given
receptor, and the algorithm (simple or complex) that gives the
highest concentration at that receptor is used to calculate the
deposition value.
1-74
-------
IL
z
m
x
FIGURE 1-1.
T-
tt
LINEAR DECAY FACTOR, A AS A FUNCTION OF
EFFECTIVE STACK HEIGHT, He. A SQUAT BUILDING IS
ASSUMED FOR SIMPLICITY.
1-75
-------
100
50
<°l
H «60
Building Ti«r
70
•2 H-
• 1
SO
Height of wake effects is HW - H + 1-5 LB
where LR is the lesser of the height of the
width.
East and west wind:
" Hw, = 60 + 1.5(50) = 135
HW2 = 80 + 1.5(10) = 95
Therefore, the lower building tier #1 width and
height
(][ = 60, W = 50) are used
North and South wind:
HW1 = 60 + 1.5(60) = 150
HW2 = 80 + 1.5(70) = 185
Therefore, the upper building tier #2
width and height
(H = 80, W = 70) are used
tltr 2 dart nates
tier 1 dominates
N-S wind
E-W wind
FIGURE 1-2.
ILLUSTRATION OF TWO TIERED BUILDING WITH
DIFFERENT TIERS DOMINATING DIFFERENT WIND
DIRECTIONS
1-76
-------
V W
FIGURE 1-3.
THE METHOD OF MULTIPLE PLUME IMAGES USED TO
SIMULATE PLUME REFLECTION IN THE ISC2 MODEL
1-77
-------
1—
Ł2
UJ
3:
P
kj
(Neutrot)^
-••'"""' /
(Stable)
*
1 r~~— _
\
(Stable)
\
\
/ Hm {ma.} \
/
.
H^mln}
i .
\
\
\
\
(Neutral) DAY,
"""""' ~"^ ^.^^^B^^MHH^^
4
/
.
/Hm
(Stable)
"T7}
" 1 "'" ,
* \ ~~— .
\
(Stable)
\
{max} \
\
DAY,^,
-—^Neutral)
^~> ;
/ '
xx
^X Hm 1
(Stable)
(Neutral)
— x"
(StabU)
rnaxX
•. j
t
TIME (LST)
(a) Urban Mixing Heights
DAY,.,
(NeutralJ_^-l
,. /
/
/
/
(Stable)
/
i- j_J
i — —
max}
DAY,
~- (Neutral)
~~ ; — 5
/
i
1 "*
/
/
(Stable)
/
/
t ^-^_
11 *•
1 .
OAY,^.,
" ^iNeutral )
*^- — -^.^^^^
/
(Stable)
/ Hm
N SR 1400 SS MN SR I4OO SS MN SR I4OO
f— '
t
mom}
SS M
TIME (UST)
(b) Rural Mixing Heights
FIGURE 1-4.
SCHEMATIC ILLUSTRATION OF (a) URBAN AND (b)
RURAL MIXING HEIGHT INTERPOLATION PROCEDURES
1-78
-------
FIGURE 1-5.
ILLUSTRATION OF PLUME BEHAVIOR IN ELEVATED
TERRAIN ASSUMED BY THE ISC2 MODEL
1-79
-------
2.0 -i
N 1.5 -
O
(fl
N
1.0 -
CF>
0.0
Depletion Foctor
Profile Correction
0.0
0.4
0.8
1.2
FIGURE 1-6.
ILLUSTRATION OF THE DEPLETION FACTOR FQ AND THE CORRESPOND!
CORRECTION FACTOR P(x,z).
1-80
-------
2.0 -i
N 1.5 -
D
C/)
N
1.0 -
cn
0.0
0.0
Depleted Profile
0.5
Original Profile
Concentration
1.5
2.0
FIGURE 1-7.
VERTICAL PROFILE OF CONCENTRATION BEFORE AND AFTER APPLYIN
P(x,z) SHOWN IN
FIGURE 1-6.
1-81
-------
cr =-*-
u 2.15
t
w
10
•9
8
(a) EXACT REPRESENTATION
2.15
W
2W-
•
2
•5
(b) APPROXIMATE REPRESENTATION
FIGURE 1-8.
EXACT AND APPROXIMATE REPRESENTATIONS OF A LINE
SOURCE BY MULTIPLE VOLUME SOURCES
1-82
-------
FIGURE 1-9
REPRESENTATION OF AN IRREGULARLY SHAPED AREA
SOURCE BY 4 RECTANGULAR AREA SOURCES
1-83
-------
Wind direction
W
effective
area
\
AL
4
L
AW
alongwind
length (I)
FIGURE 1-10. EFFECTIVE AREA AND ALONGWIND WIDTH FOR AN OPEN
PIT SOURCE
1-84
-------
Wet Scavenging Rate Coefficient (10 s )/mm—h
-1
1 10
Particle Diameter (microns)
100
FIGURE 1-11.
WET SCAVENGING RATE COEFFICIENT AS A FUNCTION OF
PARTICLE SIZE (JINDAL & HEINOLD, 1991)
1-85
-------
2.0 THE ISC LONG-TERM DISPERSION MODEL EQUATIONS
This section describes the ISC Long-Term model equations.
Where the technical information is the same, this section
refers to the ISC Short-Term model description in Section I for
details. The long-term model provides options for modeling the
same types of sources as provided by the short-term model. The
information provided below follows the same order as used for
the short-term model equations.
The ISC long-term model uses input meteorological data
that have been summarized into joint frequencies of occurrence
for particular wind speed classes, wind direction sectors, and
stability categories. These summaries, called STAR summaries
for STability ARray, may include frequency distributions over a
monthly, seasonal or annual basis. The long term model has the
option of calculating concentration or dry deposition values
for each separate STAR summary input and/or for the combined
period covered by all available STAR summaries. Since the wind
direction input is the frequency of occurrence over a sector,
with no information on the distribution of winds within the
sector, the ISC long-term model uses a Gaussian sector-average
plume equation as the basis for modeling pollutant emissions on
a long-term basis.
2.1 POINT SOURCE EMISSIONS
2.1.1 The Gaussian Sector Average Equation
The ISC long-term model makes the same basic assumption as
the short-term model. In the long-term model, the area
surrounding a continuous source of pollutants is divided into
sectors of equal angular width corresponding to the sectors of
the seasonal and annual frequency distributions of wind
direction, wind speed, and stability. Seasonal or annual
emissions from the source are partitioned among the sectors
according to the frequencies of wind blowing toward the
2-1
-------
sectors. The concentration fields calculated for each source
are translated to a common coordinate system (either polar or
Cartesian as specified by the user) and summed to obtain the
total due to all sources.
For a single stack, the mean seasonal concentration is
given by:
7 •'
K
Qf SVD
(2-1)
•R-
u
where:
K = units scaling coefficient (see Equation (1-1))
Q = pollutant emission rate (mass per unit time),
for the ith wind-speed category, the kth
stability category and the 1th season
f = frequency of occurrence of the ith wind-speed
category, the jth wind-direction category and
the kth stability category for the 1th season
• • •'• = the sector width in radians
R = radial distance from lateral virtual point
source (for building downwash) to the receptor =
[(x+xy)2 + y2]172 (m)
x = downwind distance from source center to
receptor, measured along the plume axis (m)
y = lateral distance from the plume axis to the
receptor (m)
xy = lateral virtual distance (see Equation (1-35) ) ,
equals zero for point sources without building
downwash, and for downwash sources that do not
experience lateral dispersion enhancement (m)
S = a smoothing function similar to that of the AQDM
(see Section 2.1.8)
us = mean wind speed (m/sec) at stack height for the
ith wind-speed category and kth stability
category
2-2
-------
•• = standard deviation of the vertical concentration
distribution (m) for the kth stability category
V = the Vertical Term for the ith wind-speed
category, kth stability category and 1th season
D = the Decay Term for the ith wind speed category
and kth stability category
The mean annual concentration at the point (r,»} is
calculated from the seasonal concentrations using the
expression:
4
•• "0.25 . . «r (2-2)
1 -i
The terms in Equation (2-1) correspond to the terms
discussed in Section 1.1 for the short-term model except that
the parameters are defined for discrete categories of
wind-speed, wind-direction, stability and season. The various
terms are briefly discussed in the following subsections. In
addition to point source emissions, the ISC long-term
concentration model considers emissions from volume and area
sources. These model options are discussed in Section 2.2.
The optional algorithms for calculating dry deposition are
discussed in Section 2.3.
2.1.2 Downwind and Crosswind Distances
See the discussion given in Section 1.1.2.
2.1.3 Wind Speed Profile
See the discussion given in Section 1.1.3.
2.1.4 Plume Rise Formulas
See the discussion given in Section 1.1.4.
2-3
-------
2.1.5 The Dispersion Parameters
2.1.5.1 Point Source Dispersion Parameters.
See Section 1.1.5.1 for a discussion of the procedures use
to calculate the standard deviation of the vertical
concentration distribution •• for point sources (sources
without initial dimensions). Since the long term model assumes
a uniform lateral distribution across the sector width, the
model does not use the standard deviation of the lateral
dispersion, •• (except for use with the Schulman-Scire plume
rise formulas described in Section 1.1.4.11).
2.1.5.2 Lateral and Vertical Virtual Distances.
See Section 1.1.5.2 for a discussion of the procedures
used to calculate vertical virtual distances. The lateral
virtual distance is given by:
• "r0cot (2-3)
•^ I <~> I
where r0 is the effective source radius in meters. For volume
sources (see Section 2.2.2), the program sets r0 equal to
2.15**0, where **0 is the initial lateral dimension. For area
sources (see Section 2.2.3), the program sets r0 equal to x0/y**
where x0 is the length of the side of the area source. For
plumes affected by building wakes (see Section 1.1.5.2), the
program sets r0 equal to 2.15 ••' where ••' is given for squat
buildings by Equation (1-41), (1-42), or (1-43) for downwind
distances between 3 and 10 building heights and for tall
buildings by Equation (1-44) for downwind distances between 3
and 10 building widths. At downwind distances greater than 10
building heights for Equation (1-41), (1-42), or (1-43), ••' is
held constant at the value of ••' calculated at a downwind
distance of 10 building heights. Similarly, at downwind
distances greater than 10 building widths for Equation (1-44),
2-4
-------
••' is held constant at the value of ••' calculated at a
downwind distance of 10 building widths.
2.1.5.3 Procedures Used to Account for the Effects of
Building Wakes on Effluent Dispersion.
With the exception of the equations used to calculate the
lateral virtual distance, the procedures used to account for
the effects of building wake effects on effluent dispersion are
the same as those outlined in Section 1.1.5.3 for the
short-term model. The calculation of lateral virtual distances
by the long-term model is discussed in Section 2.1.5.2 above.
2.1.5.4 Procedures Used to Account for Buoyancy-Induced
Dispersion.
See the discussion given in Section 1.1.5.4.
2.1.6 The Vertical Term
2.1.6.1 The Vertical Term for Gases and Small
Particulates.
Except for the use of seasons and discrete categories of
wind-speed and stability, the Vertical Term for gases and small
particulates corresponds to the short term version discussed in
Section 1.1.6. The user may assign a separate mixing height zi
to each combination of wind-speed and stability category for
each season.
As with the Short-Term model, the Vertical Term is changed
to the form:
(2-4)
at downwind distances where the •*/zi ratio is greater than or
equal to 1.6. Additionally, the ground-level concentration is
set equal to zero if the effective stack height he exceeds the
mixing height z±. As explained in Section 1.1.6.1, the ISC
2-5
-------
model currently assumes unlimited mixing for the E and F
stability categories.
2.1.6.2 The Vertical Term in Elevated Terrain.
See the discussion given in Section 1.1.6.2.
2.1.6.3 The Vertical Term for Large Particulates.
Section 1.1.6.3 discusses the differences in the
dispersion of large particulates and the dispersion of gases
and small particulates and provides the guidance on the use of
this option. The Vertical Term for large particulates is given
by Equation (1-53).
2.1.7 The Decay Term
See the discussion given in Section 1.1.7.
2.1.8 The Smoothing Function
As shown by Equation (2-1), the rectangular concentration
distribution within a given angular sector is modified by the
function S{*} which smooths discontinuities in the
concentration at the boundaries of adjacent sectors. The
centerline concentration in each sector is unaffected by
contribution from adjacent sectors. At points off the sector
centerline, the concentration is a weighted function of the
concentration at the centerline and the concentration at the
centerline of the nearest adjoining sector. The smoothing
function is given by:
(•••'- -|- :'•"•'!)
s •• for \* f • « • |
(2-5)
or
.. 0 for •» " • » •'
2-6
-------
where:
•?' = the angle measured in radians from north to the
centerline of the jth wind-direction sector
• •' = the angle measured in radians from north to the
receptor point (R, •} where R, defined above for
equation 2-1, is measured from the lateral virtual
source.
2.2 NON-POINT SOURCE EMISSIONS
2.2.1 General
As explained in Section 1.2.1, the ISC volume, area and
open pit sources are used to simulate the effects of emissions
from a wide variety of industrial sources. Section 1.2.2
provides a description of the volume source model, Section
1.2.3 provides a description of the area source model, and
Section 1.2.4 provides a description of the open pit model.
The following subsections give the volume, area and open pit
source equations used by the long-term model.
2.2.2 The Long-Term Volume Source Model
The ISC Long Term Model uses a virtual point source
algorithm to model the effects of volume sources. Therefore,
Equation (2-1) is also used to calculate seasonal average
ground-level concentrations for volume source emissions. The
user must assign initial lateral (**0) and vertical (**0)
dimensions and the effective emission height he. A discussion
of the application of the volume source model is given in
Section 1.2.2.
2.2.3 The Long-Term Area Source Model
The ISC Long Term Area Source Model is based on the
numerical integration algorithm for modeling area sources used
by the ISC Short Term model, which is described in detail in
Section 1.2.3. For each combination of wind speed class,
2-7
-------
stability category and wind direction sector in the STAR
meteorological frequency summary, the ISC Long Term model
calculates a sector average concentration by integrating the
results from the ISC Short Term area source algorithm across
the sector. A trapezoidal integration is used, as follows:
f ,
^
N .
[2-6a)LD (
6b)'
where:
S =
fi;j =
•(•}
the sector average concentration value for the
ith sector
the sector width
the frequency of occurrence for the jth wind
direction in the ith sector
the error term - a criterion of ••(•} < 2 percent
is used to check for convergence of the sector
average calculation
the concentration value, based on the numerical
integration algorithm using Equation (1-58) for
the jth wind direction in the ith sector
. th
th
the j wind direction in the i sector, j = 1
and N correspond to the two boundaries of the
sector.
The application of Equation (2-6a) to calculate the sector
average concentration from area sources is an iterative
process. Calculations using the ISC Short Term algorithm
(Equation (1-58)) are initially made for three wind directions,
corresponding to the two boundaries of the sector and the
centerline direction. The algorithm then calculates the
concentration for wind directions midway between the three
directions, for a total of five directions, and calculates the
2-8
-------
error term. If the error is less than 2 percent, then the
concentration based on five directions is used to represent the
sector average, otherwise, additional wind directions are
selected midway between each of the five directions and the
process continued. This process continues until the
convergence criteria, described below, are satisfied.
In order to avoid abrupt changes in the concentrations at
the sector boundaries with the numerical integration algorithm,
a linear interpolation is used to determine the frequency of
occurrence of each wind direction used for the individual
simulations within a sector, based on the frequencies of
occurrence in the adjacent sectors. This "smoothing" of the
frequency distribution has a similar effect as the smoothing
function used for the ISC Long Term point source algorithm,
described in Section 2.1.8. The frequency of occurrence of the
jth wind direction between sectors i and i+1 can be calculated
as :
f .. • »F- • •(• f ,• • f- ) - — - - — (9 -
11 i v i • i 11 ' , x v ^
where:
Fi = the frequency of occurrence for the ith sector
Fi+1 = the frequency of occurrence for the i + lth sector
• i = the central wind direction for the ith sector
• j+1 = the central wind direction for the i + lth sector
• v = the specific wind direction between »j and • j+1
fi;j = the interpolated (smoothed) frequency of
occurrence for the specific wind direction »v
The ISCLT model uses a set of three criteria to determine
whether the process of calculating the sector average
concentration has "converged." The calculation process will be
2-9
-------
considered to have converged, and the most recent estimate of
the trapezoidal integral used, if any of the following
conditions is true:
1) if the number of "halving intervals" (N) in the
trapezoidal approximation of the sector average has
reached 10, where the number of individual elements
in the approximation is given by 1 + 2N~1 = 513 for N
of 10;
2) if the estimate of the sector average has converged
to within a tolerance of 0.02 (i.e., 2 percent), for
two successive iterations, and at least 2 halving
intervals have been completed (a minimum of 5 wind
direction simulations); or
3) if the estimate of the sector average concentration
is less than l.OE-10, and at least 2 halving
intervals have been completed.
The first condition essentially puts a time limit on the
integration process, the second condition checks for the
accuracy of the estimate of the sector average, and the third
condition places a lower threshold limit that avoids
convergence problems associated with very small concentrations
where truncation error may be significant.
2-10
-------
2.2.4 The Long -Term Open Pit Source Model
The ISC Long Term Open Pit Source Model is based on the
use of the long term area source model described in Section
2.2.3. The escape fractions and adjusted mass distribution for
particle emissions from an open pit, and the determination of
the size, shape and location of the effective area source used
to model open pit emissions are described in Section 1.2.4.
For the Long Term model, a sector average value for open pit
sources is calculated by determining an effective area for a
range of wind directions within the sector and increasing the
number of wind directions used until the result converges, as
described in Section 2.2.3 for the Long Term area source model.
The contribution from each effective area used within a sector
is calculated using the Short Term area source model described
in Section 1.2.3.
2.3 THE ISC LONG-TERM DRY DEPOSITION MODEL
2.3.1 General
The concepts upon which the ISC long-term dry deposition
model are based are discussed in Sections 1.1.6.3 and 1.3.
2.3.2 Point and Volume Source Emissions
The seasonal deposition at the point located at a
particular distance (r) and direction (•} with respect to the
base of a stack or the center of a volume source for
particulates in the nth particle size category is given by:
K'n QxfSVdnD
- — (2-7)
- ••
2 • »R 2 • • * • i J . k
where the vertical term for deposition, Vdn, was defined in
Section 1.3.2. K and D are described in Equations (1-1) and
(1-63) , respectively. Q.. is the product of the total time
during the 1th season, of the seasonal emission rate Q for the
2-11
-------
1th wind-speed category, kth stability category. For example,
if the emission rate is in grams per second and there are 92
days in the summer season (June, July, and August) , Q..^ is
given by 7.95 x 10s Q1_3. It should be noted that the user need
not vary the emission rate by season or by wind speed and
stability. If an annual average emission rate is assumed, Q..
is equal to 3.15 x 107 Q for a 365-day year. For a plume
comprised of N particle size categories, the total seasonal
deposition is obtained by summing Equation (2-7) over the N
particle size categories. The program also sums the seasonal
deposition values to obtain the annual deposition.
2.3.3 Area and Open Pit Source Emissions
The area and open pit source dry deposition calculations
for the ISCLT model are based on the numerical integration
algorithm for modeling area sources used by the ISCST model.
Section 1.3.3, Equation (1-61), describes the numerical
integration for the Short Term model that is applied to
specific wind directions by the Long Term model in a
trapezoidal integration to calculate the sector average. The
process of calculating sector averages for area sources in the
Long Term model is described by Equation (2-6) in Section
2.2.3.
2-12
-------
3.0 REFERENCES
Bowers, J.F., J.R. Bjorklund and C.S. Cheney, 1979: Industrial
Source Complex (ISC) Dispersion Model User's Guide. Volume
I, EPA-450/4-79-030, U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina 27711.
Bowers, J.R., J.R. Bjorklund and C.S. Cheney, 1979: Industrial
Source Complex (ISC) Dispersion Model User's Guide. Volume
II, EPA-450/4-79-031, U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina 27711.
Briggs, G.A., 1969, Plume Rise, USAEC Critical Review Series,
TID-25075, National Technical Information Service,
Springfield, Virginia 22161.
Briggs, G.A., 1979: Some Recent Analyses of Plume Rise
Observations, In Proceedings of the Second International
Clean Air Congress, Academic Press, New York.
Briggs, G.A., 1972: Discussion on Chimney Plumes in Neutral
and Stable Surroundings. Atmos. Environ., 6., 507-510.
Briggs, G.A., 1974: Diffusion Estimation for Small Emissions.
In ERL, ARL USAEC Report ATDL-106. U.S. Atomic Energy
Commission, Oak Ridge, Tennessee.
Briggs, G.A., 1975: Plume Rise Predications. In Lectures on
Air Pollution and Environmental Impact Analysis, American
Meteorological Society, Boston, Massachusetts.
Byun, D.W. and R. Dennis, 1995: Design Artifacts in Eulerian
Air Quality Models: Evaluation of the Effects of Layer
Thickness and Vertical Profile Correction on Surface Ozone
Concentrations. Atmos. Environ., 29, 105-126.
Chico, T. and J.A. Catalano, 1986: Addendum to the User's
Guide for MPTER. Contract No. EPA 68-02-4106, U.S.
Environmental Protection Agency, Research Triangle Park,
North Carolina 27711.
Cramer, H.E., et al., 1972: Development of Dosage Models and
Concepts. Final Report Under Contract DAAD09-67-C-0020(R)
with the U.S. Army, Desert Test Center Report DTC-TR-609,
Fort Douglas, Utah.
Dumbauld, R.K. and J.R. Bjorklund, 1975: NASA/MSFC Multilayer
Diffusion Models and Computer Programs -- Version 5. NASA
Contractor Report No. NASA CR-2631, National Aeronautics
and Space Administration, George C. Marshall Space Center,
Alabama.
3-1
-------
Dyer, A.J., 1974: A review of flux-profile relationships.
Boundary-Layer Meteorol. , 7., 363-372.
Environmental Protection Agency, 1985: Guideline for
Determination of Good Engineering Practice Stack Height
(Technical Support Document for the Stack Height
Regulations) - Revised, EPA-450/4-80-023R, U.S.
Environmental Protection Agency, Research Triangle Park,
NC 27711. (NTIS No. PB 85-225241)
Environmental Protection Agency, 1992. Comparison of a Revised
Area Source Algorithm for the Industrial Source Complex
Short Term Model and Wind Tunnel Data. EPA Publication
No. EPA-454/R-92-014. U.S. Environmental Protection
Agency, Research Triangle Park, NC. (NTIS No. PB 93-
226751)
Environmental Protection Agency, 1992. Sensitivity Analysis of
a Revised Area Source Algorithm for the Industrial Source
Complex Short Term Model. EPA Publication No. EPA-454/R-
92-015. U.S. Environmental Protection Agency, Research
Triangle Park, NC. (NTIS No. PB 93-226769)
Environmental Protection Agency, 1992. Development and
Evaluation of a Revised Area Source Algorithm for the
Industrial Source Complex Long Term Model. EPA
Publication No. EPA-454/R-92-016. U.S. Environ-mental
Protection Agency, Research Triangle Park, NC. (NTIS No.
PB 93-226777)
Environmental Protection Agency, 1994. Development and Testing
of a Dry Deposition Algorithm (Revised). EPA Publication
No. EPA-454/R-94-015. U.S. Environmental Protection
Agency, Research Triangle Park, NC. (NTIS No. PB 94-
183100)
Gifford, F.A., Jr. 1976: Turbulent Diffusion - Typing Schemes:
A Review. Nucl. Saf., 17, 68-86.
Hicks, B.B., 1982: Critical assessment document on acid
deposition. ATDL Contrib. File No. 81/24, Atmos. Turb.
and Diff. Laboratory, Oak Ridge, TN.
Holzworth, G.C., 1972: Mixing Heights, Wind Speeds and
Potential for Urban Air Pollution Throughout the
Contiguous United States. Publication No. AP-101, U.S.
Environmental Protection Agency, Research Triangle Park,
North Carolina 27711.
Horst, T.W., 1983: A correction to the Gaussian source-
depletion model. In Precipitation Scavenging, Dry
Deposition and Resuspension, H.R. Pruppacher, R.G.
Semonin, W.G.N. Slinn, eds., Elsevier, NY.
3-2
-------
Huber, A.H. and W.H. Snyder, 1976: Building Wake Effects on
Short Stack Effluents. Preprint Volume for the Third
Symposium on Atmospheric Diffusion and Air Quality,
American Meteorological Society, Boston, Massachusetts.
Huber, A.H. and W.H. Snyder, 1982. Wind tunnel investigation
of the effects of a rectangular-shaped building on
dispersion of effluents from short adjacent stacks. Atmos.
Environ.. 176. 2837-2848.
Huber, A.H., 1977: Incorporating Building/Terrain Wake Effects
on Stack Effluents. Preprint Volume for the Joint
Conference on Applications of Air Pollution Meteorology,
American Meteorological Society, Boston, Massachusetts.
Jindal, M. and D. Heinold, 1991: Development of particulate
scavenging coefficients to model wet deposition from
industrial combustion sources. Paper 91-59.7, 84th Annual
Meeting - Exhibition of AWMA, Vancouver, BC, June 16-21,
1991.
McDonald, J.E., I960: An Aid to Computation of Terminal Fall
Velocities of Spheres. J. Met., 17, 463.
McElroy, J.L. and F. Pooler, 1968: The St. Louis Dispersion
Study. U.S. Public Health Service, National Air Pollution
Control Administration, Report AP-53.
National Climatic Center, 1970: Card Deck 144 WEAN Hourly
Surface Observations Reference Manual 1970, Available from
the National Climatic Data Center, Asheville, North
Carolina 28801.
Pasquill, F., 1976: Atmospheric Dispersion Parameters in
Gaussian Plume Modeling. Part II. Possible Requirements
for Change in the Turner Workbook Values.
EPA-600/4-76-030b, U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina 27711.
Perry, S.G., R.S. Thompson, and W.B. Petersen, 1994:
Considerations for Modeling Small-Particulate Impacts from
Surface Coal Mining Operations Based on Wind-Tunnel
Simulations. Proceedings Eighth Joint Conference on
Applications of Air Pollution Meteorology, January 23-28,
Nashville, TN.
Petersen, W.B. and E.D. Rumsey, 1987: User's Guide for PAL 2.0
- A Gaussian-Plume Algorithm for Point, Area, and Line
Sources, EPA/600/8-87/009, U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina.
Pleim, J., A. Venkatram and R. Yamartino, 1984: ADOM/TADAP
model development program. Volume 4. The dry deposition
3-3
-------
module. Ontario Ministry of the Environment, Rexdale,
Ontario.
Press, W., B. Flannery, S. Teukolsky, and W. Vetterling, 1986:
Numerical Recipes, Cambridge University Press, New York,
797 pp.
Schulman, L.L. and S.R. Hanna, 1986: Evaluation of Downwash
Modifications to the Industrial Source Complex Model. J.
Air Poll. Control Assoc.. 36. (3), 258-264.
Schulman, L.L. and J.S. Scire, 1980: Buoyant Line and Point
Source (BLP) Dispersion Model User's Guide. Document
P-7304B, Environmental Research and Technology, Inc.,
Concord, MA.
Scire, J.S. and L.L. Schulman, 1980: Modeling Plume Rise from
Low-Level Buoyant Line and Point Sources. Proceedings
Second Joint Conference on Applications of Air Pollution
Meteorology, 24-28 March, New Orleans, LA. 133-139.
Scire, J.S., D.G. Strimaitis and R.J. Yamartino, 1990: Model
formulation and user's guide for the CALPUFF dispersion
model. Sigma Research Corp., Concord, MA.
Slinn, W.G.N., 1982: Predictions for particle deposition to
vegetative canopies. Atmos. Environ., 16, 1785-1794.
Slinn, S.A. and W.G.N. Slinn, 1980: Predictions for particle
deposition and natural waters. Atmos. Environ., 14, 1013-
1016.
Thompson, R.S., 1994: Residence Time of Contaminants Released
in Surface Coal Mines -- A Wind Tunnel Study. Proceedings
Eighth Joint Conference on Applications of Air Pollution
Meteorology, January 23-28, Nashville, TN.
Touma, J.S., J.S. Irwin, J.A. Tikvart, and C.T. Coulter, 1995.
A Review of Procedures for Updating Air Quality Modeling
Techniques for Regulatory Modeling Programs. J. App.
Meteor.. 31, 731-737.
Turner, D.B., 1970: Workbook of Atmospheric Dispersion
Estimates. PHS Publication No. 999-AP-26. U.S.
Department of Health, Education and Welfare, National Air
Pollution Control Administration, Cincinnati, Ohio.
Yamartino, R.J., J.S. Scire, S.R. Hanna, G.R. Carmichael and
Y.S. Chang, 1992: The CALGRID mesoscale photochemical
grid model. Volume I. Model formulation. Atmos.
Environ.. 26A. 1493-1512.
3-4
-------
INDEX
Area source
deposition algorithm 1-61, 2-12
for the Long Term model 2-7
for the Short Term model 1-43, 1-46
Atmospheric resistance 1-38, 1-56
Attenuation correction factor
in complex terrain 1-68
Briggs plume rise formulas
buoyant plume rise 1-7, 1-9
momentum plume rise 1-8, 1-10
stack tip downwash 1-6
Building downwash procedures 1-23, 2-4
and buoyancy-induced dispersion 1-30
effects on dispersion parameters 1-21
for the Long Term model 1-64, 2-2, 2-4, 2-5
general 1-5, 1-22
Huber and Snyder 1-23
Schulman and Scire 1-5, 1-12, 1-28, 1-29
Schulman-Scire plume rise 1-12, 1-14
virtual distances 1-20
wake plume height 1-11
Buoyancy flux 1-6, 1-13
Buoyancy-induced dispersion 1-30, 2-5
Buoyant plume rise
stable 1-9
unstable and neutral 1-7
Cartesian receptor network 1-3
Complex terrain modeling
Short Term model 1-63, 1-69
Crossover temperature difference 1-7
Crosswind distance 1-2, 1-3, 1-4, 1-64, 1-65, 2-3
Decay coefficient 1-42
Decay term 1-3, 1-42, 1-65
for the Long Term model 1-65, 2-3, 2-6
for the Short Term model 1-42, 1-68
Depletion
for the dry deposition algorithm 1-35, 1-42
for the wet deposition algorithm 1-62
Deposition layer resistance 1-57
Deposition velocity 1-34, 1-55
Direction-specific building dimensions 1-22, 1-29
with Huber-Snyder downwash 1-29
Dispersion coefficients
see Dispersion parameters 1-14, 1-66
Dispersion parameters
for the Long Term model 2-4
McElroy-Pooler 1-15, 1-19
Pasquill-Gifford 1-14, 1-16, 1-17, 1-18
INDEX-1
-------
Distance-dependent plume rise 1-13
Downwind distance 1-2, 1-3, 1-4, 1-64, 1-65, 2-3
and virtual distance 1-20
for area sources 1-47
for building wake dispersion 1-24
for dispersion coefficients 1-14
Dry deposition 1-3, 1-65, 2-11
for the Long Term model 2-11
for the Short Term model 1-54
Elevated terrain 1-33, 1-67, 2-6
truncation above stack height 1-34
Entrainment coefficient 1-14
Final plume rise 1-30
distance to 1-7
stable 1-9, 1-10
unstable or neutral 1-7, 1-8
Flagpole receptor 1-32
Gaussian plume model 1-2,1-63
sector averages for complex terrain 1-63
sector averages for Long Term 2-1
GEP stack height 1-12, 1-29
Gradual plume rise 1-10
for buoyant plumes 1-10
for Schulman-Scire downwash 1-13
stable momentum 1-11
unstable and neutral momentum 1-11
used for wake plume height 1-11
Half life 1-43
Huber-Snyder downwash algorithm 1-5
Initial lateral dimension
for the Long Term model 2-4
for volume sources 1-45, 1-46
Initial plume length
Schulman-Scire downwash 1-12
Initial plume radius
Schulman-Scire downwash 1-13
Initial vertical dimension
for volume sources 1-46
Intermediate terrain 1-69
Jet entrainment coefficient 1-11, 1-14
Lateral dispersion parameters 1-16, 1-19, 1-30
for the Long Term model 2-4
Lateral virtual distance
for the Long Term model 1-64, 2-2, 2-4
Lateral virtual distances
for building downwash 1-26
INDEX-2
-------
Line source
approximation for Schulman-Scire sources . . . 1-12, 1-13
Line sources, modeled as volumes . 1-43, 1-44, 1-45, 1-46, 2-7
Linear decay factor
Schulman-Scire downwash 1-13, 1-29
Long-term dispersion model 2-1
McElroy-Pooler dispersion parameters
see Dispersion parameters 1-19
Mixing heights 1-33
Momentum flux 1-6, 1-13
Momentum plume rise 1-11, 1-23, 1-29
stable 1-10
unstable and neutral 1-8
Open pit source
deposition algorithm 1-61, 2-12
for the Long Term model 2-11
for the Short Term model 1-50
Open pit sources 1-43, 1-50
Pasquill-Gifford dispersion parameters
see Dispersion parameters 1-16
Plume rise
for Schulman-Scire downwash 1-12
for the Long Term model 2-3
for the Short Term model 1-5, 1-65
Point source
deposition algorithm 1-60, 2-11
dispersion parameters 1-14, 2-4
for the Long Term model 2-1
for the Short Term model 1-2
Polar receptor network 1-3
Receptors
calculation of source-receptor distances .... 1-3, 1-4
Rural
dispersion parameters 1-14
virtual distances 1-20, 1-25
Schulman-Scire downwash algorithm 1-5
Short-term dispersion model 1-1
Sigma-y 1-14, 1-66
Sigma-z 1-14, 1-66
Smoothing function
for the Long Term model 2-2,2-6
Stability parameter 1-8, 1-14
Stack-tip downwash 1-6
for wake plume height 1-12
Uniform vertical mixing 1-32
Urban
decay term for S02 1-43
INDEX-3
-------
dispersion parameters 1-15
virtual distances 1-20, 1-25
Vertical dispersion parameters 1-17, 1-18, 1-19
Vertical term 1-3, 1-47, 1-65, 2-5
for gases and small particulates 1-31
for large particulates 2-6
for the Long Term model 1-65, 2-3, 2-5
for the Short Term model 1-31, 1-66
for uniform vertical mixing 1-32
in complex terrain 1-67
in elevated terrain 1-33, 2-6
Vertical virtual distances
for building downwash 1-24, 1-25
Virtual distances 1-20, 1-21, 1-28, 1-29, 1-44, 1-47
for the Long Term model 2-4,2-5
for volume sources 1-44
Virtual point source 1-43, 1-64, 2-2, 2-7
Volume source 1-46
deposition algorithm 1-60, 2-11
for the Long Term model 2-7
for the Short Term model 1-43
Wet deposition
for the Short Term model 1-61
Wind speed
minimum wind speed for modeling 1-5
Wind speed profile 1-4, 1-65, 2-3
INDEX-4
-------
ADDENDUM
USER'S GUIDE FOR THE
INDUSTRIAL SOURCE COMPLEX (ISC3) DISPERSION MODELS
VOLUME II - DESCRIPTION OF MODEL ALGORITHMS
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air Quality Planning and Standards
Emissions, Monitoring, and Analysis Division
Research Triangle Park, North Carolina 27711
June 1999
-------
ACKNOWLEDGMENTS
The Addendum to the User's Guide for the ISC3 Models has been prepared by
Roger W. Erode of Pacific Environmental Services, Inc., Research Triangle Park, North
Carolina, under subcontract to EC/R, Inc., Chapel Hill, North Carolina. This effort has been
funded by the Environmental Protection Agency under Contract No. 68D98006, with Dennis
G. Atkinson as Work Assignment Manager.
INDEX-vi
-------
TECHNICAL DESCRIPTION FOR THE
REVISED ISCST3 MODEL (DATED 99155)
This document provides a technical description of model algorithms for recent
enhancements of the ISCST3 model, including the most recent version dated 99155. The
algorithms described in this Addendum include the gas dry deposition algorithms based on the
draft GDISCDFT model (dated 96248), and the optimizations of the area source algorithm.
Both of these enhancements are associated with the non-regulatory default TOXICS option
introduced with version 99155 of ISCST3. A brief description of the user instructions for these
new options is presented in the accompanying Addendum to Volume I of the ISC3 model
user's guide (ISC3ADD1.WPD).
Gas Dry Deposition Algorithms
The ISCST3 dry deposition algorithm for gaseous pollutants is based on the algorithm
contained in the CALPUFF dispersion model (EPA, 1995a), and has undergone limited review
and evaluation (Moore, at al. 1995).
The deposition flux, Fd, is calculated as the product of the concentration, ^ and a
deposition velocity, vd, computed at a reference height zd:
Fd'"d-vd (AD
The concentration value, Xj, used in Equation Al is calculated according to Equation 1-1 of the
ISC3 model user's guide, Volume II (EPA, 1995b), with deposition effects accounted for in
the vertical term as described in Section 1.1.6.3 of Volume n. The calculation of deposition
velocities is described below for gaseous emissions.
Deposition Velocities for Gases
At a reference height zd, the deposition velocity (vd) for gases is expressed (Wesley and
Hicks, 1977; Hicks, 1982) as the inverse of a sum of three resistances:
Vd **(ra **rd **rc)'i
where, ra = the atmospheric resistance (s/m) through the surface layer,
rd = the deposition layer resistance (s/m), and,
rc = the canopy (vegetation layer) resistance (s/m).
INDEX-1
-------
An alternative pathway that is potentially important in sparsely vegetated areas or over water is
deposition directly to the ground/water surface. Although not involving vegetation, it is
convenient to include the ground/water surface resistance as a component of rc.
The atmospheric resistance term (ra) is given by Equations 1-81 and 1-82 in Section
1.3.2 of the ISC3 model user's guide, Volume H (EPA, 1995b).
The deposition layer resistance (rd) is parameterized in terms of the Schmidt number
(EPA, 1995a) as:
rd --dcu (A3
where, Sc = the Schmidt number
u = the kinematic viscosity of air (-0.15 x 10"4 m2/s),
DM = the molecular diffusivity of the pollutant (m2/s), and,
dl3 d2 = empirical parameters; dl/k=5, d2=2/3 (Hicks, 1982)
k = the von Karman constant (-0.4)
u» = surface friction velocity (m/s)
The canopy resistance (rc) is the resistance for gases in the vegetation layer, including
the ground/water surface. There are three main pathways for uptake/reaction within the
vegetation or at the surface (EPA, 1995a):
(1) Transfer through the stomatal pore and dissolution or reaction in the mesophyll cells
(plant tissue that contains chlorophyll).
(2) Reaction with or transfer through the leaf cuticle.
(3) Transfer into the ground/water surface.
These pathways are treated as three resistances in parallel.
rc • '[LAI / rf • 'LAI / rcut • • 1 / rj * (A4 )
INDEX-2
-------
where, rf = the internal foliage resistance (s/m) (Pathway 1, Transfer through the
stomatal pore and dissolution or reaction in mesophyll cells),
rcut = the cuticle resistance (s/m), (Pathway 2, Reaction with or transfer
through the leaf cuticle, a thin film covering the surface of plants),
rg = the ground or water surface resistance (s/m), (Pathway 3, Transfer
into the ground/water surface), and,
LAI = the leaf area index (ratio of leaf surface area divided by ground
surface area). The LAI is specified as a function of wind direction
and month/season, and is included in the meteorological input file
provided by the MPRM preprocessor.
Pathway 1:
The internal foliage resistance (rf) consists of two components:
rf **rs **rm (A5)
where, rs = the resistance (s/m) to transport through the stomatal pore (see below),
and,
rm = the resistance (s/m) to dissolution or reaction of the pollutant in the
mesophyll (spongy parenchyma) cells, user input by species. For
soluble compounds (HF, SO2, CL2, NH3), set to zero; for less
soluble compounds (NO2), it could be > 0)
Stomatal opening/closing is a response to the plant's competing needs for uptake of
CO2 and prevention of water loss from the leaves. Stomatal action imposes a strong diurnal
cycle on the stomatal resistance, and has an important role in determining deposition rates for
soluble gaseous pollutants such as SO2. Stomatal resistance (rs) is given by (EPA, 1995a):
rs "ps/ (bDM) (A6)
where, ps = a stomatal constant corresponding to the characteristics of leaf
physiology (- 2.3 x 10'8 m2),
b = the width of the stomatal opening (m), and,
DM = the molecular diffusivity of the pollutant (m2/s).
INDEX-3
-------
The width of the stomatal opening (b) is a function of the radiation intensity, moisture
availability, and temperature. In ISC3, the state of vegetation is specified as one of three
states: (A) active and unstressed, (B) active and stressed, or (C) inactive. Irrigated vegetation
can be assumed to be in an active and unstressed state. The variation in stomatal opening
width during period (A) when vegetation is active and unstressed (Pleim et al., 1984) is:
( a 7 1
\*± ' I
where, bmax = the maximum width (m) of the stomatal opening (~ 2.5 x 10"6 m) (Padro
etal., 1991),
bmin = the minimum width (m) of the stomatal opening (~ 0.1 x 10"6 m),
Rj = the incoming solar radiation (W/m2) received at the ground, and is
included in the meteorological input file for the model by the
MPRM preprocessor, and,
R^ = the incoming solar radiation (W/m2) at which full opening of the
stomata occur; assume constant and equal to 600.
During periods of moisture stress, the need to prevent moisture loss becomes critical,
and the stomata close. Thus for period (B), active vegetation under moisture stress conditions,
assume that b = bmin. When vegetation is inactive (e.g., during the seasonal dry period), the
internal foliage resistance becomes very large, essentially cutting off Pathway 1.
Assuming the vegetation is in state (A), active and unstressed, ambient temperature
provides an additional bound on the value of rs. During cold periods (T<10°C), metabolic
activity slows, and b is set by the code to b^. During hot weather conditions (T>~35°C) the
stomata are fully open (b=bmax) to allow evaporative cooling of the plant.
Pathway 2:
The resistance due to reaction with or transfer through the leaf cuticle (rcut) is given by
(EPA, 1995a):
rcut ••(Aref/AR)rcut(ref) (A8)
where, A,.ef = the reference reactivity parameter of SO2 (~ 8.0),
AR = the reactivity parameter for the depositing gas, (NO2=8, O3=15,
HNO3=18, PAN=4), and,
rcut(ref) = the empirically determined reference cuticle resistance (s/m) of
SO2, set equal to 3000 s/m (Padro et al., 1991).
INDEX-4
-------
Pathway 3:
The third resistance pathway for rc is transfer into the ground/water surface (rg). In
sparsely vegetated areas, deposition directly to the surface may be an important pathway.
rg ••(Aref/AR)rg(ref) (A9)
where, rg(ref) = the reference resistance of SO2 over ground (~ 1000 s/m) (Padro et al.,
1991).
Over water, deposition of soluble pollutants can be quite rapid. The liquid phase resistance of
the depositing pollutant over water is a function of its solubility and reactivity characteristics,
and is given by (Slinn et al., 1978):
rg ••H/(«j.d3 u.) (A10)
where, H = the Henry's law constant, which is the ratio of gas to liquid phase
concentration of the pollutant, (H ~ 4 x 10'2 (SO2), 4 x 10'7 (H2O2), 8 x
ID'8 (HNO3), 2x10° (O3), 3.5x10° (NO2), 1 x 10'2 (PAN), and 4 x 10'6
(HCHO)),
a* = a solubility enhancement factor due to the aqueous phase
dissociation of the pollutant (a* ~ 103 for SO2, ~ 1 for CO2 10 for
O3), and
d3 = a constant (~ 4.8 x 10'4).
If sufficient data are not available to compute the canopy resistance term, rc, from
Equation A4, then an option for user-specified gas dry deposition velocity is provided.
Selection of this option will by-pass the algorithm for computing deposition velocities for
gaseous pollutants, and results from the ISCST3 model based on a user-specified deposition
velocity should be used with extra caution.
Optimizations for Area Sources
When the non-regulatory default TOXICS option is specified, the ISCST3 model
optimizes the area source algorithm to improve model runtimes. These optimizations are
briefly described below.
In the regulatory default mode, the ISCST3 model utilizes a Romberg numerical
integration to estimate the area source impacts, as described in Section 1.2.3 of the ISC3
model user's guide, Volume II (EPA, 1995b). While the Romberg integration performs well
INDEX-5
-------
relative to other approaches for receptors located within or adjacent to the area source, its
advantages diminish as the receptor location is moved further away from the source. The
shape of the integrand becomes less complex for the latter case, approaching that of a point
source at distances of about 15 source widths downwind. Recognizing this behavior, the
TOXICS option in ISCST3 makes use of a more computationally efficient 2-point Gaussian
Quadrature routine to approximate the numerical integral for cases where the receptor location
satisfies the following condition relative to the side of the area source being integrated:
XU-XL<5*XL
(All)
where, XL
XU
the minimum distance from the side of the area source to the receptor,
and
= the maximum distance from the side of the area source to the
receptor.
If the receptor location does not satisfy the condition in Equation All, then the
Romberg numerical integration routine is used. In addition, for receptors that are located
several source widths downwind of an area source, a point source approximation is used. The
distance used to determine if a point source approximation is applied is stability dependent,
and is determined as follows:
X > FACT * WIDTH
where, X = the downwind distance from the center of the source to the
receptor,
FACT = a stability-dependent factor (see below), and
WIDTH = the crosswind width of the area source.
(A12)
Values of FACT:
Stability Class
A
B
C
D
E
F
Rural
3.5
5.5
7.5
12.5
15.5
25.5
Urban
3.5
3.5
5.5
10.5
15.5
15.5
When area sources are modeled with dry depletion, the TOXICS option also allows the
user to specify the AREADPLT option, which applies a single effective dry depletion factor to
the undepleted value calculated for the area source. The effective dry depletion factor, which
INDEX-6
-------
replaces the application of dry depletion within the area source integration, is intended to
provide potential runtime savings to the user. Since dry depletion is distance-dependent, the
effective dry depletion factor is calculated for an empirically-derived effective distance. The
effective distance is calculated as the distance from the receptor to a point within the area
source that is one-third the distance from the downwind edge to the upwind edge. For
receptors located upwind of the downwind edge, including receptors located within the area
source, the effective distance is one-third the distance from the receptor to the upwind edge of
the source.
In addition to the area source optimizations described above, when the TOXICS option
is specified, the dry depletion integration is performed using a 2-point Gaussian Quadrature
routine rather than the Romberg integration used for regulatory applications.
References
Environmental Protection Agency, 1995a. A User's Guide for the CALPUFF Dispersion
Model. EPA-454/B-95-006. U.S. Environmental Protection Agency, Research
Triangle Park, NC.
Environmental Protection Agency, 1995b. User's Guide for the Industrial Source Complex
(ISC3) Dispersion Models, Volume n - Description of Model Algorithms.
EPA-454/B-95-003b. U.S. Environmental Protection Agency, Research Triangle Park,
NC.
Hicks, B.B., 1982: Critical assessment document on acid deposition. ATDL Contrib. File No.
81/24, Atmos. Turb. and Diff. Laboratory, Oak Ridge, TN.
Moore, G., P. Ryan, D. Schwede, and D. Strimaitis, 1995: Model performance evaluation of
gaseous dry deposition algorithms. Paper 95-TA34.02, 88th Annual Meeting &
Exhibition of the Air and Waste Management Association, San Antonio, Texas, June
18-23, 1995.
Padro, J., G.D. Hartog, and H.H. Neumann, 1991: An investigation of the ADOM dry
deposition module using summertime O3 measurements above a deciduous forest.
Atmos. Environ, 25A, 1689-1704.
Pleim, J., A. Venkatram and R. Yamartino, 1984: ADOM/TADAP model development
program. Volume 4. The dry deposition module. Ontario Ministry of the
Environment, Rexdale, Ontario.
Slinn, W.G.N., L. Hasse, B.B. Hicks, A.W. Hogan, D. Lai, P.S. Liss, K.O. Munnich, GA.
Sehmel and O. Vittori, 1978: Some aspects of the transfer of atmospheric trace
constituents past the air-sea interface. Atmos. Environ., 12, 2055-2087.
INDEX-7
-------
Wesley, M.L. and B.B. Hicks, 1977: Some factors that effect the deposition rates of sulfur
dioxide and similar gases on vegetation. J. Air Poll. Control Assoc., 27, 1110-1116.
INDEX-8
-------
EPA-454/B-95-003a
USER'S GUIDE FOR THE
INDUSTRIAL SOURCE COMPLEX (ISC3) DISPERSION MODELS
VOLUME I - USER INSTRUCTIONS
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air Quality Planning and Standards
Emissions, Monitoring, and Analysis Division
Research Triangle Park, North Carolina 27711
September 1995
-------
DISCLAIMER
The information in this document has been reviewed in its
entirety by the U.S. Environmental Protection Agency (EPA), and
approved for publication as an EPA document. Mention of trade
names, products, or services does not convey, and should not be
interpreted as conveying official EPA approval, endorsement, or
recommendation.
The following trademarks appear in this guide:
IBM, IBM/MVS, IBM VS FORTRAN, and IBM 3090 are registered
trademarks of International Business Machines Corp.
Microsoft and MS-DOS are registered trademarks of Microsoft
Corp.
VAX/VMS is a registered trademark of Digital Equipment Corp.
Lahey F77L-EM/32 is a registered trademark of Lahey Computer
Systems, Inc.
OS/386 is a registered trademark of Ergo Computing, Inc.
INTEL, 8086, 80286, 80386, 80486, 80287, and 80387 are
registered trademarks of Intel, Inc.
SunOS is a registered trademark of Sun Microelectronics, Inc.
UNIX is a registered trademark of AT&T Bell Laboratories
Cray and UNICOS are registered trademarks and CFT77, CRAY Y-MP,
and SEGLDR are trademarks of Cray Research, Inc.
11
-------
PREFACE
This User's Guide provides documentation for the
Industrial Source Complex (ISC3) models, referred to hereafter
as the Short Term (ISCST3) and Long Term (ISCLT3) models. This
volume provides user instructions for the ISCST3 and ISCLT3
models, including the new area source and dry deposition
algorithms, both of which are a part of Supplement C to the
Guideline on Air Quality Models (Revised).
This volume also includes user instructions for the
following algorithms that are not included in Supplement C:
pit retention (ISCST3 and ISCLT3), wet deposition (ISCST3
only), and COMPLEXl (ISCST3 only). The pit retention and wet
deposition algorithms have not undergone extensive evaluation
at this time, and their use is optional. COMPLEXl is
incorporated to provide a means for conducting screening
estimates in complex terrain. EPA guidance on complex terrain
screening procedures is provided in Section 5.2.1 of the
Guideline on Air Quality Models (Revised).
Volume II of the ISC3 User's Guide provides the technical
description of the ISC3 algorithms.
111
-------
ACKNOWLEDGEMENTS
The User's Guide for the ISC3 Models has been prepared by
Pacific Environmental Services, Inc., Research Triangle Park,
North Carolina. This effort has been funded by the
Environmental Protection Agency (EPA) under Contract No. 68-
D30032, with Desmond T. Bailey and Donna B. Schwede as Work
Assignment Managers (WAMs). The user instructions for the dry
deposition algorithm were developed from material prepared by
Sigma Research Corporation and funded by EPA under Contract No.
68-D90067, with Jawad S. Touma as WAM.
IV
-------
CONTENTS
PREFACE ill
ACKNOWLEDGEMENTS iv
FIGURES ix
TABLES x
1.0 INTRODUCTION 1-1
1.1 HOW TO USE THE ISC MANUALS 1-1
I.I.I Novice Users 1-1
1.1.2 Experienced Modelers 1-2
1.1.3 Management/Decision Makers 1-3
1.1.4 Programmers/Systems Analysts 1-3
1.2 OVERVIEW OF THE ISC MODELS 1-4
1.2.1 Regulatory Applicability 1-4
1.2.2 Basic Input Data Requirements 1-5
1.2.3 Computer Hardware Requirements 1-5
1.2.4 Overview of Available Modeling Options . . 1-7
1.3 RELATION TO PREVIOUS VERSIONS OF ISC 1-15
1.3.1 Brief History of the ISC Models 1-15
1.3.2 Overview of New Features in the ISC3
Models 1-15
2.0 GETTING STARTED - A BRIEF TUTORIAL 2-1
2.1 DESCRIPTION OF KEYWORD/PARAMETER APPROACH .... 2-1
2.1.1 Basic Rules for Structuring Input
Runstream Files 2-3
2.1.2 Advantages of the Keyword Approach . . . .2-5
2.2 REGULATORY DEFAULT OPTION 2-7
2.3 MODEL STORAGE LIMITS 2-8
2.4 SETTING UP A SIMPLE RUNSTREAM FILE 2-10
2.4.1 A Simple Industrial Source Application . 2-11
2.4.2 Selecting Modeling Options - CO Pathway . 2-12
2.4.3 Specifying Source Inputs - SO Pathway . . 2-16
2.4.4 Specifying a Receptor Network - RE Pathway
2-20
2.4.5 Specifying the Meteorological Input - ME
Pathway 2-21
2.4.6 Selecting Output Options - OU Pathway . . 2-24
2.4.7 Using the Error Message File to Debug the
Input Runstream File 2-26
2.4.8 Running the Model and Reviewing the
Results 2-32
2.5 MODIFYING AN EXISTING RUNSTREAM FILE 2-41
2.5.1 Modifying Modeling Options 2-41
2.5.2 Adding or Modifying a Source or Source
Group 2-43
2.5.3 Adding or Modifying a Receptor Network . 2-43
2.5.4 Modifying Output Options 2-44
v
-------
3.0 DETAILED KEYWORD REFERENCE 3-1
3.1 AN OVERVIEW OF SHORT TERM VS. LONG TERM MODEL
INPUTS 3-2
3.2 CONTROL PATHWAY INPUTS AND OPTIONS 3-2
3.2.1 Title Information 3-3
3.2.2 Dispersion Options 3-3
3.2.3 Averaging Time Options 3-8
3.2.4 Specifying the Pollutant Type 3-12
3.2.5 Modeling With Exponential Decay 3-13
3.2.6 Options for Elevated Terrain 3-13
3.2.7 Flagpole Receptor Height Option 3-15
3.2.8 To Run or Not to Run - That is the
Question 3-15
3.2.9 Generating an Input File for the Short
Term EVENT Model 3-16
3.2.10 The Model Re-start Capability 3-17
3.2.11 Performing Multiple Year Analyses for
PM-10 3-19
3.2.12 Detailed Error Listing File 3-21
3.3 SOURCE PATHWAY INPUTS AND OPTIONS 3-21
3.3.1 Identifying Source Types and Locations . 3-22
3.3.2 Specifying Source Release Parameters . . 3-24
3.3.3 Specifying Building Downwash Information 3-35
3.3.4 Using Variable Emission Rates 3-40
3.3.5 Adjusting the Emission Rate Units for
Output 3-44
3.3.6 Specifying Variables for Settling, Removal
and Deposition Calculations 3-46
3.3.7 Specifying Variables for Precipitation
Scavenging and Wet Deposition Calculations 3-47
3.3.8 Specifying an Hourly Emission Rate File . 3-49
3.3.9 Using Source Groups 3-51
3.4 RECEPTOR PATHWAY INPUTS AND OPTIONS 3-52
3.4.1 Defining Networks of Gridded Receptors . 3-53
3.4.2 Using Multiple Receptor Networks .... 3-60
3.4.3 Specifying Discrete Receptor Locations . 3-61
3.4.4 Specifying Plant Boundary Distances ... 3-64
3.5 METEOROLOGY PATHWAY INPUTS AND OPTIONS 3-65
3.5.1 Specifying the Input Data File and Format 3-65
3.5.2 Specification of Anemometer Height . . . 3-74
3.5.3 Specifying Station Information 3-75
3.5.4 Specifying the Meteorological STAR Data
(Applies Only to ISCLT) 3-76
3.5.5 Specifying a Data Period to Process
(Applies Only to ISCST) 3-78
3.5.6 Correcting Wind Direction Alignment
Problems 3-80
3.5.7 Specifying Wind Speed Categories .... 3-81
3.5.8 Specifying Wind Profile Exponents .... 3-82
3.5.9 Specifying Vertical Temperature Gradients 3-83
3.5.10 Specifying Average Wind Speeds for the
Long Term Model 3-84
VI
-------
3.5.11 Specifying Average Temperatures for the
Long Term Model 3-85
3.5.12 Specifying Average Mixing Heights for the
Long Term Model 3-86
3.5.13 Specifying Average Surface Roughness for
the Long Term Model 3-87
3.6 TERRAIN GRID PATHWAY INPUTS AND OPTIONS .... 3-90
3.7 EVENT PATHWAY INPUTS AND OPTIONS (APPLIES ONLY TO
ISCEV) 3-92
3.7.1 Using Events Generated by the ISCST Model 3-94
3.7.2 Specifying Discrete Events 3-95
3.8 OUTPUT PATHWAY INPUTS AND OPTIONS 3-96
3.8.1 Short Term Model Options 3-96
3.8.2 Short Term EVENT Model (ISCEV) Options . 3-110
3.8.3 Long Term Model Options 3-111
3.9 CONTROLLING INPUT AND OUTPUT FILES 3-115
3.9.1 Description of ISC Input Files 3-116
3.9.2 Description of ISC Output Files 3-118
3.9.3 Control of File Inputs and Outputs (I/O) 3-126
4 . 0 COMPUTER NOTES 4-1
4.1 MINIMUM HARDWARE REQUIREMENTS 4-1
4.1.1 Requirements for Execution on a PC .... 4-1
4.1.2 Requirements for Execution on a DEC VAX
Minicomputer 4-3
4.1.3 Requirements for Execution on an IBM
Mainframe 4-3
4.2 COMPILING AND RUNNING THE MODELS ON A PC 4-3
4.2.1 Microsoft Compiler Options 4-3
4.2.2 Modifying PARAMETER Statements for Unusual
Modeling Needs 4-6
4.3 PORTING THE MODELS TO OTHER HARDWARE ENVIRONMENTS
4-9
4.3.1 Non-DOS PCs 4-10
4.3.2 DEC VAX 4-10
4.3.3 IBM 3090 4-12
4.3.4 Various UNIX machines (CRAY, SUN, DEC VAX,
AT&T) 4-14
4.3.5 Advanced Topics 4-16
5.0 REFERENCES 5-1
APPENDIX A. ALPHABETICAL KEYWORD REFERENCE A-l
APPENDIX B. FUNCTIONAL KEYWORD/PARAMETER REFERENCE .... B-l
APPENDIX C. UTILITY PROGRAMS C-l
C.I CONVERTING INPUT RUNSTREAM FILES - STOLDNEW . . . C-l
C.2 CONVERTING UNFORMATTED PCRAMMET FILES TO ASCII
FORMATTED FILES - BINTOASC C-3
C.3 LISTING HOURLY METEOROLOGICAL DATA - METLIST . . . C-4
VI1
-------
APPENDIX D. BATCH FILE DESCRIPTIONS FOR COMPILING THE
MODELS ON A PC D-l
D.I MICROSOFT/DOS VERSIONS D-l
D.2 LAHEY/EXTENDED MEMORY VERSIONS D-4
APPENDIX E. EXPLANATION OF ERROR MESSAGE CODES E-l
E.I INTRODUCTION E-l
E.2 THE OUTPUT MESSAGE SUMMARY E-2
E.3 DESCRIPTION OF THE DETAILED MESSAGE LAYOUT .... E-3
E.4 DETAILED DESCRIPTION OF THE ERROR/MESSAGE CODES . E-6
APPENDIX F. DESCRIPTION OF FILE FORMATS F-l
F.I ASCII METEOROLOGICAL DATA F-l
F.2 PCRAMMET METEOROLOGICAL DATA F-3
F.3 STAR SUMMARY JOINT FREQUENCY DISTRIBUTIONS .... F-5
F.4 THRESHOLD VIOLATION FILES (MAXIFILE OPTION) . . . F-6
F.5 POSTPROCESSOR FILES (POSTFILE OPTION) F-7
F.6 HIGH VALUE RESULTS FOR PLOTTING (PLOTFILE OPTION)
F-9
F.7 TOXX MODEL INPUT FILES (TOXXFILE OPTION) .... F-10
APPENDIX G. QUICK REFERENCE FOR ISCST AND ISCLT MODELS . . G-l
APPENDIX H. QUICK REFERENCE FOR ISCEV (EVENT) MODEL . . . . H-l
GLOSSARY GLOSSARY-1
INDEX INDEX-1
VI11
-------
FIGURES
Figure Page
2-1. INPUT RUNSTREAM FILE FOR ISCST MODEL FOR SAMPLE
PROBLEM 2-11
2-2. EXAMPLE INPUT RUNSTREAM FILE FOR SAMPLE PROBLEM . . 2-26
2-3. EXAMPLE MESSAGE SUMMARY TABLE FOR RUNSTREAM SETUP . 2-31
2-4. EXAMPLE OF KEYWORD ERROR AND ASSOCIATED MESSAGE
SUMMARY TABLE 2-32
2-5. ORGANIZATION OF ISCST MODEL OUTPUT FILE 2-34
2-6. SAMPLE OF MODEL OPTION SUMMARY TABLE FROM AN ISC
MODEL OUTPUT FILE 2-38
2-7. EXAMPLE OUTPUT TABLE OF HIGH VALUES BY RECEPTOR . . 2-39
2-8. EXAMPLE OF RESULT SUMMARY TABLES FOR THE ISC SHORT
TERM MODEL 2-40
3-1. RELATIONSHIP OF AREA SOURCE PARAMETERS FOR ROTATED
RECTANGLE 3-30
E-l. EXAMPLE OF AN ISC MESSAGE SUMMARY E-3
IX
-------
TABLES
Table Page
3-1 SUMMARY OF SUGGESTED PROCEDURES FOR ESTIMATING
INITIAL LATERAL DIMENSIONS • »0 AND INITIAL VERTICAL
DIMENSIONS '10 FOR VOLUME AND LINE SOURCES 3-27
3-2 SURFACE ROUGHNESS LENGTH, METERS, FOR LAND-USE TYPES
AND SEASONS, FROM SHIEH ET AL. , 1979 3-89
B-l DESCRIPTION OF CONTROL PATHWAY KEYWORDS B-3
B-2 DESCRIPTION OF CONTROL PATHWAY KEYWORDS AND
PARAMETERS B-4
B-3 DESCRIPTION OF SOURCE PATHWAY KEYWORDS B-7
B-4 DESCRIPTION OF SOURCE PATHWAY KEYWORDS AND PARAMETERS
B-8
B-5 DESCRIPTION OF RECEPTOR PATHWAY KEYWORDS B-ll
B-6 DESCRIPTION OF RECEPTOR PATHWAY KEYWORDS AND
PARAMETERS B-12
B-7 DESCRIPTION OF METEOROLOGY PATHWAY KEYWORDS .... B-15
B-8 DESCRIPTION OF METEOROLOGY PATHWAY KEYWORDS AND
PARAMETERS B-16
B-9 DESCRIPTION OF TERRAIN GRID PATHWAY KEYWORDS .... B-l9
B-10 DESCRIPTION OF TERRAIN GRID PATHWAY KEYWORDS AND
PARAMETERS B-20
B-ll DESCRIPTION OF EVENT PATHWAY KEYWORDS B-21
B-12 DESCRIPTION OF EVENT PATHWAY KEYWORDS AND PARAMETERS B-22
B-13 DESCRIPTION OF OUTPUT PATHWAY KEYWORDS B-23
B-14 DESCRIPTION OF OUTPUT PATHWAY KEYWORDS AND PARAMETERS
B-24
x
-------
1.0 INTRODUCTION
This section provides an overall introduction to the ISC
models and to the ISC User's Guide. It also serves
specifically as an introduction to the user instructions
contained in this volume for setting up and running the ISC
models. Some suggestions are offered on how various users
would best benefit from using the manuals. Also provided is an
overview of the model's applicability, range of options, basic
input data and hardware requirements, and a discussion of the
history of the ISC models. The input file needed to run the
ISC models is based on an approach that uses descriptive
keywords and allows for a flexible structure and format.
1.1 HOW TO USE THE ISC MANUALS
The ISC Model User's Guide has been designed in an attempt
to meet the needs of various types of users, depending on their
level of experience with the models. This section describes
briefly how different types of users would benefit most from
their use of the manual.
I.I.I Novice Users
Novice users are those whose exposure to or experience
with the ISC models has been limited. They may be new to
dispersion modeling applications in general, or new to the ISC
models and therefore unfamiliar with the keyword/parameter
approach utilized for the input file. These users should
review the remainder of this Introduction to gain an overall
perspective of the use of ISC models, particularly for
regulatory modeling applications. They should then concentrate
their review on Section 2, which provides a brief tutorial on
setting up an input file that illustrates the most commonly
used options of the ISC Short Term model. Section 2 provides a
basic description of the input file structure and explains some
1-1
-------
of the advantages of the keyword/parameter approach to
specifying modeling options and inputs. As the user becomes
more familiar with the operation of the models and encounters
the need to use more advanced features of the models, he/she
will want to review the contents of Section 3, which provides a
more detailed and complete reference of the various options for
running the models.
1.1.2 Experienced Modelers
Experienced modelers will have had considerable experience
in applying the ISC models in a variety of situations. They
should have basic familiarity with the overall goals and
purposes of regulatory modeling in general, and with the scope
of options available in the ISC models in particular.
Experienced modelers who are new to the ISC models will benefit
from first reviewing the contents of Section 2 of this volume,
which will give them a basic orientation to the structure,
organization and philosophy of the keyword/parameter approach
used for the input runstream file. Once they have a basic
grasp of the input file structure and syntax rules, they will
benefit most from using Section 3 of this volume as a reference
to learn the overall capabilities of the models, or to
understand the mechanics for implementing particular options.
The information in Section 3 is organized by pathway, with
detailed descriptions of each of the individual keyword options
by pathway. Once they are familiar with most or all of the
keywords, they may find the functional keyword reference
provided in Appendix B useful to quickly review the proper
syntax and available options/parameters for a particular
keyword. They may also find the Quick Reference available at
the end of the user's guide sufficient as a simple reminder of
the available keywords for each pathway and to ensure the
proper order of parameters for each input image.
1-2
-------
Experienced modelers may also have occasion to peruse the
contents of Volume II, which describes the technical details of
the dispersion modeling algorithms utilized in the ISC models.
They may also have an interest in or need to review the
contents of Volume III to learn about the structure and
organization of the computer code, particularly if they are
involved with installing the code on another computer system,
or with compiling the code to meet the memory storage
requirements for a particular application.
1.1.3 Management/Decision Makers
Those involved in a management or decision-making role for
dispersion modeling applications will be especially interested
in the remainder of this section, which provides an overview of
the models, including their role in various regulatory
programs, a brief description of the range of available
options, and basic input data and computer hardware
requirements needed to run the models. From this information
they should understand the basic capabilities of the ISC models
well enough to judge the suitability of the models for
particular applications. They may also want to review the
brief tutorial provided in Section 2 to learn about the nature
and structure of the input runstream file, in order to better
be able to review the modeling results.
1.1.4 Programmers/Systems Analysts
Programmers and systems analysts, specifically those
involved with installing the ISC code on other computer systems
or charged with maintaining the code, should review the
contents of Volume III. This will acquaint them with the
structure and organization of the computer code, give specific
details on compiling and linking the code for various
situations, and explain in detail the memory storage
requirements and control of input and output (I/O). They may
1-3
-------
also wish to review the remainder of this Introduction and the
brief tutorial in Section 2 of this volume in order to have a
basic understanding of the nature and overall capabilities of
the models, and to understand the basic input runstream file
structure and organization.
1.2 OVERVIEW OF THE ISC MODELS
This section provides an overview of the ISC models,
including a discussion of the regulatory applicability of the
models, a description of the basic options available for
running the models, and an explanation of the basic input data
and hardware requirements needed for executing the models.
1.2.1 Regulatory Applicability
The U.S. Environmental Protection Agency (EPA) maintains
the Guideline on Air Quality Models (Revised) (hereafter
referred to as the "Guideline"1) which provides the agency's
guidance on regulatory applicability of air quality dispersion
models in the review and preparation of new source permits and
State Implementation Plan (SIP) revisions. Regulatory
application of the ISC models should conform to the guidance
set forth in the Guideline, including the most recent
Supplements. Any non-guideline application of the models
should meet the requirements of the applicable reviewing
agency, such as an EPA Regional Office, a State or a local air
pollution control agency. In general, regulatory modeling
applications should be carried out in accordance with a
modeling protocol that is reviewed and approved by the
appropriate agency prior to conducting the modeling. The
modeling protocol should identify the specific model, modeling
options and input data to be used for a particular application.
The Guideline is published as Appendix W to 40 CFR Part 51.
1-4
-------
1.2.2 Basic Input Data Requirements
There are two basic types of inputs that are needed to run
the ISC models. They are (1) the input runstream file, and (2)
the meteorological data file. The runstream setup file
contains the selected modeling options, as well as source
location and parameter data, receptor locations, meteorological
data file specifications, and output options. The ISC models
offer various options for file formats of the meteorological
data. These are described briefly later in this section, and
in more detail in Sections 2 and 3. A third type of input may
also be used by the models when implementing the dry deposition
and depletion algorithm. The user may optionally specify a
file of gridded terrain elevations that are used to integrate
the amount of plume material that has been depleted through dry
deposition processes along the path of the plume from the
source to the receptor. The optional terrain grid file is
described in more detail in Section 3. The user also has the
option of specifying a separate file of hourly emission rates
for the ISCST model.
1.2.3 Computer Hardware Requirements
1.2.3.1 PC Hardware Requirements.
Given the rapid increase in speed and capacity of personal
computers (PCs) available for modeling in recent years, and
their relative ease of use and access, the PC has become the
most popular environment for performing dispersion modeling
applications within the modeling community (Bauman and Dehart,
1988; Rorex, 1990). This trend can be expected to continue in
the future. The current versions of the ISC models were
developed on an IBM-compatible PC using the Microsoft FORTRAN
Optimizing Compiler (Version 5.1), and have been designed to
run on such machines with a minimum of 640K bytes of RAM and
MS-DOS Version 3.2 or higher. In order to handle the input
1-5
-------
data files (runstream setup and meteorology) and the output
files, it is highly recommended that the system have a hard
disk drive. The amount of storage space required on the hard
disk for a particular application will depend greatly on the
output options selected. Some of the optional output files of
concentration data can be rather large. More information on
output file products is provided in Sections 2 and 3.
While a math coprocessor chip is optional for execution of
the ISC models on a PC, it is highly recommended, especially
for the Short Term model, due to the large increase in
execution speed that will be experienced. The model may be
expected to run about five to ten times faster with a math
coprocessor than without one.
For particularly large applications, involving a large
number of sources, source groups, receptors and averaging
periods, the user may find that the 640K RAM limit available
with DOS is not enough. In addition to the DOS executable
versions of the models, extended memory versions are available
for use on 80386, 80486 or higher PCs with at least 8 MB of RAM
for the ISCST model and at least 4 MB of RAM for the ISCLT
model. The extended memory versions of the models were
developed using the Lahey F77L/EM-32 Fortran Compiler (Version
5.2), and also require a math co-processor to be present. For
larger application scenarios, a Lahey-compiled ISCST executable
and 8 MB of RAM are recommended. Section 4.2.2 of this volume
of the ISC User's Guide contains information on increasing the
capacity of the model and setting it up to run on systems (with
80386 processors and higher) that make use of extended memory
beyond the 640K limit of DOS. There are special requirements
for the operating system and Fortran language compiler needed
to utilize the extended memory on these machines.
1-6
-------
1.2.3.2 DEC VAX Requirements.
The models have also been uploaded and tested on a DEC VAX
minicomputer. As with the IBM 3090, the VAX has some
advantages of speed and greater memory capacity over the PC
environment. There are no particular hardware requirements for
running the models on the VAX. The user must be familiar with
the operating system and Fortran language compiler being
utilized on the VAX in order to properly setup and run the
model and control the input and output files. Instructions for
setting up and running the models on the DEC VAX are included
in this volume and in more detail in Volume III of the User's
Guide.
1.2.3.3 IBM 3090 Requirements.
While the models were developed on the PC, they have been
uploaded and tested on EPA's IBM 3090 mainframe computer. The
mainframe has advantages of speed and greater memory capacity
over the PC environment. There are no particular hardware
requirements for running the models on the IBM 3090. However,
the user must be familiar with the IBM Job Control Language
(JCL) and the VS FORTRAN Version 2.0 compiler in order to
properly setup and run the models and control the input and
output files in the mainframe environment. Instructions for
setting up and running the models on the IBM 3090 are included
in this volume and in Volume III of the User's Guide.
1.2.4 Overview of Available Modeling Options
The ISC models include a wide range of options for
modeling air quality impacts of pollution sources, making them
popular choices among the modeling community for a variety of
applications. The following sections provide a brief overview
of the options available in the ISC models.
1-7
-------
1.2.4.1 Dispersion Options.
Since the ISC models are especially designed to support
the EPA's regulatory modeling programs, the regulatory modeling
options, as specified in the Guideline on Air Quality Models
(Revised), are the default mode of operation for the models.
These options include the use of stack-tip downwash,
buoyancy-induced dispersion, final plume rise (except for
sources with building downwash), a routine for processing
averages when calm winds occur, default values for wind profile
exponents and for the vertical potential temperature gradients,
and the use of upper bound estimates for super-squat buildings
having an influence on the lateral dispersion of the plume. The
user can easily ensure the use of the regulatory default
options by selecting a single keyword on the modeling option
input card. To maintain the flexibility of the model, the
non-regulatory default options have been retained, and by using
descriptive keywords to specify these options it is evident at
a glance from the input or output file which options have been
employed for a particular application.
The Short Term model also incorporates the COMPLEXl
screening model dispersion algorithms for receptors in complex
terrain, i.e., where the receptor elevation is above the
release height of the source. The user has the option of
specifying only simple terrain (i.e., ISCST) calculations, only
complex terrain (i.e., COMPLEXl) calculations, or of using both
simple and complex terrain algorithms. In the latter case, the
model will select the higher of the simple and complex terrain
calculations on an hour-by-hour, source-by-source and receptor-
by-receptor basis for receptors in intermediate terrain, i.e.,
terrain between release height and plume height.
The user may select either rural or urban dispersion
parameters, depending on the characteristics of the source
location. The user also has the option of calculating
1-8
-------
concentration values or deposition values for a particular run.
For the Short Term model, the user may select more than one
output type (concentration and/or deposition) in a single run,
depending on the setting for one of the array storage limits.
The user can specify several short term averages to be
calculated in a single run of the ISC Short Term model, as well
as requesting the overall period (e.g. annual) averages.
1.2.4.2 Source Options.
The model is capable of handling multiple sources,
including point, volume, area and open pit source types. Line
sources may also be modeled as a string of volume sources or as
elongated area sources. Several source groups may be specified
in a single run, with the source contributions combined for
each group. This is particularly useful for Prevention of
Significant Deterioration (PSD) applications where combined
impacts may be needed for a subset of the modeled background
sources that consume increment, while the combined impacts from
all background sources (and the permitted source) are needed to
demonstrate compliance with the National Ambient Air Quality
Standards (NAAQS). The models contain algorithms for modeling
the effects of aerodynamic downwash due to nearby buildings on
point source emissions, and algorithms for modeling the effects
of settling and removal (through dry deposition) of
particulates.
The Short Term model also contains an algorithm for
modeling the effects of precipitation scavenging for gases or
particulates. For the Short Term model, the user may specify
for the model to output dry deposition, wet deposition and/or
total deposition.
Source emission rates can be treated as constant
throughout the modeling period, or may be varied by month,
season, hour-of-day, or other optional periods of variation.
1-9
-------
These variable emission rate factors may be specified for a
single source or for a group of sources. For the Short Term
model, the user may also specify a separate file of hourly
emission rates for some or all of the sources included in a
particular model run.
1.2.4.3 Receptor Options.
The ISC models have considerable flexibility in the
specification of receptor locations. The user has the
capability of specifying multiple receptor networks in a single
run, and may also mix Cartesian grid receptor networks and
polar grid receptor networks in the same run. This is useful
for applications where the user may need a coarse grid over the
whole modeling domain, but a denser grid in the area of maximum
expected impacts. There is also flexibility in specifying the
location of the origin for polar receptors, other than the
default origin at (0,0) in x,y, coordinates.
The user can input elevated receptor heights in order to
model the effects of terrain above (or below) stack base, and
may also specify receptor elevations above ground level to
model flagpole receptors. For simple terrain calculations, any
terrain heights input above the release height for a particular
source are "chopped-off" at the release height for that
source's calculations. The Short Term model includes the
complex terrain algorithms from the COMPLEXl screening model.
If these algorithms are used, the model will calculate impacts
for terrain above the release height. The Long Term model does
not include any complex terrain algorithms.
1-10
-------
1.2.4.4 Meteorology Options.
The Short Term model can utilize the unformatted,
sequential files of meteorological data generated by the
PCRAMMET and the MPRM preprocessors, provided the data file was
generated by the same Fortran compiler as was used for the
model, and provided the deposition algorithms are not being
used. The meteorology options for the deposition algorithms in
the ISC models are described later in this section.
The user also has considerable flexibility to utilize
formatted ASCII files that contain sequential hourly records of
meteorological variables. For these hourly ASCII files, the
user may use a default ASCII format, may specify the ASCII read
format, or may select free-formatted reads for inputting the
meteorological data. A utility program called BINTOASC is
provided with the ISC models to convert unformatted
meteorological data files of several types to the default ASCII
format used by ISCST and ISCEV. This greatly improves the
portability of applications to different computer systems. The
BINTOASC program is described in Appendix C. The model will
process all available meteorological data in the specified
input file by default, but the user can easily specify selected
days or ranges of days to process.
The Short Term model includes a dry deposition algorithm
and a wet deposition algorithm. The dry deposition algorithm
requires additional meteorological input variables, such as
Monin-Obukhov length and surface friction velocity, that are
provided by the PCRAMMET and MPRM preprocessor. The wet
deposition algorithm in the Short Term model also needs
precipitation data, which is optionally available in the
PCRAMMET preprocessed data. When using the dry deposition or
wet deposition algorithms in ISCST, the meteorological data
must be a formatted ASCII file.
1-11
-------
The Long Term model uses joint frequency distributions of
wind speed class, by wind direction sector, by stability
category, known as STAR (STability ARray) summaries. These
STAR summaries are available from the National Climatic Data
Center in Asheville, North Carolina. They may also be
generated from sequential data files using the STAR utility
program available on EPA's SCRAM Bulletin Board System or by
the MPRM meteorological processor for on-site data. The
meteorological data for ISCLT are read in from a separate data
file, and the user may use a default ASCII format or may
specify the ASCII read format for the data.
1.2.4.5 Output Options.
The basic types of printed output available with the Short
Term model are:
• Summaries of high values (highest, second highest,
etc.) by receptor for each averaging period and source
group combination;
• Summaries of overall maximum values (e.g., the maximum
50) for each averaging period and source group
combination; and
• Tables of concurrent values summarized by receptor for
each averaging period and source group combination for
each day of data processed. These "raw" concentration
values may also be output to unformatted (binary)
files, as described below.
For the Long Term model, the user can also select output
tables of values for each receptor, and/or tables of overall
maximum values. The tables by receptor and maximum value
tables can be output for the source group values or for the
individual source values, or both. In addition, when maximum
values for individual sources are output, the user has the
option of specifying whether the values are to be the maximum
values for each source independently, or the contribution of
each source to the maximum group values, or both.
1-12
-------
In addition to the tabular printed output products
described above, the ISC models provide options for several
types of file output products. One of these options for ISCST
is to output an unformatted ("binary") file of all
concentration and/or deposition values as they are calculated.
These files are often used for special postprocessing of the
data. In addition to the unformatted concentration files,
ISCST provides options for three additional types of file
outputs. One option is to generate an ASCII formatted file
with the same results that are included in the unformatted
postprocessing file. Another option is to generate a file of
(X,Y) coordinates and design values (e.g., the second highest
values at each receptor for a particular averaging period and
source group combination) that can be easily imported into many
graphics plotting packages to generate contour plots of the
concentration and/or deposition values. Separate files can be
specified for each of the averaging period and source group
combinations of interest to the user.
Another output file option of the ISCST model is to
generate a file of all occurrences when a concentration or
deposition value equals or exceeds a user-specified threshold.
Again, separate files are generated for only those combinations
of averaging period and source group that are of interest to
the user. These files include the date on which the threshold
exceedance occurred, the receptor location, and the
concentration value.
1.2.4.6 Source Contribution Analyses.
In air quality dispersion modeling applications, the user
may have a need to know the contribution that a particular
source makes to an overall concentration value for a group of
sources. This section provides a brief introduction to how
these types of source contribution (sometimes referred to as
source culpability) analyses are performed using the ISC
1-13
-------
models. More detailed information about exercising these
options is provided in Section 3.
Recognizing that source contribution information is
important to many short term modeling analyses, the ISCST model
has been designed to facilitate performing this type of
analysis. This is accomplished with an additional model,
referred to as the ISC Short Term - EVENT model (ISCEV). The
ISCST model treats source groups independently. The ISCEV
(EVENT) model is set up specifically to provide the
contributions from individual sources to the concentration
values for particular events. These events may be the design
concentrations (e.g., the high-second-high 24-hour average
concentration for a particular group of sources) that were
generated from an execution of the ISCST model. Other events
of interest might be occurrences of violations of a particular
standard, for which it is necessary to determine whether the
source being permitted contributes above a significance level.
The models are set up in such a way that both of these types of
events can be passed directly from an execution of the ISCST
model to an input file for the EVENT model. The user is thus
able to run the models in a batch mode to obtain the overall
design value results from ISCST and the source contribution
information from ISCEV in a single step. The EVENT model can
also be run separately and accepts user-specified events for
source contribution processing.
In the ISCLT model, the user has an option to have the
highest 10 values for each source and source group reported
independently, or to have the 10 highest values from the
combined source group and the contributions from the individual
sources to those highest group values.
1-14
-------
1.3 RELATION TO PREVIOUS VERSIONS OF ISC
1.3.1 Brief History of the ISC Models
The ISC3 models are based on revisions to the algorithms
contained in the ISC2 models. The latter came about as a
result of a major effort to restructure and reprogram the ISC
models that began in April 1989, and was completed in March
1992. The reprogramming effort was largely motivated by the
need to improve the quality, reliability, and maintainability
of the code when numerous "bugs" were discovered after the
implementation of the revised downwash algorithms for shorter
stacks. It became widely recognized that the code, originally
developed in the 1970's and modified numerous times since, had
become impossible to reliably modify, debug or maintain.
However, the goals of the reprogramming effort also included
improving the user interface by modifying the input file
structure and the output products, and to provide better "end
user" documentation for the revised models. The ISC2 models
were developed as replacements for and not updates to the
previous versions of the models.
1.3.2 Overview of New Features in the ISC3 Models
The ISC3 models include several new features. A revised
area source algorithm and revised dry deposition algorithm have
been incorporated in the models. The ISC3 models also include
an algorithm for modeling impacts of particulate emissions from
open pit sources, such as surface coal mines. The Short Term
model includes a new wet deposition algorithm, and also
incorporates the COMPLEXl screening model algorithms for use
with complex and intermediate terrain. When both simple and
complex terrain algorithms are included in a Short Term model
run, the model will select the higher impact from the two
algorithms on an hour-by-hour, source-by-source, and receptor-
by-receptor basis for receptors located on intermediate
1-15
-------
terrain, i.e., terrain located between the release height and
the plume height. A more detailed technical description of
these new features of the ISC models is included in Volume II
of the ISC User's Guide. The Long Term model does not include
wet deposition or complex terrain algorithms.
Some of the model input options have changed as a result
of the new features contained in the ISC3 models. There are
new options available on the CO MODELOPT card for both the
Short Term and Long Term models. The source deposition
parameters have changed somewhat with the new dry deposition
algorithm, and there are new source parameters needed for the
wet deposition algorithm in the Short Term model. Both models
include a new optional pathway for specifying a terrain grid
file that may be used in calculating the effects of plume
depletion due to dry removal mechanisms in elevated terrain.
There are also new meteorology input requirements for use of
the new deposition algorithms. The option for specifying
elevation units has been extended to source elevations and
terrain grid elevations, in addition to receptor elevations.
The CO ELEVUNIT card used to specify receptor elevations in the
previous version of ISC is now obsolescent, and is being
replaced by a new RE ELEVUNIT card. These new input options
are described in Section 3 and summarized in Appendix B.
The utility programs, STOLDNEW, BINTOASC, and METLIST,
described in Appendix C, have not been updated. While they may
continue to be used as before, they are not applicable to the
new deposition algorithms in the ISC3 models.
1-16
-------
2.0 GETTING STARTED - A BRIEF TUTORIAL
This section provides a brief tutorial for setting up a
simple application problem with the ISC Short Term model, which
serves as an introduction for novice users to the ISC models.
The example illustrates the usage of the most commonly used
options in the ISC models for regulatory applications. A more
complete description of the available options for setting up
the ISC models is provided in Section 3.
The example problem presented in this section is a simple
application of the ISCST model to a single point source. The
source is a hypothetical stack at a small isolated facility in
a rural setting. Since the stack is below the Good Engineering
Practice (GEP) stack height, the emissions from the source are
subject to the influence of aerodynamic downwash due to the
presence of nearby buildings. The tutorial leads the user
through selection and specification of modeling options,
specification of source parameters, definition of receptor
locations, specification of the input meteorological data, and
selection of output options. Since this discussion is aimed at
novice users of the ISC models, a general description of the
input file keyword/parameter approach is provided first.
2.1 DESCRIPTION OF KEYWORD/PARAMETER APPROACH
The input file for the ISC models makes use of a
keyword/parameter approach to specifying the options and input
data for running the models. The descriptive keywords and
parameters that make up this input runstream file may be
thought of as a command language through which the user
communicates with the model what he/she wishes to accomplish
for a particular model run. The keywords specify the type of
option or input data being entered on each line of the input
file, and the parameters following the keyword define the
specific options selected or the actual input data. Some of
2-1
-------
the parameters are also input as descriptive secondary
keywords.
The runstream file is divided into six functional
"pathways." These pathways are identified by a two-character
pathway ID placed at the beginning of each runstream image. The
pathways and the order in which they are input to the model are
as follows:
CO - for specifying overall job COntrol options;
SO - for specifying SOurce information;
RE - for specifying REceptor information;
ME - for specifying MEteorology information;
TG - for specifying Terrain Grid information; and
OU - for specifying Output options.
The TG pathway is an optional pathway that is only used for
implementing the dry depletion algorithm in elevated terrain.
Each line of the input runstream file consists of a
pathway ID, an 8-character keyword, and a parameter list. An
example of a line of input from a runstream file, with its
various parts identified, is shown below:
Column: 12345678901234567890123456789012345678901234567890123456789
CO MODELOPT DFAULT RURAL CONC
# # * * *
# # * * *
.))))))2)))))2))))))))) Parameters
# #
.))))))))))))))))))))))))))))))))))) 8-Character Keyword
#
.)))))))))))))))))))))))))))))))))))))))))))) 2-Character Pathway ID
2-2
-------
The following sections describe the rules for structuring
the input runstream file, and explain some of the advantages of
the keyword/parameter approach.
2.1.1 Basic Rules for Structuring Input Runstream Files
While the input runstream file has been designed to
provide the user with considerable flexibility in structuring
the input file, there are some basic syntax rules that need to
be followed. These rules serve to maintain some consistency
between input files generated by different users, to simplify
the job of error handling performed by the models on the input
data, and to provide information to the model in the
appropriate order wherever order is critical to the
interpretation of the inputs. These basic rules and the
various elements of the input runstream file are described in
the paragraphs that follow.
One of the most basic rules is that all inputs for a
particular pathway must be contiguous, i.e., all inputs for the
CO pathway must come first, followed by the inputs for the SO
pathway, and so on. The beginning of each pathway is
identified with a "STARTING" keyword, and the ending of the
pathway with the "FINISHED" keyword. Thus the first functional
record of each input file must be "CO STARTING" and the last
record of each input file must be "OU FINISHED." The rest of
the input images will define the options and input data for a
particular run.
Each record in the input runstream file is referred to as
a runstream "image." These records are initially read into the
model as 132-character images. The information on each input
image consists of a "pathway," a "keyword," and one or more
"parameters." Each of these "fields" on the runstream image
must be separated from other fields by at least one blank
space. To simplify the interpretation of the runstream image
2-3
-------
by the model, the runstream file must be structured with the
two-character pathway in columns 1 and 2, the eight-character
keyword in columns 4 through 11, followed by the parameters in
columns 13 through 132, as necessary. (For reasons that are
explained in Section 2.4.8, the models will accept input files
where all inputs are shifted by up to three columns to the
right.) For most keywords, the order of parameters following
the keyword is important -- the exact spacing of the parameters
is not important, as long as they are separated from each other
by at least one blank space and do not extend beyond the 132
character limit. The example of a runstream image from the CO
pathway shown above is repeated here:
Column: 12345678901234567890123456789012345678901234567890123456789
CO MODELOPT DFAULT RURAL CONC
# # * * *
# # * * *
.))))))2)))))2))))))))) Parameters
# #
.))))))))))))))))))))))))))))))))))) 8-Character Keyword
#
.)))))))))))))))))))))))))))))))))))))))))))) 2-Character Pathway ID
Alphabetical characters can be input as either lower case
or upper case letters. The models convert all character input
to upper case letters internally, with the exception of the
title fields and file names to be discussed later. Throughout
this document, the convention of using upper case letters is
followed. For numeric input data, it should be noted that all
data are assumed to be in metric units, i.e., length units of
meters, speed units of meters per second, temperature units of
degrees Kelvin, and emission units of grams per second. In a
few instances, the user has the option of specifying units of
feet for length and the model will perform the conversion to
meters. These exceptions are the input of receptor heights for
elevated terrain and the specification of anemometer height,
2-4
-------
since these values are often more readily available in feet
than in meters.
Certain keywords are mandatory and must be present in
every runstream file, such as the MODELOPT keyword shown in the
example above which identifies the modeling options. Other
keywords are optional and are only needed to exercise
particular options, such as the option to allow for the input
of flagpole receptor heights. Some of the keywords are
repeatable, such as the keywords to specify source parameters,
while other keywords may only appear once. The keyword
references in Section 3, Appendices A and B and the Quick
Reference at the end of this volume identify each keyword as to
its type, either mandatory or optional, and either repeatable
or non-repeatable.
With a few exceptions that are described below, the order
of keywords within each pathway is not critical. For the CO
pathway, an exception is that the MODELOPT and POLLUTID
keywords must be specified before the DCAYCOEF or HALFLIFE
keyword because of the link between the urban default option
and the decay coefficient for S02. For the SO pathway, the
LOCATION keyword must be specified before other keywords for a
particular source, and the SRCGROUP keyword must be the last
keyword before SO FINISHED. For keywords on the SO pathway
that accept a range of source IDs, the source parameters
specified by those keywords will only be applied to the sources
already defined, and will exclude any sources that are
specified latter in the input file.
2.1.2 Advantages of the Keyword Approach
The keyword approach provides some advantages over the
type of input file that uses non-descriptive numeric option
switches and requires rigidly formatted inputs. One advantage
is that the keywords are descriptive of the options and inputs
2-5
-------
being used for a particular run, making it easier for a
reviewer to ascertain what was accomplished in a particular run
by reviewing the input file. Another advantage is that the
user has considerable flexibility in structuring the inputs to
improve their readability and understandability, as long as
they adhere to the few basic rules described above.
Some special provisions have been made to increase the
flexibility to the user in structuring the input files. One
provision is to allow for blank records in the input file.
This allows the user to separate the pathways from each other,
or to separate a group of images, such as source locations,
from the other images. Another provision is for the use of
"comment cards," identified by a "**" in the pathway field. Any
input image that has "**" for the pathway ID will be ignored by
the model. This is especially useful for labeling the columns
in the source parameter input images, as illustrated in the
example problem later in this section. It may also be used to
"comment out" certain options for a particular run without
deleting the options and associated data (e.g., elevated
terrain heights) completely from the input file. Because of
the descriptive nature of the keyword options and the
flexibility of the inputs it is generally much easier to make
modifications to an existing input runstream file to obtain the
desired result.
Another aspect of the "user-friendliness" of the ISC
models is that detailed error-handling has been built into the
models. The model provides descriptions of the location and
nature of all of the errors encountered for a particular run.
Rather than stopping execution at each occurrence of an input
error, the new model will read through and attempt to process
all input records and report all errors encountered. If a
fatal error occurs, then the model will not attempt to execute
the model calculations.
2-6
-------
2.2 REGULATORY DEFAULT OPTION
The regulatory default option is controlled from the
MODELOPT keyword on the CO pathway. As its name implies, this
keyword controls the selection of modeling options. It is a
mandatory, non-repeatable keyword, and it is an especially
important keyword for understanding and controlling the
operation of the ISC models. As noted in Section 1, the
regulatory default options, as specified in the Guideline on
Air Quality Models, are truly the default options for the ISC
models. That is to say that, unless specified otherwise
through the available keyword options, the ISC models implement
the following regulatory options:
• Use stack-tip downwash (except for Schulman-Scire
downwash);
• Use buoyancy-induced dispersion (except for
Schulman-Scire downwash);
• Do not use gradual plume rise (except for building
downwash);
• Use the calms processing routines;
• Use upper-bound concentration estimates for sources
influenced by building downwash from super-squat
buildings;
• Use default wind profile exponents; and
• Use default vertical potential temperature gradients.
Rather than specifying options with numeric switches, the
parameters used for the MODELOPT keyword are character strings,
called "secondary keywords," that are descriptive of the option
being selected. For example, to ensure that the regulatory
default options be used for a particular run, the user would
include the secondary keyword "DFAULT" on the MODELOPT input.
The presence of this secondary keyword tells the model to
override any attempt to use a non-regulatory default option.
The model will warn the user if a non-regulatory option is
2-7
-------
selected along with the DFAULT option, but will not halt
processing. For regulatory modeling applications, it is
strongly suggested that the DFAULT switch be set, even though
the model defaults to the regulatory options without it.
For any application in which a non-regulatory option is to
be selected, the DFAULT switch must not be set, since it would
otherwise override the non-regulatory option. The
non-regulatory options are also specified by descriptive
secondary keywords, such as "NOBID" to specify the option not
to use buoyancy-induced dispersion. (A programmer note: these
modeling option keywords also correspond to the Fortran logical
variable names used to control the options in the ISC computer
code. This is one reason why they are limited to six
characters, .e.g., DFAULT instead of DEFAULT, since the
standard Fortran language (ANSI, 1978) only allows variable
names up to six characters in length).
The MODELOPT keyword, which is also used to specify the
selection of rural or urban dispersion parameters, and
concentration or deposition values, is described in more detail
in the Section 3.2.2.
2.3 MODEL STORAGE LIMITS
The ISC models have been designed using a static storage
allocation approach, where the model results are stored in data
arrays, and the array limits are controlled by PARAMETER
statements in the Fortran computer code. These array limits
also correspond to the limits on the number of sources,
receptors, source groups and averaging periods that the model
can accept for a given run. Depending on the amount of memory
available on the particular computer system being used, and the
needs for a particular modeling application, the storage limits
can easily be changed by modifying the PARAMETER statements and
recompiling the model. Section 4.2.2 of this volume and Volume
2-8
-------
Ill of the User's Guide provide more information about
modifying the storage limits of the models.
The limits on the number of receptors, sources, source
groups, averaging periods, and events (for ISCEV model) are
initially set as follows for the three models for the DOS and
extended memory (EM) versions on the PC:
PARAMETER
Name
NREC
NSRC
NGRP
NAVE
NEVE
Limit
Controlled
Number of
Receptors
Number of
Sources
Number of
Source
Groups
Number of
Short Term
Averages
Number of
Events
ISCST
500 (DOS)
1200 (EM)
100 (DOS)
300 (EM)
2 (DOS)
4 (EM)
2 (DOS)
4 (EM)
-
ISCEV
-
100 (DOS)
500 (EM)
25 (DOS)
50 (EM)
4 (DOS)
4 (EM)
2500 (DOS)
5000 (EM)
ISCLT
500 (DOS)
1200 (EM)
50 (DOS)
300 (EM)
3 (DOS)
5 (EM)
-
-
Fortran PARAMETER statements are also used to specify the
array limits for the number of output types (CONG, DEPOS, DDEP,
and/or WDEP) available with the ISCST model (NTYP, initially
set to 2 for the DOS version and 4 for the EM version); the
number of high short term values by receptor to store for the
ISCST model (NVAL, initially set to 2 for the DOS version and 6
for the EM version); the number of overall maximum values to
store (NMAX, initially set to 50 for ISCST and to 10 for Long
Term); and the number of x-coordinates and y-coordinates that
may be included in the optional terrain grid file (MXTX and
MXTY, initially set to 101 for the DOS version of Short Term,
201 for the DOS version of Long Term, and 601 for the EM
version of both models).
2-9
-------
In addition to the parameters mentioned above, parameters
are used to specify the number of gridded receptor networks in
a particular run (NNET), and the number of x-coordinate (or
distance) and y-coordinate (or direction) values (IXM and IYM)
for each receptor network. Initially, the models allow up to 5
receptor networks (of any type), and up to 50 x-coordinates (or
distances) and up to 50 y-coordinates (or directions). The
source arrays also include limits on the number of variable
emission rate factors per source (NQF, initially set to 24 for
the DOS version of Short Term and 96 for the EM version of
Short Term, and to 36 for the DOS version of Long Term and 144
for the EM version of Long Term), the number of sectors for
direction-specific building dimensions (NSEC, initially set to
36 for Short Term and 16 for Long Term), and the number of
settling and removal categories (NPDMAX, initially set to 10
for the DOS version of Short Term and 20 for the EM version of
Short Term and both versions of Long Term).
2.4 SETTING UP A SIMPLE RUNSTREAM FILE
This section goes through a step-by-step description of
setting up a simple application problem, illustrating the most
commonly used options of the ISCST model. The ISCST input
runstream file for the example problem is shown in Figure 2-1.
The remainder of this section explains the various parts of the
input file for the ISCST model, and also illustrates some of
the flexibility in structuring the input file.
2-10
-------
CO
CO
CO
CO
CO
CO
CO
so
so
so
so
so
so
so
so
so
so
so
so
so
RE
RE
RE
RE
RE
RE
RE
ME
ME
ME
ME
ME
ME
OU
OU
OU
OU
STARTING
TITLEONE
MODELOPT
AVERTIME
POLLUTID
RUNORNOT
FINISHED
STARTING
LOCATION
SRCPARAM
BUILDHGT
BUILDHGT
BUILDHGT
BUILDWID
BUILDWID
BUILDWID
BUILDWID
BUILDWID
SRCGROUP
FINISHED
STARTING
GRIDPOLR
GRIDPOLR
GRIDPOLR
GRIDPOLR
GRIDPOLR
FINISHED
STARTING
INPUTFIL
ANEMHGHT
SURFDATA
UAIRDATA
FINISHED
STARTING
RECTABLE
MAXTABLE
FINISHED
A Simp!
DFAULT
3 24
S02
RUN
STACK1
STACK1
STACK1
STACK1
STACK1
STACK1
STACK1
STACK1
STACK1
STACK1
ALL
POL1
POL1
POL1
POL1
POL1
PREPIT
e Example
RURAL
PERIOD
POINT
1.00
34 34
34 34
34 34
35.43
15.00
35.43
25.50
36.37
STA
ORIG 0
0
35
36
20
33
20
36
.0
Problem for the ISCST Model
CONC
.0
.0
34
34
34
.45
.56
.33
.56
.45
DIST 100.
GDIR 36
END
0.0 0.0
432.0 11.7 2.4
34. 34. 34. 34. 34. 34.
34. 34. 34. 34. 34. 34.
34 34 34 34 34 34
36.37 35.18 32.92 29.66
25.50 29.66 32.92 35.18
35.43 36.45 0.00 35.18
15.00 20.56 25.50 29.66
35.43 33.33
0.0
200. 300. 500. 1000.
10. 10.
34. 34. 34.
34. 34. 34.
34 34 34
25.50 20.56
36.37 36.45
32.92 29.66
32.92 35.18
.BIN UNFORM
20 FEET
94823
94823
ALLAVE
ALLAVE
1964
1964
FIRST
50
PITTSBURGH
PITTSBURGH
SECOND
FIGURE 2-1. INPUT RUNSTREAM FILE FOR ISCST MODEL FOR SAMPLE
PROBLEM
2.4.1 A Simple Industrial Source Application
For this simple tutorial, an application is selected
involving a single point source of S02 that is subject to the
2-11
-------
influences of building downwash. The source consists of a
35-meter stack with a buoyant release that is adjacent to a
building. We will assume that the stack is situated in a rural
setting with relatively flat terrain within 50 kilometers of
the plant. A polar receptor network will be placed around the
stack location to identify areas of maximum impact.
2.4.2 Selecting Modeling Options - CO Pathway
The modeling options are input to the model on the Control
pathway. The mandatory keywords for the CO pathway are listed
below. A complete listing of all keywords is provided in
Appendix B.
STARTING - Indicates the beginning of inputs for the
pathway; this keyword is mandatory on each of
the pathways.
TITLEONE - A user-specified title line (up to 68
characters) that will appear on each page of
the printed output file (an optional second
title line is also available with the keyword
TITLETWO).
MODELOPT - Controls the modeling options selected for a
particular run through a series of secondary
keywords.
AVERTIME - Identifies the averaging periods to be
calculated for a particular run.
POLLUTID - Identifies the type of pollutant being modeled.
At the present time, this option only
influences the results if S02 is modeled with
urban dispersion in the regulatory default
mode, when a half-life of 4 hours is used to
model exponential decay.
RUNORNOT - A special keyword that tells the model whether
to run the full model executions or not. If
the user selects not to run, then the runstream
setup file will be processed and any input
errors reported, but no dispersion calculations
will be made.
2-12
-------
FINISHED - Indicates that the user is finished with the
inputs for this pathway; this keyword is also
mandatory on each of the other pathways.
The first two keywords are fairly self-explanatory. As
discussed above in Section 2.2, the MODELOPT keyword on the CO
pathway is pivotal to controlling the modeling options used for
a particular run. For this example, we intend to use the
regulatory default options, so we will include the "DFAULT"
keyword on our MODELOPT input image. We also need to identify
whether the source being modeled is in a rural or an urban
environment (see Section 8.2.8 of the Guideline on Air Quality
Models for a discussion of rural/urban determinations). For
this example we are assuming that the facility is in a rural
setting. We also need to identify on this input image whether
we want the model to calculate concentration values or
deposition values. For this example, we are calculating
concentration values. After the first three input records our
input file will look something like this:
CO STARTING
CO TITLEONE A Simple Example Problem for the ISCST Model
CO MODELOPT DFAULT RURAL CONC
Note that the title parameter field does not need to be in
quotations, even though it represents a single parameter. The
model simply reads whatever appears in columns 13 through 80 of
the TITLEONE card as the title field, without changing the
lower case to upper case letters. Leading blanks are therefore
significant if the user wishes to center the title within the
field. Note also that the spacing and order of the secondary
2-13
-------
keywords on the MODELOPT card are not significant. A MODELOPT
card that looked like this:
CO MODELOPT RURAL CONC DFAULT
would have an identical result as the example above. It is
suggested that the user adopt a style that is consistent and
easy to read. A complete description of the available modeling
options that can be specified on the MODELOPT keyword is
provided in Section 3.
Since the pollutant in this example is S02, we will
probably need to calculate average values for 3-hour and
24-hour time periods, and we also need to calculate averages
for the full annual time period. Our runstream file might
therefore look something like this after adding two more
keywords:
CO STARTING
CO TITLEONE A Simple Example Problem for the ISCST Model
CO MODELOPT DFAULT RURAL CONC
CO AVERTIME 3 24 PERIOD
CO POLLUTID S02
Note again that the order of the parameters on the AVERTIME
keyword is not critical, although the order of the short term
averages given on the AVERTIME keyword will also be the order
in which the results are presented in the output file. The
order of the keywords within each pathway is also not critical
in most cases, although the intent of the input runstream file
may be easier to decipher if a consistent and logical order is
followed. It is suggested that users follow the order in which
the keywords are presented in Section 3, in Appendix B, and in
the Quick Reference, unless there is a clear advantage to doing
otherwise.
2-14
-------
The only remaining mandatory keywords for the CO pathway
are RUNORNOT and FINISHED. We will set the RUNORNOT switch to
RUN for this example. If a user is unsure about the operation
of certain options, or is setting up a complex runstream file
to run for the first time, it may be desirable to set the model
NOT to run, but simply to read and analyze the input file and
report any errors or warning messages that are generated. Once
the input file has been debugged using these descriptive
error/warning messages, then the RUNORNOT switch can be set to
RUN, avoiding a possible costly waste of resources generating
erroneous results. Even if the model is set NOT to run, all of
the inputs are summarized in the output file for the user to
review.
Our complete runstream file for the CO pathway may look
something like this:
CO STARTING
CO TITLEONE A Simple Example Problem for the ISCST2 Model
CO MODELOPT DFAULT RURAL CONC
CO AVERTIME 3 24 PERIOD
CO POLLUTID S02
CO RUNORNOT RUN
CO FINISHED
The following set of runstream images has a more structured
look, but it is equivalent to the example above:
CO STARTING
TITLEONE A Simple Example Problem for the ISCST2 Model
MODELOPT DFAULT RURAL CONC
AVERTIME 3 24 PERIOD
POLLUTID S02
RUNORNOT RUN
CO FINISHED
Since the pathway ID is required to begin in column 1 (see
Section 2.4.8 for a discussion of this restriction), the model
2-15
-------
will assume that the previous pathway is in effect if the
pathway field is left blank. The model will do the same for
blank keyword fields, which will be illustrated in the next
section.
In addition to these mandatory keywords on the CO pathway,
the user may select optional keywords to specify that elevated
terrain heights will be used (the default is flat terrain), to
allow the use of receptor heights above ground-level for
flagpole receptors, to specify a decay coefficient or a
half-life for exponential decay, and to generate an input file
containing events for processing with the EVENT model. The
user also has the option of having the model periodically save
the results to a file for later re-starting in the event of a
power failure or other interruption of the model's execution.
These options are described in more detail in Section 3 of this
volume.
2.4.3 Specifying Source Inputs - SO Pathway
Besides the STARTING and FINISHED keywords that are
mandatory for all pathways, the Source pathway has the
following mandatory keywords:
LOCATION - Identifies a particular source ID and specifies
the source type and location of that source.
SRCPARAM - Specifies the source parameters for a
particular source ID identified by a previous
LOCATION card.
SRCGROUP - Specifies how sources will be grouped for
calculational purposes. There is always at
least one group, even though it may be the
group of ALL sources and even if there is only
one source.
Since the hypothetical source in our example problem is
influenced by a nearby building, we also need to include the
optional keywords BUILDHGT and BUILDWID in our input file.
2-16
-------
The input file for the SO pathway for this example will
look something like this:
SO
so
so
sn
sn
sn
so
so
so
so
so
so
so
STARTING
LOCATION
SRCPARAM
BUILDHGT
BUILDHGT
BUILDHGT
BUILDWID
BUILDWID
BUILDWID
BUILDWID
BUILDWID
SRCGROUP
FINISHED
STACK1
STACK1
STACK1
STACK1
STACK1
STACK1
STACK1
STACK1
STACK1
STACK1
ALL
POINT 0.0
1.00 35.0
34
34
34
35
15
35
25
36
34. 34.
34 34
34 34
43 36.45
00 20.56
43 33.33
50 20.56
37 36.45
0
0 0.0
432.0 11.7
34
34
34
36
25
35
15
35
34. 34.
34 34
34 34
37 35.18
50 29.66
43 36.45
00 20.56
43 33.33
2
34
34
34
32
32
0
25
4
34. 34.
34 34
34 34
92 29.66
92 35.18
00 35.18
50 29.66
34
34
34
25
36
32
32
34
34
34
50
37
92
92
. 34.
34
34
20.56
36.45
29.66
35.18
There are a few things to note about these inputs.
Firstly, the source ID (STACK1 in this example) is an
alphanumeric parameter (up to eight characters) that identifies
the inputs for different keywords with a particular source. It
is crucial that the source be identified with a LOCATION card
before any other keyword makes reference to that source, since
this identifies the source type (POINT in this case), and
therefore which parameters the model will allow. Besides POINT
sources, the ISC models also allow VOLUME, AREA, and OPENPIT
sources to be specified.
Another thing to note is that there are 36 building
heights and 36 building widths entered on the appropriate
keywords, one value for each 10 degree sector beginning with
the 10 degree flow vector (direction toward which the wind is
blowing), and continuing clockwise. Since the user could not
fit all 36 values on a single record, the pathway, keyword and
source ID were repeated as many times as were necessary. In
this case there were 12 values given on each of three lines for
the building heights, and eight values on each of four lines
plus a line of four values for building widths. There could
have been fewer or more lines as long as exactly 36 values were
2-17
-------
entered before starting with a new keyword. Since all of the
building heights were the same across the sectors (fairly
realistic for the height but not for widths, unless the
structure was circular), there is a short cut available for
specifying numeric input in the runstream files for the new
models. The user can specify "repeat values" by entering a
field such as "36*34.0" as a parameter for the BUILDHGT
keyword. The model will interpret this as "36 separate
entries, each with a value of 34.0," and store the values in
the appropriate arrays within the model. Since the model must
identify this as a single parameter field, there must not be
any spaces between the repeat-value and the value to be
repeated.
The final keyword before finishing the SO pathway must be
the SRCGROUP keyword. In this example, since there is only one
source, we have taken advantage of a short cut provided by the
model by specifying a source group ID (which may be up to eight
characters) of ALL. Whenever this card appears in an input
file, it will generate a source group with a source-group ID of
ALL, consisting of all sources defined for that run. The
sources do not have to be explicitly identified. In a run
involving multiple sources, the user may specify multiple
source groups by repeating the SRCGROUP keyword. The use of
the SRCGROUP card is explained in more detail in Section 3.
2-18
-------
Using some of the formatting options discussed above, the
SO pathway for our example may look like this, with the same
result as above:
SO STARTING
LOCATION STACK1
** Point Source
** Parameters:
SRCPARAM STACK1
BUILDHGT STACK1
BUILDWID STACK1
STACK1
STACK1
STACK1
STACK1
SRCGROUP ALL
SO FINISHED
POINT
OS
1.00
0.0
HS
35.0
0.0
TS
432.
0.0
VS
11
7
DS
2.4
36*34.
35
15
35
25
36
43
00
43
50
37
36.45
20.56
33.33
20.56
36.45
36.37
25.50
35.43
15.00
35.43
35
29
36
20
33
18
66
45
56
33
32
32
0
25
92
92
00
50
29
35
35
29
66 25
18 36
18 32
66 32
50
37
92
92
20.56
36.45
29.66
35.18
This version of the SO pathway inputs illustrates the use of
the comment card to label the stack parameters on the SRCPARAM
card, i.e., QS for emission rate (g/s), HS for stack height
(m), TS for stack exit temperature (K), VS for exit velocity
(m/s), and DS for stack diameter (m). A complete description
of the source parameter card, with a list of parameters for
each source type, is provided in Section 3.3 and in Appendix B.
Other optional inputs that may be entered on the SO
pathway include specifying variable emission rate factors for
sources whose emissions vary as a function of month, season,
hour-of-day, STAR category, or season and hour-of-day (see
Section 3.3.4 for more details) . The number of factors entered
depends on the option selected, and factors may be input for
single sources or for a range of sources. Other keywords allow
the user to specify settling velocity categories, mass
fractions, and reflection coefficients for sources of large
particulates that experience settling and removal of the
pollutant as it is dispersed and transported downwind. This
option is also explained in more detail in Section 3.
2-19
-------
2.4.4 Specifying a Receptor Network - RE Pathway
As mentioned above, this example will illustrate the use
of a single polar receptor network centered on the stack
location. Other options available on the REceptor pathway
include specifying a Cartesian grid receptor network,
specifying discrete receptor locations in either a polar or a
Cartesian system, and specifying the location of receptors
along the boundary around a particular source. These other
options are described in more detail in Section 3.4.
For this example we will specify a polar network with
receptors located at five downwind distances for every
10-degree flow vector around the plant. There will be a total
of 180 receptors. The RE pathway for this example will look
something like this:
RE STARTING
GRIDPOLR POL1 STA
POL1 ORIG 0.0 0.0
POL1 DIST 100. 200. 300. 500. 1000.
POL1 GDIR 36 10. 10.
POL1 END
RE FINISHED
The first thing to note about these inputs is that there
is a new set of keywords, including something that looks like a
STArting and ENDing. In fact the GRIDPOLR keyword can be
thought of as a "sub-pathway," in that all of the information
for a particular polar network must be in contiguous records,
and that the starting and ending of the sub-pathway are
identified. The order of secondary keywords within the
sub-pathway is not critical, similar to the main pathways. Each
card must be identified with a network ID (up to eight
alphanumeric characters), in this case it is "POL1." Multiple
networks may be specified in a single model run. The model
waits until the END secondary keyword is encountered to set the
2-20
-------
variables, which may include terrain heights for receptors on
elevated terrain or flagpole receptor heights if those options
are being exercised by the user. The use of these optional
secondary keywords is described in detail in Section 3.4.
For this example, the ORIG secondary keyword specifies the
location of the origin, in (X,Y) coordinates, for the polar
network being defined. This network is centered at the same
(X,Y) location as the source specified above. The ORIG keyword
is optional, and the model will default to an origin of (0.0,
0.0) if it is omitted. The DIST keyword identifies the
distances along each direction radial at which the receptors
will be located. In this case there are five distances. More
could be added by adding values to that input card or by
including a continuation card, if needed. The GDIR keyword
specifies that the model will Generate DIRection radials for
the network, in this case there will be 36 directions,
beginning with the 10 degree flow vector and incrementing every
10 degrees clockwise. The user may elect to define Discrete
DIRection radials instead by using the DDIR keyword in place of
the GDIR keyword.
2.4.5 Specifying the Meteorological Input - ME Pathway
The MEteorolgy pathway has the following three mandatory
keywords (besides STARTING and FINISHED, of course):
INPUTFIL - Specifies the filename and format for the input
meteorological data file.
ANEMHGHT - Specifies the anemometer height for the wind
data to be used for the modeling run.
SURFDATA - Specifies information about the surface
meteorological data which will be used in the
modeling.
UAIRDATA - Specifies information about the upper air
meteorological data (i.e. mixing heights) which
will be used in the modeling.
2-21
-------
For the purposes of this example we will assume that the
meteorological data file is a formatted ASCII file in the
default format for ISCST3 that was generated by the PCRAMMET
meteorological preprocessor program. The filename is
PREPIT.ASC (the sample file that is provided on the SCRAM BBS
with the ISCST3 model), and it consists of twenty days of data
for Pittsburgh, PA from 1964. The runstream images for the
MEteorology pathway would look something like this:
ME STARTING
INPUTFIL PREPIT.ASC
ANEMHGHT 20 FEET
SURFDATA 94823 1964 PITTSBURGH
UAIRDATA 94823 1964 PITTSBURGH
ME FINISHED
The first parameter on the INPUTFIL keyword is the
filename, which can be entered as a full DOS pathname,
including the drive specification and subdirectories, up to a
total of 40 characters. The second parameter is the format of
the meteorology data file. In this case the secondary keyword
is blank, indicating that the meteorological data file is an
ASCII file in the default format for the model. Another option
would be to place the secondary keyword UNFORM following the
filename, in which case the model will assume an unformatted
meteorological data file of the type generated by PCRAMMET.
The order of variables assumed for the ASCII file input is as
follows: year, month, day, hour, flow vector, wind speed
(m/s), temperature (K), stability category, rural mixing height
(m), and urban mixing height (m). Other user options for
specifying the format for ASCII meteorology files are described
more fully in Section 3.5.1.
The ANEMHGHT keyword is important because the input wind
speed data are adjusted from the anemometer height to the
release height for model calculations, so that differences in
anemometer height can significantly effect the modeled results.
2-22
-------
For National Weather Service (NWS) data, the user should check
records (e.g. the Local Climatological Data summary report) for
the particular station to determine the correct anemometer
height for the data period used in the modeling, since the
anemometer location and height may change over time. The model
will assume that the anemometer height is in meters, unless the
secondary keyword "FEET" is included in the runstream image, as
illustrated in this example. The model will convert inputs in
feet to meters.
The final two mandatory inputs identify the location and
data period of the input meteorological data. A separate
keyword is used for the surface meteorological data and for the
upper air (mixing height) data. The parameters on these cards
are the station number (e.g. WBAN number for NWS stations), the
data period (year), and a station name. It is important that
these inputs be provided correctly since the model compares the
station number and year from the runstream input file with
values provided in the first record of the meteorology file.
The user may also optionally input the (X,Y) coordinates for
the location of the station(s), although these values are not
currently used by the model.
Other optional keywords available on the ME pathway
provide the user with options to specify selected days to
process from the meteorological data file, a wind direction
rotation correction term, and user-specified wind speed profile
exponents and/or vertical potential temperature gradients. The
wind profile exponents and potential temperature gradients are
ignored (and a warning message generated) if the regulatory
default option is selected. These optional inputs are
described in more detail in Section 3.5.
2-23
-------
2.4.6 Selecting Output Options - OU Pathway
All of the keywords on the Output pathway are optional,
although the model will warn the user if no printed outputs are
requested and will halt processing if no outputs (printed
results or file outputs) are selected. The printed table
keywords are:
RECTABLE - Specifies the selection of high value by
receptor table output options.
MAXTABLE - Specifies the selection of overall maximum
value table output options.
DAYTABLE - Specifies the selection of printed results (by
receptor) for each day of data processed (this
option can produce very large files and such be
used with care).
The RECTABLE keyword corresponds to the option for
highest, second-highest and third-highest values by receptor
available in the old ISCST model. The MAXTABLE keyword
corresponds to the maximum 50 table option available in the old
ISCST model. For both of these keywords, the user has
additional flexibility to specify for which short term
averaging periods the outputs are selected. For the MAXTABLE
keyword the user can also specify the number of overall maximum
values to summarize for each averaging period selected, up to a
maximum number controlled by a parameter in the computer code.
For this example problem we will select the highest and
second-highest values by receptor, and the maximum 50 values
for all averaging periods. These OU pathway inputs will look
something like this:
OU STARTING
RECTABLE ALLAVE FIRST SECOND
MAXTABLE ALLAVE 50
OU FINISHED
2-24
-------
To simplify the input for users who request the same
printed table output options for all averaging periods, these
keywords recognize the secondary keyword "ALLAVE" as the first
parameter for that purpose. In order to obtain the overall
maximum 10 values for the 24-hour averages only, then the OU
pathway images would look like this:
OU STARTING
RECTABLE ALLAVE FIRST SECOND
MAXTABLE 24 10
OU FINISHED
It should also be noted that these output table options apply
only to the short-term averaging periods, such as the 3-hour
and 24-hour averages used in our example. If the user has
selected that PERIOD averages be calculated (on the CO AVERTIME
keyword), then the output file will automatically include a
table of period averages summarized by receptor (the RECTABLE
option does not apply since there is only one period value for
each receptor). In addition, the printed output file will
include tables summarizing the highest values for each
averaging period and source group.
Other options on the OU pathway include several keywords
to produce output files for specialized purposes, such as
generating contour plots of high values, identifying
occurrences of violations of a particular threshold value (e.g.
a NAAQS), and for postprocessing of the raw concentration data.
These options are described in detail in Section 3.6.
The complete input runstream file for this simple example
is shown in Figure 2-2. Note that a consistent style has been
used for formatting and structuring the file in order to
improve its readability. This input file is comparable to the
version shown earlier in Figure 2-1, which used a somewhat
different style.
2-25
-------
CO STARTING
TITLEONE
MODELOPT
AVERTIME
POLLUTID
RUNORNOT
CO FINISHED
A Simple Example Problem for the ISCST2 Model
DFAULT RURAL CONC
3 24 PERIOD
S02
RUN
SO STARTING
LOCATION STACK1 POINT 0.0
** Point Source
** Parameters:
SRCPARAM STACK1
OS
1.00
HS
35.0
TS
432.
VS
11.7
DS
2.4
BUILDHGT
BUILDWID
SRCGROUP
SO FINISHED
STACK1
STACK1
STACK1
STACK1
STACK1
STACK1
ALL
36*34.
35.43 36.45 36.37 35.18 32.92 29.66 25.50 20.56
15.00 20.56 25.50 29.66 32.92 35.18 36.37 36.45
35.43 33.33 35.43 36.45 0.00 35.18 32.92 29.66
25.50 20.56 15.00 20.56 25.50 29.66 32.92 35.
36.37 36.45 35.43 33.33
18
RE STARTING
GRIDPOLR POL1 STA
POL1 ORIG 0.0 0.0
POL1 DIST 100. 200.
POL1 GDIR 36 10.
POL1 END
RE FINISHED
300. 500. 1000.
10.
ME STARTING
INPUTFIL PREPIT.ASC
ANEMHGHT 20 FEET
SURFDATA 94823 1964 PITTSBURGH
UAIRDATA 94823 1964 PITTSBURGH
ME FINISHED
OU STARTING
RECTABLE ALLAVE FIRST SECOND
MAXTABLE ALLAVE 50
OU FINISHED
FIGURE 2-2. EXAMPLE INPUT RUNSTREAM FILE FOR SAMPLE PROBLEM
2.4.7 Using the Error Message File to Debug the Input Runstream
File
The previous sections in this tutorial have lead through
the step-by-step construction of a sample runstream input file
for ISCST. This simple example problem illustrated the usage
2-26
-------
of the most commonly used options of the ISCST model. However,
many real-time applications of the model will be much more
complex than this example, perhaps involving multiple sources
and source groups, multiple receptor networks, the addition of
discrete receptor locations, and/or elevated terrain heights.
Since humans are prone to make errors from time to time, an
effort has been made to develop improved error handling
capabilities for the ISC models.
The error handling capabilities of the ISC models are
designed to accomplish two things for the user. First, the
model should read through the complete input file and report
all occurrences of errors or suspect entries before stopping,
rather than stopping on the first instance (and every instance
thereafter) of an error in the input file. Second, the model
should provide error and warning messages that are detailed and
descriptive enough that they will help the user in his/her
effort to debug the input file. The remainder of this section
provides of brief introduction to the use of the model's error
handling capabilities. Appendix E of this volume provides more
details about the error handling provided by the ISC models,
including a listing and explanation of all error and other
types of messages generated by the models.
The ISC models generate messages during the processing of
the input data and during the execution of model calculations.
These messages inform the user about a range of possible
conditions including:
• Errors that will halt any further processing, except to
identify additional error conditions;
• Warnings that do not halt processing but indicate a
possible errors or suspect conditions; and
• Informational messages that may be of interest to the
user but have no direct bearing on the validity of the
results.
2-27
-------
As the model encounters a condition for which a message is
generated, the model writes the message to a temporary storage
file. At the completion of the setup processing for a run, and
at the completion of the model calculations, the model rereads
the message file and generates a summary of the messages which
is included in the main printed output file. If the processing
of the model setup information indicates no errors or warnings,
and the user has selected the option to RUN the model
calculations on the CO RUNORNOT card, then the model will
simply write a statement to the print file that the model setup
was completed successfully. Otherwise, the model will report a
summary of the messages encountered. The summary of model
setup messages that would be generated for the example problem
if the option NOT to run was chosen is shown in Figure 2-3.
This summary table reports the total number of occurrences for
each of the message types, and lists the detailed message for
any fatal errors or warning messages that were generated. In
this case, since there were no errors or suspicious conditions
in the setup file, there are no error or warning messages
listed.
An example of the warning message that would have been
generated had we left out the card on the RE pathway that
specifies the origin of the polar receptor network is shown
below:
2-28
-------
RE W220 39 REPOLR: Missing Origin (Use Default = 0,0) In GRIDPOLR POL1
***** *
***** *
* * * * * Hints
* * * * *
* * * * Detailed error/warning message
* * * *
* * * Subroutine from which message is generated
* * *
* * Line number of file where message occurred
* *
* Message code - including message type (E, W, I) and message number
*
Pathway ID where message originated
Since this is a warning message, it would have appeared at the
end of the message summary table in the output file, but it
would not have halted processing of the data. The last item on
the message line, "Hints," may include such information as the
keyword or parameter name causing the error, the source ID,
group ID or (as in this case) the network ID involved, or
perhaps the date variable identifying when the message occurred
during the processing of the meteorological data, such as an
informational message identifying the occurrence of a calm
wind.
For new users and for particularly complex applications,
it is strongly recommended that the model first be run with the
RUNORNOT keyword (on the CO pathway) set NOT to run. In this
way, the user can determine if the model is being setup
properly by the runstream file before committing the resources
to perform a complete run. The user should make a point of
examining any warning messages carefully to be sure that the
model is operating as expected for their application, since
these messages will not halt processing by the model. In most
cases, the detailed messages will provide enough information
for the user to determine the location and nature of any errors
in the runstream setup file. If the intent of the message is
not immediately clear, then the user should refer to the more
detailed descriptions provided in Appendix E for the particular
error code generated.
2-29
-------
In deciphering the error and warning messages, the line
number provided as part of the message may be particularly
helpful in locating the error within the input file. However,
if it is an error of omission that is caught by the error
checking performed at the completion of inputs for a pathway,
then the line number will correspond to the last record for
that pathway. The user may need to examine all of the messages
carefully before locating the error or errors, especially since
a single occurrence of certain types of errors may lead to
other error conditions being identified later in the input file
which do not really constitute errors in themselves. An
example of this is provided in Figure 2-4, which shows some
inputs for the SO pathway where the building dimension keywords
have been typed incorrectly, and the associated list of error
messages. Since continuation cards were being used for the
building width inputs, and the keyword was entered incorrectly
on the first line, the subsequent records were also taken by
the model to be invalid keyword inputs. While the error
messages are the same for these records, the message originates
from a different part of the model (SUBROUTINE SOCARD) for the
records with the blank keyword.
Since the detailed error and warning messages are listed
in the output file as part of the message summary table, there
will generally not be a need for the user to examine the
contents of the detailed message file. For this reason, the
default operation of the model is to write the messages that
are generated by a particular run to a temporary file that is
deleted when the run is completed. If the user wishes to
examine the complete list of detailed messages (of all types),
there is an optional keyword available on the CO pathway for
that purpose. The ERRORFIL keyword, which is described in
2-30
-------
detail in Section 3.2.7, allows the user to save the complete
list of detailed messages to a user-specified filename.
Message Summary For ISC3 Model Setup ***
Summary of Total Messages
A Total of 0 Fatal Error Message(s)
A Total of 0 Warning Message(s)
A Total of 0 Information Message(s)
******** FATAL ERROR MESSAGES ********
*** NONE ***
******** WARNING MESSAGES ********
*** NONE ***
***********************************
*** SETUP Finishes Successfully ***
FIGURE 2-3. EXAMPLE MESSAGE SUMMARY TABLE FOR RUNSTREAM SETUP
2-31
-------
SO STARTING
LOCATION STACK1
** Point Source
** Parameters:
SRCPARAM STACK1
BUILDHTS STACK1
BUILDWTS STACK1
STACK1
STACK1
STACK1
STACK1
SRCGROUP ALL
SO FINISHED
POINT
OS
0.0
HS
0.0
TS
0.0
VS
DS
1.00 35.0 432.0 11.7 2.4
36*34.
35.43 36.45 36.37 35.18 32.92 29.66 25.50 20.56
15.00 20.56 25.50 29.66 32.92 35.18 36.37 36.45
35.43 33.33 35.43 36.45 0.00 35.18 32.92 29.66
25.50 20.56 15.00 20.56 25.50 29.66 32.92 35.18
36.37 36.45 35.43 33.33
*** Message Summary For ISC3 Model Setup ***
Summary of Total Messages
A Total of 6 Fatal Error Message(s)
A Total of 0 Warning Message(s)
A Total of 0 Information Message(s)
******** FATAL ERROR MESSAGES ********
17 EXKEY : Invalid Keyword Specified.
SO E105
SO E105
SO E105
SO E105
SO E105
SO E105
18 EXKEY : Invalid Keyword Specified.
19 SOCARD: Invalid Keyword Specified.
20 SOCARD: Invalid Keyword Specified.
21 SOCARD: Invalid Keyword Specified.
22 SOCARD: Invalid Keyword Specified.
The Troubled Keyword is BUILDHTS
The Troubled Keyword is BUILDWTS
The Troubled Keyword is BUILDWTS
The Troubled Keyword is BUILDWTS
The Troubled Keyword is BUILDWTS
The Troubled Keyword is BUILDWTS
******** WARNING MESSAGES ********
*** NONE ***
**************************************
*** SETUP Finishes UN-successfully ***
FIGURE 2-4. EXAMPLE OF KEYWORD ERROR AND ASSOCIATED MESSAGE
SUMMARY TABLE
2.4.8 Running the Model and Reviewing the Results
Now that we have a complete and error-free runstream input
file, we are ready to run the model and then review the
results. The PC-executable files available on the SCRAM BBS
2-32
-------
open the runstream input and printed output files explicitly
within the model, so there is no need to "redirect" the I/O on
the command line using the DOS redirection symbols '<' and '>'.
The command line to run the sample problem might look something
like this on the PC:
C:\>ISCST3 TEST-ST.INP TEST-ST.OUT
The "c-prompt" of DOS has been represented by the characters
"C:\>", but may appear different on different machines. The
important points are that the ISCST3.EXE file either be in the
directory from which you are attempting to run the model, or in
a directory that is included on the DOS PATH command when the
system is "booted-up." The runstream input filename must
appear first (without any DOS "redirection" symbol), followed
by the desired output filename (also without the DOS
redirection symbol), and these files must also be located in
the directory from which the model is being executed, unless a
complete DOS pathname is provided on the command line.
As mentioned above, the SCRAM PC-executable files for ISC
open the input and output files explicitly. One reason for
this is to allow for the models to write an update on the
status of processing to the PC terminal screen. For the ISCST
model, the model first indicates that setup information is
being processed and then gives the Julian day currently being
processed. If no status message is seen then the model did not
load into memory properly. If the model stops after completing
the setup processing, then either the RUNORNOT option was set
NOT to run, or a fatal error was encountered during the setup
processing. Another reason for not sending the printed output
to the default output device (i.e., to the screen or redirected
to a file), is so that any DOS error messages will be visible
on the screen and not be written to the printed file. One such
message might be that there is insufficient memory available to
run the program. Handling of DOS error messages may require
2-33
-------
some knowledge of DOS, unless the meaning of the message is
obvious.
The order of contents and organization of the main output
file for the ISC models is presented in Figure 2-5.
Echo of Input Runstream Images
Summary of Runstream Setup Messages
Summary of Inputs
Summary of Modeling Options
Summary of Source Data
Summary of Receptor Data
Summary of Meteorology Data
Model Results
Daily Results for Each Averaging Period and Output Type
Selected for Each Day Processed (If Applicable)
- DAYTABLE Keyword
PERIOD or ANNUAL Results for Each Source Group and
Output Type (If Applicable)
- PERIOD or ANNUAL Parameter on AVERTIME Keyword
Short Term Average Results (High, Second High, etc.) by
Receptor for Each Source Group and Output Type (If
Applicable)
- RECTABLE Keyword
Overall Maximum Short Term Average Results for Each
Source Group and Output Type (If Applicable)
- MAXTABLE Keyword
Summary Tables of High Values for Each Averaging Period,
Source Group and Output Type (Always provided if PERIOD
or ANNUAL averages or the RECTABLE keyword are used)
Summary of Complete Model Execution Messages
FIGURE 2-5. ORGANIZATION OF ISCST MODEL OUTPUT FILE
The references to "Output Type" in Figure 2-5 refer to the
option with the Short Term model to output concentration, total
deposition, dry deposition, and/or wet deposition in a single
model run. Each page of the output file, except for the echo
2-34
-------
of the input file images, is labeled with the model name and
version number, user-specified title(s), page number, and, for
the PC version of the model, the date and time of the
particular run. Also included as part of the header
information for each page is a one-line summary of the modeling
options used for that particular run. The modeling options are
listed as the secondary keywords used to control the options,
such as URBAN or RURAL, CONG or DEPOS, DFAULT, NOCALM, etc.
(Details about the date/time routines and other PC-specific
features of the computer code are discussed in Section 4.0 of
this Volume and in Volume III.)
Since the complete input file is normally echoed back as
part of the output file, and since processing of the inputs
stops when the OU FINISHED card is reached, the run can be
duplicated by simply specifying the output filename as the
input runstream file. Alternatively, the input records could
be "cut and pasted" from the output file to a separate file
using a text editor. This allows for the model run to be
duplicated even if the original runstream file is not
available.
By default, the models will echo each line of the input
runstream file to the printed output file. This provides a
convenient record of the inputs as originally read into the
model, without any rounding of numerical values that may appear
in the input summary tables. As noted above, it also means
that the output file can be used as an input file to the model
to reproduce a particular application. However, for some
applications, the length of the input runstream file may be too
cumbersome to include the entire set of inputs at the beginning
of each output file. This may happen, for example, if a large
number of sources are being defined or if a large number of
discrete receptor locations are used. For this reason, the
user is provided with the option to "turn off" the echoing of
the input file at any point within the runstream file. This is
2-35
-------
accomplished by entering the keywords "NO ECHO" in the first
two fields anywhere within the runstream file. In other words,
place NO in the pathway field, followed by a space and then
ECHO. None of the input runstream images after the NO ECHO
will be echoed to the output file. Thus, a user may choose to
place NO ECHO after the Control pathway in order to keep the
control options echoed, but suppress echoing the rest of the
input file.
The details of the message summary tables were discussed
in the previous section. A portion of the summary of modeling
option inputs is shown in Figure 2-6 for the simple example
described in this section. For the new model, the summary of
source parameter input data includes separate tables for each
source type, rather than combining all sources onto a single
table. In this way the column headings are specific to the
source type.
Figure 2-7 presents an example of the results output for
the second highest values by receptor for our sample problem.
These values are the second highest 24-hour averages at each
receptor location. Note that several of the numbers are
followed by a 'c.' This flag indicates that the average
included at least one calm hour during the averaging period.
The number in parentheses following each concentration value is
the date corresponding to each value. The date is given as an
eight digit integer variable that includes the year (2-digits),
month, day, and hour corresponding to the end of the averaging
period. Since these are 24-hour averages and are based on
block (end-to-end) rather than running averages, all of the
dates end on hour 24.
For each of the different types of model result tables,
the controlling keyword is identified above at the end of the
description. All of the outputs of the same type, e.g. high
values by receptor, are printed together, and the order of
2-36
-------
tables loops through all source groups for a particular
averaging period, and then loops through all averaging periods.
The summary tables of high values at the end of the model
results follow the same order of loops. An example of the
summary tables for our sample problem is shown in Figure 2-8.
The summaries for all averaging periods have been combined onto
a single figure, but would appear on separate pages of the
actual output file.
2-37
-------
*** ISCST3 - VERSION 95250 *** *** A Simple Example Problem for the ISCST Model
09/07/95
***
12:00:00
PAGE 1
**MODELOPTs: CONC RURAL FLAT DFAULT
*** MODEL SETUP OPTIONS SUMMARY
**Intermediate Terrain Processing is Selected
**Model Is Setup For Calculation of Average CONCentration Values.
-- SCAVENGING/DEPOSITION LOGIC --
**Model Uses NO DRY DEPLETION. DDPLETE = F
**Model Uses NO WET DEPLETION. WDPLETE = F
**NO WET SCAVENGING Data Provided.
**Model Does NOT Use GRIDDED TERRAIN Data for Depletion Calculations
**Model Uses RURAL Dispersion.
**Model Uses Regulatory DEFAULT Options:
1. Final Plume Rise.
2. Stack-tip Downwash.
3. Buoyancy-induced Dispersion.
4. Use Calms Processing Routine.
5. Not Use Missing Data Processing Routine.
6. Default Wind Profile Exponents.
7. Default Vertical Potential Temperature Gradients.
8. "Upper Bound" Values for Supersquat Buildings.
9. No Exponential Decay for RURAL Mode
**Model Assumes Receptors on FLAT Terrain.
**Model Assumes No FLAGPOLE Receptor Heights.
**Model Calculates 2 Short Term Average(s) of: 3-HR 24-HR
and Calculates PERIOD Averages
**This Run Includes: 1 Source(s); 1 Source Group(s); and 180 Receptor(s)
2-38
-------
**The Model Assumes A Pollutant Type of: S02
**Model Set To Continue RUNning After the Setup Testing.
'"Output Options Selected:
Model Outputs Tables of PERIOD Averages by Receptor
Model Outputs Tables of Highest Short Term Values by Receptor (RECTABLE Keyword)
Model Outputs Tables of Overall Maximum Short Term Values (MAXTABLE Keyword)
FIGURE 2-6.
*** ISCST3 - VERSION 95250 ***
09/07/95
12:00:30
PAGE 13
**MODELOPTs: CONC
SAMPLE OF MODEL OPTION SUMMARY TABLE FROM AN ISC MODEL OUTPUT FILE
*** A Simple Example Problem for the ISCST Model
***
RURAL FLAT
DFAULT
*** THE 2ND HIGHEST 24-HR AVERAGE CONCENTRATION
INCLUDING SOURCE(S): STACK1 ,
VALUES FOR SOURCE GROUP: ALL
"** NETWORK ID: POL1
** CONC OF S02
; NETWORK TYPE: GRIDPOLR ***
IN MICROGRAMS/M**3
DIRECTION I
(DEGREES) 1
1000.00
10.0 1
(64010224)
20.0 1
(64010324)
30.0 1
(64010224)
40.0 1
(64010524)
50.0 1
(64010524)
60.0 1
(64010224)
0
0
0
2
17
9
100.
.00038
.00032
.06544
.24546
.05618
.40921
00
(64010524)
(64010224)
(64010324)
(64010524)
(64010524)
(64010224)
0.
0.
3.
7.
12.
6.
200.
,00759
,73597
,09471
,13027
,96035
,06938
DISTANCE (METERS)
00 300.00
(64010324)
(64010324)
(64010224)
(64010324)
(64010524)
(64010224)
0
0
2
4
8
4
.00223
.46271
.05010
.90821
.87260
.17845
(64010224)
(64010324)
(64010224)
(64010324)
(64010524)
(64010224)
0
0
1
2
4
2
500.
.00058
.22714
.00969
.56813
.40116
.05521
00
(64010224)
(64010324)
(64010224)
(64010524)
(64010524)
(64010224)
0.00012
0.08851
0.46573
1.20217
2.17334
0.94001
2-39
-------
70.0 I 4
(64011024)
80.0 I 1
(64010124)
90.0 I 0
(64011024)
100.0 I 1
(64011024)
110.0 I 1
(64010424)
120.0 I 1
0.06649c(64010724)
130.0 I 0
(64010924)
140.0 I 0
(64010924)
150.0 I 0
(64010124)
160.0 I 0
(64010124)
170.0 I 0
(64010124)
180.0 I 0
(64010124)
190.0 I 0
(64010124)
200.0 I 0
(64010124)
210.0 I 0
0.00315c(64010724)
220.0 I 0
0.00000c(64010724)
230.0 I 0
(64010824)
240.0 I 0
(64010824)
250.0 I 2
(64010824)
260.0 I 0
(64010824)
270.0 I 0
(64010824)
280.0 I 0
(64010924)
98424 (64011024)
10668 (64010424)
33531 (64010424)
14289 (64011024)
38580 (64010424)
46832c(64010724)
73820 (64010924)
00385 (64010924)
00000 (64010924)
00000 (64010924)
00000 ( 0)
00000 ( 0)
00000 (64010124)
00000 (64010124)
00000c(64010724)
00000c(64010724)
00017 (64010824)
82936 (64010824)
85290 (64010124)
93134 (64010824)
01273 (64010824)
44666 (64010924)
4.83446 (64011024)
1.32557 (64010124)
0.89549 (64010424)
1.66369 (64011024)
1.41520 (64010424)
0.72598c(64010724)
0.50974 (64010924)
0.00152 (64010924)
0.00000 (64010124)
0.00203 (64010124)
0.12191 (64010124)
0.04481 (64010124)
0.00008 (64010124)
0.00000 (64010124)
0.00014 (64010124)
0.00021c(64010724)
0.00004 (64010824)
0.52206 (64010824)
2.16804 (64010124)
0.90262 (64010824)
0.02553 (64010824)
0.36178 (64010924)
3.64057 (64011024)
0.99239 (64010124)
0.76865 (64010424)
1.30464 (64011024)
1.09491 (64010424)
0.41049c(64010724)
0.36027 (64010924)
0.00072 (64010924)
0.00000 (64010124)
0.00054 (64010124)
0.04290 (64010124)
0.01473 (64010124)
0.00002 (64010124)
0.00000 (64010124)
0.00003 (64010124)
0.00005c(64010724)
0.00001 (64010824)
0.34721 (64010824)
1.36673 (64010824)
0.43338 (64010824)
0.02055 (64010824)
0.24921 (64010924)
1.93861 (64011024) 0.96955
0.55702 (64010124) 0.38055
0.55710 (64010424) 0.68970
0.77602 (64010924) 0.45574
0.59547 (64010424) 0.32417
0.12771c(64010724)
0.16093 (64010924) 0.07651
0.00020 (64010924) 0.00004
0.00000 (64010124) 0.00000
0.00005 (64010124) 0.00003
0.00504 (64010124) 0.00702
0.00161 (64010124) 0.00183
0.00000 (64010124) 0.00000
0.00003 (64010124) 0.00087
0.00020 (64010124)
0.00000c(64010724)
0.00000 (64010824) 0.00000
0.10982 (64010824) 0.06490
0.46188 (64010824) 0.34731
0.15206 (64010824) 0.12491
0.01631 (64010824) 0.05295
0.10171 (64010924) 0.05489
2-40
-------
290.0 1
(64010924)
300.0 1
(64010924)
310.0 1
(64010824)
320.0 1
(64010924)
330.0 1
(64010924)
340.0 1
(64010924)
350.0 1
(64010624)
360.0 1
1.99281 (64010924)
3.26315 (64010924)
1.61856 (64010824)
1.08368 (64010624)
0.00133 (64010124)
0.00000 (64010124)
0.01162 (64010524)
2.22860c(64010724)
1.57520 (64010924)
2.22510 (64010924)
1.02108 (64010824)
2.99288 (64010924)
1.34910 (64010924)
0.18219 (64010924)
0.01620 (64010624)
1.25950 (64010924)
1.11347 (64010924)
1.53359 (64010924)
0.68047 (64010824)
2.00757 (64010924)
0.95774 (64010924)
0.11241 (64010924)
0.00568 (64010624)
0.83449c(64010724)
0.47971 (64010924)
0.69664 (64010924)
0.27048 (64010824)
0.98393 (64010924)
0.49932 (64010524)
0.05144 (64010924)
0.00109 (64010624)
0.67738c(64010724)
0.25976
0.33747
0.11362
0.44887
0.26437
0.01881
0.00032
0.38261c(64010724)
*** ISCST3 -
09/07/95
12:00:00
PAGE 16
**MODELOPTs:
GROUP ID
ALL 1ST
2ND
3RD
4TH
5TH
6TH
FIGURE
VERSION 95250 *** ***
***
CONC
AVERAGE
HIGHEST VALUE IS 5.
HIGHEST VALUE IS 4.
HIGHEST VALUE IS 3.
HIGHEST VALUE IS 3.
HIGHEST VALUE IS 2.
HIGHEST VALUE IS 2.
2-7. EXAMPLE OUTPUT
TABLE OF HIGH VALUES BY RECEPTOR
A Simple Example Problem for the ISCST Model ***
RURAL FLAT
*** THE SUMMARY
** CONC OF S02
DFAULT
OF MAXIMUM PERIOD ( 240
IN MICROGRAMS/M**3
CONC RECEPTOR (XR, YR, ZELEV,
59843 AT ( 76.60
46934 AT ( 153.21
96137 AT ( 86.60
17067 AT ( 229.81
88217 AT ( 128.56
72413 AT ( 173.21
64.28, 0.00,
128.56, 0.00,
50.00, 0.00,
192.84, 0.00,
153.21, 0.00,
100.00, 0.00,
***
HRS) RESULTS ***
**
NETWORK
ZFLAG) OF TYPE GRID-ID
0.00) GP POL1
0.00) GP POL1
0.00) GP POL1
0.00) GP POL1
0.00) GP POL1
0.00) GP POL1
*** THE SUMMARY OF HIGHEST 3-HR RESULTS *"
2-41
-------
** CONC OF S02 IN MICROGRAMS/M**3
DATE
NETWORK
GROUP ID
GRID-ID
ALL HIGH
POL1
HIGH
POL1
NETWORK
GROUP ID
GRID-ID
ALL HIGH
POL1
HIGH
AVERAGE CONC (YYMMDDHH) RECEPTOR (XR, YR, ZELEV, ZFLAG)
1ST HIGH VALUE IS 58.49796 ON 64010524: AT ( 0.00, 100.00, 0.00,
2ND HIGH VALUE IS 42.91793 ON 64010218: AT ( 76.60, 64.28, 0.00,
*** THE SUMMARY OF HIGHEST 24-HR RESULTS ***
** CONC OF S02 IN MICROGRAMS/M**3 **
DATE
AVERAGE CONC (YYMMDDHH) RECEPTOR (XR, YR, ZELEV, ZFLAG)
1ST HIGH VALUE IS 19.16219 ON 64010224: AT ( 76.60, 64.28, 0.00,
2ND HIGH VALUE IS 17.05618 ON 64010524: AT ( 76.60, 64.28, 0.00,
OF TYPE
0.00) GP
0.00) GP
OF TYPE
0.00) GP
0.00) GP
POL1
** RECEPTOR TYPES: GC = GRIDCART
GP = GRIDPOLR
DC = DISCCART
DP = DISCPOLR
BD = BOUNDARY
FIGURE 2-8. EXAMPLE OF RESULT SUMMARY TABLES FOR THE ISC SHORT TERM MODEL
2-42
-------
2.5 MODIFYING AN EXISTING RUNSTREAM FILE
As noted earlier, one of the advantages of the keyword/parameter approach and the
flexible format adopted for the input runstream file is that it will be easier for the
user to make modifications to the runstream file and obtain the desired result. This
section briefly illustrates some examples of how a runstream file can be modified. It
is assumed that the reader is familiar with the operation of and basic editing commands
for a text editor (i.e., a program that edits ASCII files), and is familiar with the
previous sections of this tutorial.
2.5.1 Modifying Modeling Options
Depending on the type of analysis being performed, the user may need to modify the
modeling options and run the model again. Because of the descriptive nature of the
keywords and the secondary keywords used to control the modeling options, this can
easily be done with the new runstream file, and usually without having to refer back to
the user's guide each time a modification is attempted.
One example where a modeling option might need to be changed is if a modeler wanted
to obtain both concentration estimates and estimates of dry deposition for a source or
sources of large particulates. The only change needed to accomplish this is to replace
the secondary keyword of CONG (for CONCentration) with the secondary keyword of DEPOS
(for DEPOSition) on the MODELOPT input card. None of the source information needs to be
changed since the model automatically converts the emission rates to the proper units
for deposition calculations. For an ISCST run, both concentration and deposition can be
2-43
-------
estimated in the same model run. It is equally easy to modify a run to use urban
dispersion instead of rural dispersion (or vice versa) by replacing the RURAL secondary
keyword with URBAN on the MODELOPT card. As noted earlier, the order and exact spacing
of the secondary keywords on the MODELOPT card is not important.
Another modeling option change that will be discussed here is switching between
flat and elevated terrain modeling. As noted earlier, the model assumes flat terrain,
i.e., all receptors are assumed to be at the same elevation as the base elevation for
the source as the default mode of operation. If the user wishes to model receptors on
elevated terrain, then the TERRHGTS keyword must be included on the CO pathway. This
keyword, which is described in more detail in Section 3.2.3, accepts one of two possible
secondary keywords, either FLAT or ELEV. Their meaning should be obvious. Note that
the input runstream image:
CO TERRHGTS FLAT
has the same effect as having no TERRHGTS keyword at all. If the user elects to perform
FLAT terrain modeling for a particular application, the model will ignore any elevated
terrain height information given on the RE pathway. Processing will continue as flat
terrain, and warning messages will be generated to warn the user that elevated terrain
heights were present in the file, but ignored for processing. The advantage of this
approach is that if an application is setup for elevated terrain modeling, a simple
change of the secondary keyword on the TERRHGTS card from ELEV to FLAT is all that is
needed to run the model in flat terrain mode. The terrain height information does not
need to be removed from the input file.
2-44
-------
2.5.2 Adding or Modifying a Source or Source Group
Modifying the input file to add a source or a source group, or to add a source to a
source group, is as simple as just adding it. There is no need to specify the total
number of sources in the run, which would then have to be changed if more sources were
added. The same applies to the number of groups, or the number of sources per group.
If the user attempts to input more than the total number of sources or groups allowed
for a particular run, an error message will be generated to that effect. Also,
modifying a source group to delete a source is as easy as just deleting it from the
input card, without having to change any other inputs.
Another way of "deleting" a source or a group from an input file is to place a "**"
in the pathway field of the card or cards which define the source or group to "comment
out" those inputs. This approach, which was discussed above in Section 2.1.2, has the
advantage of leaving the input data for the source or group in the input file for
possible later use. It doesn't matter whether the "**" is entered with the text editor
in "insert" mode, in which case the other inputs of that line are moved over, or if it
is in "overtype" mode, which would replace the pathway ID that was already there.
2.5.3 Adding or Modifying a Receptor Network
As with source data, adding to or modifying the receptor information in the ISC
models is relatively straight forward. The problem of having to make several changes to
accomplish one small modification, such as adding a distance to a polar receptor
network, has been avoided in the new model. All that the user needs to do is to add the
new distance on the appropriate input card, which is easily identifiable because of the
2-45
-------
use of descriptive keywords. The model checks to ensure that the user does not attempt
to specify more than the maximum number of receptors for a particular run, and generates
an appropriate message if too many are input.
2.5.4 Modifying Output Options
Modifying the output options involves many of the same principles that are
described above. In addition, all of the output options are structured in a way that
allows the user to select options for specific averaging periods, so that the user may
find it useful to copy a record or group of records set up for one averaging period and
simply change the averaging period parameter. The other important short cut that is
available for the printed table output options is to use the secondary keyword ALLAVE to
indicate that the option applies to all averaging periods that are calculated. In this
way, there will be no need to change the output options if a new averaging period is
added to a run or if one is deleted.
2-46
-------
3.0 DETAILED KEYWORD REFERENCE
This section of the ISC User's Guide provides a detailed reference for all of the
input keyword options for the ISC Short Term and Long Term models. The information
provided in this section is more complete and detailed than the information provided in
the Brief Tutorial in Section 2. Since this section is intended to meet the needs of
experienced modelers who may need to understand completely how particular options are
implemented in the model, the information for each keyword should stand on its own.
This section assumes that the reader has a basic understanding of the keyword/parameter
approach used by the new models for specification of input options and data. Novice
users should first review the contents of Section 2 in order to obtain that
understanding.
The information in this section is organized by function, i.e., the keywords are
grouped by pathway, and are in a logical order based on their function within the model.
The order of keywords presented here is the same as the order used in the functional
keyword reference in Appendix B, and the Quick Reference section at the end of the
volume. The syntax for each keyword is provided, and the keyword type is specified -
either mandatory or optional and either repeatable or non-repeatable. Unless noted
otherwise, there are no special requirements for the order of keywords within each
pathway, although the order in which the keywords are presented here and in Appendix B
is recommended. Any keyword which has special requirements for its order within the
pathway is so noted following the syntax and type description.
The syntax descriptions in the following sections use certain conventions.
Parameters that are in all capital letters and underlined in the syntax description are
3-1
-------
secondary keywords that are to be entered as indicated for that keyword. Other
parameters are given descriptive names to convey the meaning of the parameter, and are
listed with an initial capital letter. Many of the parameter names used correspond to
variable names used in the computer code of the models. Parentheses around a parameter
indicate that the parameter is optional for that keyword. The default that is taken
when an optional parameter is left blank is explained in the discussion for that
keyword.
3.1 AN OVERVIEW OF SHORT TERM VS. LONG TERM MODEL INPUTS
One of the goals of the ISC reprogramming effort was to make the inputs for the new
Short Term and Long Term models as consistent as possible. As a result, the majority of
keywords are the same for both models. Because of this similarity, and because the
Short Term model is the more widely used of the two, the discussions in the following
sections are oriented toward the Short Term model. Any differences in the parameters
for a keyword for the Long Term model are highlighted so that they are easily
distinguishable. Also, any keyword that applies to only one of the models includes a
note to that effect. There is no separate reference for the Long Term model inputs in
the user's guide.
Also, unless otherwise noted, the input keywords described below apply to both the
ISCST and the ISCEV (EVENT) Short Term models. In addition to the isolated keywords
noted that apply to only one or the other model, the entire REceptor pathway applies
only to ISCST, and the EVent pathway applies only to the ISCEV model.
3-2
-------
3.2 CONTROL PATHWAY INPUTS AND OPTIONS
The COntrol pathway contains the keywords that provide the overall control of the
model run. These include the dispersion options, averaging time options, terrain height
options, and others that are described below. The CO pathway must be the first pathway
in the runstream input file.
3.2.1 Title Information
There are two keywords that allow the user to specify up to two lines of title
information that will appear on each page of the main output file from the model. The
first keyword, TITLEONE, is mandatory, while the second keyword, TITLETWO, is optional.
The syntax and type for the keywords are summarized below:
Syntax: CO TITLEONE Titiei
CO TITLETWO TitleZ
Type: TITLEONE - Mandatory, Non-repeatabl e
TITLETWO - Optional, Non-repeatable
The parameters Titlel and Title2 are character parameters of length 68, which are read
as a single field from columns 13 to 80 of the input record. The title information is
taken as it appears in the runstream file without any conversion of lower case to upper
case letters. If the TITLETWO keyword is not included in the runstream file, then the
second line of the title in the output file will appear blank.
3-3
-------
3.2.2 Dispersion Options
The dispersion options are controlled by the MODELOPT keyword on the CO pathway.
The syntax, type, and order of the MODELOPT keyword are summarized below:
Syntax: Short Term model:
CO MODELOPT DFAULT CONG DRYDPLT WETDPLT RURAL GRDRIS NOSTD NOBID NOCALM MSGPRO NOSMPL
DEPPS or or
DDEP URBAN NOCMPL
and/or
WDEP
Long Term model:
CO MODELOPT DFAULT CONG DRYDPLT RURAL GRDRIS NOSTD NOBID
DEPPS or
or URBAN
DDEP
Type: Mandatory, Non-repeatabl e
Order: Must precede POLLUTID, HALFLIFE and DCAYCOEF
where the secondary keyword parameters are described below (the order and spacing of
these parameters is not critical):
DFAULT - Specifies that the regulatory default options will be used;
CONG - Specifies that concentration values will be calculated;
DEPPS - Specifies that total deposition flux values (both dry and wet) will be
calculated for Short Term and dry deposition flux values for Long Term;
DDEP - Specifies that dry deposition flux values only will be calculated (same as
DEPPS for Long Term);
3-4
-------
WDEP - Specifies that wet deposition flux values only will be calculated (Short
Term only);
DRYDPLT -Specifies that plume depletion due to dry removal mechanisms will be
included in calculations;
WETDPLT -Specifies that plume depletion due to wet removal mechanisms will be
included in calculations (Short Term only);
RURAL - Specifies that rural dispersion parameters will be used;
URBAN - Specifies that urban dispersion parameters will be used;
GRDRIS - Specifies that the non-default option of gradual plume rise will be used;
NOSTD - Specifies that the non-default option of no stack-tip downwash will be
used;
NOBID - Specifies that the non-default option of no buoyancy-induced dispersion
will be used;
NOCALM - Specifies that the non-default option to bypass the calms processing
routine will be used (Short Term only);
MSGPRO - Specifies that the non-default option of the missing data processing
routine will be used (Short Term only);
NOSMPL - Specifies that no simple terrain calculations will be made, i.e., uses
COMPLEXl algorithms only (Short Term only);
NOCMPL - Specifies that no complex terrain calculations will be made, i.e., uses
ISCST algorithms only (Short Term only).
If the DFAULT secondary keyword is included among the parameter fields, then any
non-default option will be overridden. This includes the non-default options that may
be specified on the MODELOPT keyword, and also any attempt to enter non-default values
3-5
-------
of the wind profile exponents (see keyword WINDPROF on the ME pathway) or vertical
potential temperature gradients (see keyword DTHETADZ on the ME pathway). If the DFAULT
parameter is not specified, then the regulatory default options will still be used
unless a non-default option is specified in the input runstream. The model will also
assume RURAL dispersion if neither the RURAL or URBAN keywords are present, and will
assume CONCentration calculations if neither the CONG, DEPPS, DDEP or WDEP keywords are
used. Non-fatal warning messages are generated in either case. For the Short Term
model, the user may select any or all of the output types (CONG, DEPOS, DDEP and/or
WDEP) to be generated in a single model run (up to the number of output types available,
which is controlled by the NTYP parameter in the MAIN1.INC file). The order of these
secondary keywords on the MODELOPT card has no effect on the order of results in the
output files. If both the NOCMPL and the NOSMPL keywords are omitted from the MODELOPT
card, then the model will implement both simple and complex terrain algorithms and also
apply intermediate terrain processing.
The regulatory default options are identified in Appendix A of the Guideline on Air
Quality Models (Revised) (EPA, 1987b), and include the following:
• Use stack-tip downwash (except for Schulman-Scire downwash);
• Use buoyancy-induced dispersion (except for Schulman-Scire downwash);
• Do not use gradual plume rise (except for building downwash);
• Use the calms processing routines;
• Use upper-bound concentration estimates for sources influenced by building
downwash from super-squat buildings;
• Use default wind speed profile exponents; and
3-6
-------
• Use default vertical potential temperature gradients.
Other model options, such as complex terrain, are not affected by the regulatory default
options.
The default wind profile exponents and vertical potential temperature gradients are
provided below:
Pasquill
Stability
Category
A
B
C
D
E
F
Rural
Wind
Profile
Exponent
0.07
0.07
0.10
0.15
0.35
0.55
Urban
Wind
Profile
Exponent
0.15
0.15
0.20
0.25
0.30
0.30
Rural
Temperature
Gradient
(K/m)
0.0
0.0
0.0
0.0
0.020
0.035
Urban
Temperature
Gradient
(K/m)
0.0
0.0
0.0
0.0
0.020
0.035
The depletion options (DRYDPLT and WETDPLT) may be used with CONG, DEPOS, DDEP or
WDEP, but particle information must be specified in the SO pathway (see Section 3.3.6)
if DRYDPLT is included, and scavenging coefficients must be specified on the SO pathway
if WETDPLT is included. When particles are modeled, a settling velocity and a
deposition velocity are calculated for each size category. The settling velocity causes
the plume to "tilt" towards the surface (if the plume is elevated) as it travels
downwind, while the deposition velocity is used to calculate the flux of matter
deposited at the surface. If the depletion parameters (DRYDPLT and WETDPLT) are not
3-7
-------
included as model options, then the mass of particles deposited on the surface from
gravitational settling and/or precipitation scavenging is not removed from the plume.
However, plume settling is still modeled if particle information is included on the SO
pathway, and wet deposition is still modeled if scavenging coefficients are included on
the SO pathway. The no depletion option may be acceptable if deposition is weak, and it
will result in an overestimate of both concentrations and deposition. When DRYDPLT
and/or WETDPLT are included, particle mass is removed from the plume as it is deposited
on the surface, thereby conserving mass. However, the additional calculations required
for dry depletion corrections will result in significantly longer execution times for
the model, since the model must integrate along the plume path between the source and
receptor. The amount of increase in execution time will vary depending on source
characteristics and the terrain grid option used, but could be a factor of 10 or more
for typical applications.
The missing data processing routines, that are included in the ISC Short Term model
as a non-regulatory option, allow the model to handle missing meteorological data in the
processing of short term averages. With this option selected, the model treats missing
meteorological data in the same way as the calms processing routine, i.e., it sets the
concentration (or deposition) values to zero for that hour, and calculates the short
term averages according to EPA's calms policy. Calms and missing values are tracked
separately for the purpose of flagging the short term averages. An average that
includes a calm hour is flagged with a 'c', an average that includes a missing hour is
flagged with an 'm', and an average that includes both calm and missing hours is flagged
with a 'b'. If missing meteorological data are encountered without the missing data
processing option, then the model will continue to read through and check the
meteorological data, but will not perform any dispersion calculations.
3-8
-------
3.2.3 Averaging Time Options
The averaging periods for both the Short Term and Long Term models are selected
using the AVERTIME keyword. Since the averaging period options are different between
the Short Term and Long Term models, the syntax for the AVERTIME keyword is somewhat
different.
3.2.3.1 Short Term Model Options.
The syntax and type of the Short Term AVERTIME keyword are summarized below:
Syntax: CO AVERTIME Timel TimeZ TimeS Time4 MONTH PERIOD
or
ANNUAL
Type: Mandatory, Non-repeatabl e
where the parameters Timel . . . Time4 refer to the user-specified short term averaging
periods of 1, 2, 3, 4, 6, 8, 12, or 24 hours, the secondary keyword MONTH refers to
monthly averages (for calendar months), the secondary keyword PERIOD refers to the
average for the entire data period, and the secondary keyword ANNUAL refers to an annual
average. Any of the short term averaging periods listed above may be selected for a
given run, up to the maximum number of short term averages set in the computer code by
the parameter NAVE. The initial values for NAVE are given in Sections 2.3 and 4.2.2.
The monthly averages are treated as short term averages, and selection of the MONTH
average counts toward the limit of NAVE. Since the monthly averages are treated as
short term averages, the user can select appropriate output options, such as the second
3-9
-------
highest values by receptor, on the Output pathway. The user may specify either the
PERIOD keyword or the ANNUAL keyword, but not both. For concentration calculations, the
PERIOD and ANNUAL keywords produce the same results. They both may be used to calculate
the annual average for a full year of meteorological data, or to calculate the period
average for a period other than a year. For deposition calculations, the PERIOD keyword
will provide a total deposition flux for the full period of meteorological data that is
modeled in units of g/m2, including multiple-year data files, whereas the ANNUAL keyword
will provide an annualized rate of the deposition flux in units of g/m2/yr. For
meteorological periods of less than a year, the ANNUAL deposition rate is determined by
dividing by the length of the period in years. For meteorological periods of longer
than a year, the model will assume that full years of data are provided and divide by
the number of years, rounded to the nearest whole number. The treatment of short term
averages with multiple-year data files is comparable to their treatment when the CO
MULTYEAR option is used (see Section 3.2.11).
The location of the PERIOD or ANNUAL keyword in the parameter list is not critical.
The order of the short term averaging periods (including MONTH) is also not critical,
although it does control the order of the averaging period result tables in the main
output file. Generally, it is recommended that the short term averaging periods be
input in increasing order, unless there is a clear advantage in doing otherwise.
3.2.3.2 Long Term Model Options.
The syntax and type of the Long Term AVERTIME keyword are summarized below:
3-10
-------
Syntax :
Type:
CO AVERTIME JAN FEB MAR APR MAY JUN JUL AUG
WINTER SPRING SUMMER FALL
QUART1 QUARTZ QUARTS QUART4
MONTH SEASON QUARTR ANNUAL
PERIOD
Mandatory, Non-repeatabl e
SEP OCT NOV DEC
where all of the parameters are secondary keywords that relate to an averaging period or
periods associated with a single STAR data summary or a group of STAR summaries. The
keywords for individual months, seasons and quarters are fairly self-explanatory. If
the secondary keyword of SEASON is used, then it is assumed that all four seasons are
present in the STAR data file, and averages are calculated for each season. Similarly,
if the keyword MONTH or QUARTR is used, then the model assumes that all twelve months or
all four quarters are present in the STAR data file, and averages are calculated for
each averaging period. The MONTH and SEASON keywords or the MONTH and QUARTR keywords
can also be used together in the same run. However, seasonal STAR summaries and
quarterly STAR summaries cannot be used together in the same run, since the seasons and
quarters occupy the same locations in data storage. It is assumed that the STAR
summaries for the individual seasons, months or quarters are in the order listed in
above. Thus, the following two cards produce the same result:
3-11
-------
CO AVERTIME WINTER SPRING SUMMER FALL
and
CO AVERTIME SEASON
The ANNUAL secondary keyword indicates that averages are to be calculated for an
annual STAR summary. This differs from the PERIOD secondary keyword, which refers to an
average calculated for all STAR summaries included in the data file. The PERIOD keyword
may be used to calculate the annual average from a data file consisting of STAR
summaries for each of the four seasons or for each of the twelve months. Thus, the
ANNUAL and PERIOD keywords cannot both be present on the AVERTIME card. The PERIOD
average cannot be used when monthly STARs are included with seasonal or quarterly STARs
in the same data file.
The following card can be used to calculate the averages for each of the four
seasons and for the annual period from a data file consisting of five STAR summaries,
one for each season and one for the annual period:
CO AVERTIME SEASON ANNUAL
3-12
-------
whereas the following card will calculate the averages for each of the four seasons, and
will then rewind the meteorology file and calculate the averages for the annual period
from the four seasonal STAR summaries:
CO AVERTIME SEASON PERIOD
The AVERTIME keyword works in conjunction with the STARDATA keyword on the ME
pathway to control which averaging periods are calculated. Both of these keywords
recognize the same set of secondary keywords. The CO AVERTIME card defines which
averaging periods are to be calculated, and is a mandatory keyword. The ME STARDATA
card defines which STAR summaries are included in the data file. The STARDATA keyword
is optional, unless the AVERTIME card includes only the PERIOD average, in which case
the STARDATA keyword is mandatory in order to define which STAR summaries are included
in the period average. If the ME STARDATA keyword is omitted, then the ISCLT model
assumes that the meteorological data file contains only the STAR summaries identified on
the CO AVERTIME card.
3.2.4 Specifying the Pollutant Type
The POLLUTID keyword is used to identify the type of pollutant being modeled for a
particular run. The syntax, type, and order of the POLLUTID keyword are summarized
below:
Syntax; CO POLLUTID Poll Lit
3-13
-------
Type:
Mandatory, Non-repeatable
Order:
Must follow MODELOPT and precede HALFLIFE and DCAYCOEF
where the Pollut parameter may be name of up to eight characters. Examples include S02,
NOX, CO, PM10, TSP, and OTHER. The only choices that currently have any impact on the
results are the selection of S02 in conjunction with URBAN dispersion and the regulatory
default option, and the selection of PM10 (or PM-10) with the multi-year option for
generating the high-sixth-high in five years. For the urban S02 default case, the model
uses a half life of 4 hours for exponential decay of the S02.
3.2.5 Modeling With Exponential Decay
The models provide the option to use exponential decay of the pollutant being
modeled. Two keywords are available for this purpose, the HALFLIFE and DCAYCOEF
keywords. The syntax, type, and order of these keywords are summarized below:
Syntax: CO HALFLIFE Haflif
CO DCAYCOEF Decay
Type:
Optional, Non-repeatable
Order: Must follow MODELOPT and POLLUTID
where the Haflif parameter is used to specify the half life for exponential decay in
seconds, and the parameter Decay is used to specify the decay coefficient in units of
s"1. The relationship between these parameters is DECAY = 0.693/HAFLIF.
3-14
-------
Only one of these keywords may be specified in a given run. If more than one is
encountered, a non-fatal warning message is generated and the first specification is
used in the modeling. Also, since the regulatory default option includes a half life of
4 hours for exponential decay of S02 in urban settings, any HALFLIFE or DCAYCOEF input
conflicting with that option will be overridden if the DFAULT option is selected on the
CO MODELOPT card.
3.2.6 Options for Elevated Terrain
Two optional keywords are available on the Control pathway to control the receptor
options for modeling elevated terrain - the TERRHGTS and ELEVUNIT keywords.
The TERRHGTS keyword controls whether the model assumes flat or elevated terrain.
For elevated terrain, the terrain height should be specified for each receptor. The
syntax and type of the TERRHGTS keyword are summarized below:
Syntax: CO TERRHGTS FLAT or ELEV
Type: Optional, Non-repeatable
where the FLAT secondary keyword forces flat terrain calculations to be used throughout,
regardless of any terrain heights that may be entered on the Receptor pathway. Any
terrain heights that are entered on the Receptor pathway are ignored if FLAT terrain is
specified, and a non-fatal warning message is generated. The ELEV secondary keyword
indicates that terrain heights are allowed/expected on the Receptor pathway. The
default terrian height of 0.0 meters is used if no value is given. For terrain above
the release height (i.e., complex terrain), the models automatically truncate ("chop")
3-15
-------
the terrain to the physical release height(s) when modeling impacts at those receptors
using the simple terrain (ISC) algorithm. Terrain above the release height is not
truncated when the COMPLEXl algorithm is used in ISCST. The models assume flat terrain
as the default if no TERRHGTS keyword is present in the input runstream.
The ELEVUNIT keyword for the CO pathway is obsolescent. It has been replaced by
ELEVUNIT keywords on the SO, RE and TG pathways. The new RE ELEVUNIT card is equivalent
to the CO ELEVUNIT card, and should be used in its place. For compatibility with
existing input files, the ISC models will process the CO ELEVUNIT keyword in the same
way as done by the previous version of the models, but will write a warning message to
indicate that it is obsolescent. The CO ELEVUNIT keyword specifies the units for
terrain elevation data included in the RE pathway. The syntax and type of the ELEVUNIT
keyword are summarized below:
Syntax: CO ELEVUNIT METERS or FEET
Type: Optional, Non-repeatable
The default units for terrain elevation data is meters.
3.2.7 Flagpole Receptor Height Option
The FLAGPOLE keyword specifies that receptor heights above local ground level (i.e.
flagpole receptors) are allowed on the REceptor pathway. The FLAGPOLE keyword may also
be used to specify a default flagpole receptor height other than 0.0 meters. The syntax
and type of the FLAGPOLE keyword are summarized below:
3-16
-------
Syntax: CO FLAGPOLE (Flagdf)
Type: Optional, Non-repeatable
where Flagdf is an optional parameter to specify a default flagpole receptor height. If
no parameter is provided, then a default flagpole receptor height of 0.0 meters is used.
Any flagpole receptor heights that are entered on the Receptor pathway are ignored if
the FLAGPOLE keyword is not present on the Control pathway, and a non-fatal warning
message is generated.
3.2.8 To Run or Not to Run - That is the Question
Because of the improved error handling and the "defensive programming" that has
been employed in the design of the ISC model, it is intended that the model will read
through all of the inputs in the runstream file regardless of any errors or warnings
that may be encountered. If a fatal error occurs in processing of the runstream
information, then further model calculations will be aborted. Otherwise, the model will
attempt to run. Because of the great many options available in the ISC models, and the
potential for wasted resources if a large run is performed with some incorrect input
data, the RUNORNOT keyword has been included on the Control pathway to allow the user to
specify whether to RUN the model and perform all of the calculations, or NOT to run and
only process the input runstream data and summarize the setup information. The syntax
and type of the RUNORNOT keyword are summarized below:
Syntax: CO RUNORNOT RUN or NOT
Type: Mandatory, Non-repeatable
3-17
-------
3.2.9 Generating an Input File for the Short Term EVENT Model (ISCEV)
The Short Term model consists of two executable files - one is used for routine
processing (ISCST) and the other is used for EVENT processing (ISCEV). The EVENTFIL
keyword controls whether or not the ISCST model will generate an input file for use with
the EVENT model, and applies only to the ISCST model. The syntax and type of the
EVENTFIL keyword are summarized below:
Syntax: CO EVENTFIL (Evfile) (Evopt)
Type: Optional, Non-repeatable
where the optional Evfile parameter specifies the name of the EVENT input file to be
generated (up to 40 characters), and the optional parameter, Evopt, specifies the level
of detail to be used in the EVENT output file. Valid inputs for the Evopt parameter are
the secondary keywords of SOCONT and DETAIL (see the EVENTOUT keyword on the Output
pathway, Section 3.7.2). The default filename used if no parameters are specified is
PASSTWO.INP, and the default for the level of detail is DETAIL. If only one parameter
is present, then it is taken to be the Evfile, and the default will be used for Evopt.
The primary difference between routine ISCST and EVENT processing is in the
treatment of source group contributions. The ISCST model treats the source groups
independently. The EVENT model is designed to provide source contributions to
particular events, such as the design concentrations determined from ISCST, or user
specified events. The user may specify the "events" to process using the EVent pathway,
which lists specific combinations of receptor location, source group, and averaging
3-18
-------
period. By specifying the EVENTFIL keyword, an input runstream file will be generated
that can be used directly with the EVENT model. The events included in the generated
EVENT model input file are the design concentrations defined by the RECTABLE keyword and
the threshold violations identified by the MAXIFILE keyword on the OU pathway. If more
than one output type (CONG, DEPOS, DDEP, and/or WDEP) is selected for the ISCST model,
only the events associated with the first output type, in the order stated above, will
be included in the EVENT model input file. This is because the EVENT model can only
process one type of output at a time.
3.2.10 The Model Re-start Capability
The ISCST model has an optional capability to store intermediate results into an
unformatted file, so that the model run can be continued later in case of a power
failure or a user interrupt. This re-start option is controlled by the SAVEFILE and
INITFILE keywords on the CO pathway. The syntax and type of these keywords are
summarized below:
Syntax: CO SAVEFILE (Savfil) (Dayinc) (SavflZ)
CO INITFILE (Inifil)
Type: Optional, Non-repeatable
The SAVEFILE keyword instructs the model to save the intermediate results to a
file, and controls the save options. All three parameters for this keyword are optional.
If the user specifies only the Savfil parameter, then the intermediate results are saved
to the same file (and overwritten) each time. If the user specifies both the Savfil and
the Savfl2 parameters, then the model alternates between the two files for storing
3-19
-------
intermediate results. The latter approach requires additional disk space to handle two
storage files. However, selecting two files avoids the potential problem that the power
failure or interrupt might occur while the temporary file is open and the intermediate
results are being copied to it. In such a case, the temporary results file would be
lost.
The optional Dayinc parameter allows the user to specify the number of days between
successive dumps. The default is to dump values at the end of each day, i.e., Dayinc =
1. For larger modeling runs, where the SAVEFILE option is most useful, the additional
execution time required to implement this option is very small compared to the total
runtime. To be most effective, it is recommended that results be saved at least every 5
days.
If no parameters are specified for the SAVEFILE keyword, then the model will store
intermediate results at the end of each day using a default filename of SAVE.FIL.
The INITFILE keyword works in conjunction with the SAVEFILE keyword, and instructs
the model to initialize the results arrays from a previously saved file. The optional
parameter, Inifil, identifies the unformatted file of intermediate results to use for
initializing the model. If no Inifil parameter is specified, then the model assumes the
default filename of SAVE.FIL. If the file doesn't exist or if there are any errors
encountered in opening the file, then a fatal error message is generated and processing
is halted.
Note: It is important to note that if both the SAVEFILE and INITFILE keywords are
used in a the same model run, then different filenames must be specified for the Savfil
3-20
-------
and Inifil parameters. Otherwise, the model will encounter an error in opening the
files, and further processing will be halted.
3.2.11 Performing Multiple Year Analyses for PM-10
The MULTYEAR keyword on the CO pathway provides an option for the user to perform a
multiple year analysis such as would be needed to determine the "high-sixth-high in five
years" design value for determining PM-10 impacts. In the past, such modeling would
require extensive postprocessing of ISCST binary concentration files. Since the
multiple year option makes use of the model re-start capabilities described in the
previous section, the MULTYEAR keyword is not compatible with the SAVEFILE or INITFILE
keywords. The model will generate a fatal error message if the user attempts to
exercise both options in a single run. The syntax and type of this keyword is
summarized below:
Syntax: CO MULTYEAR Savfil (inifil)
Type: Optional, Non-repeatable
where the Savfil parameter specifies the filename for saving the results arrays at the
end of each year of processing, and the Inifil parameter specifies the filename to use
for initializing the results arrays at the beginning of the current year. The Inifil
parameter is optional, and should be left blank for the first year in the multi-year
series of runs.
The MULTYEAR option works by accumulating the high short term average results from
year to year through the mechanism of the re-start save file. The model may be setup to
3-21
-------
run in a batch file with several years of meteorological data, and at the end of each
year of processing, the short term average results reflect the cumulative high values
for the years that have been processed. The PERIOD average results are given for only
the current year, but the model carries the highest PERIOD values from year to year and
includes the cumulative highest PERIOD averages in the summary table at the end of the
run.
When setting up a batch file to perform a multiple year analysis, the user would
first create an input runstream file for the first year with all of the applicable
modeling options, the source inventory data, the receptor locations, the meteorology
options for the first year and the output file options. To obtain the PM-10 design
value, be sure to include the SIXTH highest value on the OU RECTABLE card (see Section
3.8.1). For the CO MULTYEAR card for the first year, the user would only specify the
Savfil parameter, and may use a card such as:
CO MULTYEAR YEAR1.SAV
For the subsequent years, the user could copy the input file created for Year-1, and
edit the files to change the year parameters and meteorology filename on the ME pathway
(and possibly in the title information), and edit the MULTYEAR cards. For the
subsequent years, both the Savfil and Inifil parameters must be specified, with the
3-22
-------
Savfil for Year-1 becoming the Inifil for Year-2, and so on. The MULTYEAR cards (one
for each ISCST run) might look like this:
CO MULTYEAR YEAR1.SAV (First year)
CO MULTYEAR YEAR2.SAV YEAR1.SAV (Second year)
CO MULTYEAR YEARS.SAV YEAR2.SAV (Third year)
CO MULTYEAR YEAR4.SAV YEARS.SAV (Fourth year)
CO MULTYEAR YEAR5.SAV YEAR4.SAV (Sixth year)
The MULTYEAR keyword option is separate from the ability of the ISCST model to
process a multiple-year meterological data file in a single model run. The latter
capability is primarily intended for applications of the model to long term risk
assessments where the average impacts over a long time period are of concern rather than
the maximum annual average determined from five individual years. The use of the ISCST
model with multiple-year data sets is discussed in more detail in Section 3.5.1.1.
3.2.12 Detailed Error Listing File
The ERRORFIL keyword on the CO pathway allows the user to request a detailed
listing file of all the messages generated by the model. This includes the error and
warning messages that are listed as part of the message summaries provided in the main
output file, and also any informational messages (such as occurrences of calm winds) and
quality assurance messages that are generated. The syntax and type of the ERRORFIL
keyword are summarized below:
3-23
-------
Syntax: CO ERRORFIL (Errfil) (DEBUG)
Type: Optional, Non-repeatable
where the Errfil parameter is the name of the detailed message file, and the DEBUG
secondary keyword implements an option to obtain detailed output results including plume
heights, sigmas, etc., for each hour calculated for debugging purposes. Note: The DEBUG
option generates very large files and should be used with CAUTION! If the optional
Errfil parameter is left blank, then the model will use a default filename of
ERRORS.LST. A complete description of the error and other types of messages generated
by the models is provided in Appendix E.
3.3 SOURCE PATHWAY INPUTS AND OPTIONS
The SOurce pathway contains the keywords that define the source information for a
particular model run. The model currently handles four source types, identified as
point, volume, area or open pit sources. The input parameters vary depending on the
source type. For point sources, the user can also identify building dimensions for
nearby structure that cause aerodynamic downwash influences on the source. The user can
also identify groups of sources for which the models will combine the results. With the
exception of the variable emission rate options on the EMISFACT keyword, all of the SO
pathway inputs are identical between the Short Term and Long Term models.
The LOCATION keyword, which identifies the source type and location, must be the
first card entered for each source. The only other requirement for order of the
keywords is that the SRCGROUP keyword must be the last keyword before the SO FINISHED
card. The user may group all of the LOCATION cards together, then group the source
3-24
-------
parameter cards together, or they may want to group all input cards for a particular
source together as was done in the old ISC input file. All sources are given a source
ID by the user, which is used to link the source parameter inputs to the correct source
or sources. The source ID can be any alphanumeric string of up to eight characters.
The number of sources allowed in a given run is controlled by a Fortran PARAMETER
statement in the computer code. The initial storage limits for each of the models is
given in Section 2.3, which discusses storage allocation in general. These limits can
easily be modified by the user and the code recompiled to accommodate different user
needs.
3.3.1 Identifying Source Types and Locations
The LOCATION keyword is used to identify the source type and the location of each
source to be modeled. The LOCATION card must be the first card entered for each source
since it identifies the source type, and dictates which parameters are needed and/or
accepted. The syntax, type and order of the LOCATION keyword are summarized below:
Syntax: SO LOCATION Srcid Srctyp Xs Ys (Zs)
Type: Mandatory, Repeatable
Order: Must be first card for each source input
where the Srcid parameter is the alphanumeric source ID defined by the user (up to eight
characters), Srctyp is the source type, which is identified by one of the secondary
keywords - POINT, VOLUME, AREA, or OPENPIT - and Xs, Ys, and Zs are the x, y, and z
coordinates of the source location in meters. Note that the source elevation, Zs, is an
3-25
-------
optional parameter. If the source elevation is omitted, it will be given a default
value of 0.0, but the source elevation is only used if the CO TERRHGTS ELEV option is
selected. While the default units of Zs are meters, the user may also specify source
elevations to be in feet by adding the SO ELEVUNIT FEET card immediately following the
SO STARTING card. The x (east-west) and y (north-south) coordinates are for the center
of the source for POINT and VOLUME sources, and are for the southwest corner of the
source for AREA and OPENPIT sources. The source coordinates may be input as Universal
Transverse Mercator (UTM) coordinates, or may be referenced to a user-defined origin.
Certain types of line sources can be handled in ISC using either a string of volume
sources, or as an elongated area source. The volume source algorithms are most
applicable to line sources with some initial plume depth, such as conveyor belts and
rail lines. Section 1.2.2 of Volume II provides technical information on how to model a
line source with multiple volume sources. The use of the ISC area source algorithm for
elongated rectangles would be most applicable to near ground level line sources, such as
a viaduct. Also, as shown in Section 1.2.3 of Volume II, irregularly shaped areas may
be modeled with the ISC Models by subdividing the area.
The source ID entered on the LOCATION card identifies that source for the remainder
of the SO pathway inputs. Since the model accepts alphanumeric strings of up to eight
characters for the source ID, the sources can be identified with descriptive names, such
as STACK1, STACK2, BOILER3, SLAGPILE, etc. This may also be useful if line sources or
irregularly-shaped area sources are being modeled as multiple volume or areas, as
discussed above. Since they are part of the same physical source, they can be given
names that will identify them as being related, such as LINE1A, LINE1B, LINE1C, etc.
3-26
-------
3.3.2 Specifying Source Release Parameters
The main source parameters are input on the SRCPARAM card, which is a mandatory
keyword for each source being modeled. Since the input parameters vary depending on the
source type, the four source types handled by the ISC models (POINT, VOLUME, AREA and
OPENPIT) are discussed separately.
3.3.2.1 POINT Source Inputs.
The ISC POINT source algorithms are used to model releases from stacks and isolated
vents, as well as other kinds of sources. The syntax, type and order for the SRCPARAM
card for POINT sources are summarized below:
Syntax: SO SRCPARAM Srcid Ptemis Stkhgt Stktmp Stkvel Stkdia
Type: Mandatory, Repeatable
Order: Must follow the LOCATION card for each source input
where the Srcid parameter is the same source ID that was entered on the LOCATION card
for a particular source, and the other parameters are as follows:
3-:
-------
Ptemis - point emission rate in g/s,
Stkhgt - release height above ground in meters,
Stktmp - stack gas exit temperature in degrees K,
Stkvel - stack gas exit velocity in m/s, and
Stkdia - stack inside diameter in meters.
It should be noted that the same emission rate is used for both concentration and
deposition calculations in the ISC models. An example of a valid SRCPARAM input card for
a point source is given below:
SO SRCPARAM STACK1 16.71 35.0 444.0 22.7 2.74
where the source ID is STACK1, the emission rate is 16.71 g/s, the release height is
35.0 m, the exit temperature is 444.0 K, the exit velocity is 22.7 m/s, and the inside
stack diameter is 2.74 m. All of the parameters must be present on the input card.
Since the ISC models use direction-specific building dimensions for all sources
subject to building downwash, there are no building parameters entered on the SRCPARAM
card. Building dimensions are entered on the BUILDHGT and BUILDWID cards described below
in Section 3.3.3.
3-28
-------
3.3.2.2 VOLUME Source Inputs.
The ISC VOLUME source algorithms are used to model releases from a variety of
industrial sources, such as building roof monitors, multiple vents, and conveyor belts
The syntax, type and order for the SRCPARAM card for VOLUME sources are summarized
below:
Syntax: SO SRCPARAM Srcid Vlemis Relhgt Syinit Szinit
Type: Mandatory, Repeatable
Order: Must follow the LOCATION card for each source input
where the Srcid parameter is the same source ID that was entered on the LOCATION card
for a particular source, and the other parameters are as follows:
Vlemis - volume emission rate in g/s,
Relhgt - release height (center of volume) above ground, in meters,
Syinit - initial lateral dimension of the volume in meters, and
Szinit - initial vertical dimension of the volume in meters.
It should be noted that the same emission rate is used for both concentration and
deposition calculations in the ISC models. The following table, which is explained in
more detail in Section 1.2.2 of Volume II of the User's Guide, summarizes the suggested
procedures to be used for estimating the initial lateral and vertical dimensions for
various types of volume and line sources.
3-29
-------
TABLE 3 -1.
SUMMARY OF SUGGESTED PROCEDURES FOR ESTIMATING
INITIAL LATERAL DIMENSIONS • »0 AND
INITIAL VERTICAL DIMENSIONS • »0 FOR VOLUME AND LINE SOURCES
Procedure for Obtaining
Type of Source Initial Dimension
(a) Initial Lateral Dimensions (**0)
Single Volume Source • *0 = length of side divided by 4.3
Line Source Represented by Adjacent Volume **0 = length of side divided by 2.15
Sources (see Figure 1-8(a) in Volume II)
Line Source Represented by Separated Volume • *0 = center to center distance divided
Sources (see Figure 1-8(b) in Volume II) by 2.15
(b) Initial Vertical Dimensions (**0)
Surface-Based Source (he ~ 0) • *0 = vertical dimension of source
divided by 2.15
Elevated Source (he > 0) on or Adjacent to • *0 = building height divided by 2.15
a Building
Elevated Source (he > 0) not on or Adjacent • *0 = vertical dimension of source
to a Building divided by 4.3
3.3.2.3 AREA Source Inputs
The ISC AREA source algorithms are used to model low level or ground level releases
with no plume rise (e.g., storage piles, slag dumps, and lagoons). The ISC models use a
numerical integration approach for modeling impacts from area sources. The ISC models
accept rectangular areas that may also have a rotation angle specified relative to a
3-30
-------
north-south orientation. The rotation angle is specified relative to the vertex used to
define the source location on the SO LOCATION card (e.g., the southwest corner). The
syntax, type and order for the SRCPARAM card for AREA sources are summarized below:
Syntax: SO SRCPARAM Srcid Aremis Relhgt Xinit (Yinit) (Angle)
(Szinit)
Type: Mandatory, Repeatable
Order: Must follow the LOCATION card for each source input
where the Srcid parameter is the same source ID that was entered on the LOCATION card
for a particular source, and the other parameters are as follows:
Aremis - area emission rate in g/(s-m2),
Relhgt - release height above ground in meters,
Xinit - length of X side of the area (in the east-west direction if Angle is 0
degrees) in meters,
Yinit - length of Y side of the area (in the north-south direction if Angle is 0
degrees) in meters (optional),
Angle - orientation angle for the rectangular area in degrees from North, measured
positive in the clockwise direction (optional), and
Szinit - initial vertical dimension of the area source plume in meters (optional).
The same emission rate is used for both concentration and deposition calculations in the
ISC models. It should also be noted that the emission rate for the area source is an
emission rate per unit area, which is different from the point and volume source
emission rates, which are total emissions for the source.
3-31
-------
If the optional Yinit parameter is omitted, then the model assumes that the area is
a square, i.e., Yinit = Xinit. If the optional Angle parameter is omitted, then the
model assumes that the area is oriented in the north-south and east-west directions,
i.e., Angle = 0.0. If the Angle parameter is input, and the value does not equal 0.0,
then the model will rotate the area clockwise around the vertex defined on the SO
LOCATION card for this source. Figure 3-1 illustrates the relationship between the
Xinit, Yinit, and Angle parameters and the source location, (Xs,Ys), for a rotated
rectangle. The Xinit dimension is measured from the side of the area that is
counterclockwise along the perimeter from the vertex defined by (Xs,Ys), while the Yinit
dimension is measured from the side of the area that is clockwise from (Xs,Ys). The
Angle parameter is measured as the orientation relative to North of the side that is
clockwise from (Xs,Ys), i.e. the side with length Yinit. The Angle parameter may be
positive (for clockwise rotation) or negative (for counterclockwise rotation), and a
warning message is generated if the absolute value of Angle is greater than 180 degrees.
The selection of the vertex to use for the source location is not critical, as long as
the relationship described above for the Xinit, Yinit, and Angle parameters is
maintained. However, for consistency with the previous versions of ISCST and ISCLT, it
is recommended that the user select the vertex that occurs in the southwest quadrant as
the location of the area source.
3-32
-------
Y
t
(Xs.Ys)
(Xs.Ys)
X
0
3-33
-------
FIGURE 3-1. RELATIONSHIP OF AREA SOURCE PARAMETERS FOR ROTATED RECTANGLE
3-34
-------
By making the Yinit and Angle parameters optional, the area source input data for
the previous versions of ISC that were limited to square areas with a north-south
orientation can still be used with the new algorithm. The aspect ratio (i.e.,
length/width) for area sources should be less than 10 to 1. If this is exceeded, then
the area should be subdivided to achieve a 10 to 1 aspect ratio (or less) for all
subareas.
The optional Szinit parameter may be used to specify an initial vertical dimension
to the area source plume, similar to the use of the Szinit parameter for volume sources.
This parameter may be important when the area source algorithm is used to model
mechanically generated emission sources, such as mobile sources. In these cases, the
emissions may be turbulently mixed near the source by the process that is generating the
emissions, and therefore occupy some initial depth. For more passive area source
emissions, such as evaporation or wind erosion, the Szinit parameter may be omitted,
which is equivalent to using an initial sigma-z of zero.
An example of a valid SRCPARAM input card for a rectangular area source is given
below:
SO SRCPARAM SLAGPILE 0.0015 5.0 50.0 100.0 30.0
where the source ID is SLAGPILE, the emission rate is 0.0015 g/(s-m2), the release
height is 5.0 m, the X-dimension is 50.0 m, the Y-dimension is 100.0 m, and the
orientation angle is 30.0 degrees clockwise from North. Note that if the orientation
3-35
-------
angle is zero, the Y-dimension is North and the X-dimension is east, which is the
standard convention.
In order to model irregularly-shaped areas, the user may have to subdivide the area
into smaller areas of varying shapes, sizes, and orientations. However, with the
ability to specify rectangular shapes and orientation angles, the user has considerable
flexibility in subdividing the area. Since the numerical integration algorithm can
handle elongated areas with aspect ratios of up to 10 to 1, the ISC area source
algorithm may be useful for modeling certain types of line sources. There are no
restrictions on the placement of receptors relative to area sources for the ISC models.
Receptors may be placed within the area and at the edge of an area. The ISC models will
integrate over the portion of the area that is upwind of the receptor. However, since
the numerical integration is not performed for portions of the area that are closer than
1.0 meter upwind of the receptor, caution should be used when placing receptors within
or adjacent to areas that are less than a few meters wide. More technical information
about the application of the ISC area source algorithm is provided in Sections 1.2.3 and
2.2.3 of Volume II of the User's Guide.
3.3.2.4 OPENPIT Source Inputs
The ISC OPENPIT source algorithms are used to model particulate emissions from open
pits, such as surface coal mines and rock quarries. The OPENPIT algorithm uses an
effective area for modeling pit emissions, based on meteorological conditions, and then
utilizes the numerical integration area source algorithm to model the impact of
emissions from the effective area sources. The ISC models accept rectangular pits with
an optional rotation angle specified relative to a north-south orientation. The
3-36
-------
rotation angle is specified relative to the vertex used to define the source location on
the SO LOCATION card (e.g., the southwest corner). The syntax, type and order for the
SRCPARAM card for OPENPIT sources are summarized below:
Syntax: SO SRCPARAM Srcid Opemis Relhgt Xinit Yinit Pitvol (Angle)
Type:
Mandatory, Repeatable
Order: Must follow the LOCATION card for each source input
where the Srcid parameter is the same source ID that was entered on the LOCATION card
for a particular source, and the other parameters are as follows:
Opemis -open pit emission rate in g/(s-m2),
Relhgt -average release height above the base of the pit in meters,
Xinit -length of X side of the open pit (in the east-west direction if Angle is 0
degrees) in meters,
Yinit -length of Y side of the open pit (in the north-south direction if Angle is
0 degrees) in meters,
Pitvol -volume of open pit in cubic meters, and
Angle -orientation angle for the rectangular open pit in degrees from North,
measured positive in the clockwise direction (optional).
The same emission rate is used for both concentration and deposition calculations in the
ISC models. It should also be noted that the emission rate for the open pit source is
an emission rate per unit area, which is different from the point and volume source
emission rates, which are total emissions for the source. The Relhgt parameter cannot
3-37
-------
exceed the effective depth of the pit, which is calculated by the model based on the
length, width and volume of the pit. A Relhgt of 0.0 indicates emissions that are
released from the base of the pit.
If the optional Angle parameter is input, and the value does not equal 0.0, then
the model will rotate the open pit clockwise around the vertex defined on the SO
LOCATION card for this source. The relationship between the Xinit, Yinit, and Angle
parameters and the source location, (Xs,Ys), for a rotated pit is the same as that shown
in Figure 3-1 for area sources. The Xinit dimension is measured from the side of the
area that is counterclockwise along the perimeter from the vertex defined by (Xs,Ys),
while the Yinit dimension is measured from the side of the open pit that is clockwise
along the perimeter from (Xs,Ys). Unlike the area source inputs, the Yinit parameter is
not optional for open pit sources. The Angle parameter is measured as the orientation
relative to North of the side that is clockwise from (Xs,Ys), i.e. the side with length
Yinit. The Angle parameter may be positive (for clockwise rotation) or negative (for
counterclockwise rotation), and a warning message is generated if the absolute value of
Angle is greater than 180 degrees. The selection of the vertex to use for the source
location is not critical, as long as the relationship described above for the Xinit,
Yinit, and Angle parameters is maintained.
The aspect ratio (i.e., length/width) of open pit sources should be less than 10 to
1. However, since the pit algorithm generates an effective area for modeling emissions
from the pit, and the size, shape and location of the effective area is a function of
wind direction, an open pit cannot be subdivided into a series of smaller sources.
Aspect ratios of greater than 10 to 1 will be flagged by a warning message in the output
file, and processing will continue. Since open pit sources cannot be subdivided, the
3-38
-------
user should characterize irregularly-shaped pit areas by a rectangular shape of equal
area. Receptors should not be located within the boundaries of the pit; concentration
and/or deposition at such receptors will be set to zero. Such receptors will be
identified during model setup and will be flagged in the summary of inputs.
An example of a valid SRCPARAM input card for an open pit source is given below:
SO SRCPARAM NORTHPIT 1.15E-4 0.0 150.0 500.0 3.75E+6 30.0
where the source ID is NORTHPIT, the emission rate is 1.15E-4 g/(s-m2), the release
height is 0.0 m, the X-dimension is 150.0 m, the Y-dimension is 500.0 m, the pit volume
is 3.75E+6 cubic meters (corresponding to an effective pit depth of about 50 meters) and
the orientation angle is 30.0 degrees clockwise from North.
Since the OPENPIT algorithm is applicable for particulate emissions, the particle
categories for an open pit source must be defined using the PARTDIAM, MASSFRAX, and
PARTDENS keywords on the SO pathway.
3.3.3 Specifying Building Downwash Information
As noted above, the ISC models include algorithms to model the effects of buildings
downwash on emissions from nearby or adjacent point sources. The building downwash
algorithms do not apply to volume, area or open pit sources. For a technical
description of the building downwash algorithms, the user is referred to Volume II of
3-39
-------
the ISC User's Guide. The ISC models use direction-specific information for all
building downwash cases.
There are three keywords that are used to specify building downwash information,
BUILDHGT, BUILDWID, and LOWBOUND. The syntax, type and order for the BUILDHGT keyword,
used to input direction specific building heights, are summarized below:
Syntax: SO BUILDHGT Srcid (or Srcrng) Dsbh( i) , i=l ,36 (16 for LT)
Type: Optional, Repeatable
Order: Must follow the LOCATION card for each source input
where the Srcid parameter is the same source ID that was entered on the LOCATION card
for a particular source. The user also has the option of specifying a range of sources
(the Srcrng parameter) for which the building heights apply, instead of identifying a
single source. This is accomplished by two source ID character strings separated by a
dash, e.g., STACK1-STACK10. Since the model reads the source range as a single input
field there must not be any spaces between the source IDs. The model then places the
building heights that follow (the Dsbh(i) parameter) into the appropriate arrays for all
Srcid's that fall within that range, including STACK1 and STACK10.
When comparing a source ID to the range limits for a Srcrng parameter, the model
separates the source IDs into three parts: an initial alphabetical part, a numerical
part, and then the remainder of the string. Each part is then compared to the
corresponding parts of the source range, and all three parts must satisfy the respective
ranges in order for the source ID to be included. If there is no numeric part, then the
ID consists of only one alphabetical part. If the ID begins with a numeric character,
3-40
-------
then the initial aphabetical part defaults to a single blank. If there is no trailing
alphabetical part, then the third part also defaults to a single blank part. If the
trailing part consists of more than one alphabetical or numeric field, it is all lumped
into one character field. For example, the source ID 'STACK2' consists of the parts
'STACK' plus '2' plus a single trailing blank, ' '. By comparing the separate parts of
the source IDs, it can be seen that STACK2 falls between the range 'STACK1-STACK10.'
For a three-part example, it can also be seen that VENT1B falls within the range of
VENT1A-VENT1C. However, VENT2 does not fall within the range of VENT1A to VENT3B, since
the third part of VENT2 is a single blank, which does not fall within the range of A to
C. This is because a blank character will preceed a normal alphabetical character.
Normally, the source ranges will work as one would intuitively expect for simple source
names. Most importantly, for names that are made up entirely of numeric characters,
such as for old input files converted using STOLDNEW (see Appendix C), the source ranges
will be based simply on the relative numerical values. The user is strongly encouraged
to check the summary of model inputs to ensure that the source ranges were interpreted
as expected, and also to avoid using complex source names in ranges, such as
AA1B2C-AB3A3C. Since the order of keywords within the SO pathway is quite flexible, it
is also important to note that the building heights will only be applied to those
sources that have been defined previously in the input file.
Following the Srcid or the Srcrng parameter, the user inputs 36 direction-specific
building heights (Dsbh parameter) in meters for the Short Term model, beginning with the
10 degree flow vector (wind blowing toward 10 degrees from north), and incrementing by
10 degrees in a clockwise direction. For the Long Term model, the Dsbh parameter
consists of 16 direction-specific building heights beginning with the flow vector for
3-41
-------
the north sector, and proceeding clockwise to north-northwest. Some examples of
building height inputs are presented below:
SO BUILDHGT STACK1 34. 34. 34. 34. 34. 34. 34. 34. 34. 34. 34. 34.
SO BUILDHGT STACK1 34. 34. 34. 34. 34. 34. 34. 34. 34. 34. 34. 34.
SO BUILDHGT STACK1 34. 34. 34. 34. 34. 34. 34. 34. 34. 34. 34. 34.
SO BUILDHGT STACK1 36*34.0
SO BUILDHGT STACK1-STACK10 33*34.0 3*0.0
SO BUILDHGT STACK1 35.43 36.45 36.37 35.18 32.92 29.66 25.50 20.56
SO BUILDHGT STACK1 15.00 20.56 25.50 29.66 32.92 35.18 36.37 36.45
SO BUILDHGT STACK1 35.43 33.33 35.43 36.45 0.00 35.18 32.92 29.66
SO BUILDHGT STACK1 25.50 20.56 15.00 20.56 25.50 29.66 32.92 35.18
SO BUILDHGT STACK1 36.37 36.45 35.43 33.33
The first example illustrates the use of repeat cards if more than one card is needed to
input all of the values. The values are processed in the order in which they appear in
the input file, and are identified as being repeat cards by repeating the Srcid
parameter. The first and second examples produce identical results within the model.
The second one illustrates the use of a repeat value that can simplify numerical input
in some cases. The field "36*34.0" is interpreted by the model as "repeat the value
34.0 a total of 36 times." This is also used in the third example where the building
height is constant for directions of 10 degrees through 330 degrees, and then is set to
0.0 (e.g. the stack may be outside the region of downwash influence) for directions 340
through 360. The third example also uses a source range rather than a single source ID.
The last example illustrates building heights which vary by direction, and shows that
the number of values on each card need not be the same. For improved readability of the
3-42
-------
input file, the user may want to put the numerical inputs into "columns," but there are
no special rules regarding the spacing of the parameters on this keyword.
The BUILDWID keyword is used to input direction-specific building widths for
downwash analyses. The syntax for this keyword, which is very similar to the BUILDHGT
keyword, is summarized below, along with the type and order information:
Syntax: SO BUILDWID Srcid (or Srcrng) Dsbw( i) , i=l ,36 (16 for LT)
Type: Optional, Repeatable
Order: Must follow the LOCATION card for each source input
For a description of the Srcid and Srcrng parameters, and for a discussion and examples
of the numeric input options, refer to the BUILDHGT keyword above. The Dsbw(i)
parameter contains the direction-specific building widths, 36 for the Short Term model,
and 16 for the Long Term model. The directions proceed in a clockwise direction,
beginning with the 10 degree flow vector for the Short Term model and beginning with the
flow vector for the north sector for the Long Term model.
The LOWBOUND keyword is used to exercise the non-regulatory default option of
calculating "lower bound" concentration or deposition values for downwash sources
subject to enhanced lateral plume spread by super-squat buildings (width is more than
five times the height). The syntax, type and order of this keyword is summarized below:
3-43
-------
Syntax: SO LOWBOUND Srcid (or Srcrng) IdswakC i) , i=l ,36 (16 for LT)
Type: Optional, Repeatable
Order: Must follow the LOCATION card for each source input
where the Srcid and Srcrng parameters are described above for the BUILDHGT keyword, and
the Idswak(i) parameter is an array of lower bound wake option switches beginning with
the 10 degree flow vector and incrementing by 10 degrees clockwise for the Short Term
model and beginning with the flow vector for the north sector for the Long Term model.
A value of 0 means to use the upper bound (regulatory default) for that sector, and a
value of 1 means to use the lower bound for that sector. The use of repeat values is
permitted for inputting the Idswak array, e.g., a field of '36*1' indicates to use the
lower bound for all 36 sectors. Since this is a non-regulatory default option, if the
DFAULT option has been selected on the MODELOPT keyword (CO pathway), then any LOWBOUND
inputs will be ignored, and the model will calculate the upper bound estimates. The
model will generate a non-fatal warning message in such a case.
For a technical description of the "lower bound" option, the reader is referred to
Section 1.1.5.3 of Volume II.
3.3.4 Using Variable Emission Rates
The ISC models provide the option of specifying variable emission rate factors for
individual sources or for groups of sources. The factors may vary on different time
scales, such as by season, hour-of-day, etc. Since the Short Term and Long Term models
work on different averaging periods, the variable emission rate factors are somewhat
3-44
-------
different. Therefore the models are discussed separately. See Section 3.3.8 for ISCST.
3.3.4.1 Short Term Model Options.
The EMISFACT keyword provides the user the option of specifying variable emission
rate factors for sources modeled by the Short Term model. The syntax, type and order of
this keyword are summarized below:
Syntax: SO EMISFACT Srcid (or Srcrng) Qflag Qf act( i) , i=l ,n
Type: Optional, Repeatable
Order: Must follow the LOCATION card for each source input
where the Srcid parameter is the same source ID that was entered on the LOCATION card
for a particular source. The user also has the option of using the Srcrng parameter for
specifying a range of sources for which the emission rate factors apply, instead of
identifying a single source. This is accomplished by two source ID character strings
separated by a dash, e.g., STACK1-STACK10. The use of the Srcrng parameter is explained
in more detail in Section 3.3.3 above for the BUILDHGT keyword.
The parameter Qflag is the variable emission rate flag, and is one of the following
secondary keywords:
SEASON - emission rates vary seasonally (n=4),
MONTH - emission rates vary monthly (n=12),
HROFDY - emission rates vary by hour-of-day (n=24),
3-45
-------
STAR - emission rates vary by speed and stability
category (n=36), and
SEASHR - emission rates vary by season and hour-of-day
(n=96)
The Qfact array is the array of factors, where the number of factors is shown above for
each Qflag option. The EMISFACT card may be repeated as many times as necessary to
input all of the factors, and repeat values may be used for the numerical inputs. An
example of each of these options is presented below, with column headers to indicate the
order in which values are to be input.
**
so
•A- -A-
sn
•A- -A-
SO
l.(
•A- -A-
sn
•A- -A-
SO
**
SO
so
EMISFACT STACK1 SEASON
EMISFACT STACK1 MONTH
EMISFACT STACK1 HROFDY
)
EMISFACT STACK1 HROFDY
or, equi val ently :
EMISFACT STACK1 HROFDY
Stab. Cat. :
EMISFACT STACK1 STAR
EMISFACT STACK1 SEASHR
WINTER
0.50
JAN FEB M
0.1 0.2 0
1 2
0.0 0.0
13 14
1010
1-5
5*0.0 0
A
6*0.5 6*
enter 24
seasons
SPRING SUMMER FALL
0.50 1.00 0.75
AR APR MAY JUN JUL AUG SE
.3 0.4 0.5 0.5 0.5 0.6 0
3456789
0.0 0.0 0.0 0.5 1.0 1.0
15 16 17 18 19 20 21
101010050000(
6 7-17 18 19-24
.5 11*1.0 0.5 6*0.0
B C D E F
0.6 6*0.7 6*0.8 6*0.9
hourly scalars for each
(winter, spring, summer,
:P OCT NOV DEC
7 1.0 1.0 1.0
10 11 12
1.0 1.0 1.0
L 22 23 24
50000000
(6 WS Cat.)
6*1.0
of the four
fall )
The ISCST model also has the option of specifying hourly emission rates in a
separate file, as described in Section 3.3.8.
3-46
-------
3.3.4.2 Long Term Model Options.
The EMISFACT keyword provides the user the option of specifying variable emission
rate factors for sources modeled by the Long Term model. The syntax, type and order of
this keyword are summarized below:
Syntax: SO EMISFACT Srcid (or Srcrng) Qflag Qf act( i) , i=l ,n
Type: Optional, Repeatable
Order: Must follow the LOCATION card for each source input
where the Srcid parameter is the same source ID that was entered on the LOCATION card
for a particular source. The user also has the option of specifying a range of sources
for which the emission rate factors apply, instead of identifying a single source. This
is accomplished by two source ID character strings separated by a dash, e.g.,
STACK1-STACK10. The use of the Srcrng parameter is explained in more detail in Section
3.3.3 above for the BUILDHGT keyword.
The parameter Qflag is the variable emission rate flag, and is one of the following
secondary keywords:
SEASON - emission rates vary seasonally (n=4),
QUARTR - emission rates vary by quarter (n=4),
MONTH - emission rates vary monthly (n=12),
SSTAB - emission rates vary by season and stability
(n=24),
SSPEED - emission rates vary by season and speed (n=24),
3-47
-------
STAR - emission rates vary by speed and stability
only(n=36), and
SSTAR - emission rates vary by season, speed and
stability (n=144),
The Qfact array is the array of factors, where the number of factors is shown above for
each Qflag option. The EMISFACT card may be repeated as many times as necessary to
input all of the factors, and repeat values may be used for the numerical inputs. An
example of each of these options is presented below, with column headers to indicate the
order in which values are to be input.
•A- -A-
so
**
so
•A- -A-
sn
**
SO
**
so
•A- -A-
SO
**
EMISFACT
EMISFACT
STACK1
STACK1
SEASON
QUARTR
WINTER SPRING SUMMER
0.50 0.50 1.00 0
QUART1 QUART2 QUARTS
0.50 0.50 1.00 0
JAN FEB MAR
EMISFACT
STACK1
MONTH
0.1 0
2 0.3
WINTER
EMISFACT
STACK1
SSTAB
6*0
50
WINTER
EMISFACT
EMISFACT
STACK1
Stab
STACK1
Stab
SSPEED
Cat. :
STAR
Cat. :
6*0
A
6*0.5
A
50
B
6*0.6
B
APR MAY
0.4 0.5
SPRING
6*0.50
SPRING
6*0.50
C
6*0.7
C
JUN
0.5
FALL
.75
QUART4
.75
JUL
0 5
SUMMER
6*
1.00
SUMMER
6*
D
6*0
D
1.00
E
.8 6
E
AUG SEP
0.6 0
FALL
6*0.
FALL
6*0.
F
*0.9
F
7
(6
75
(6
75
(6
6*
(6
OCT NOV DEC
1.0 1.0 1.0
Stab Cat. )
WS Cat.)
WS Cat.)
1.0
WS Cat.)
** Season 1:
SO
**
SO
•A- -A-
SO
**
so
EMISFACT
STACK1
SSTAR
6*0.5
6*0.6
6*0.7
6*0
.8 6
*0.9
6*
1.0
Season 2:
EMISFACT
STACK1
SSTAR
6*0.5
6*0.6
6*0.7
6*0
.8 6
*0.9
6*
1.0
Season 3:
EMISFACT
STACK1
SSTAR
6*0.5
6*0.6
6*0.7
6*0
.8 6
*0.9
6*
1.0
Season 4:
EMISFACT
STACK1
SSTAR
6*0.5
6*0.6
6*0.7
6*0
.8 6
*0.9
6*
1.0
3-48
-------
If a monthly emission rate variation is selected, then the factors will only to
apply to monthly STAR summaries. A warning message will be generated if no monthly
averages are to be calculated. For the other variable emission rate choices, the model
will determine the correct season or quarter and apply that factor to any monthly STAR
summaries for which calculations are made. Also, if quarterly averages are being
calculated, then none of the emission rate factors involving seasonal variation may be
used (SEASON, SSTAB, SSPEED, or SSTAR). If a seasonal variation of emission rates is
needed in the calculation of quarterly averages, then it must be implemented through the
use of the MONTHly variable emission rate option.
3.3.5 Adjusting the Emission Rate Units for Output
The default emission rate units for the ISC models are grams per second for point
and volume sources, and grams per second per square meter for area sources. By default,
the models convert these input units to output units of micrograms per cubic meter for
concentration calculations and grams per square meter for deposition calculations. This
is accomplished by applying a default emission rate unit factor of 1.0E06 for
concentration and 3600 for deposition. The deposition factor essentially converts the
emission rate to grams per hour for total deposition calculations. For the Long Term
model, an additional factor is applied for deposition calculations to adjust the
emissions for the number of hours in the STAR data period. This is done automatically
by the ISCLT model, which allows the user to use the same set of source parameter inputs
whether the model is calculating concentration or deposition in either model.
The EMISUNIT keyword on the SO pathway allows the user to specify a different unit
conversion factor, and to specify the appropriate label for the output units for either
3-49
-------
concentration or deposition calculations. The syntax and type of the EMISUNIT keyword
are summarized below:
Syntax: SO EMISUNIT Emifac Emilbl Conlbl (orDeplbl)
Type: Optional, Non-repeatable
where the parameter Emifac is the emission rate unit factor, Emilbl is the label for the
emission units (up to 40 characters), and Conlbl and Deplbl are the output unit labels
(up to 40 characters) for concentration and deposition calculations, respectively. For
example, to produce output concentrations in milligrams per cubic meter, assuming input
units of grams per sec, the following card could be input:
SO EMISUNIT 1.0E3 GRAMS/SEC MILLIGRAMS/M**3
since there are 1.0E3 milligrams per gram. The emission rate unit factor applies to all
sources for a given run. Since the model uses one or more spaces to separate different
fields on the input runstream images, it is important that there not be any spaces
within the label fields on this card. Thus, instead of entering 'GRAMS PER SECOND' for
the emission label, a label of 'GRAMS/SECOND', or 'GRAMS-PER-SECOND' or an equivalent
variation should be used.
Since the ISCST model allows for both concentration and deposition to be output in
the same model run, the EMISUNIT keyword cannot be used to specify emission unit factors
if more than one output type is being generated. The ISCST model therefore allows for
concentration and deposition units to be specified separately through the CONCUNIT and
3-50
-------
DEPOUNIT keywords, respectively. The syntax and type of the CONCUNIT keyword are
summarized below:
Syntax: SO CONCUNIT Emifac Emilbl Conlbl
Type: Optional, Non-repeatable
where the parameter Emifac is the emission rate unit factor, Emilbl is the label for the
emission units (up to 40 characters), and Conlbl is the output unit label (up to 40
characters) for concentration calculations. The syntax and type of the DEPOUNIT keyword
are summarized below:
Syntax: SO DEPOUNIT Emifac Emilbl Deplbl
Type: Optional, Non-repeatable
where the parameter Emifac is the emission rate unit factor, Emilbl is the label for the
emission units (up to 40 characters), and Deplbl is the output unit label (up to 40
characters) for deposition calculations.
3.3.6 Specifying Variables for Settling, Removal and Deposition Calculations
The ISC models include algorithms to handle the gravitational settling and removal
by dry deposition of particulates. The input of source variables for settling and
removal are controlled by three keywords on the SO pathway, PARTDIAM, MASSFRAX, and
PARTDENS. As with building dimensions and variable emission rate factors described
above, the settling and removal variables may be input for a single source, or may be
applied to a range of sources.
3-51
-------
The syntax, type and order for these three keywords are summarized below:
Syntax: SO PARTDIAM Srcid (or Srcrng) PdiamCi ) ,i = l,Npd
SO MASSFRAX Srcid (or Srcrng) Phi(i),i=l,Npd
SO PARTDENS Srcid (or Srcrng) Pdens(i),i=l,Npd
Type: Optional, Repeatable
Order: Must follow the LOCATION card for each source input
where the Srcid or Srcrng identify the source or sources for which the inputs apply, and
where the Pdiam array consists of the particle diameter (microns) for each of the
particle size categories (up to a maximum of 20 set by the NPDMAX PARAMETER in the
computer code), the Phi array is the corresponding mass fractions (between 0 and 1) for
each of the categories, and the Pdens array is the corresponding particle density
(g/cm3)
for each of the categories. The use of the Srcrng parameter is explained in more detail
in Section 3.3.3 above for the BUILDHGT keyword.
The number of categories for a particular source is Npd. The user does not
explicitly tell the model the number of categories being input, but if continuation
cards are used all inputs of a keyword for a particular source or source range must be
contiguous, and the number of categories must agree for each of the three keywords input
for a particular source. As many continuation cards as needed may be used to define the
inputs for a particular keyword. The model checks the inputs to ensure that the mass
fractions sum to 1.0 (within 2 percent) for each source input, and that the mass
fractions are within the proper range (between 0 and 1) .
3-52
-------
For a technical description of the ISC dry deposition algorithms, refer to Sections
1.3 and 2.3 of Volume II of the User's Guide.
3.3.7 Specifying Variables for Precipitation Scavenging and Wet Deposition Calculations
The ISC Short Term (ISCST) model also includes algorithms to handle the scavenging
and removal by wet deposition (i.e., precipitation scavenging) of gases and
particulates. For wet deposition of particulates, the user must input the source
particle variables controlled by the PARTDIAM, MASSFRAX, and PARTDENS keywords on the SO
pathway. As with building dimensions and variable emission rate factors described
above, the scavenging coefficients may be input for a single source, or may be applied
to a range of sources. A separate scavenging coefficient is input for liquid
precipitation and for frozen precipitation.
For particulates, the scavenging coefficients are input through the PARTSLIQ and
PARTSICE keywords for liquid and frozen precipitation, respectively. The syntax, type
and order for these two keywords are summarized below:
Syntax: SO PARTSLIQ Srcid (or Srcrng) Scavcoef (i) ,i=l ,Npd
SO PARTSICE Srcid (or Srcrng) Scavcoef(i),i=l,Npd
Type: Optional, Repeatable
Order: Must follow the LOCATION card for each source input
where the Srcid or Srcrng identify the source or sources for which the inputs apply, and
where the Scavcoef array consists of the scavenging coefficients (s-mm/hr)"1 for each of
3-53
-------
the particle size categories defined on the SO PARTDIAM card (up to a maximum of 20 set
by the NPDMAX PARAMETER in the computer code).
The scavenging coefficients for gaseous emissions are specified by a single
keyword, GAS-SCAV, which uses a secondary keyword, LIQ or ICE, to distinguish between
liquid and frozen precipitation scavenging. The syntax, type and order for this keyword
are summarized below:
Syntax: SO GAS-SCAV Srcid (or Srcrng) LIQ or ICE Scavcoef
Type:
Optional, Repeatable
Order: Must follow the LOCATION card for each source input
where the Srcid or Srcrng identify the source or sources for which the inputs apply, and
where the Scavcoef parameter is the scavenging coefficient (s-mm/hr)"1 for either liquid
precipitation (secondary keyword of LIQ) or for frozen precipitation (secondary keyword
of ICE).
3-54
-------
3.3.8 Specifying an Hourly Emission Rate File
The source (SO) pathway includes an option for inputting hourly emission rates for
the ISCST model, controlled by the HOUREMIS keyword. ISCST allows for a single hourly
emission file to be used with each model run. The syntax, type and order for this
keyword are summarized below:
Syntax: SO HOUREMIS Emifil Srcid's (and/or Srcrng's)
Type: Optional, Repeatable
Order: Must follow the LOCATION card for each source input
where the Emifil parameter specifies the filename (up to 40 characters) for the hourly
emission file, and Srcid or Srcrng identify the source or sources for which hourly
emission rates are included. Source ranges, which are described in more detail in the
description of the BUILDHGT keyword (Section 3.3.3), are input as two source IDs
separated by a dash, e.g., STACK1-STACK10. The user may include more than one HOUREMIS
card in a runstream file, if needed to specify additional sources, but there can be only
one hourly emissions file, and therefore the filename must be the same on all HOUREMIS
cards.
The format of each record of the hourly emissions file includes a pathway and
keyword (SO HOUREMIS), followed by the Year, Month, Day, Hour, Source ID, emission rate
(in the appropriate units), and for point sources the stack gas exit temperature (K),
and stack gas exit velocity (m/s). The hourly emissions file is processed using the
same routines used to process the runstream input file, therefore each of the parameters
must be separated by at least one space, but otherwise the format is variable
3-55
-------
(parameters are not required to be specific columns). It is also not necessary to
include the SO HOUREMIS on each line, as long as the parameters (Year, Month, etc.) do
not begin before column 13.
The data in the hourly emission file must include the exact same dates as are
included in the meteorological input files, and the source IDs must correspond to the
source IDs defined on the SO LOCATION cards and be in the same order. Multiple records
are required to define the emissions for one hour if more than one source is referenced.
The model will check for a date mismatch between the hourly emissions file and the
meteorological data, and also for a source ID mismatch. An error will occur if a data
or ID mismatch is found. However, it is not necessary to process the entire hourly
emissions file on each model run, i.e., the correct emissions data will be read if the
ME DAYRANGE or the ME STARTEND cards (see Section 3.5.5) are used, as long as all the
dates (including those that are processed and those that are skipped) match the
meteorological data files. An example of several lines from an hourly emissions file
for two point sources is provided below:
so
so
so
so
so
so
so
so
HOUREMIS
HOUREMIS
HOUREMIS
HOUREMIS
HOUREMIS
HOUREMIS
HOUREMIS
HOUREMIS
88
88
88
88
88
88
88
88
8
8
8
8
8
8
8
8
16
16
16
16
16
16
16
16
1
1
2
2
3
3
4
4
STACK1
STACK2
STACK1
STACK2
STACK1
STACK2
STACK1
STACK2
52
44
22
42
51
41
36
43
467
327
321
166
499
349
020
672
382
432
377
437
373
437
374
437
604
326
882
682
716
276
827
682
12.27
22.17
9.27
19.67
11.87
18.77
9.63
18.23
3-56
-------
The model will use the stack release height and stack inside diameter defined on the SO
SRCPARAM card, but will use the emission rate, exit temperature and exit velocity from
the hourly emission file. If the emission rate, exit temperature and exit velocity are
not included for a particular hour, i.e, any or all of those fields are blank, the model
will interpret emissions data for that hour as missing and will set the parameters to
zero. Since the emission rate will be zero, there will be no calculations made for that
hour and that source.
3.3.9 Using Source Groups
The ISC models allow the user to group contributions from particular sources
together. Several source groups may be setup in a single run, and they may, for
example, be used to model impacts from the source being permitted, the group of
increment consuming PSD sources, and the group of all sources for comparison to a NAAQS
in a single run. There is always at least one source group in a run, which may consist
of all sources, so the SRCGROUP keyword has been made mandatory in the ISC models. The
syntax, type and order of the SRCGROUP keyword are summarized below:
Syntax: SO SRCGROUP Grpid Srcid's and/or Srcrng's
Type: Mandatory, Repeatable
Order: Must be the last keyword in the SO pathway before FINISHED
where the Grpid parameter is an alphanumeric string of up to eight characters that
identifies the group name. The Srcid's and Srcrng's are the individual source IDs
and/or source ranges that make up the group of sources. Source ranges, which are
described in more detail in the description of the BUILDHGT keyword (Section 3.3.3), are
3-57
-------
input as two source IDs separated by a dash, e.g., STACK1-STACK10. Individual source
IDs and source ranges may be used on the same card. If more than one input card is
needed to define the sources for a particular group, then additional cards may be input,
repeating the pathway, keyword and group ID.
A special group ID has been reserved for use in specifying the group of all
sources. When Grpid = ALL, the model will automatically setup a source group called ALL
that includes all sources modeled for that particular run. If desired, the user can
setup a group of all sources with a different group ID by explicitly specifying all
sources on the input card(s).
As described in Section 2.3, the maximum number of source groups is controlled by a
Fortran PARAMETER statement in the computer code. If the user attempts to define more
than the allowable number of source groups, the model will generate an appropriate
error message.
As discussed in Sections 1.2.4.6 and 3.2.9, it is sometimes important for a user to
know the contribution of a particular source to the total result for a group. These
source contribution analyses are facilitated in the Short Term model by the introduction
of the EVENT model. The EVENT model uses the same source groups that are identified by
ISCST (when the input file is generated using the CO EVENTFIL option), but the model is
structured in a way that it retains individual source results for particular events.
The Long Term model is able to provide source contribution information in the first
pass, because of the different data structures and memory requirements for that model.
Refer to the sections noted above for a more complete description of the EVENT model and
its uses.
3-58
-------
3.4 RECEPTOR PATHWAY INPUTS AND OPTIONS
The REceptor pathway contains keywords that define the receptor information for a
particular model run. The receptor pathway inputs are identical between the ISCST model
and the ISCLT model. The RE pathway is not used at all by the ISCEV (EVENT) model,
since the receptor locations are defined on the EVent pathway in combination with
particular time periods.
The RE pathway contains keywords that allow the user to define Cartesian grid
receptor networks and/or polar grid receptor networks, with either uniform or
non-uniform grid spacing, as well as discrete receptor locations referenced to a
Cartesian or a polar system. The program is initially setup to allow five (5) gridded
receptor networks of either (or both) types in a single run, plus discrete receptors of
either type, up to a maximum limit on the total number of receptors. The limit on the
number of receptors in a given run is controlled by a Fortran PARAMETER in the computer
code (see Sections 2.3 and 4.2.2). The number of receptor networks allowed is also
controlled by a PARAMETER statement and may be easily changed by the user.
The default units for receptor elevations for the ISC models are in meters,
however, the user may specify receptor elevations to be in units of feet by adding the
RE ELEVUNIT FEET card immediately after the RE STARTING card. This optional card has
the same effect as the obsolescent CO ELEVUNIT FEET card.
3-59
-------
3.4.1 Defining Networks of Gridded Receptors
Two types of receptor networks are allowed by the ISC models. A Cartesian grid
network, defined through the GRIDCART keyword, includes an array of points identified by
their x (east-west) and y (north-south) coordinates. A polar network, defined by the
GRIDPOLR keyword, is an array of points identified by direction and distance from a
user-defined origin. Each of these keywords has a series of secondary keywords
associated with it that are used to define the network, including any receptor
elevations for elevated terrain and flagpole receptor heights. The GRIDCART and
GRIDPOLR keywords can be thought of as "sub-pathways," since their secondary keywords
include a STArt and an END card to define the start and end of inputs for a particular
network.
3.4.1.1 Cartesian Grid Receptor Networks.
Cartesian grid receptor networks are defined by use of the GRIDCART keyword. The
GRIDCART keyword may be thought of as a "sub-pathway," in that there are a series of
secondary keywords that are used to define the start and the end of the inputs for a
particular network, and to select the options for defining the receptor locations that
make up the network. The syntax and type of the GRIDCART keyword are summarized below:
3-60
-------
Syntax :
Type:
RE GRIDCART Netid
Ydelta
or
Gridxn, and
Gridyn
Zel evn
Zf 1 agn
Opti onal , Repeatabl
STA
XYINC
XPNTS
YPNTS
ELEV
FLAG
END
e
Xinit
Gridxl
Gridyl
Row Zel
Row Zfl
Xnum Xdelta Yinit Ynum
GridxZ GridxS
GridyZ GridyS ....
evl Zel ev2 Zel ev3
agl ZflagZ ZflagS . . .
where the parameters are defined as follows:
Netid
STA
XYINC
Xinit
Xnum
Xdelta
Yinit
Ynum
Ydelta
XPNTS
Gridxl
Gridxn
Receptor network identification code (up to eight
al phanumeri c
characters )
Indicates the STArt of GRIDCART inputs for a particular
network,
repeated for each new Netid
Keyword identifying uniform grid network generated from x
and y
increments
Starting x-axis grid location in meters
Number of x-axis receptors
Spacing in meters between x-axis receptors
Starting y-axis grid location in meters
Number of y-axis receptors
Spacing in meters between y-axis receptors
Keyword identifying grid network defined by a series
of discrete x and y coordinates (used with YPNTS)
Value of first x-coordinate for Cartesian grid (m)
Value of 'nth' x-coordinate for Cartesian grid (m)
3-61
-------
YPNTS
Gridyl
Gridyn
ELEV
Row
Zel ev
FLAG
Row
Zflag
END
Keyword identifying grid network defined by a series
of discrete x and y coordinates (used with XPNTS)
Value of first y-coordinate for Cartesian grid (m)
Value of 'nth' y-coordinate for Cartesian grid (m)
Keyword to specify that receptor elevations follow
(optional )
Indicates which row (y-coordinate fixed) is being
input (Row=l means first, i.e., southmost row)
An array of receptor terrain elevations (m) for a
particular Row (default units of meters may be changed
to feet by
use of RE ELEVUNIT or CO ELEVUNIT keyword), number of
entries per
row equals the number of x-coordinates for that network
Keyword to specify that flagpole receptor heights
fol1ow (opti onal )
Indicates which row (y-coordinate fixed) is being
input (Row=l means first, i.e., southmost row)
An array of receptor heights (m) above local terrain
elevation for a particular Row (flagpole receptors),
number of
entries per row equals the number of x-coordinates for
that
network
Indicates the END of GRIDCART inputs for a particular
network,
repeated for each new Netid
The ELEV and FLAG keywords are optional inputs, and are only needed if elevated
terrain or flagpole receptor heights are to be used. If the ELEV keyword is used and
the model is being run with the flat terrain option (see Section 3.2.6), then the
elevated terrain height inputs will be ignored by the model, and a non-fatal warning
message will be generated. If the elevated terrain option is selected, and no elevated
terrain heights are entered, the elevations will default to 0.0 meters, and warning
messages will also be generated. The model handles flagpole receptor height inputs in a
similar manner.
3-62
-------
The order of cards within the GRIDCART subpathway is not important, as long as all
inputs for a particular network are contiguous and start with the STA secondary keyword
and end with the END secondary keyword. It is not even required that all ELEV cards be
contiguous, although the input file will be more readable if a logical order is
followed. The network ID is also not required to appear on each runstream image (except
for the STA card). The model will assume the previous ID if none is entered, similar to
the use of continuation cards for pathway and keywords. Thus, the following two
examples produce the same 8X4 Cartesian grid network:
RE GRIDCART
RE GRIDCART
RE GRIDCART
RE GRIDCART
RE GRIDCART
RE GRIDCART
RE GRIDCART
RE GRIDCART
RE GRIDCART
RE GRIDCART
RE GRIDCART
RE GRIDCART
RE GRIDCART
RE GRIDCART
CAR1
CAR1
CAR1
CAR1
CAR1
CAR1
CAR1
CAR1
CAR1
CAR1
CAR1
CAR1
CAR1
CAR1
STA
XPNTS
YPNTS
ELEV
ELEV
ELEV
ELEV
FLAG
FLAG
FLAG
FLAG
END
STA
XPNTS
YPNTS
ELEV
FLAG
ELEV
FLAG
ELEV
FLAG
ELEV
FLAG
END
1
2
3
4
1
2
3
4
1
1
2
2
3
3
4
4
-500.
-500.
10.
20.
30.
40.
10.
20.
30.
40.
-500.
-500.
8*10
8*10
8*20
8*20
8*30
8*30
8*40
8*40
-400
-250
10.
20.
30.
40.
10.
20.
30.
40.
-400
-250
10.
20.
30.
40.
10.
20.
30.
40.
200.
250.
10.
20.
30.
40.
10.
20.
30.
40.
200.
250.
-100.
500.
10.
20.
30.
40.
10.
20.
30.
40.
-100.
500.
100
10.
20.
30.
40.
10.
20.
30.
40.
100
10
20
30
40
10
20
30
40
200. 400. 500.
. 10.
. 20.
. 30.
. 40.
. 10.
. 20.
. 30.
. 40.
200. 400. 500.
3-63
-------
The Row parameter on the ELEV and FLAG inputs may be entered as either the row
number, i.e., 1, 2, etc., or as the actual y-coordinate value, e.g., -500., -250., etc.
in the example above. The model sorts the inputs using Row as the index, so the result
is the same. The above example could therefore be entered as follows, with the same
result:
RE GRIDCART CAR1 STA
XPNTS
YPNTS
ELEV
FLAG
ELEV
FLAG
ELEV
FLAG
ELEV
FLAG
RE GRIDCART CAR1 END
-500.
-500.
-500.
-500.
-250.
-250.
250.
250.
500.
500.
-400. -200. -100. 100. 200. 400. 500.
-250. 250. 500.
8*10.
8*10.
8*20.
8*20.
8*30.
8*30.
8*40.
8*40.
Of course, one must use either the row number or y-coordinate value consistently within
each network to have the desired result.
The following simple example illustrates the use of the XYINC secondary keyword to
generate a uniformly spaced Cartesian grid network. The resulting grid is 11 x 11, with
3-64
-------
a uniform spacing of 1 kilometer (1000. meters), and is centered on the origin (0., 0.
No elevated terrain heights or flagpole receptor heights are included in this example.
RE GRIDCART CGI STA
XYINC -5000. 11 1000. -5000. 11 1000.
RE GRIDCART CGI END
3.4.1.2 Polar Grid Receptor Networks.
Polar receptor networks are defined by use of the GRIDPOLR keyword. The GRIDPOLR
keyword may also be thought of as a "sub-pathway," in that there are a series of
secondary keywords that are used to define the start and the end of the inputs for a
particular network, and to select the options for defining the receptor locations that
make up the network. The syntax and type of the GRIDPOLR keyword are summarized below:
Syntax
Type:
: RE GRIDPOLR Netid STA
ORIG
or ORIG
DIST
DDIR
or GDIR
ELEV
Zel evn
FLAG
Zfl agn
END
Optional, Repeatable
Xi
nit
Yinit,
S r c i d
Ringl
Di
Di
Di
Di
rl
RingZ
Dir2
rnum Dirini
r
r
Zel evl
Zflagl
Ri
Di
Di
ng3 . . .
r3 .. .
rinc
ZelevZ Zel
Zfl
ag2 Zfl
Ri
ngn
Dim,
ev3
ag3
3-65
-------
where the parameters are defined as follows:
3-66
-------
Netid
Receptor network identification code (up to eight
alphanumeri c
characters)
STA
Indicates STArt of GRIDPOLR inputs for a particular
network,
repeat for each new Netid
PRIG
Xinit
Yinit
Srcid
Keyword to specify the origin of the polar network
(optional )
x-coordinate for origin of polar network
y-coordinate for origin of polar network
Source ID of source used as origin of polar network
DIST
Ringl
Ri ngn
Keyword to specify distances for the polar network
Distance to the first ring of polar coordinates
Distance to the 'nth' ring of polar coordinates
DDIR
Dirl
Di rn
Keyword to specify discrete direction radials for the
polar network
First direction radial in degrees (1 to 360)
The 'nth' direction radial in degrees (1 to 360)
Di rnum
D i r i n i
D i r i n c
Keyword to specify generated direction radials for
the polar network
Number of directions used to define the polar system
Starting direction of the polar system
Increment (in degrees) for defining directions
ELEV
Di r
Zel ev
Keyword to specify that receptor elevations follow
(optional )
Indicates which direction is being input
An array of receptor terrain elevations for a
particular direction radial (default units of meters may
be
changed to feet by use of RE ELEVUNIT or CO ELEVUNIT
keyword),
number of entries per radial equals the number of
distances for
that network
Di r
Zflag
Keyword to specify that flagpole receptor heights
fol1ow (opti onal )
Indicates which direction is being input
An array of receptor heights above local terrain
elevation for a particular direction (flagpole
receptors)
3-67
-------
END
Indicates END of GRIDPOLR subpathway, repeat for each
new Netid
The PRIG secondary keyword is optional for the GRIDPOLR inputs. If omitted, the
model assumes a default origin of (0., 0.,) in x,y coordinates. The ELEV and FLAG
keywords are also optional inputs, and are only needed if elevated terrain or flagpole
receptor heights are to be used. If the ELEV keyword is used and the model is being run
with the flat terrain option (see Section 3.2.6), then the elevated terrain height
inputs will be ignored by the model, and a non-fatal warning message will be generated.
If the elevated terrain option is selected, and no elevated terrain heights are entered,
the elevations will default to 0.0 meters, and warning messages will also be generated.
The model handles flagpole receptor height inputs in a similar manner.
As with the GRIDCART keyword described above, the order of cards within the
GRIDPOLR subpathway is not important, as long as all inputs for a particular network are
contiguous and start with the STA secondary keyword and end with the END secondary
keyword. It is not even required that all ELEV cards be contiguous, although the input
file will be more readable if a logical order is followed. The network ID is also not
required to appear on each runstream image (except for the STA card). The model assumes
the previous ID if none is entered, similar to the use of continuation cards for pathway
and keywords.
The following example of the GRIDPOLR keyword generates a receptor network
consisting of 180 receptor points on five concentric distance rings centered on an
assumed default origin of (O.,0.). The receptor locations are placed along 36 direction
3-68
-------
radials, beginning with 10. degrees and incrementing by 10. degrees in a clockwise
fashion.
RE GRIDPOLR POL1 STA
DIST 100. 300. 500. 1000. 2000.
GDIR 36 10. 10.
RE GRIDPOLR POL1 END
Another example is provided showing the use of a non-zero origin, discrete
direction radials and the specification of elevated terrain and flagpole receptor
heights:
3-69
-------
RE GRIDPOLR POL1 STA
ORIG 500. 500.
DIST 100. 300. 500. 1000. 2000.
DDIR 90. 180. 270. 360.
ELEV 90. 5. 10. 15. 20. 25.
ELEV 180. 5. 10. 15. 20. 25.
ELEV 270. 5. 10. 15. 20. 25.
ELEV 360. 5. 10. 15. 20. 25.
FLAG 90. 5. 10. 15. 20. 25.
FLAG 180. 5. 10. 15. 20. 25.
FLAG 270. 5. 10. 15. 20. 25.
FLAG 360. 5. 10. 15. 20. 25.
RE GRIDPOLR POL1 END
As with the GRIDCART keyword described above, the user has the option of specifying the
radial number (e.g. 1, 2, 3, etc.) on the ELEV and FLAG inputs, or the actual direction
associated with each radial.
For purposes of model calculations, all receptor locations, including those
specified as polar, are stored in the model arrays as x, y and z coordinates and
flagpole heights. For the purposes of reporting the results by receptor in the main
print file, the tables are labeled with the polar inputs, i.e., directions and
distances.
3.4.2 Using Multiple Receptor Networks
For some modeling applications, the user may need a fairly coarsely spaced network
covering a large area to identify the area of significant impacts for a plant, and a
denser network covering a smaller area to identify the maximum impacts. To accommodate
this modeling need, the ISC models allow the user to specify multiple receptor networks
3-70
-------
in a single model run. The user can define either Cartesian grid networks or polar
networks, or both. With the use of the PRIG option in the GRIDPOLR keyword, the user
can easily place a receptor network centered on the facility being permitted, and also
place a network centered on another background source known to be a significant
contributor to high concentrations. Alternatively, the polar network may be centered on
a receptor location of special concern, such as a nearby Class I area.
As noted in the introduction to this section (3.4), the model initially allows up
to 5 receptor networks in a single run. This limit can be changed by modifying the
Fortran PARAMETER statement and recompiling the model. The variables that define each
array, e.g., the distances and directions for a polar network, are stored in arrays, so
that results can be presented for each network separately in the main output file of the
model. Thus, increasing the number of networks allowed will increase the amount of
memory needed to run the model, although the increase is relatively small. There are
also limits on the number of distances or directions (or the number of x-points and the
number of y-points for Cartesian grids) that can be specified for each network. These
are initially set to 50 distances or x-points and 50 directions or y-points. These
limits are also controlled by Fortran PARAMETER statements, and may be modified. More
information on controlling the storage limits of the models is provided in Section
4.2.2.
3.4.3 Specifying Discrete Receptor Locations
In addition to the receptor networks defined by the GRIDCART and GRIDPOLR keywords
described above, the user may also specify discrete receptor points for modeling impacts
at specific locations of interest. This may be used to model critical receptors, such
3-71
-------
as the locations of schools or houses, nearby Class I areas, or locations identified as
having high concentrations by previous modeling analyses. The discrete receptors may be
input as either Cartesian x,y points (DISCCART keyword) or as polar distance and
direction coordinates (DISCPOLR keyword). Both types of receptors may be identified in
a single run. In addition, for discrete polar receptor points the user specifies the
source whose location is used as the origin for the receptor.
A special option has been included in the ISC models, controlled by the BOUNDARY
keyword, which simplifies the input of plant boundary distances in a polar framework.
This option is described in Section 3.4.4 below.
3.4.3.1 Discrete Cartesian Receptors.
Discrete Cartesian receptors are defined by use of the DISCCART keyword. The
syntax and type of this keyword are summarized below:
Syntax: RE DISCCART Xcoord Ycoord (Zelev) (Zflag)
Type: Optional, Repeatable
where the Xcoord and Ycoord parameters are the x-coordinate and y-coordinate (m),
respectively, for the receptor location. The Zelev parameter is an optional terrain
elevation (m) for the receptor for use in elevated terrain modeling. The Zflag
parameter is the optional receptor height above ground (m) for modeling flagpole
receptors. All of the parameters are in units of meters, except for Zelev, which
defaults to meters but may be specified in feet by use of the RE ELEVUNIT or CO ELEVUNIT
keyword.
3-72
-------
If neither the elevated terrain option (Section 3.2.6) nor the flagpole receptor
height option (Section 3.2.7) are used, then the optional parameters are ignored if
present. If only the elevated terrain height option is used (no flagpoles), then the
third parameter (the field after the Ycoord) is read as the Zelev parameter. If only
the flagpole receptor height option is used (no elevated terrain), then the third
parameter is read as the Zflag parameter. If both options are used, then the parameters
are read in the order indicated for the syntax above. If the optional parameters are
left blank, then default values will be used. The default value for Zelev is 0.0, and
the default value for Zflag is defined by the CO FLAGPOLE card (see Section 3.2.7) .
Note: If both the elevated terrain and flagpole receptor height options are used, then
the third parameter will always be used as Zelev, and it is not possible to use a
default value for Zelev while entering a specific value for the Zflag parameter.
3.4.3.2 Discrete Polar Receptors.
Discrete polar receptors are defined by use of the DISCPOLR keyword. The syntax
and type of this keyword are summarized below:
Syntax: RE DISCPOLR Srcid Dist Direct (Zelev) (Zflag)
Type: Optional, Repeatable
where the Srcid is the alphanumeric source identification for one of the sources defined
on the SO pathway which will be used to define the origin for the polar receptor
location. The Dist and Direct parameters are the distance in meters and direction in
degrees for the discrete receptor location. Degrees are measured clockwise from north.
The Zelev parameter is an optional terrain elevation for the receptor for use in
3-73
-------
elevated terrain modeling. The units of Zelev are in meters, unless specified as feet
by the RE ELEVUNIT or CO ELEVUNIT keyword. The Zflag parameter is the optional receptor
height above ground (meters) for modeling flagpole receptors.
If neither the elevated terrain option (Section 3.2.6) nor the flagpole receptor
height option (Section 3.2.7) are used, then the optional parameters are ignored if
present. If only the elevated terrain height option is used (no flagpoles), then the
third parameter (the field after the Ycoord) is read as the Zelev parameter. If only
the flagpole receptor height option is used (no elevated terrain), then the third
parameter is read as the Zflag parameter. If both options are used, then the parameters
are read in the order indicated for the syntax above. If the optional parameters are
left blank, then default values will be used. The default value for Zelev is 0.0, and
the default value for Zflag is defined by the CO FLAGPOLE card (see Section 3.2.7).
Note: If both the elevated terrain and flagpole receptor height options are used, then
fourth parameter will always be used as Zelev, and it is not possible to use a default
value for Zelev while entering a specific value for the Zflag parameter.
3.4.4 Specifying Plant Boundary Distances
The ISC models include a special option to simplify the input of discrete receptor
locations for plant boundary distances. This option is controlled by the BOUNDARY
keyword. The syntax and type of this keyword are summarized below:
Syntax: RE BOUNDARY Srcid DistCi ),i = l,36
Type: Optional, Repeatable
3-74
-------
where the Srcid is the alphanumeric source identification for one of the sources defined
on the SO pathway for which the boundary distances are to be defined. The location of
the source will serve as the origin for 36 discrete polar receptors located at every 10
degrees around the source. The Dist array includes the distances (in meters) for each
of the directions, beginning with the 10 degree radial and incrementing every 10
degrees clockwise. While the BOUNDARY keyword generates 36 discrete polar receptors,
the results for these receptors are summarized separately from receptors defined by the
DISCPOLR keyword in the main output file. The RE BOUNDARY card may be repeated for the
source as many times as needed to input the 36 distances.
A related keyword, BOUNDELV, is used to define terrain elevations for the receptor
locations identified with the BOUNDARY keyword. The BOUNDELV keyword defines the
terrain elevations in meters (or feet if the RE ELEVUNIT or CO ELEVUNIT FEET card
appears) for each of the 36 boundary receptor points. The syntax and type for this
keyword are summarized below:
Syntax: RE BOUNDELV Srcid Zelev(i) ,1=1,36
Type: Optional, Repeatable
The purpose of the BOUNDARY and BOUNDELV keywords is to provide a short-cut for
inputting the discrete polar receptors for the plant boundary. There is no
corresponding keyword for inputting boundary receptor flagpole heights. The easiest way
to input boundary receptors with flagpole receptor heights is to define them as discrete
polar receptors using the DISCPOLR keyword. This method provides better assurance that
the flagpole heights are associated with the correct receptor, and makes it easier to
check and debug the input file. For applications where a uniform flagpole receptor
3-75
-------
height is used for all receptors, which can be specified as a parameter on the CO
FLAGPOLE input card, those flagpole receptor heights will also apply to any boundary
receptors identified through the BOUNDARY keyword.
3.5 METEOROLOGY PATHWAY INPUTS AND OPTIONS
The MEeteorology pathway contains keywords that define the input meteorological
data for a particular model run. Because of differences in the meteorological data
needs for the Short Term and Long Term models, some of the ME pathway inputs are
different between the two models. These differences are highlighted in the discussions
below. An effort has been made to keep the inputs as similar as possible between the
Short Term and Long Term models.
3.5.1 Specifying the Input Data File and Format
The input meteorological data filename and format are identified by the INPUTFIL
keyword on the ME pathway. The syntax of this keyword is very similar between the Short
Term and Long Term models, but there are some differences due to the different formats
of data available for the two types of models. Therefore the Short Term and Long Term
model inputs are described separately.
3-76
-------
3.5.1.1 Short Term Model Options.
3-77
-------
The ISC Short Term model uses hourly meteorological data as one of the basic model
inputs. The user has several options for specifying the format of the meteorological
data using the INPUTFIL keyword. The syntax and type of this keyword are summarized
below:
Syntax: ME INPUTFIL Metfil (Format)
Type: Mandatory, Non-repeatabl e
where the Metfil parameter is a character field of up to 40 characters that identifies
the filename for the meteorological data file. For running the model on an
IBM-compatible PC, the Metfil parameter may include the complete DOS pathname for the
file, or will assume the current directory if only the filename is given. The optional
Format parameter specifies the format of the meteorological data file. The user has the
following five options for specifying the Format:
1) Use the default ASCII format for a sequential hourly file (if Format is left
blank);
2) Specify the Fortran READ format for an ASCII sequential hourly file;
3) Use free-formatted READs for an ASCII sequential hourly file, by inputting the
secondary keyword of FREE;
4) Use unformatted file generated by the PCRAMMET or MPRM preprocessors, by
inputting the secondary keyword of UNFORM; or
5) Use "card image" data using a default ASCII format by specifying the secondary
keyword of CARD - this option differs from option 1) by the addition of hourly
wind profile exponents and hourly vertical potential temperature gradients in
the input file.
3-78
-------
Since the deposition algorithms require additional meteorological variables, the exact
format of ASCII meteorological data will depend on whether the dry and/or wet deposition
algorithms are being used. If the deposition algorithms are being used, then the
unfomatted data file (option 4 above) cannot be used.
The first record of the meteorological data input file contains the station number
and year for both the surface station and the upper air (mixing height) station. For
the formatted ASCII files, these four integer variables are read using a free-format
READ, i.e., the variables must be separated by either a comma or by one or more blank
spaces. For the UNFORMatted files, the four variables are read as integers without any
format specification. The order of these variables is as follows:
Surface Station Number, e.g., WBAN Number for NWS Stations
Year for Surface Data (2 or 4 digits)
Upper Air Station Number (for Mixing Height Data)
Year for Upper Air Data (2 or 4 digits)
The model checks these variables against the values input by the user on the ME SURFDATA
and ME UAIRDATA cards (see Section 3.5.3 below).
The rest of the records in the file include the sequential meteorological data.
The order of the meteorological variables for the formatted ASCII files and the default
ASCII format are as follows:
3-79
-------
Variable
Year (last 2 digits)
Month
Day
Hour
Flow Vector (deg.)
Wind Speed (m/s)
Ambient Temperature (K)
Stability Class
(A=l, B=2, . . . F=6)
Rural Mixing Height (m)
Urban Mixing Height (m)
Wind Profile Exponent
(CARD only)
Vertical Potential
Temperature Gradient (K/m)
(CARD only)
Friction Velocity (m/s)
(Dry or Wet Deposition Only)
Monin-Obukhov Length (m)
(Dry or Wet Deposition Only)
Surface Roughness Length (m)
(Dry or Wet Deposition Only)
Precipitation Code (00-45)
(Wet Deposition Only)
Fortran Format
12
12
12
12
F9.4
F9.4
F6.1
12
F7.1
F7.1
F8.4
F8.4
F9.4
F10.1
F8.4
14
Columns
1-2
3-4
5-6
7-8
9-17
18-26
27-32
33-34
35-41
42-48
49-56
57-65
49-57
(66-74
for CARD)
58-67
(75-84
for CARD)
68-75
(85-92
for CARD)
76-79
(93-96
for CARD)
3-80
-------
Precipitation Rate (mm/hr)
(Wet Deposition Only)
F7.2
80-86
(97-103
for CARD)
Thus the following two cards would have the same effect, one using the default read
format (Format parameter left blank) and the other explicitly providing the ASCII read
format described above:
ME INPUTFIL C:\DATA\METDATA.INP
ME INPUTFIL C:\DATA\METDATA.INP (412,2F9.4,F6.1,12,2F7.1,F9.4,F10.1,F8.4,14,F7.2)
The user-specified ASCII format is input as a character field of up to 60 characters,
and may be used to specify the READ format for files that differ from the default
format. The variables are identified in the READ format in the order given above, but
by using the Fortran tab edit descriptor (Tx, where x is the column number), the order
of variables within the file may be different. A utility program, BINTOASC, is
available for converting unformatted PCRAMMET meteorological files to the default ASCII
format for applications that do not involve dry deposition. The BINTOASC utility
program is described in Appendix C.
For FREE-formatted reads, the model uses a Fortran free-format READ statement,
meaning that the variables in the meteorological data file must be in the order listed
above, and must be separated from each other by a comma or at least one blank. The
3-81
-------
format does not need to be the same on each record as long as the variables are
appropriately delimited.
The UNFORM secondary keyword indicates to the model that the meteorological data
are in an unformatted (sometimes called a "binary") file that was generated by the
RAMMET or the MPRM preprocessor. The preprocessed data files consist of unformatted
records that include 24 hours of meteorology per record. The variables are read from
the unformatted records in the following order:
Year
Month
Julian Day (1-366)
Stability Class (hours 1 to 24)
Wind Speed, m/s (hours 1 to 24)
Ambient Temperature, K (hours 1 to 24)
Flow Vector, deg. (hours 1 to 24)
Randomized Flow Vector, deg. (hours 1 to 24)
Mixing Heights, m (hr 1 rural, hr 1 urban, ... to hr 24)
The following example illustrates the use of the unformatted file option:
ME INPUTFIL C:\BIN\PREPIT.BIN UNFORM
where the Metfil parameter has been used to identify a complete DOS pathname.
The ASCII file input options on the INPUTFIL card allow the user to read the "card
image" meteorological data. This includes the option for inputting hourly wind profile
3-82
-------
exponents and vertical potential temperature gradients through use of the CARD format
option. If the CARD format is not used, then the default values of wind profile
exponents and vertical potential temperature gradients are used unless the user
specifies non-default inputs using the ME WINDPROF or ME DTHETADZ keyword options.
The meteorological data file for the Short Term model normally consists of a single
complete year of meteorological data, beginning with hour 0100 of January 1 and ending
with hour 2400 of December 31. For certain applications, such as long term risk
assessments, it may be desirable to obtain averages calculated over a period longer than
a single year. For these applications, the Short Term model is able to read multiple-
year meteorological data files in any of the ASCII formats described above. At the
present time, the model is not able to read multiple-year UNFORMatted meteorological
data files.
The simplest way to obtain these multiple-year data files is by using the DOS COPY
command to concatenate preprocessed ASCII data files. An example of using the DOS COPY
command for this purpose is shown below for concatenating five years of meteorological
data:
COPY RDU86.ASC+RDU87.ASC+RDU88.ASC+RDU89.ASC+RDU90.ASC RDU86-90.ASC
To use this five-year ASCII data file, simply include the new file name on the ME
INPUTFIL card with the appropriate ASCII file format, and include the year corresponding
to the first data file on the ME SURFDATA and ME UAIRDATA cards, described below in
3-83
-------
Section 3.5.3. By using the DOS COPY command, the header record at the beginning of
each yearly data file will be included within the multiple-year data file. The model
will read the embedded header records if they are present, and check for agreement of
the surface and upper air station IDs with the values input on the SURFDATA and UAIRDATA
cards. The model is also able to read the multiple-year data file if the header records
for subsequent years have been removed. See Section 3.2.3.1 for a discussion of how
different averaging time options are handled when multiple-year data files are used with
the Short Term model.
3.5.1.2 Long Term Model Options.
The ISC Long Term model uses a standard STability ARray (STAR) meteorological data
file in place of sequential hourly meteorological data used in the Short Term model.
The meteorological data in the STAR file consists of a joint frequency distribution of
wind speed and wind direction by stability category. The input of other variables to
the Long Term model, (temperature, mixing height, and surface roughness (z0) ) are
controlled by separate ME pathway keywords described later in this section. The Monin-
Obukhov lenght (L) and friction velocity (u,) are calculated internally when needed for
dry deposition modeling.
The ISCLT model reads the STAR meteorological data from a separate data file. The
STAR data filename and format are specified following the INPUTFIL keyword. The
following syntax is used:
3-84
-------
Syntax: ME INPUTFIL Metfil (Format)
Type: Optional, Non-repeatable
where the Metfil parameter is a character field of up to 40 characters that identifies
the filename for the meteorological data file. For running the model on an
IBM-compatible PC, the Metfil parameter may include the complete DOS pathname for the
file; the current directory is assumed if only the filename is given. The optional
FORMAT parameter specifies the format
for the STAR data. The user has the following three options for specifying the Format:
1) Use the default ASCII format for the STAR file (if Format is left blank) ;
2) Specify the Fortran READ format for the ASCII STAR file; or
3) Use free-formatted READs for the ASCII STAR file, by inputting the secondary
keyword of FREE.
The default ASCII format corresponds to the format of the data files generated by
EPA's STAR utility program for the ISCLT model. Each record of STAR meteorological data
consists of six values (default format of 6F10.0) corresponding to the six wind speed
classes for a particular wind direction and stability category. The program reads
stability category A first, and the first record contains the six values for the north
wind direction. There are 16 cards for each stability category corresponding to the 16
wind direction categories entered clockwise from north (north, north-northeast, etc.).
This pattern is repeated for each of the six stability categories, A through F.
The frequency data may be input as normalized frequencies, in which case the total
of all frequencies for a particular STAR summary will add up to 1.0, or as the number of
3-85
-------
occurrences for each combination. If the total of normalized frequencies is not within
2 percent of 1.0, then the model will generate a non-fatal warning message. If the
total adds up to 2.0 or more and is a whole number, then the model divides the number of
occurrences for each STAR category by the total number to obtain the normalized
frequency.
Without the optional STARDATA keyword (described in Section 3.5.4), it is assumed
that the STAR summaries in the input file corresponds to the averaging periods selected
on the CO AVERTIME card (see Section 3.2.3.1). If SEASON averages are selected, then
the model will assume that the meteorological data file consists of four seasons in the
order of WINTER, SPRING, SUMMER, and FALL. If an ANNUAL average is to be calculated
from an annual STAR summary, then the annual STAR should follow any seasonal STAR
summaries to be used. For example, the following runstream image calculates averages
for each of the four seasons and the annual average from a data file consisting of five
STAR summaries (winter, spring, summer, fall, and annual):
CO AVERTIME SEASON ANNUAL
3-86
-------
The following example calculates averages for the four seasons, and then calculates an
annual average as a period average for the four seasons combined:
CO AVERTIME SEASON PERIOD
and the input meteorological file for this example would include only the four seasonal
STAR summaries.
3.5.2 Specification of Anemometer Height
An important input for both the Short Term and the Long Term models is the
specification of the anemometer height, i.e., the height above ground at which the wind
speed data were collected. Since the models adjust the input wind speeds from the
anemometer height to the release height (see Section 1.1.3 of Volume II), the accurate
specification of anemometer height is important to obtaining the correct model results.
The syntax and type of the ANEMHGHT keyword are summarized below:
Syntax: ME ANEMHGHT Zref (Zrunit)
Type: Mandatory, Non-repeatabl e
where the parameter Zref is the height of the anemometer measurement above ground, and
the optional parameter Zrunit is used to specify the units of Zref. Valid inputs for
Zrunit are the secondary keywords METERS or FEET. The default units for Zref are in
meters if Zrunit is left blank.
3-87
-------
3.5.3 Specifying Station Information
Two keywords are used to specify information about the meteorological stations,
SURFDATA for the surface meteorological station, and UAIRDATA for the upper air station
used in the determination of mixing heights. The syntax and type of these keywords are
summarized below:
Syntax: ME SURFDATA Stanum Year (Name) (Xcoord) (Ycoord)
Syntax: ME UAIRDATA Stanum Year (Name) (Xcoord) (Ycoord)
Type:
Mandatory, Non-repeatable
where Stanum is the station number, e.g. the 5-digit WBAN number for NWS stations, Year
is the year of data being processed (either 2 or 4 digits), Name is an optional
character field (up to 40 characters with no blanks) specifying the name of the station,
and Xcoord and Ycoord are optional parameters for specifying the x and y coordinates
for the location of the stations. At the present time, the station locations are not
utilized in the models. Therefore, no units are specified for Xcoord and Ycoord at this
time, although meters are suggested in order to be consistent with the source and
receptor coordinates.
3.5.4 Specifying the Meteorological STAR Data (Applies Only to ISCLT)
The STARDATA keyword is used to define what STAR meteorological data summaries are
actually included in the data file. The syntax and type of this keyword is summarized
below:
3-88
-------
Syntax: ME STARDATA JAN FEB MAR APR MAY JUN JUL AUG SEP ocr NOV DEC
WINTER SPRING SUMMER FALL
QUART1 QUARTZ QUARTS QUART4
MONTH SEASON QUARTR ANNUAL
PERIOD
Type:
Optional, Non-repeatable
This keyword works is conjunction with the CO AVERTIME keyword (Section 3.2.3) to
determine which STAR summaries are processed for a particular run. If the STARDATA
keyword is omitted, then the model assumes that the meteorological data file consists
only of the STAR summaries identified on the CO AVERTIME keyword. While the STARDATA
keyword is identified as being optional, it is required in the case where the CO
AVERTIME card specifies only the PERIOD average to be calculated. In this case, the
model needs the STARDATA input in order to determine what STAR summaries are included in
the data file to properly calculate the PERIOD average. A fatal error message will be
generated (and processing aborted) if the STARDATA card is omitted for cases with only
PERIOD averages being calculated.
The STARDATA keyword allows the user considerable flexibility in controlling which
averaging periods to calculate from one run to another. As an example, suppose that the
user has a STAR data file consisting of 12 monthly STAR summaries. This would be
identified to the model by including the following card on the ME pathway:
ME STARDATA MONTH
3-89
-------
The user could then generate annual average results by specifying only PERIOD on the CO
AVERTIME card. The emission rate factor may be varied by month in the process. With
the same meteorological data file, the user could also calculate results for the first
quarter only by changing the AVERTIME card to read:
CO AVERTIME JAN FEB MAR PERIOD
This would result in results being produced for each of the first three months of the
year and for the combined period of JAN-MAR. Each quarter could be calculated in turn
simply by changing the AVERTIME card as follows:
CO
CO
CO
AVERTIME
AVERTIME
AVERTIME
APR
JUL
OCT
MAY
AUG
NOV
JUN
SEP
DEC
PERIOD
PERIOD
PERIOD
(for
(for
(for
Quarter
Quarter
Quarter
2)
3)
4)
3-90
-------
By specifying MONTH on the ME STARDATA card, the model will be able to retrieve the
correct STAR summary for each of these cases. The only requirement is that STAR
summaries always be included in the following order within the meteorological data file:
JAN, FEB, MAR DEC, WINTER (or QUART1), SPRING (or QUARTZ),
SUMMER (or QUARTS), FALL (or QUART4), and ANNUAL
Any number of STAR summaries may be included, up to a maximum of 17 (for 12 months, plus
4 seasons or quarters, plus 1 annual.
3.5.5 Specifying a Data Period to Process (Applies Only to ISCST)
There are two keywords that allow the user to specify particular days or ranges of
days to process from the sequential meteorological file input for the ISCST model. The
STARTEND keyword controls which period within the meteorological data file is read by
the model, while the DAYRANGE keyword controls which days or ranges of days (of those
that are read) for the model to process. The default for the model is to read the
entire meteorological data file (up to a full year) and to process all days within that
period.
The syntax and type for the STARTEND keyword are summarized below:
3-91
-------
Syntax: ME STARTEND Strtyr Strtmn Strtdy (Strthr) Endyr Endmn Enddy
(Endhr)
Type: Optional, Non-repeatable
where the Strtyr Strtmn Strtdy parameters specify the year, month and day of the first
record to be read (e.g., 87 01 31 for January 31, 1987), and the parameters Endyr Endmn
Enddy specify the year, month and day of the last record to be read. The Strthr and
Endhr are optional parameters that may be used to specify the start and end hours for
the data period to be read. If either Strthr or Endhr is to be specified, then both
must be specified. Any records in the data file that occur before the start date are
ignored, as are any records in the data file that occur after the end date. In fact,
once the end date has been reached, the model does not read any more data from the
meteorological file. If Strthr and Endhr are not specified, then processing begins with
hour 1 of the start date, and ends with hour 24 of the end date, unless specific days
are selected by the DAYRANGE card described below.
Any PERIOD or ANNUAL averages calculated by the model will apply only to the period
of data actually processed. Therefore, if someone wanted to calculate a six-month
average, they could select PERIOD averages on the CO AVERTIME card, and then specify the
period as follows:
ME STARTEND 87 01 01 87 06 30
3-92
-------
for the period January 1, 1987 through June 30, 1987. The difference between the PERIOD
and ANNUAL averages in the Short Term model is described in Section 3.2.3.1.
The syntax and type for the DAYRANGE keyword are summarized below:
Syntax: ME DAYRANGE Rangel RangeZ RangeS ... Rangen
Type: Optional, Repeatable
where the Range parameters specify particular days or ranges of days to process. The
days may be specified as individual days (e.g. 1 2 3 4 5) or as a range of days (e.g.
1-5). The user also has the option of specifying Julian day numbers, from 1 to 365 (366
for leap years), or specifying month and day (e.g., 1/31 for January 31). Any
combination of these may also be used. For example the following card will tell the
model to process the days from January 1 (Julian day 1) through January 31 (1/31) :
ME DAYRANGE 1-1/31
The DAYRANGE keyword is also repeatable, so that as many cards as needed may be included
in the ME pathway.
As with the STARTEND keyword, any PERIOD or ANNUAL averages calculated by the model
will apply only to the period of data actually processed. If the STARTEND keyword is
also used, then only those days selected on the DAYRANGE cards that fall within the
3-93
-------
period from the start date to the end date will be processed. Thus, if the ME pathway
included the following two cards:
ME STARTEND 87 02 01 87 12 31
ME DAYRANGE 1-31
then no data would be processed, since the days 1 through 31 fall outside the period 2/1
to 12/31.
3.5.6 Correcting Wind Direction Alignment Problems
The WDROTATE keyword allows the user to correct the input meteorological data for
wind direction alignment problems. All input wind directions or flow vectors are
rotated by a user-specified amount. Since the model results at particular receptor
locations are often quite sensitive to the transport wind direction, this optional
keyword should be used only with extreme caution and with clear justification.
The syntax and type of this keyword are summarized below:
Syntax: ME WDROTATE Rotang
Type: Optional, Non-repeatable
where the Rotang parameter specifies the angle in degrees to rotate the input wind
direction measurements. The value of Rotang is subtracted from the wind direction
3-94
-------
measurements. It may be used to correct for known (and documented) calibration errors,
or to adjust for the alignment of a valley if the meteorological station is located in a
valley with a different alignment than the source location. Since the Short Term models
use the flow vector (direction toward which the wind is blowing) as the basic input, the
WDROTATE keyword may also be used to convert input data as wind direction (from which
the wind is blowing) to flow vector by setting the parameter Rotang = 180.
3.5.7 Specifying Wind Speed Categories
Some of the parameters that may be input to the models are allowed to vary by wind
speed category. Examples of such inputs are user-specified wind speed profile
exponents, vertical potential temperature gradients, and variable emission rate factors.
The models use six wind speed categories, and these are defined by the upper bound wind
speed for the first five categories (the sixth category is assumed to have no upper
bound). The default values for the wind speed categories are as follows: 1.54, 3.09,
5.14, 8.23, and 10.8 m/s. The syntax and type of the WINDCATS keyword, which may be
used to specify different category boundaries, are summarized below:
Syntax: ME WINDCATS Wsl Ws2 Ws3 Ws4 Ws5
Type: Optional, Non-repeatable
where the Wsl through Ws5 parameters are the upper bound wind speeds of the first
through fifth categories in meters per second. The upper bound values are inclusive,
i.e., a wind speed equal to the value of Wsl will be placed in the first wind speed
category.
3-95
-------
3.5.8 Specifying Wind Profile Exponents
While the model uses default wind profile exponents if the regulatory default
option is selected (see the CO MODELOPT description in Section 3.2.2), for
non-regulatory default applications the user can specify wind profile exponents through
use of the WINDPROF keyword on the ME pathway. The syntax and type of this keyword are
summarized below:
Syntax: ME WINDPROF Stab Profl ProfZ ProfS Prof4 ProfB Prof6
Type: Optional, Repeatable
where the Stab parameter specifies the stability category for the following six values,
and Profl through Prof6 are the wind profile exponents for each of the six wind speed
categories. The Stab parameter may be input either alphabetically (A through F) or
numerically (1 for A through 6 for F). The WINDPROF cards do not need to be input in
any particular order.
The wind speed categories are either the default categories used by the model (with
upper bound speeds of 1.54, 3.09, 5.14, 8.23, and 10.8 m/s for the first five categories
- the sixth category is assumed to have no upper bound), or the categories specified by
the user on the optional ME WINDCATS keyword (Section 3.5.6).
3-96
-------
The following example will input the default exponents for the rural mode, and
illustrates the use of a repeat value for applying the exponents to all six wind speed
categories:
ME
ME
ME
ME
ME
ME
WINDPROF
WINDPROF
WINDPROF
WINDPROF
WINDPROF
WINDPROF
A
B
C
D
E
F
6*0
6*0
6*0
6*0
6*0
6*0
07
07
10
15
35
55
If the regulatory default option has been selected, then any inputs on the WINDPROF
keyword are ignored by the model, and a non-fatal warning message is generated.
3.5.9 Specifying Vertical Temperature Gradients
While the model uses default vertical potential temperature gradients if the
regulatory default option is selected (see the CO MODELOPT description in Section
3.2.2), for non-regulatory default applications the user can specify vertical potential
temperature gradients through use of the DTHETADZ keyword on the ME pathway. The
syntax and type of this keyword are summarized below:
Syntax: ME DTHETADZ Stab Dtdzl DtdzZ DtdzS Dtdz4 Dtdz5 Dtdz6
Type: Optional, Repeatable
3-97
-------
where the Stab parameter specifies the stability category for the following six values,
and Dtdzl through Dtdz6 are the vertical potential temperature gradients for each of the
six wind speed categories. The Stab parameter may be input either alphabetically (A
through F) or numerically (1 for A through 6 for F). The DTHETADZ cards do not need to
be input in any particular order.
The wind speed categories are either the default categories used by the model (with
upper bound speeds of 1.54, 3.09, 5.14, 8.23, and 10.8 m/s for the first five categories
- the sixth category is assumed to have no upper bound), or the categories specified by
the user on the optional ME WINDCATS keyword (Section 3.5.6) .
The following example will input the default values of DTDZ, and illustrates the
use of a repeat value for applying the inputs to all six wind speed categories:
ME
ME
ME
ME
ME
ME
DTHETADZ
DTHETADZ
DTHETADZ
DTHETADZ
DTHETADZ
DTHETADZ
A
B
C
D
E
F
6*0
6*0
6*0
6*0
6*0
6*0
00
00
00
00
020
035
If the regulatory default option has been selected, then any inputs on the DTHETADZ
keyword are ignored by the model, and a non-fatal warning message is generated.
3-98
-------
3.5.10 Specifying Average Wind Speeds for the Long Term Model
The ISC Long Term model uses joint frequencies of wind speed class by wind
direction sector by stability category as the basic meteorological input to the model.
These STAR summaries (for STability ARray) are described in more detail in Section
3.5.1.2. The optional AVESPEED keyword on the ME pathway allows the user to specify the
median wind speed for each of the wind speed categories in the STAR summary. The syntax
and type of this keyword are summarized below:
Syntax: ME AVESPEED wsi Ws2 Ws3 Ws4 Ws5 Ws6
Type: Optional, Non-repeatable
where the Wsl through Ws6 parameters are the median wind speeds (m/s) for each of the
six wind speed categories. The default values used by the model in the absence of the
AVESPEED keyword are as follows: 1.50, 2.50, 4.30, 6.80, 9.50, and 12.50 m/s.
3.5.11 Specifying Average Temperatures for the Long Term Model
For the ISC Long Term model, the user must specify average values of ambient
temperature following the AVETEMPS keyword. The following syntax is used:
Syntax: ME AVETEMPS Aveper Tal Ta2 Ta3 Ta4 Ta5 Ta6
Type: Mandatory, Repeatable
where the Aveper parameter specifies the long term averaging period for the following
inputs, and must be one of the secondary keywords used on the Long Term AVERTIME card
3-99
-------
described in Section 3.2.3.2 (e.g., JAN, WINTER, ANNUAL, etc.). The Tal through Ta6
parameters are the average ambient temperatures (K) for each of the six stability
categories, A through F. The AVETEMPS keyword is repeated for each of the averaging
periods being processed. Common practice is to apply the average daily maximum
temperature for the time period being modeled to stability classes A, B and C, the
average daily minimum temperature to stability classes E and F, and the average daily
temperature to stability class D. These average temperatures may be obtained from
various climatological summaries, including the Local Climatological Data - Annual
Summary published for major National Weather Service stations by the National Climatic
Data Center in Asheville, North Carolina.
The following example illustrates the use of the AVETEMPS keyword:
ME AVETEMPS WINTER 3*280.0 275.0 2*270.0
ME AVETEMPS SPRING 3*285.0 280.0 2*275.0
ME AVETEMPS SUMMER 6*293.0
ME AVETEMPS FALL 280. 280. 275. 270. 265. 265.
where repeat values have been used for the unstable and stable classes for winter and
spring, and for all classes for summer.
3.5.12 Specifying Average Mixing Heights for the Long Term Model
Fop the ISC Long Term model, the user must specify average values of mixing height
following the AVEMIXHT keyword. The following syntax is used:
3-100
-------
Syntax: ME AVEMIXHT Aveper Stab Mixhtl MixhtZ MixhtS Mixht4 MixhtB
Mixht6
Type: Mandatory, Repeatable
where the Aveper parameter specifies the long term averaging period for the following
inputs, and must be one of the secondary keywords used on the Long Term AVERTIME card
described in Section 3.2.3.2 (i.e., JAN, WINTER, ANNUAL, etc.) The Stab parameter
specifies the stability category (A through F or 1 through 6). The Mixhtl through
Mixht6 parameters are the average mixing heights (m) for each of the six wind speed
categories. The AVEMIXHT keyword is repeated for each stability category and for each
of the averaging periods being processed. For mixing heights in rural areas, the common
practice is to apply the mean afternoon mixing height given by Holzworth (1972) to
stability classes B, C and D, and 1.5 times the mean afternoon mixing height to
stability class A. For mixing heights in urban areas, the common practice is to apply
the mean afternoon mixing height given by Holzworth (1972) to stability classes B and C,
1.5 times the mean afternoon mixing height to stability class A, and the average of the
mean early morning and afternoon mixing heights to stability class D. The ISCLT model
assumes unlimited mixing for stability classes E and F for both rural and urban
conditions, and a large value such as 10,000 meters may be input for those classes. It
is also common practice to apply the average mixing height to all wind speed classes for
a particular stability class, although if better information is available, separate
values may be input by wind speed class.
The following example illustrates the use of the AVEMIXHT keyword:
3-101
-------
ME
ME
ME
ME
ME
ME
AVEMIXHT
AVEMIXHT
AVEMIXHT
AVEMIXHT
AVEMIXHT
AVEMIXHT
WINTER
WINTER
WINTER
WINTER
WINTER
WINTER
A
B
C
D
E
F
6
6
6
6
6
6
•A-
•A-
•A-
•A-
•A-
•A-
2250
2000
1500
1000
500.
300.
.0
.0
.0
.0
0
0
where repeat values have been used to apply the mixing heights to each of the wind speed
categories.
3.5.13 Specifying Average Surface Roughness for the Long Term Model
When using the dry deposition algorithms in ISCLT, the user must specify average
values of surface roughness length following the AVEROUGH keyword. The following syntax
is used:
Syntax: ME AVEROUGH Aveper Z0
Type: Optional, Repeatable
where the Aveper parameter specifies the long term averaging period for the following
input, and must be one of the secondary keywords used on the Long Term AVERTIME card
described in Section 3.2.3.2 (e.g., JAN, WINTER, ANNUAL, etc.). The Z0 parameter is the
average surface roughness length in meters for the specified averaging period. Only one
roughness length is supplied for each averaging period. Surface roughness lengths
representative of several land-use types are given in Table 3-2 by season. Depending on
the land-use type and climate, surface roughness may vary considerable by season, as
shown for deciduous forests in Table 3-2.
3-102
-------
TABLE 3-2
SURFACE ROUGHNESS LENGTH, METERS, FOR LAND-USE TYPES AND
SEASONS, FROM SHIEH ET AL., 1979
1.
2 .
3.
4.
5.
6.
7.
8.
Land-Use
Type
Water Surface
Deciduous
Coniferous
Swamp
Cultivated
Grassland
Urban
Forest
Forest
Land
Desert Shrubland
Spring
0.
1
1
0
0
0
1
0
0001
.00
.30
.20
.03
.05
.00
.30
Summer
0.
1
1
0
0
0
1
0
0001
.30
.30
.20
.20
.10
.00
.30
Autumn
0.
0
1
0
0
0
1
0
0001
.80
.30
.20
.05
.01
.00
.30
Winter
0.0001
0
1
0
0
0.
1
0
.50
.30
.05
.01
001
.00
.15
Definitions of Seasons:
Spring: Periods when vegetation is emerging or partially
green. This is a transitional situation that applies
for 1-2 months after the last killing frost in
spring.
Summer: Periods when vegetation is lush and healthy, typical
of mid-summer, but also of other seasons where frost
is less common.
Autumn: Periods when freezing conditions are common,
deciduous trees are leafless, crops are not yet
planted or are already harvested (bare soil exposed),
3-103
-------
3.6 TERRAIN GRID PATHWAY INPUTS AND OPTIONS
The Terrain Grid pathway contains keywords that define the input terrain grid data
used in calculating dry depletion in elevated or complex terrain. The TG pathway is an
optional pathway for the ISC models. If dry depletion is not being calculated, then the
TG pathway may be omitted. If dry depletion is being calculated and the TG pathway is
omitted, then the model will linearly interpolate between the source base elevation and
the receptor elevation when calculating dry depletion.
The TG pathway includes two mandatory, non-repeatable keywords, and one optional
keyword. The INPUTFIL keyword identifies the name of the input file containing the TG
data. The syntax and type of the TG INPUTFIL keyword are summarized below:
Syntax: TG INPUTFIL Tgfile
Type: Mandatory, Non-repeatable
where the Tgfile parameter is a character field of up to 40 characters that identifies
the filename for the terrain grid data file. The Tgfile parameter may include the
complete DOS pathname for the file when running the model on an IBM-compatible PC.
The TG LOCATION keyword is used to specify the location of the terrain grid data
relative to the coordinate system used to define the source and receptor locations.
The terrain grid data file must be in UTM coordinates, while the source/receptor
coordinates may be in a user specified coordinate system, such as plant coordinates.
The syntax and type of the TG LOCATION keyword are summarized below:
3-104
-------
Syntax: TG LOCATION Xorig Yorig (Units)
Type: Mandatory, Non-repeatabl e
where the Xorig and Yorig parameters are the values needed to transform the locations
given in user-specified coordinates for sources and receptors to UTM coordinates. The
user coordinates are transformed by adding Xorig and Yorig to the x-coordinates and y-
coordinates, respectively, of the sources and receptors. The optional Units parameter
is used to specify the units for the Xorig and Yorig parameters only. The units may be
specified as FEET, KM, or METERS. The default units for Xorig and Yorig is in meters if
the Units parameter is omitted. For example, if the source and receptor coordinates in
the runstream file are in UTM coordinates, then the TG LOCATION card should have a value
of 0.0 for Xorig and Yorig, since no conversion is needed to match up the
source/receptor locations to the terrain grid data. If the source and receptor
coordinates in the runstream file are in a different (non-UTM) coordinate system, such
as a plant-based system, then the Xorig and Yorig parameters should be the UTM
coordinates for the origin (x=0, y=0) of the source/receptor coordinate system. The
values of Xorig and Yorig are added to the source and receptor coordinates to convert
them to UTM coordinates. An example of the TG pathway is shown below:
TG STARTING
TG INPUTFIL C:\TERRAIN\GRIDELEV.MSL
TG LOCATION 532.2 4391.74 KM
TG FINISHED
3-105
-------
The terrain grid file contains 1 header record, followed by any number of data
records. The file is read as a free-format ASCII file. The header record contains the
following information:
nx, ny, xllm, yllm, xurm, yurm, sizem
where:
nx, ny number of data points in x (Easting) and y (Northing) directions;
xllm, yllm
xurm, yurm
sizem
UTM coordinates (in meters) of the point at the lower left corner of
the grid;
UTM coordinates (in meters) of the point at the upper right corner
of the grid; and
spacing between grid points in both the x and y directions, in
meters.
The data records are ordered by rows. The first row contains nx terrain elevations
ordered from west to east, starting at point (XLLM, YLLM). Row 2 contains the data for
the next row to the north in the grid. There are a total of ny rows of data in the
terrain grid file. The default units for terrain elevations in the terrain grid file
are meters MSL. However, the user may specify terrain elevations to be in units of feet
by adding the optional TG ELEVUNIT FEET card. The order of the ELEVUNIT card on the TG
pathway is not important. The maximum number of points in the terrain grid file is
controlled by the MXTX and MXTY parameters in the DEPVAR.INC file.
3-106
-------
3.7 EVENT PATHWAY INPUTS AND OPTIONS (APPLIES ONLY TO ISCEV)
The ISCEV (EVENT) model is specifically designed to facilitate analysis of source
contributions to specific events for short term averages (less than or equal to 24
hours). These events may be design concentrations generated by the ISCST model,
occurrences of violations of an air quality standard, or user-specified events. These
events are input to the ISCEV model through the EVent pathway. Each event is defined
by an averaging period and specific data period, a source group, and a receptor
location. Since the locations are only of interest in combination with particular
averaging and data periods, the REceptor pathway is not used by the EVENT model.
There are two keywords that are used to define the events on the EV pathway. The
EVENTPER keyword defines the averaging period, data period and source group, while the
EVENTLOC keyword defines the receptor location for the event. Each event is also given
an alphanumeric name that links the two input cards for that event.
The syntax and type of the EVENTPER and EVENTLOC keywords are summarized below:
Syntax: EV EVENTPER Evname Aveper Grpid Date
Syntax: EV EVENTLOC Evname XR= Xr YR= Yr (Zelev) (Zflag)
or Evname RNG= Rng DIR= Dir (Zelev) (Zflag)
Type: Mandatory, Repeatable
where the parameters are as follows:
Evname - event name (an alphanumeric string of up to 8 characters),
3-107
-------
Aveper - averaging period for the event (e.g. I, 3_, 8., 24 hr)
Grpid - source group ID for the event (must be defined on SO pathway),
Date - date for the event, input as an eight digit integer for the ending hour of
the data period (YYMMDDHH), e.g. 84030324 defines a data period ending at
hour 24 on March 3, 1984. The length of the period corresponds to Aveper.
XR= - X-coordinate (m) for the event location, referenced to a Cartesian
coordinate system
YR= - Y-coordinate (m) for the event location, referenced to a Cartesian
coordinate system
RNG= - distance range (m) for the event location, referenced to a polar
coordinate system with an origin of (0., 0.)
DIR= - radial direction (deg.) for the event location, referenced to a polar
coordinate system with an origin of (0., 0.)
Zelev - optional terrain elevation for the event location (m)
Zflag - optional receptor height above ground (flagpole receptor) for the event
location (m)
Each event is defined by the two input cards EVENTPER and EVENTLOC, and these inputs are
linked by the event name, which must be unique among the events being processed in a
given run. There is no particular requirement for the order of cards on the EV pathway.
Note that the location for the event may be specified by either Cartesian coordinates or
by polar coordinates, however, the polar coordinates must be relative to an origin of
(0,0) .
3-108
-------
3.7.1 Using Events Generated by the ISCST Model
Since the ISCEV (EVENT) model was designed to work in conjunction with the ISCST
model, the ISCST model has an option (CO EVENTFIL described in Section 3.2.9) to
generate an input file for the ISCEV model. When this option is used, the ISCST model
copies relevant inputs from the ISCST runstream input file to the ISCEV model input
file, and generates the inputs for the EVent pathway from the results of the modeling
run. These events are the design concentrations identified by the OU RECTABLE keyword
(see Section 3.8.1.1), such as the highest and high-second-high 24-hour averages, etc.,
and any threshold violations identified by the OU MAXIFILE keyword (see Section
3.8.1.2). The inputs generated by the ISCST model correspond to the syntax described
above for the EVENTPER and EVENTLOC keywords. The locations for events generated by the
ISCST model are always provided as Cartesian coordinates.
To easily identify the events generated by the ISCST model, and to provide a
mechanism for the ISCST model to manage the events generated from the model run, a
naming convention is used for the EVNAME parameter. The following examples illustrate
the event names used by the ISCST model:
H1H01001 - High-first-high 1-hour average for source group number 1
H2H24003 - High-second-high 24-hour average for source group number 3
TH030010 - Threshold violation number 10 for 3-hour averages
TH240019 - Threshold violation number 19 for 24-hour averages
3-109
-------
The high value design concentrations are listed first in the ISCEV model input file,
followed by the threshold violations (grouped by averaging period). To make it easier
for the user to review the ISCEV model input file generated by the ISCST model, and
determine which events are of most concern, the actual concentration or deposition value
associated with the event is included as the last field on the EVENTPER card. This
field is ignored by the ISCEV model, and is included only for informational purposes.
The user should be aware that the same event may appear in the ISCEV model input file as
both a design value and as a threshold violation, depending on the options selected and
the actual results. Since the model processes the events by date sequence and outputs
the results for each event as it is processed, the order of events in the output file
will generally not follow the order of events in the input file, unless all of the
events were generated by the MAXIFILE option.
3.7.2 Specifying Discrete Events
The user can specify discrete events by entering the EVENTPER and EVENTLOC cards as
described above. The averaging period and source group selected for the event must be
among those specified on the CO AVERTIME and SO SRCGROUP cards. If the ISCEV model
input file was generated by the ISCST model, the user may include additional events for
those averaging periods and source groups used in the original ISCST model run. They
may also add averaging periods or define new source groups in the ISCEV model input file
in order to define additional events.
3-110
-------
3.8 OUTPUT PATHWAY INPUTS AND OPTIONS
The Output pathway contains keywords that define the output options for the model
runs. Since the output options are somewhat different for each of the three models, the
OU pathway options for the models are discussed separately.
3.8.1 Short Term Model Options
The ISCST model has three keywords that control different types of tabular output
for the main output file of the model, and four keywords that control separate output
file options for specialized purposes. The user may select any combination of output
options for a particular application. For each tabular output option specified by the
user, the model will cycle through the selected output types in the following order -
CONG, DEPOS, DDEP, and/or WDEP. For the POSTFILE and PLOTFILE output options, the model
will list the selected output types in the order given above, as described below for
each file option. For the MAXIFILE and TOXXFILE output options, the output will only
include the first output type selected from the list given above, since outputs from
these options are based on a value exceeding a threshold.
3.8.1.1 Selecting Options for Tabular Printed Outputs.
The three tabular printed output options are controlled by the following keywords:
RECTABLE - Controls output option for high value summary tables by receptor;
MAXTABLE - Controls output option for overall maximum value summary tables; and
3-111
-------
DAYTABLE - Controls output option for tables of concurrent values summarized by
receptor for each day processed.
The keywords are described in more detail in the order listed above.
The syntax and type for the RECTABLE keyword are summarized below:
Syntax: ou RECTABLE Aveper FIRST SECOND ... SIXTH or IST ZND
6TH
Type: Optional, Repeatable
where the Aveper parameter is the short term averaging period (e.g. JL, 3., 8. or 24 hr or
MONTH) for which the receptor table is selected, and the secondary keywords, FIRST,
SECOND, etc., indicate which high values are to be summarized by receptor for that
averaging period. The RECTABLE card may be repeated for each averaging period. For
cases where the user wants the same RECTABLE options for all short term averaging
periods being modeled, the input may be simplified by entering the secondary keyword
ALLAVE for the Aveper parameter. The following example will select summaries of the
highest, second highest and third highest values by receptor for all averaging periods:
OU RECTABLE ALLAVE FIRST SECOND THIRD
3-112
-------
The model will also recognize a range of high values on the RECTABLE input card, and
therefore the following card will have the effect:
OU RECTABLE ALLAVE FIRST-THIRD
The output file will include tables for only the high values selected. Tables for
all source groups for a particular averaging period are grouped together, and the
averaging periods are output in the order that they appear the CO AVERTIME card. For
each averaging period and source group combination, the tables of high values for the
receptor networks (if any) are printed first, followed by any discrete Cartesian
receptors, any discrete polar receptors, and any boundary receptors.
The number of high values per receptor that the model can store is controlled by
the NVAL PARAMETER in the Fortran computer code. The value of NVAL is initially set at
2 for the DOS version of ISCST and 6 for the extended memory version. The NVAL
PARAMETER can be changed (up to 10), and the model recompiled in order to meet other
modeling needs, such as the highest of the sixth highest values by receptor for PM-10
modeling, assuming sufficient memory is available for the model's storage requirements.
Changing the model storage limits is discussed in more detail in Section 4.2.2.
If the CO EVENTFIL keyword has been used to generate an input file for the ISCEV
(EVENT) model, then the design values identified by the RECTABLE options, e.g., the
high-second-high 24-hour average, are included in the events that are defined in the
ISCEV model input file.
3-113
-------
The syntax and type for the MAXTABLE keyword are summarized below:
Syntax: OU MAXTABLE Aveper Maxnum
Type: Optional, Repeatable
where the Aveper parameter is the short term averaging period (e.g. JL, 3., 8. or 24 hr or
MONTH) for which the receptor table is selected, and the Maxnum parameter specifies the
number of overall maximum values to be summarized for each averaging period. The
MAXTABLE card may be repeated for each averaging period. As with the RECTABLE keyword,
for cases where the user wants the same MAXTABLE options for all short term averaging
periods being modeled, the input may be simplified by entering the secondary keyword
ALLAVE for the Aveper parameter. The following example will select the maximum 50 table
for all averaging periods:
OU MAXTABLE ALLAVE 50
A separate maximum overall value table is produced for each source group. The
maximum value tables follow the RECTABLE outputs in the main print file. All source
group tables for a particular averaging period are grouped together, and the averaging
periods are output in the order that they appear on the CO AVERTIME card.
The number of overall maximum values that the model can store for each averaging
period and source group is controlled by the NMAX PARAMETER in the Fortran computer
code. The value of NMAX is initially set at 50. The NMAX PARAMETER can be changed (up
3-114
-------
or down), and the model recompiled in order to meet other modeling needs, assuming
sufficient memory is available for the model's storage requirements. Changing the model
storage limits is discussed in more detail in Section 4.2.2.
The syntax and type for the DAYTABLE keyword are summarized below:
Syntax: OU DAYTABLE Avperl AvperZ AvperS
Type: Optional, Non-repeatable
where the Avpern parameters are the short term averaging periods (e.g. JL, 3., 8. or 24 hr
or MONTH) for which the daily tables are selected. The DAYTABLE card is non-repeatable,
but as with the RECTABLE and MAXTABLE keywords, for cases where the user wants daily
tables for all short term averaging periods being modeled, the input may be simplified
by entering the secondary keyword ALLAVE for the first parameter. The following example
will select the daily tables for all averaging periods:
OU DAYTABLE ALLAVE
For each averaging period for which the DAYTABLE option is selected, the model will
print the concurrent averages for all receptors for each day of data processed. The
receptor networks (if any) are printed first, followed by any discrete Cartesian
receptors, discrete polar receptors, and boundary receptors. Results for each source
group are output. For example, if 1, 3, and 24-hour averages are calculated, and the OU
DAYTABLE ALLAVE option is used, then for the first day of data processed, there will be
3-115
-------
24 sets of tables of hourly averages (one for each hour in the day), eight sets of
3-hour averages (one for each 3-hour period in the day), and one set of 24-hour
averages. The averages are printed as they are calculated by the model, but for hours
where more than one averaging period is calculated (e.g., hour 24 is the end of an
hourly average, a 3-hour average, and a 24-hour average), the order in which the
averages are output will follow the order used on the CO AVERTIME card. Note: This
option can produce very large output files, especially when used with a full year of
data and very short period averages, such 1-hour and 3-hour. It should therefore be used
with CAUTION.
3.8.1.2 Selecting Options for Special Purpose Output Files.
The ISCST model provides options for four types of output files for specialized
purposes. One option produces files of all occurrences of violations of user-specified
threshold values (MAXIFILE keyword), another option produces files of concurrent (raw)
results at each receptor suitable for post-processing (POSTFILE keyword), and a third
option produces files of design values that can be imported into graphics packages in
order to produce contour plots (PLOTFILE keyword), and a fourth option produces
unformatted files of raw results above a threshold value with a special structure for
use with the TOXX model component of TOXST (TOXXFILE keyword). Each of these options is
described in detail below.
The syntax and type for the MAXIFILE keyword are summarized below:
3-116
-------
Syntax: OU MAXIFILE Aveper Grpid Thresh Filnam (Funit)
Type: Optional, Repeatable
where the Aveper parameter is the short term averaging period (e.g. 3., 8., 24 for 3, 8
and 24-hour averages, or MONTH for monthly averages) and Grpid is the source group ID
for which the MAXIFILE option is selected. The Thresh parameter is the user-specified
threshold value, and Filnam is the name of the file where the MAXIFILE results are to be
written. The optional Funit parameter allows the user the option of specifying the
Fortran logical file unit for the output file. The user-specified file unit must be in
the range of 20-100, inclusive. By specifying the same filename and unit for more than
one MAXIFILE card, results for different source groups and/or averaging periods may be
combined into a single file. If the Funit parameter is omitted, then the model will
dynamically allocate a unique file unit for this file (see Section 3.9.2).
The MAXIFILE card may be repeated for each combination of averaging period and
source group, and a different filename should be used for each file. The resulting
maximum value file will include several header records identifying the averaging period,
source group and the threshold value for that file, and a listing of every occurrence
where the result for that averaging period/source group equals or exceeds the threshold
value. Each of these records includes the averaging period, source group ID, date for
the threshold violation (ending hour of the averaging period), the x, y, z and flagpole
receptor height for the receptor location where the violation occurred, and the
concentration or deposition value. If more than one output type is selected in a model
run, then the MAXIFILE threshold will only apply to the first output type selected among
the list of CONG, DEPOS, DDEP, and/or WDEP, and only the corresponding value will be
output in the maximum value file.
3-117
-------
Each of the threshold violations, except for monthly averages, identify events that
may be modeled for source contribution information with the ISCEV (EVENT) model by
selecting the CO EVENTFIL option (see Sections 3.2.9 and 3.7). Each of the threshold
violations is included as an event on the EV pathway, and is given a name of the form
THxxyyyy, where xx is the averaging period, and yyyy is the violation number for that
averaging period. For example, an event name of TH240019 identifies the 19th threshold
violation for 24-hour averages. Monthly average threshold violations are included in the
file specified on the MAXIFILE card, but are not included in the ISCEV model input file
since the ISCEV model currently handles only averaging periods of up to 24 hours.
The following examples illustrate the use of the MAXIFILE option:
ou
ou
ou
ou
ou
MAXIFILE
MAXIFILE
MAXIFILE
MAXIFILE
MAXIFILE
24
24
3
3
MONTH
ALL
PSD
PSD
PLANT
ALL
364
91
365
25
10
0
0
0
0
0
MAX24ALL.OUT
MAXPSD.OUT 50
MAXPSD.OUT 50
C:\OUTPUT\MAXI3HR
MAXMONTH.OUT
FIL
where the 3-hour example illustrates the use of a DOS pathname for the PC, and the last
example illustrates the use of monthly averages. The FILNAM parameter may be up to 40
characters in length. It should also be noted that only one MAXIFILE card may be used
for each averaging period/source group combination. Note: The MAXIFILE option may
produce very large files for runs involving a large number of receptors if a
significant percentage of the results exceed the threshold value.
3-118
-------
The syntax and type for the POSTFILE keyword are summarized below:
Syntax: OU POSTFILE Aveper Grpid Format Filnam (Funit)
Type: Optional, Repeatable
where the Aveper parameter is the averaging period (e.g. 3., 8., 24 for 3, 8 and 24-hour
averages, MONTH for monthly averages, PERIOD for period averages, or ANNUAL for annual
averages) and Grpid is the source group ID for which the POSTFILE option is selected.
The Format parameter specifies the format of the POSTFILE output, and may either be the
secondary keyword UNFORM for unformatted concentration files, or the secondary keyword
PLOT to obtain formatted files of receptor locations (x- and y-coordinates) and
concentrations suitable for plotting contours of concurrent values. The Filnam
parameter is the name of the file where the POSTFILE results are to be written. The
optional Funit parameter allows the user the option of specifying the Fortran logical
file unit for the output file. The user-specified file unit must be in the range of
20-100, inclusive. By specifying the same filename and unit for more than one POSTFILE
card, results for different source groups and/or averaging periods may be combined into
a single file. If the Funit parameter is omitted, then the model will dynamically
allocate a unique file unit for this file (see Section 3.9.2).
The POSTFILE card may be repeated for each combination of averaging period and
source group, and a different filename should be used for each file. If UNFORM is
specified for the Format parameter, then the resulting unformatted file includes a
constant-length record for each of the selected averaging periods calculated during the
model run. The first variable of each record is an integer variable (4 bytes)
containing the ending date (YYMMDDHH) for the averages on that record. The second
3-119
-------
variable for each record is an integer variable (4 bytes) for the number of hours in the
averaging period. The third variable for each record is a character variable of length
eight containing the source group ID. The remaining variables of each record contain
the calculated average concentration or total deposition values for all receptors, in
the order in which they were defined in the input runstream.
The following examples illustrate the use of the POSTFILE option:
OU POSTFILE 24 ALL UNFORM PST24ALL.BIN
OU POSTFILE 24 PSD UNFORM PST24PSD.BIN
OU POSTFILE 3 PLANT UNFORM C:\BINOUT\PST3HR.FIL
OU POSTFILE MONTH ALL PLOT PSTMONTH.PLT
OU POSTFILE PERIOD ALL PLOT PSTANN.PLT
where the 3-hour example illustrates the use of a DOS pathname for the PC, and the last
example illustrates the use of monthly averages. The Filnam parameter may be up to 40
characters in length. The use of separate files for each averaging period/source group
combination allows the user flexibility to select only those results that are needed for
post-processing for a particular run, and also makes the resulting unformatted files
manageable. Note: The POSTFILE option can produce very large files, and should be used
with some caution. For a file of hourly values for a full year (8760 records) and 400
receptors, the resulting file will use about 14 megabytes of disk space. To estimate
the size of the file (in bytes), use the following equation:
(# of Hrs/Yr)
3-120
-------
File Size (bytes) = * (# of Rec + 4) * 4
(# of Hrs/Ave)
Divide the result by 1000 to estimate the number of kilobytes (KB) and divide by 1.OE6
to estimate the number of megabytes (MB).
When more than one output type is selected among the list of CONG, DEPOS, DDEP,
and/or WDEP, the post-processing output file will include all of the output types
selected, in the order listed here. For the unformatted post-processing file, the
results for each output type will be included on a single record for each averaging
period and source group. For the PLOT-formatted post-processing file, the results for
each output type will be printed in separate columns, one record per receptor, in the
order given above.
The syntax and type for the PLOTFILE keyword are summarized below:
Syntax: OU PLOTFILE Aveper Grpid Hivalu Filnam (Funit), or
OU PLOTFILE PERIOD Grpid Filnam (Funit)
OU PLOTFILE ANNUAL Grpid Filnam (Funit)
Type: Optional, Repeatable
where the Aveper parameter is the averaging period (e.g. 3_, 8., 24 for 3, 8 and 24-hour
averages, MONTH for monthly averages, PERIOD for period averages, or ANNUAL for annual
averages), Grpid is the source group ID for which the PLOTFILE option is selected, and
Hivalu specifies which short term high values are to be output (FIRST for the first
highest at each receptor, SECOND for the second highest at each receptor, etc.) Note
that the Hivalu parameter is not specified for PERIOD or ANNUAL averages, since there is
3-121
-------
only one period or annual average for each receptor. The Filnam parameter is the name
of the file where the PLOTFILE results are to be written. The optional Funit parameter
allows the user the option of specifying the Fortran logical file unit for the output
file. The user-specified file unit must be in the range of 20-100, inclusive. By
specifying the same filename and unit for more than one PLOTFILE card, results for
different source groups and/or averaging periods may be combined into a single file. If
the Funit parameter is omitted, then the model will dynamically allocate a unique file
unit for this file (see Section 3.9.2) .
The PLOTFILE card may be repeated for each combination of averaging period, source
group, and high value, and a different filename should be used for each file. The
resulting formatted file includes several records with header information identifying
the averaging period, source group and high value number of the results, and then a
record for each receptor which contains the x and y coordinates for the receptor
location, the appropriate high value at that location, and the averaging period, source
group and high value number. The data are written to the file in the order of x-coord,
y-coord, concentration (or deposition) so that the file can easily be imported into a
graphics package designed to generate contour plots. Many such programs will read the
PLOTFILEs directly without any modification, ignoring the header records, and produce
the desired plots.
The following examples illustrate the use of the PLOTFILE option:
3-122
-------
OU PLOTFILE 24 ALL
OU PLOTFILE 24 ALL
OU PLOTFILE 24 PSD
OU PLOTFILE 3 PSD
OU PLOTFILE 3 PLANT
OU PLOTFILE MONTH ALL
OU PLOTFILE PERIOD ALL
FIRST PLT24ALL.FST
SECOND PLT24ALL.SEC
2ND PLTPSD.OUT 75
2ND PLTPSD.OUT 75
1ST C:\PLOTS\PLT3HR.FIL
THIRD PLTMONTH.OUT
PSTANN.PLT
where the 3-hour example illustrates the use of a DOS pathname for the PC, and the last
example illustrates the use of monthly averages. As illustrated by the second and third
examples, the high value parameter may also be input as secondary keywords using the
standard abbreviations of 1ST, 2ND, 3RD, . . . 10TH. The Filnam parameter may be up to
40 characters in length. The use of separate files for each averaging period, source
group, high value combination allows the user flexibility to select only those results
that are needed for plotting from a particular run.
When more than one output type is selected among the list of CONG, DEPOS, DDEP,
and/or WDEP, the PLOTFILE output file will include all of the output types selected, in
the order listed here. The results for each output type will be printed in separate
columns, one record per receptor, in the order given above.
The syntax and type for the TOXXFILE keyword are summarized below:
Syntax: OU TOXXFILE Aveper Cutoff Filnam (Funit)
Type:
Optional, Repeatable
3-123
-------
where the Aveper parameter is the short term averaging period (e.g. I, 3_, 8., 24 for 1,
3, 8 and 24-hour averages, or MONTH for monthly averages) for which the TOXXFILE option
has been selected. The Cutoff (threshold) parameter is the user-specified threshold
cutoff value in g/m3, and Filnam is the name of the file where the TOXXFILE results are
to be written. It is important to note that the units of the Cutoff parameter are g/m3,
regardless of the input and output units selected with the SO EMISUNIT card. The
optional Funit parameter allows the user the option of specifying the Fortran logical
file unit for the output file. The user-specified file unit must be in the range of
20-100, inclusive. If the Funit parameter is omitted, then the model will dynamically
allocate a unique file unit for this file (see Section 3.8.2). While the TOXXFILE
option may be specified for any of the short term averaging periods that are identified
on the CO AVERTIME card for a particular run, a non-fatal warning message will be
generated if other than 1-hour averages are specified. This is because the TOXST model
currently supports only 1-hour averages.
The TOXXFILE card may be repeated for each averaging period, but a different
filename should be used for each file since the structure of the output file generated
by the TOXXFILE option does not allow for a clear way to distinguish between results for
different averaging periods. The resulting output file for the Short Term model is an
unformatted file with several header records identifying the title, averaging period,
receptor information, and the threshold value for that file, followed by records listing
every occurrence where the result for any source group for that averaging period equals
or exceeds the threshold value. When one of the source groups exceeds the threshold
value, the results for all source groups for that averaging period and receptor location
are output. Each concentration that is output through the TOXXFILE option is paired
with an integer ID variable that identifies the averaging period (hour number of the
3-124
-------
year), the source group number, and the receptor number corresponding to that value.
The concentration values and corresponding ID variables are stored in buffer arrays, and
the arrays are then written to the unformatted output file when full. The size of the
arrays is controlled by the NPAIR PARAMETER defined in the MAIN1.INC file, and is
initially set at 100. At the end of the modeling run, any values remaining in the
buffer arrays are written to the file, padded to the right with zeroes. The structure
of the output file generated by the TOXXFILE option is described in more detail in
Section 3.8.2 and in Appendix F. When using the TOXXFILE option, the user will normally
place a single source in each source group, and may need to modify the array storage
PARAMETERS in MAIN1.INC to accommodate certain modeling needs. The user should refer to
the user's guide for TOXST for further instructions on the application of the TOXXFILE
option of the ISCST model.
The following examples illustrate the use of the TOXXFILE option:
OU TOXXFILE 1 l.OE-5 TOXX1HR.BIN
OU TOXXFILE 24 2.5E-3 TOXX24HR.BIN 50
The Filnam parameter may be up to 40 characters in length. It should be noted that only
one TOXXFILE card may be used for each averaging period. Note: The TOXXFILE option may
produce very large files for runs involving a large number of receptors if a significant
percentage of the results exceed the threshold value. If more than one output type is
selected in a model run, then the TOXXFILE threshold will only apply to the first output
type selected among the list of CONG, DEPOS, DDEP, and/or WDEP, and only the
corresponding value will be output in the TOXXFILE output file.
3-125
-------
3.8.2 Short Term EVENT Model (ISCEV) Options
The ISC Short Term EVENT model (ISCEV) is designed specifically to perform source
contribution analyses for short term average (less than or equal to 24-hour) events.
The events may either be generated by the ISCST model, or they may be user-specified
events, or both. Because of this rather narrow focus of applications for the ISCEV
model, the output options are limited to a single keyword. The EVENTOUT keyword
controls the level of detail in the source contribution output from the EVENT model.
The syntax and type of the EVENTOUT keyword are summarized below:
Syntax: ou EVENTOUT SOCONT DETAIL
Type: Mandatory, Non-repeatabl e
where the SOCONT secondary keyword specifies the option to produce only the source
contribution information in the output file, and the DETAIL secondary keyword specifies
the option to produce more detailed summaries in the output file. The SOCONT option
provides the average concentration (or total deposition) value (i.e., the contribution)
from each source for the period corresponding to the event for the source group. The
basic source contribution information is also provided with the DETAIL option. In
addition, the DETAIL option provides the hourly average concentration (or total
deposition) values for each source for every hour in the averaging period, and a summary
of the hourly meteorological data for the event period. In general, the DETAIL option
produces a larger output file than the SOCONT file, especially if there are a large
number of sources. There is no default setting for the EVENTOUT options.
3-126
-------
3.8.3 Long Term Model Options
The ISCLT model has three keywords available on the OU pathway to specify the
output options. The RECTABLE and MAXTABLE keywords are similar to the corresponding
keywords for the ISCST model in that RECTABLE specifies the options for tabular
summaries of results by receptor, and MAXTABLE specifies options for tabular summaries
of overall maximum results. The third keyword, PLOTFILE, is also similar to the
corresponding keyword for ISCST, and allows the user to generate separate output files
suitable for importing into graphics packages to generate contour plots. However, the
parameters on these keywords differ between the two models because of the different data
structures of the models.
For the Short Term model there are several short term averages during the data
period, from which the model sorts and stores the highest, second highest and third
highest values at each location, whereas for the Long Term model, there is only one long
term average result at each location. Because of these differences in the data
structure, the Long Term model is able to store the results for all sources at each
receptor location, in addition to the combined source group values. Therefore, the
output keywords for Long Term include options to summarize results for each source or
for the source groups, and also to provide source contribution information for the
maximum source group values (thereby eliminating the need for a Long Term EVENT model).
The syntax and type for the Long Term RECTABLE keyword are summarized below:
3-127
-------
Syntax: ou RECTABLE INDSRC and/or SRCGRP
Type: Optional, Non-repeatable
where the INDSRC secondary keyword specifies that summaries of individual sources for
each receptor are to be output, and the secondary keyword SRCGRP specifies that
summaries of source group values for each receptor are to be provided. The user may
select either option or both options in a given run. The individual source values are
presented first in the output file, with the results by receptor network followed by
any discrete Cartesian receptors, discrete polar receptors and boundary receptors. The
source group results follow the same pattern as the individual source tables. A
complete set of summary tables is output for each STAR summary processed, and for the
PERIOD averages, if calculated.
The syntax and type for the Long Term MAXTABLE keyword are summarized below:
Syntax: OU MAXTABLE Maxnum INDSRC and/or SRCGRP and/or SOCONT
Type: Optional, Non-repeatable
where the Maxnum parameter specifies the number of maximum values to summarize, and
where the INDSRC and SRCGRP secondary keywords specify that summaries of maximum values
for individual sources and for source groups, respectively, are to be provided. The
individual source maximum values are treated independently of the source group maxima
with the INDSRC option. To obtain the contribution from each source to the maximum
source group values (similar to the information obtained from ISCEV), the user may
select the SOCONT option. The user may select any combination of these options in a
given run. If the SOCONT option is selected, and the SRCGRP option has not been
3-128
-------
selected, the model will automatically determine the maximum source group values so that
the source contribution analysis can be performed, but the maximum source group values
will not be included in the output file. The individual source values are presented
first in the output file, followed by the maximum source group values, and the source
contribution results, according to the options selected. A complete set of maximum
value summary tables is output for each STAR summary processed, and for the PERIOD
averages, if calculated.
The number of overall maximum values that the model can store for each source and
source group is controlled by the NMAX PARAMETER in the Fortran computer code. The
value of NMAX is initially set at 10 for the Long Term model. The NMAX PARAMETER can be
changed (up or down), and the model recompiled in order to meet other modeling needs,
assuming sufficient memory is available for the model's storage requirements. Changing
the model storage limits is discussed in more detail in Section 4.2.2.
The syntax and type for the Long Term PLOTFILE keyword are summarized below:
Syntax: OU PLOTFILE Aveper Grpid Filnam (Funit)
Type: Optional, Repeatable
where the Aveper parameter is the long term averaging period (e.g. WINTER, SPRING, etc.)
and Grpid is the source group ID for which the PLOTFILE option is selected. The Filnam
parameter is the name of the file where the PLOTFILE results are to be written. The
optional Funit parameter allows the user the option of specifying the Fortran logical
file unit for the output file. The user-specified file unit must be in the range of
20-100, inclusive. If the Funit parameter is omitted, then the model will dynamically
3-129
-------
allocate a unique file unit for this file (see Section 3.8.2) . The PLOTFILE card may be
repeated for each combination of averaging period and source group, and a different
filename should be used for each file. The resulting formatted file includes several
records with header information identifying the averaging period and source group of the
results, and then a record for each receptor which contains the x and y coordinates for
the receptor location, the long term average value at that location, the averaging
period and the source group ID. The data are written to the file in the order of
x-coord, y-coord, concentration (or deposition) so that the file can easily be imported
into a graphics package designed to generate contour plots. Many such programs will
read the PLOTFILEs directly without any modification, although the user may have to
delete the header records to produce the desired plots.
The syntax and type for the Long Term TOXXFILE keyword are summarized below:
Syntax: OU TOXXFILE Aveper Grpid Filnam (Funit)
Type: Optional, Repeatable
where the Aveper parameter is the long term averaging period (e.g. WINTER, SPRING, etc.)
and Grpid is the source group ID for which the TOXXFILE option is selected. The PERIOD
average, if selected on the CO AVERTIME card, may also be specified for the Aveper
parameter for period averages. The optional Funit parameter allows the user the option
of specifying the Fortran logical file unit for the output file. The user-specified
file unit must be in the range of 20-100, inclusive. If the Funit parameter is omitted,
then the model will dynamically allocate a unique file unit for this file (see Section
3.8.2). The TOXXFILE card may be repeated for each combination of averaging period and
3-130
-------
source group, and a different filename should normally be used for each file. The
resulting formatted file includes several records with header information identifying
the averaging period and source group of the results, and then a record for each
receptor which contains the x and y coordinates for the receptor location, the long term
average value at that location, the averaging period and the source group ID. The data
are written to the file in the order of x-coord, y-coord, concentration (or deposition)
so that the file can easily be imported into a graphics package designed to generate
contour plots. Many such programs will read the TOXXFILEs directly without any
modification, although the user may have to delete the header records to produce the
desired plots. Each TOXXFILE output file includes the results for each source in the
specified source group, in the order in which they are defined on the SO pathway.
The example below illustrates the use of various Long Term model output options:
OU RECTABLE
OU MAXTABLE
OU PLOTFILE
OU PLOTFILE SPRING
OU PLOTFILE ANNUAL
OU TOXXFILE WINTER ALL
OU TOXXFILE PERIOD GROUP1
INDSRC SRCGRP
10 INDSRC SRCGRP SOCONT
WINTER ALL PLTWINT.OUT
PSD PSDSPRG.PLT
PLANT C:\PLOTS\PLANT.ALL
WINTTOXX.OUT 25
PERTOX.OUT
where all of the tabular printed output options have been selected, and several PLOTFILE
and TOXXFILE options have also been selected.
3-131
-------
3.9 CONTROLLING INPUT AND OUTPUT FILES
This section describes the various input and output files used by the ISC models,
and discusses control of input and output (I/O) on the IBM-compatible PC environment.
Much of this discussion also applies to operating the models in other environments.
3.9.1 Description of ISC Input Files
The two basic types of input files needed to run all of the ISC models are the
input runstream file containing the modeling options, source data and receptor data, and
the input meteorological data file. Each of these is discussed below, as well as a
special file that may be used to initialize the ISCST model with intermediate results
from a previous run.
3.9.1.1 Input Runstream File.
The input runstream file contains the user-specified options for running the
various ISC models, includes the source parameter data and source group information,
defines the receptor locations, specifies the location and parameters regarding the
meteorological data, and specifies the output options. The basic structure of the input
runstream file is the same for all three models, although the list of available
keywords for defining options, and the exact syntax for certain keywords are slightly
different between the Short Term and Long Term models. Details regarding the keywords
and parameters used in the input runstream file are provided in Section 3, and Appendix
B.
3-132
-------
For the PC-executable versions of the models available on the SCRAM BBS, the
runstream file is explicitly opened by the models using a Fortran OPEN statement, and
the integer variable, INUNIT, specifies the unit number for the file. The variable
INUNIT is initialized to a value of 5 in a BLOCK DATA subprogram of the model, which
corresponds to the default input unit for Fortran. The INUNIT variable is included in a
named COMMON block (FUNITS) in the MAIN1.INC include file, and is therefore available to
all of the necessary subroutines.
Since the input runstream file is opened explicitly by the PC-executable versions
of the models, the model will take the first parameter on the command line when running
the model as the input filename. No DOS redirection symbol should be used preceding the
runstream filename.
3.9.1.2 Meteorological Data File.
The input meteorological data is read into the models from a separate data file for
all three models. The meteorological filename and format are specified within the input
runstream file using the ME INPUTFIL keyword. The Short Term models accept
meteorological data from unformatted sequential files generated by the PCRAMMET and MPRM
preprocessors, and also accept a wide range of formatted ASCII files of hourly
sequential records. The Long Term model accepts STability ARray (STAR) meteorological
data from sequential ASCII files using either a default READ format, a user-specified
READ format or free-formatted READs.
The meteorological data file is explicitly opened by the models using a Fortran
OPEN statement, and the integer variable, MFUNIT, specifies the unit number for the
3-133
-------
file. The variable MFUNIT is initialized to a value of 19 in a BLOCK DATA subprogram of
the model. The MFUNIT variable is included in a named COMMON block (FUNITS) in the
MAIN1.INC include file, and is therefore available to all of the necessary subroutines.
3.9.1.3 Initialization File for Model Re-start.
The ISCST model has an optional capability to store intermediate results to an
unformatted (sometimes called binary) file for later re-starting of the model in the
event of a power failure or user interrupt. This unformatted file may therefore be used
as an input file to initialize the model. This option is controlled by the SAVEFILE
(saves intermediate results to a file) and the INITFILE (initialize result arrays from
a previously saved file) keywords on the CO pathway.
When initializing the model for the re-start option, the user specifies the name of
the unformatted results file on the INITFILE keyword. The default filename used if no
parameter is provided is SAVE.FIL. The initialization file is explicitly opened by the
ISCST model, and the integer variable, IRSUNT, specifies the unit number for the file.
The variable IRSUNT is initialized to a value of 15 in a BLOCK DATA subprogram of the
model. The IRSUNT variable is included in a named COMMON block (FUNITS) in the
MAIN1.INC include file, and is therefore available to all of the necessary subroutines.
3.9.2 Description of ISC Output Files
The ISC models produce a variety of output files, including the main print file of
model results, an unformatted file of intermediate results for later re-start of the
3-134
-------
model (ISCST only), and several output data files for specialized purposes. These files
are described in detail below.
3.9.2.1 Output Print File.
Each of the ISC models produces a main output print file of model results. The
contents and organization of this file for the ISCST model were shown in Figure 2-5.
This file includes an echo of the input runstream images at the beginning of the file
(up until a NO ECHO input is encountered). A summary of runstream setup messages and a
summary of the inputs follow the echo of inputs. The input summary includes a summary
of modeling options, source data, receptor data, and meteorological data, following the
same order as the pathways in the runstream file. If model calculations are performed,
then the model results are summarized next. The content and order of the model result
summaries depend on the output options selected and on the particular model being run.
Following the detailed model results are summary tables of the high values for each
averaging period and source group (ISCST only). The final portion of the main output
print file is the summary of messages for the complete model run.
For the PC-executable versions of the models available on the SCRAM BBS, the main
print output file is explicitly opened by the models using a Fortran OPEN statement, and
the integer variable, IOUNIT, specifies the unit number for the file. The variable
IOUNIT is initialized to a value of 6 in a BLOCK DATA subprogram of the model, which
corresponds to the default output unit for Fortran. The IOUNIT variable is included in
a named COMMON block (FUNITS) in the MAIN1.INC include file, and is therefore available
to all of the necessary subroutines.
3-135
-------
Since the main print output file is opened explicitly, the model will take the
second parameter on the command line when running the model as the output filename. No
DOS redirection symbol should be used preceding the output filename. If an output file
is not given on the command line, then the model will return an error message and abort
execution.
By opening the printed output file explicitly, the outputs are not automatically
formatted for the printer. This formatting is accomplished using the CARRIAGE CONTROL
specifier in the OPEN statement for the Lahey extended memory version of the models, and
by explicitly writing the ASCII form feed character to the file for the Microsoft DOS
version.
3.9.2.2 Detailed Error Message File.
The user may select an option for the model to save a separate file of detailed
error and other messages, through use of the CO ERRORFIL keyword. The format and syntax
of these messages is described in Appendix E. The order of messages within the file is
the order in which they were generated by the model. The file includes all types of
messages that were generated.
The error message file is explicitly opened by the model using a Fortran OPEN
statement, and the integer variable, IERUNT, specifies the unit number for the file.
The variable IERUNT is initialized to a value of 10 in a BLOCK DATA subprogram of the
model. The IERUNT variable is included in a named COMMON block (FUNITS) in the
MAIN1.INC include file, and is therefore available to all of the necessary subroutines.
3-136
-------
3.9.2.3 Intermediate Results File for Model Re-start.
The ISCST model has an optional capability to store intermediate results to an
unformatted (sometimes called binary) file for later re-starting of the model in the
event of a power failure or user interrupt. This unformatted file may therefore be used
as an input file to initialize the model. This option is controlled by the SAVEFILE
(saves intermediate results to a file) and the INITFILE (initialize result arrays from a
previously saved file) keywords on the CO pathway.
When saving the intermediate results for the re-start option, the user specifies
the name of the unformatted results file on the SAVEFILE keyword. The user has the
option of specifying a single filename, two filenames (for alternate saves), or
specifying no filename. The default filename used if no parameter is provided is
SAVE.FIL. If a single file is used, then the intermediate results file is overwritten
on each successive dump, with the chance that the file will be lost if the interrupt
occurs during the time that the file is opened. If two filenames are provided, then the
model also saves to the second file on alternate dumps, so that the next most recent
dump will always be available. The main save file is explicitly opened by the ISCST
model, and the integer variable, IDPUNT, specifies the unit number for the file. The
variable IDPUNT is initialized to a value of 12 in a BLOCK DATA subprogram of the model.
If a second save file is used, then it is also opened explicitly, and the integer
variable IDPUN2, initialized to a value of 14, specifies the unit number.
3-137
-------
3.9.2.4 Maximum Value/Threshold File.
The user may select an option for the ISCST model to generate a file or files of
concentration (or deposition) values exceeding a user-specified threshold. The OU
MAXIFILE keyword controls this option. The user may select separate files for each
averaging period and source group combination for which a list of threshold violations
may be needed. Each file includes several records with header information identifying
the averaging period, source group and threshold value, and then a record for every
occurrence where the result for that averaging period/source group equals or exceeds the
threshold value. Each of these records includes the averaging period, source group ID,
date for the threshold violation (ending hour of the averaging period), the x, y, z and
flagpole receptor height for the receptor location where the violation occurred, and the
concentration or deposition value.
The structure of the threshold violation file is described in more detail in
Appendix F. Each of the files selected by the user is opened explicitly by the model as
an formatted file. The filenames are provided on the input runstream image. The user
may specify the file unit on the MAXIFILE card through the optional FUNIT parameter.
User-specified units must be greater than or equal to 20, and are recommended to be less
than or equal to 100. If no file unit is specified, then the file unit is determined
internally according to the following formula:
IMXUNT = 100 + IGRP*10 + IAVE
where IMXUNT is the Fortran unit number, IGRP is the source group number (the order in
which the group is defined in the runstream file), and IAVE is the averaging period
3-138
-------
number (the order of the averaging period as specified on the CO AVERTIME card). This
formula will not cause any conflict with other file units used by the model for up to 9
source groups and up to 9 short term averaging periods.
3.9.2.5 Sequential Results File for Postprocessing.
The user may select an option for the ISCST model to generate a file or files of
concentration (or deposition) values suitable for postprocessing. The OU POSTFILE
keyword controls this option. The user may select separate files for each averaging
period and source group combination for which postprocessing may be needed. For each
file requested, the user has the option of specifying whether to use unformatted files
suitable for postprocessing or to use a plot format which could allow for inporting the
x,y,conc files into a graphics package for plotting. For the unformatted file option,
each file consists of sequential unformatted records of values at each receptor location
for every averaging period calculated. For the plot file format option, each file
consists of formatted records listing the x-coordinate, y-coordinate and concurrent
concentration (or deposition) values for each receptor and for all averaging periods
calculated. For certain applications, these files may become quite large, and should
only be used when needed, especially when using the plot format.
The structure of both types of postprocessing file is described in more detail in
Appendix F. Each of the postprocessing files selected by the user is opened explicitly
by the model as either an unformatted or a formatted file, depending on the option
selected. The filenames are provided on the input runstream image. The user may
specify the file unit on the POSTFILE card through the optional FUNIT parameter.
User-specified units must be greater than or equal to 20, and are recommended to be less
3-139
-------
than or equal to 100. If no file unit is specified, then the file unit is determined
internally according to the following formulas:
IPSUNT = 200 + IGRP*10 + IAVE for short term averages
IAPUNT = 300 + IGRP*10 - 5 for PERIOD averages
where IPSUNT and IAPUNT are the Fortran unit numbers, IGRP is the source group number
(the order in which the group is defined in the runstream file), and IAVE is the
averaging period number (the order of the averaging period as specified on the CO
AVERTIME card). This formula will not cause any conflict with other file units used by
the model for up to 9 source groups and up to 9 short term averaging periods.
3.9.2.6 High Value Summary File for Plotting.
The user may select an option for the ISCST model to generate a file or files of
the highest concentration (or deposition) values at each receptor suitable for importing
into a graphics package in order to generate contour plots. The OU PLOTFILE keyword
controls this option. The user may select separate files for each averaging period,
source group and high value combination for which a plot file may be needed. Each file
includes several records with header information identifying the averaging period,
source group and high value number of the results, and then a record for each receptor
which contains the x and y coordinates for the receptor location, the appropriate high
value at that location, and the averaging period, source group and high value number.
The structure of the plot file is described in more detail in Appendix F. Each of
the plot files selected by the user is opened explicitly by the model as an formatted
3-140
-------
file. The filenames are provided on the input runstream image. The user may specify
the file unit on the PLOTFILE card through the optional FUNIT parameter. User-specified
units must be greater than or equal to 20, and are recommended to be less than or equal
to 100. If no file unit is specified, then the file unit is determined internally
according to the following formulas:
IPLUNT = (IVAL+3)*100 + IGRP*10 + IAVE for short term aver.
IPPUNT = 300 + IGRP*10 for PERIOD averages
where IPLUNT and IPPUNT are the Fortran unit numbers, IVAL is the high value number (1
for FIRST highest, 2 for SECOND highest, etc.), IGRP is the source group number (the
order in which the group is defined in the runstream file), and IAVE is the averaging
period number (the order of the averaging period as specified on the CO AVERTIME card).
This formula will not cause any conflict with other file units used by the model for up
to 9 source groups and up to 9 short term averaging periods.
3.9.2.7 TOXX Model Input Files
The user may select an option for the ISCST model to generate an unformatted file
or files of concentration (or deposition) values exceeding a user-specified threshold
for use with the TOXX model component of TOXST. The OU TOXXFILE keyword controls this
option. The user may select separate files for each averaging period for which a
threshold violation file may be needed. Each file includes several records with header
information identifying the title, averaging period, threshold value, and receptor
network information, and then records including every occurrence where the result of any
source group for that averaging period equals or exceeds the threshold value. Records
3-141
-------
are also output that identify the averaging period (hour number of the year), source
group number and receptor number corresponding to the concentration values.
The structure of the threshold exceedance file for use with the TOXX model
component of TOXST is described in more detail in Appendix F. Each of the files
selected by the user is opened explicitly by the model as an unformatted file. The
filenames are provided on the input runstream image. The user may specify the file unit
on the TOXXFILE card through the optional Funit parameter. User-specified units must be
greater than or equal to 20, and are recommended to be less than or equal to 100. If no
file unit is specified, then the file unit is determined internally according to the
following formula:
ITXUNT = 300 + IAVE
where ITXUNT is the Fortran unit number, and IAVE is the averaging period number (the
order of the averaging period as specified on the CO AVERTIME card). This formula will
not cause any conflict with other file units used by the model for up to 4 short term
averaging periods.
The user may also select an option for the ISCLT model to generate an output for
use with the RISK model component of TOXLT. The OU TOXXFILE keyword also controls this
option. The user can specify a separate TOXXFILE for each long term averaging period
and source group combination. The TOXXFILE option may also be used for PERIOD averages
with the ISCLT model. The structure of the TOXXFILE output for ISCLT is very similar to
the long term PLOTFILE output, except that results are output for each individual source
in the specified source group. The structure of the long term TOXXFILE is described in
3-142
-------
more detail in Appendix F. Each of the files selected by the user is opened explicitly
by the model as a formatted file. The filenames are provided on the input runstream
image. The user may specify the file unit on the TOXXFILE card through the optional
Funit parameter. User-specified units must be greater than or equal to 20, and are
recommended to be less than or equal to 100. If no file unit is specified, then the
file unit is determined internally according to the following formulas:
ITXUNT = 500 + IAVE*10 + IGRP for long term averages
IPXUNT = 700 + IGRP*10 for PERIOD averages
where ITXUNT and IPXUNT are the Fortran unit numbers, IAVE is the averaging period
number (in the order of months, seasons or quarters, and annual), and IGRP is the source
group number (in the order is which the groups are defined in the SO pathway). This
formula will not cause any conflict with other file units used by the model for up to 9
source groups.
3.9.3 Control of File Inputs and Outputs (I/O)
3.9.3.1 Control of I/O on DOS PCs.
The main input runstream file and the main output print file are both specified on
the command line when running the models on a PC. Since the PC-executable file provided
explicitly opens these two files, there is no need to use DOS redirection of input and
output. Therefore, a standard command line to execute the ISCST model might look
something like this:
3-143
-------
C:\>ISCST3 TEST-ST.INP TEST-ST.OUT
where the "DOS prompt" has been given as "C:\>", but may look different on different
systems, or may include a subdirectory specification. Since DOS redirection is not used
for the output file, an output filename must be specified or the model will not execute
properly. This is done to allow for the model to write an update to the PC terminal on
the status of processing. The output file generated by the DOS version includes page
feeds that are written directly to the file as part of the header for each page, rather
than using the Fortran carriage control of '1'.
3.9.3.2 Controlling I/O on Other Computer Systems.
The PC-executable versions of the models that are available on the SCRAM BBS
includes certain features that are specific to operating the models in a PC environment.
These include specifying the input and output file names on the command line and writing
an update on the status of the processing to the computer screen. In order to
accomplish the latter, the output file is opened explicitly. The PC versions also
include writing a date and time for the run on each page of the printed output file.
The Fortran computer code that is used to implement these PC-specific features has been
commented out in the source code files available on SCRAM. This is done in order to
make the most use of the features available for the PC while at the same time making the
Fortran source code as "portable" to other computer systems as reasonably possible. This
section briefly addresses the control of model input and output for non-PC computer
systems.
3-144
-------
With the PC-specific code commented out in the ISC source code, the models will use
the default input unit (Fortran unit 5) for reading the input runstream file, and the
default output unit (Fortran unit 6) for writing the printed output file. These files
are not opened explicitly by the models with the PC code commented out. These files
have to be defined, using the $DEFINE command in VAX/VMS and using the DD statement in
the JCL for the IBM/MVS. Refer to Section 4.3 for additional information about running
the models in other environments.
3-145
-------
4.0 COMPUTER NOTES
This section provides information regarding the computer aspects of the ISC models,
including the minimum hardware requirements for executing the models on a PC,
instructions regarding compiling and running the models on a PC, and information
regarding porting the models to other computer systems. A more detailed Programmer's
Guide is provided in Volume III of the ISC Model User's Guide, including details
regarding the design of the computer code.
4.1 MINIMUM HARDWARE REQUIREMENTS
4.1.1 Requirements for Execution on a PC
The ISC models were developed on an IBM-compatible PC, and were designed to run on
PCs with certain minimum hardware requirements. The basic requirements are as follows:
• 80x86 processor (e.g., 8086, 80286, 80386, 80486)
• 640 K of RAM
• Hard Disk with sufficient storage space to handle the executable file, input
data files, and output files (file sizes will vary, generally about 2 MB will be
sufficient for routine applications)
While a math coprocessor (80x87 chip) is optional for execution of the DOS versions
of the ISC models on a PC, it is highly recommended, especially for the ISCST model, due
to the large increase in execution speed that will be experienced. The model may be
expected to run about five to ten times faster with a math coprocessor than without one.
4-1
-------
The DOS models are compiled using an emulator library, meaning that a math coprocessor
will be used if one is present, but the models will also run without one.
The ISC models were designed assuming a PC with a minimum of 640 K of RAM, with the
minimum amount of available RAM for loading the various models (as provided on the SCRAM
BBS) of about 510 K. Because additional memory is needed (for buffers) when the models
open files (such as the input runstream file, the printed output file, the error message
file, etc.), the amount of memory needed to actually run the models will be somewhat
larger than the minimum load size for the executable file. Depending on the number of
externally files being used for a particular application, an additional 10K of memory
may be required.
The amount of available memory on a particular machine will depend on the machine
configuration including the amount of memory used by the operating system, memory used
by any special device drivers, and any memory-resident utility programs. Generally, a
640K PC with minimal memory overhead will have about 550 to 580K of RAM available for
applications, such as the ISC models. The amount of available RAM can be determined by
executing the DOS CHKDSK command. This is done by entering the command 'CHKDSK C:' to
check the C: drive. Refer to the DOS manual for more information about CHKDSK.
For particularly large applications, involving a large number of sources, source
groups, receptors and averaging periods, the user may find that the 640K RAM limit
available with DOS is not enough. This section contains information on increasing the
capacity of the model and setting it up to run on systems (with 80386 processors and
higher) that make use of extended memory beyond the 640K limit of DOS. There are
4-2
-------
special requirements for the operating system and Fortran language compiler needed to
utilize the extended memory on these machines.
4.1.2 Requirements for Execution on a DEC VAX Minicomputer
ISCST will run on any DEC VAX minicomputer or workstation which has enough main
memory to do the real application run. More than 5 MBytes user disk space is
recommended.
4.1.3 Requirements for Execution on an IBM Mainframe
ISCST will run on any IBM 3090 or above mainframe as long as the machine supports
enough memory. The size of the desired memory depends on the size of the application
case run. At least 5 MBytes user disk space is recommended.
4.2 COMPILING AND RUNNING THE MODELS ON A PC
As mentioned earlier, the ISC models were developed on an IBM-compatible PC, using
the Microsoft Optimizing FORTRAN Compiler (Version 5.1). This section provides details
regarding compiling and running the models on a PC.
4.2.1 Microsoft Compiler Options
The DOS versions of the executable files (.EXE) of the models provided on the SCRAM
BBS were compiled with the Microsoft Optimizing FORTRAN Compiler (Version 5.1) using the
following command line:
4-3
-------
FL /c /FPi /AH /DMICRO *.FOR
where /c instructs the compiler to compile without linking; the /FPi option instructs
the compiler to use in-line instructions for floating point operations and link with an
emulator library (uses 80x87 coprocessor if present); the /AH option that the
huge memory model be used, allowing arrays or common blocks to exceed 64K; and the
/DMICRO option instructs the compiler to use the conditional compilation blocks defined
for the Microsoft compiler. These conditional blocks of code implement the PC-specific
features of the model including writing the date and time fields on each page of the
printed output file and writing an update to the screen on the status of processing.
The *.FOR parameter tells the compiler to compile all files in the default directory
ending with an extension of *.FOR. This assumes that all of the source code modules and
the include files are in a single directory, or that the compiler has been setup to
search for the include files in the appropriate directory. This command line for the
compiler makes full use of the compiler's optimization routines to speed up the code.
To disable optimization, the /Od option would be added.
The source modules for the ISCST model are as follows:
ISCST3.FOR - Main program, error handling and other utilities
PCCODE.FOR - PC-specific code for command line, date and time
SETUP.FOR - Main SETUP subroutines and initialization module
INPSUM.FOR - Subroutines to summarize the input data
COSET.FOR - Subroutines to process CO pathway inputs
SOSET.FOR - Subroutines to process SO pathway inputs
RESET.FOR - Subroutines to process RE pathway inputs
MESET.FOR - Subroutines to process ME pathway inputs
TGSET.FOR - Subroutines to process TG pathway inputs
OUSET.FOR - Subroutines to process OU pathway inputs
4-4
-------
METEXT.FOR - Extracts and checks the meteorological data
CALC1.FOR - Main calculation subroutines, including source-type specific
CALC2.FOR - Secondary group of calculation subroutines for hourly values
CALC3.FOR - Group of subroutines to process and sort averages
CALC4.FOR - Group of subroutines to output results as calculated (e.g. DAYTABLE
and POSTFILE results)
PRISE.FOR - Plume rise subroutines
SIGMAS.FOR - Dispersion parameter subroutines
PITAREA.FOR - Open pit and area source subroutines
OUTPUT.FOR - Model output subroutines
DEPFLUX.FOR - Group of subroutines to perform dry deposition calculations
MAIN1.INC - First INCLUDE file, used throughout model
MAIN2.INC - Second INCLUDE file, used for MODNAM variable only
MAIN3.INC - Third INCLUDE file, contains only results arrays
DEPVAR.INC - INCLUDE file for common variables used with the DEPFLUX block of
subroutines
Once the source files have been compiled successfully, and object (.OBJ) files have
been generated for each source file, the model is ready to be linked and an executable
file created. The Microsoft executable file on the SCRAM BBS was linked using a memory
overlay manager so that only certain portions of the code are resident in memory at any
given time. This allows for a more efficient use of available memory by the model, and
therefore allows for larger runs to be performed than would be possible without using
overlays. This is accomplished with the following command line for the linker provided
with the Microsoft compiler:
LINK /E /SE:256 ISCST3+PCCODE+SETUP+CINPSUM)+(COSET)+(SOSET)+(RESET)+(MESET)+(TGSET)+(OUSET)+(METEXT+
CALC1+CALC2+CALC3+PRISE+SIGMAS+CALC4+DEPFLUX+PITAREA)+(OUTPUT)
The /E option instructs the linker to produce a packed executable file that occupies
less disk space. The /SE:256 option increases the number of segments allowed to 256.
The ISCST3, PCCODE and SETUP modules are always memory resident, and any module or group
4-5
-------
of modules within parentheses are overlayed into the same area of memory only when
needed. Linking without the overlay manager will increase the minimum load size for the
executable file by about 200K for the ISCST model. Since most of the overlay swapping
occurs during the setup processing, which is only a very small fraction of the execution
time for normal sized applications, the use of overlays does not significantly effect
the execution time of the model. The load size of the model can be reduced somewhat by
placing the SETUP and CALC4 modules in separate overlays. Placing SETUP in an overlay
will only effect performance (execution speed) for the setup processing stage, and will
only be significant for relatively long input runstream files (e.g. with a large number
of sources or with many discrete receptors). If the application does not make use of
the SAVEFILE, DAYTABLE, MAXIFILE and/or POSTFILE keyword options (where results are
output as their are calculated), then moving the CALC4 module to a separate overlay will
not effect performance at all, since it is only called if one of those options is used.
An example of the LINK command to minimize the load size of the model is as follows:
LINK /E /SE:256 ISCST3+PCCODE+CSETUP)+(INPSUM)+(COSET)+(SOSET)+(RESET)+(MESET)+(TGSET)+(OUSET)+(METEXT+
CALC1+CALC2+CALC3+PRISE+SIGMAS+DEPFLUX+PITAREA)+(CALC4)+(OUTPUT)
This overlay structure will reduce the load size by about 24K for the ISCST model.
4.2.2 Modifying PARAMETER Statements for Unusual Modeling Needs
As discussed in Section 2.3, the ISC models make use of a static storage allocation
design, where the model results are stored in explicitly dimensioned data arrays, and
the array limits are controlled by PARAMETER statements in the Fortran computer code.
These array limits also correspond to the limits on the number of sources, receptors,
source groups and averaging periods that the model can accept for a given run. Depending
4-6
-------
on the amount of memory available on the particular computer system being used, and the
needs for a particular modeling application, the storage limits can easily be changed by
modifying the PARAMETER statements and recompiling the model.
The limits on the number of receptors, sources, source groups, averaging periods,
and events (for ISCEV model) are initially set as follows for the three models for the
DOS and extended memory (EM) versions on the PC:
PARAMETER
Name
NREC
NSRC
NGRP
NAVE
NEVE
Limit
Controlled
Number of
Receptors
Number of
Sources
Number of
Source
Groups
Number of
Short Term
Averages
Number of
Events
ISCST
500 (DOS)
1200 (EM)
100 (DOS)
300 (EM)
2 (DOS)
4 (EM)
2 (DOS)
4 (EM)
-
ISCEV
-
100 (DOS)
500 (EM)
25 (DOS)
50 (EM)
4 (DOS)
4 (EM)
2500 (DOS)
5000 (EM)
ISCLT
500 (DOS)
1200 (EM)
50 (DOS)
300 (EM)
3 (DOS)
5 (EM)
-
-
Fortran PARAMETER statements are also used to specify the array limits for the
number of output types (CONG, DEPOS, DDEP, and/or WDEP) available with the ISCST model
(NTYP, initially set to 2 for the DOS version and 4 for the EM version), the number of
high short term values by receptor to store for the ISCST model (NVAL, initially set to
2 for the DOS version and 6 for the EM version), the number of overall maximum values to
store (NMAX, initially set to 50 for ISCST and to 10 for Long Term), and the number of
4-7
-------
x-coordinates and y-coordinates that may be included in the optional terrain grid file
(MXTX and MXTY, initially set to 101 for the DOS version of Short Term, 201 for the DOS
version of Long Term, and 601 for the EM version of both models).
In addition to the parameters mentioned above, parameters are used to specify the
number of gridded receptor networks in a particular run (NNET), and the number of
x-coordinate (or distance) and y-coordinate (or direction) values (IXM and IYM) for each
receptor network. Initially, the models allow up to 5 receptor networks (of any type),
and up to 50 x-coordinates (or distances) and up to 50 y-coordinates (or directions).
The source arrays also include limits on the number of variable emission rate factors
per source (NQF, initially set to 24 for the DOS version of Short Term and 96 for the EM
version of Short Term, and to 36 for the DOS version of Long Term and 144 for the EM
version of Long Term), the number of sectors for direction-specific building dimensions
(NSEC, initially set to 36 for Short Term and 16 for Long Term), and the number of
settling and removal categories (NPDMAX, initially set to 10 for the DOS version of
Short Term and 20 for the EM version of Short Term and both versions of Long Term).
To modify the array limits for the model, the user must first edit the appropriate
PARAMETER values in the MAIN1.INC file for that model. Once the array limits have been
customized to a particular application's needs, then the entire model must be recompiled
and linked (see Section 4.2.1 above). Because the high value arrays in the ISCST model
are 5-dimensional arrays (NREC,NVAL,NGRP,NAVE,NTYP) and there are three arrays with
these dimensions (the sorted high values, the data period for each value, and the calm
and missing value flag for each value), the model's storage requirements are
particularly sensitive to increasing the number of source groups or the number of high
values to store at each receptor location. For example, the amount of storage space
4-8
-------
required to store these three arrays with the initial PARAMETER values for the DOS
version is about 72K. To increase the number of source groups from 2 to 4 would double
the storage requirement, adding at least another 72K to the load size of the model.
The user should first determine the types of applications for which they most
typically use the models, and then modify the appropriate PARAMETER values accordingly.
If someone never (or very rarely) uses variable emission rate factors, then modifying
the NQF parameter could free up some memory. Changing NQF from 24 to 1 will free up
about 9K for a model using 100 sources. The user may also wish to reduce the NPDMAX
parameter if particulate categories are rarely used.
Often, when a larger number of source groups has been used with the ISCST model, it
has been for the purpose of performing source contribution (or source culpability)
analyses. Since the ISCEV (EVENT) model provides this type of information without
having to specify a separate source group for each source, the need for large numbers of
source groups in the ISCST model should be lessened. If the storage limits available on
the 640K PC environment are too restrictive for particular applications, then the user
should examine the possibility of using a different hardware environment or a different
operating system where the 640K barrier will not be limiting. Such systems are
available for PCs with 80386 and 80486 processors. The extended memory (EM) versions of
the models provided on the SCRAM BBS require an 80386 or 80486 processor with at least 8
MB of RAM (7 MB of available extended memory) for the Short Term model and at least 4 MB
of RAM (3 MB of available extended memory) for the Long Term model. The setup and
application of the models on the DEC VAX minicomputer and the IBM 3090 mainframe
computer are also described in the next section of this User's Guide, and in more
detail in Volume III of the ISC User's Guide.
4-9
-------
4.3 PORTING THE MODELS TO OTHER HARDWARE ENVIRONMENTS
The ISC models are designed and coded to allow them to run on most operating
environments, including DOS, UNICOS, UNIX, SunOS, VAX/VMS, and TSO/MVS. The ISC models
use ANSI Standard FORTRAN 77 with the exception of two widely supported language
extensions, namely the INCLUDE statement and the DO WHILE ... END DO loop construct.
Although the users do not need to make major changes, they may experience some minor
differences from machine to machine on the exact syntax of the INCLUDE statement. These
common language extensions may not be supported on older versions of some compilers as
well. The following sections address portability of the models to various systems in
more detail.
4.3.1 Non-DOS PCs
The only requirement for porting the models to non-DOS PC environments is the
availability of a Fortran compiler capable of operating in and compiling for the non-DOS
operating system. The extended memory (EM) versions of the models available on the SCRAM
BBS were compiled using the Lahey F77L-EM/32 Fortran Compiler, which uses the Ergo
Computing OS/386 operating system to access extended memory in 32-bit protected mode.
The EM executable files are bound with the Ergo OS/386 operating system and a load
module to allow the models to be run on DOS machines.
One significant advantage to installing and running the models in 32-bit protected
mode on PCs is the ability to address a much larger memory storage area. This allows
for the data storage limits controlled by the Fortran PARAMETER statements to be set
much higher than is possible for the DOS versions. By using the 32-bit instruction set,
4-10
-------
the protected mode versions also tend to run about 20 to 30 percent faster than the DOS
versions. More information about compiling the models with the Lahey F77L-EM/32
compiler is provided in Appendix D.
4.3.2 DEC VAX
4.3.2.1 Compiler/System Dependent Preprocessing.
The ISC codes as provided on the SCRAM BBS are compatible with VAX-11 FORTRAN
Version 2 and above, except that the PC-specific features contained in PCCODE.FOR must
be replaced with equivalent system-specific functions for the VAX (which may be called
VAXCODE.FOR), or commented out. These features include writing the date and time on
each page of the printed output file and writing an update to the screen on the status
of processing.
4.3.2.2 Creating An Executable ISCST.
Although the users can specify any way they want to group and store the code and
data files, the easiest way is to copy all the source codes modules, INCLUDE files and
meteorology data into a subdirectory. The user can then write a .COM file to compile,
link and create an executable.
The files needed to make the ISCST executable are the following:
MAIN1.INC, MAIN2.INC, MAIN3.INC, DEPVAR.INC, ISCST3.FOR, (VAXCODE.FOR), SETUP.FOR,
COSET.FOR, SOSET.FOR, RESET.FOR, MESET.FOR, TGSET.FOR, OUSET.FOR, INPSUM.FOR,
4-11
-------
METEXT.FOR, CALC1.FOR, CALC2.FOR, PRISE.FOR, SIGMAS.FOR, CALC3.FOR, CALC4.FOR,
DEPFLUX.FOR, PITAREA.FOR, OUTPUT.FOR
The following is a sample command file named MAKEISC.COM:
$SET DEF [USERNAME.ISCST3]
$ FOR ISCST3.FOR
$ FOR VAXCODE.FOR
$ FOR SETUP.FOR
$ FOR COSET.FOR
$ FOR SOSET.FOR
$ FOR RESET.FOR
$ FOR MESET.FOR
$ FOR TGSET.FOR
$ FOR OUSET.FOR
$ FOR INPSUM.FOR
$ FOR METEXT.FOR
$ FOR CALC1.FOR
$ FOR CALC2.FOR
$ FOR PRISE.FOR
$ FOR SIGMAS.FOR
$ FOR CALC3.FOR
$ FOR CALC4.FOR
$ FOR DEPFLUX.FOR
$ FOR PITAREA.FOR
$ FOR OUTPUT.FOR
SLINK ISCST3,VAXCODE,SETUP,COSET,SOSET,RESET,MESET,TGSET,OUTSET,-
INPSUM,METEXT,CALC1,CALC2,PRISE,SIGMAS,CALC3,CALC4,DEPFLUX,PITAREA,OUTPUT
$ EXIT
To make the executable file, the users should run the MAKEISC.COM file by typing
©makeisc after the command line prompt and pressing ENTER.
4-12
-------
4.3.2.3 Running ISCST.
The VAX/VMS operating system is somewhat different from the DOS and UNIX operating
environments. The users are not able to direct system I/O on the command line prompt.
Instead, the users need to generate a .COM file first, and then run the .COM file online
or submit the .COM file to a system batch queue.
Here is an example of the .COM runfile named RUNISC.COM:
$SET DEF [USERNAME.ISCST3]
$DEFINE/USER_MODE SYS$INPUT TEST-ST.INP
$DEFINE/USER_MODE SYS$OUTPUT TEST-ST.OUT
$RUN ISCST3
$EXIT
The users can either type in @runisc ENTER to run the model online or SUBMIT runisc on
the command line prompt to submit a batch job.
4.3.3 IBM 3090
4.3.3.1 Compiler/System Dependent Preprocessing.
The ISC codes as provided on the SCRAM BBS are compatible with the IBM VS FORTRAN
(Version 2), except that the PC-specific features contained in PCCODE.FOR must be
replaced with equivalent system-specific functions for the IBM (which may be called
IBMCODE.FOR), or commented out. These features include writing the date and time on
each page of the printed output file and writing an update to the screen on the status
4-13
-------
of processing. The syntax for the INCLUDE statement is different on the IBM VS FORTRAN,
and the user will have to replace the statements such as:
INCLUDE 'MAIN1.INC'
with a corresponding statement such as:
INCLUDE (MAIN1)
throughout the ISC source code. This can easily be accomplished with the editor, and
there are three INCLUDE files used in most of the models. For the ISCST model, the
INCLUDE file names are MAIN1.INC, MAIN2.INC, and MAIN3.INC. The deposition routines in
DEPFLUX.FOR use one INCLUDE file, named DEPVAR.INC.
4.3.3.2 Creating An Executable ISCST.
The ISCST model can be compiled and linked in one step under VS FORTRAN by
executing the appropriate procedure (e.g., VSF2CG to compile and load) in the JCL for
the compile job. It is easiest to concatenate all of the source (*.FOR) files into a
single partitioned data set member, and identify that file name with a DD statement in
the JCL. Special procedures may be needed to access the INCLUDE files, where each
INCLUDE file should be a member in a partitioned data set.
4-14
-------
4.3.3.3 Running ISCST.
When running the ISCST model under IBM/MVS, special attention is needed to defining
and controlling the file I/O. The input runstream file is read from the default input
unit, Fortran unit number 5, and the output print file is written to the default output
unit, Fortran unit number 6. The input meteorological data file is read from Fortran
unit 19. Other system files include the temporary error/message file (unit 10) and the
temporary event file for ISCST (unit 18). These files, as well as any user-specified
optional output files, must be defined with DD statements in the JCL.
4.3.4 Various UNIX machines (CRAY. SUN. DEC VAX. AT&T)
4.3.4.1 Compiler/System Dependent Preprocessing.
The ISC codes as provided on the SCRAM BBS are compatible with any ANSI Standard
FORTRAN 77 Compiler operating under UNICOS, UNIX, and SUN OS, except that the
PC-specific features contained in PCCODE.FOR must be replaced with equivalent system-
specific functions for UNIX (which may be called UNIXCODE.FOR), or commented out. These
features include writing the date and time on each page of the printed output file and
writing an update to the screen on the status of processing.
4.3.4.2 Creating An Executable ISCST.
Although the users can specify any way they want to group and store the code and
data files, the easiest way is to copy all the source codes modules, INCLUDE files and
meteorology data into a subdirectory. The users should make sure that every source file
4-15
-------
has suffix .f and the file name should be a lower case ASCII character string, because
the UNICOS, UNIX, and SUN OS is case-sensitive. Also, for the same reason, all of the
.INC file should be in UPPER CASE. The user can then write a make file to compile, link
and create an executable.
The files needed to make the ISCST executable are the following:
MAIN1.INC, MAIN2.INC, MAIN3.INC, DEPVAR.INC, iscstB.f, (unixcode.f), setup.f,
coset.f, soset.f, reset.f, meset.f, tgset.f, ouset.f, inpsum.f, metext.f, calcl.f,
calc2.f, prise.f, sigmas.f, calcB.f, calc4.f, depflux.f, pitarea.f, output.f
Compiling ISCST is relatively easy under UNIX operating environment due to the
similarity between DOS and UNIX. For a DEC VAX workstation running Utrix 4.3, the
command:
f77 -o iscst3 *.f
will generate an ISCST executable. For a CRAY running UNICOS 5.1, the following
commands will generate an ISCST executable under UNICOS:
4-16
-------
cft77 iscstS.f
cft77 unixcode.f
cft77 setup.f
cft77 coset.f
cft77 soset.f
cft77 reset.f
cft77 meset.f
cft77 tgset.f
cft77 ouset.f
cft77 inpsum.f
cft77 metext.f
cft77 calcl.f
cft77 calcZ.f
cft77 prise.f
cft77 sigmas.f
cft77 calcS.f
cft77 calc4.f
cft77 depflux.f
cft77 pitarea.f
cft77 output.f
segldr -o iscstS *.o
The command for compiling ISCST under the SUN OS environment is similar to the one
for VAX Ultrix 4.3.
4.3.4.3 Running ISCST.
Before running ISCST, the users need to check the meteorology data file and make
sure the file name matches the one in the input file. File names in UNIX are case
sensitive, so the characters in the file name need to match the ones in the input file.
Then the user can type:
iscst3 outputfile
4-17
-------
to run the executable.
4.3.5 Advanced Topics.
For more detailed information about porting and installing the ISC models to other
computer environments, refer to Volume III of the ISC User's Guide. Volume III provides
a more detailed description of the design and structure of the computer code, including
module calling trees, data dictionary, and a description of the model loop structures.
Volume III also includes instructions for compiling the ISC models with compilers that
do not support the INCLUDE and DO WHILE ... ENDDO Fortran language extensions.
4-18
-------
5.0 REFERENCES
Bowers, J.F., J.R. Bjorklund and C.S. Cheney, 1979: Industrial Source Complex (ISC)
Dispersion Model User's Guide. Volume I, EPA-450/4-79-030, U.S. Environmental
Protection Agency, Research Triangle Park, North Carolina 27711.
Bowers, J.R., J.R. Bjorklund and C.S. Cheney, 1979: Industrial Source Complex (ISC)
Dispersion Model User's Guide. Volume II, EPA-450/4-79-031, U.S. Environmental
Protection Agency, Research Triangle Park, North Carolina 27711.
Baumann, E.R. and R.K. Dehart, 1988: Evaluation and Assessment of UNAMAP.
EPA/600/3-88/009, U.S. Environmental Protection Agency, Research Triangle Park,
North Carolina 27711.
Environmental Protection Agency, 1986: Guideline for Determination of Good Engineering
Practice Stack Height (Technical Support Document for the Stack Height Regulations)
- Revised EPA-450/4-80-023R, U.S. Environmental Protection Agency, Research
Triangle Park, North Carolina 27711.
Environmental Protection Agency, 1987a: Industrial Source Complex (ISC) Dispersion
Model User's Guide - Second Edition (Revised) Volume I. EPA-450/4-88-002a, U.S.
Environmental Protection Agency, Research Triangle Park, North Carolina 27711.
Environmental Protection Agency, 1995: Guideline on Air Quality Models (Revised) and
Supplements. EPA-450/2-78-027R et seq., published as Appendix W to 40 CFR Part 51.
U.S. Environmental Protection Agency, Research Triangle Park, North Carolina
27711.
Environmental Protection Agency, 1992: User's Guide for the Industrial Source Complex
(ISC2) Dispersion Models - Volume I. EPA-450/4-92-008a, U.S. Environmental
Protection Agency, Research Triangle Park, North Carolina 27711.
Rorex, H.W., 1990: Operational Review of the Support Center for Regulatory Air Models
Bulletin Board Service. U.S. Environmental Protection Agency, Research Triangle
Park, North Carolina 27711.
5-1
-------
Hanna, S.R. and J.C. Chang 1991. Modification of the Hybrid Plume Dispersion Model
(HPDM) for Urban Conditions and its Evaluation Using the Indianapolis Data Set.
Vol. I. User's Guide for HPDM-Urban. Sigma Research Corporation, Concord, MA,
01742 .
Holtslag, A.A.M. and A.P. van Ulden 1983. A Simple Scheme for Daytime Estimates of the
Surface Fluxes from Routine Weather Data. J. dim. and Meteor., 22, 517-529.
Holzworth, G.C., 1972: Mixing Heights, Wind Speeds and Potential for Urban Air
Pollution Throughout the Contiguous United States. Publication No. Ap-101, U.S.
Environmental Protection Agency, Research Triangle Park, NC.
Iqbal, M. 1983. An Introduction to Solar Radiation. Academic Press, 286 pp.
Oke, T.R. 1978. Boundary Layer Climates. John Wiley & Sons, New York, NY.
Oke, T.R. 1982. The Energetic Basis of the Urban Heat Island. Quart. J.R. Meteor. Soc.,
108, 1-24.
Sheih, C.M., M.L. Wesley, and B.B. Hicks 1979. Estimated Dry Deposition Velocities of
Sulfur Over the Eastern U.S. and Surrounding Regions. Atmos. Environ., 13, 361-
368.
5-2
-------
APPENDIX A. ALPHABETICAL KEYWORD REFERENCE
This appendix provides an alphabetical listing of all of the keywords used by the
ISC models. Each keyword is identified as to the pathway for which it applies, the
keyword type (either mandatory or optional, and either repeatable or non-repeatable),
and with a brief description of the function of the keyword. For a more complete
description of the keywords, including a list of associated parameters, refer to the
Detailed Keyword Reference in Section 3 or the Functional Keyword/Parameter Reference in
Appendix B.
A-l
-------
Keyword
ANEMHGHT
AVERTIME
AVEMIXHT
AVEROUGH
AVESPEED
AVETEMPS
BOUNDARY
BOUNDELV
BUILDHGT
BUILDWID
CONCUNIT
DAYRANGE
DAYTABLE
DCAYCOEF
Path
ME
CO
ME
ME
ME
ME
RE
RE
SO
SO
SO
ME
OU
CO
Type
M - N
M - N
M - R
0 - R
0 - N
M - R
0 - R
0 - R
0 - R
0 - R
0 - N
0 - R
0 - N
0 - N
Keyword Description
Height of anemometer above stack base
Averaging time(s) to process (up to NAVE short term
plus PERIOD or ANNUAL averages)
Average mixing height for each wind speed, stability
category and season (Applies Only to Long Term)
Roughness length (m) for each season (Applies Only to
Lonq Term)
Average (median) wind speed for each speed category in
the STAR summary (Applies Only to Long Term)
Average ambient temperature for each stability
category and season (Applies Only to Long Term)
Defines discrete polar receptor locations
corresponding to minimum plant boundary distances for
each 10 degree sector
Defines terrain elevations for discrete receptors
specified with BOUNDARY keyword
Building height values for each wind sector
Building width values for each wind sector
Optional conversion factors for emission input units
and concentration output units
Specifies days or ranges of days to process (default
is to process all data read in) , applies only to ISCST
processinq
Option to provide summaries for each averaging period
for each day processed. (Applies to ISCST Only)
Optional decay coefficient for exponential decay
A-2
-------
Type: M - Mandatory
0 - Optional
N - Non-repeatable
R - Repeatable
Keyword
DEPOUNIT
DISCCART
DISCPOLR
DTHETADZ
ELEVUNIT
EM IS FACT
EMISUNIT
ERRORFIL
EVENTFIL
EVENTOUT
EVENTPER
EVENTLOC
FINISHED
FLAGPOLE
Path
CO
RE
RE
ME
CO
SO
RE
TG
SO
SO
CO
CO
OU
EV
EV
ALL
CO
Type
0 - N
0 - R
0 - R
0 - R
0 - N
0 - N
0 - N
0 - N
0 - R
0 - N
0 - N
0 - N
M - N
M - R
M - R
M - N
0 - N
Keyword Description
Optional conversion factors for emission input units
and deposition output units
Defines the discretely placed receptor locations
referenced to a Cartesian system
Defines the discretely placed receptor locations
referenced to a polar system
Input optional vertical potential temperature gradients
Defines input units for receptor elevations (RE or CO
path) , source elevations (SO path) or terrain grid
elevations (TG path) (defaults to meters)
Optional input for variable emission rate factors
Optional conversion factors for emission units, and
concentration or deposition output units
Option to generate detailed error listing file
Specifies whether to generate an input file for EVENT
model (Applies only to ISCST)
Specifies the level of output information provided by
the EVENT model
Describes data and averaging period for an event
Describes receptor location for an event
Identifies the end of inputs for a particular pathway
Specifies whether to accept receptor heights above
local terrain (m) for use with flagpole receptors, and
A-3
-------
NOTE: See ELEVUNIT footnote on p. B-5.
A-4
-------
Keyword
GAS-SCAV
GRIDCART
GRIDPOLR
HALFLIFE
HOUREMIS
INITFILE
INPUTFIL
LOCATION
LOWBOUND
MASSFRAX
MAXIFILE
MAXTABLE
MODELOPT
MULTYEAR
Path
SO
RE
RE
CO
SO
CO
ME
TG
SO
TG
SO
SO
OU
OU
CO
CO
Type
0 - R
0 - R
0 - R
0 - N
0 - R
0 - N
M - N
M - N
M - R
M - N
0 - R
0 - R
0 - R
0 - R
M - N
0 - N
Keyword Description
Optional input of precipitation scavenging
coefficients for gaseous pollutants
Defines a Cartesian grid receptor network
Defines a polar receptor network
Optional half life used for exponential decay
Option for specifying hourly emission rates in a
separate file
Option to initialize model from file of intermediate
results generated by SAVEFILE option
Describes input meteorological data file (ME path) and
terrain grid file (TG path)
Identifies coordinates for particular source (SO path)
or for the terrain grid location (TG path)
Switch to use non-D FAULT option for "lower bound" wake
calculations, controlled by sector
Optional input of mass fraction for each particle size
category
Option to list events exceeding a threshold value to
file (if CO EVENTFIL option is used, these events are
included in the input file generated for the EVENT
model)
Option to summarize the overall maximum values
Job control and dispersion options
Specifies that run is part of a multi-year run, e.g.,
for PM-10 H6H in five years
A-5
-------
PARTDENS
PARTDIAM
SO
SO
0 - R
0 - R
Optional input of particle density for each size
category
Optional input of particle diameter for each size
category
Keyword
PARTSLIQ
PARTS ICE
PLOTFILE
POLLUTID
POSTFILE
RECTABLE
RUNORNOT
SAVE FILE
SRCGROUP
SRCPARAM
Path
SO
SO
OU
CO
OU
OU
CO
CO
SO
SO
Type
0 - R
0 - R
0 - R
M - N
0 - R
0 - R
M - N
0 - N
M - R
M - R
Keyword Description
Optional input of scavenging coefficients of
particulate emissions for liquid precipitation
Optional input of scavenging coefficients of
particulate emissions for frozen precipitation
Option to write certain results to a storage file
suitable for input to plotting routines
Identifies pollutant being modeled
Option to write results to a mass storage file for
postprocessing
Option to output value (s) by receptor
Identifies whether to run model or process setup
information only
Option to store intermediate results for later
restart of the model after user or system interrupt
(ST Only)
Identification of source groups
Identifies source parameters for a particular source
A-6
-------
STARDATA
STARTEND
STARTING
SURFDATA
TERRHGTS
TITLEONE
TITLETWO
ME
ME
ALL
ME
CO
CO
CO
0 - N
0 - N
M - N
M - N
0 - N
M - N
0 - N
Identifies which STAR summaries are included in
meteorological data file
Specifies start and end dates to be read from input
meteorological data file (default is to read entire
file) , applies only to ISCST processinq
Identifies the start of inputs for a particular
pathway
Surface meteorological station
Specifies whether to assume flat terrain (default) or
to allow use of receptors on elevated terrain
First line of title for output
Optional second line of output title
A-7
-------
Keyword
TOXXFILE
UAIRDATA
WDROTATE
WINDCATS
WINDPROF
Path
OU
ME
ME
ME
ME
Type
0 - R
M - N
0 - N
0 - N
0 - R
Keyword Description
Creates output file formatted for use with TOXX model
component of TOXST or the RISK model component of
TOXLT
Uper air meteorological station
Wind direction rotation adjustment
Upper bound of wind speed categories
Input optional wind profile exponents
A-!
-------
APPENDIX B. FUNCTIONAL KEYWORD/PARAMETER REFERENCE
This appendix provides a functional reference for the keywords and parameters used
by the input runstream files for the ISC models. The keywords are organized by
functional pathway, and within each pathway the order of the keywords is based on the
function of the keyword within the models. The pathways used by the models are as
follows, in the order in which they appear in the runstream file and in the tables that
follow:
CO - for specifying overall job COntrol options;
SO - for specifying SOurce information;
RE - for specifying REceptor information (ISCST and ISCLT models only);
ME - for specifying MEteorology information and options;
TG - for specifying Terrain Grid information and options (optional);
EV - for specifying EVent information (ISCEV model only); and
OU - for specifying Output options.
The pathways and keywords are presented in the same order as in the Detailed Keyword
Reference in Section 3, and in the Quick Reference at the end of the manual.
Two types of tables are provided for each pathway. The first table lists all of
the keywords for that pathway, identifies each keyword as to its type (either mandatory
or optional and either repeatable or non-repeatable), and provides a brief description
of the function of the keyword. The second type of table, which takes up more than one
page for most pathways, presents the parameters for each keyword, in the order in which
they should appear in the runstream file where order is important, and describes each
parameter in detail. Also indicated for certain keywords or parameter descriptions are
cases where the inputs apply on to a certain model, either ISCST, ISCEV, or ISCLT.
B-l
-------
The following convention is used for identifying the different types of input
parameters. Parameters corresponding to secondary keywords which should be input "as
is" are listed on the tables with all capital letters and are underlined. Other
parameter names are given with an initial capital letter and are not input "as is." In
all cases, the parameter names are intended to be descriptive of the input variable
being represented, and they often correspond to the Fortran variable names used in the
model code. Parentheses around a parameter indicate that the parameter is optional for
that keyword. The default that is taken when an optional parameter is left blank is
explained in the discussion for that parameter.
B-2
-------
TABLE B-l
DESCRIPTION OF CONTROL PATHWAY KEYWORDS
CO Keywords
STARTING
TITLEONE
TITLETWO
MODELOPT
AVERTIME
POLLUTID
HALFLIFE
DCAYCOEF
TERRHGTS
ELEVUNIT2
FLAGPOLE
RUNORNOT
EVENTFIL3
SAVEFILE4
INITFILE4
MULTYEAR4
ERRORFIL
FINISHED
Type
M
M
0
M
M
M
0
0
0
0
0
M
0
0
0
0
0
M
- N
- N
- N
- N
- N
- N
- N1
- N1
- N
- N
- N
- N
M
- N
- N
- N
- N
- N
Keyword Description
Identifies the start of CONTROL pathway inputs
First line of title for output
Optional second line of title for output
Job control and dispersion options
Averaging time(s) to process
Identifies type of pollutant being modeled
Optional half life used for exponential decay
Optional decay coefficient
Specifies whether to assume flat terrain (default) or to allow use of receptors on elevated
terrai n
Defines input units for receptor elevations (defaults to meters)
Specifies whether to accept receptor heights above local terrain (m) for use with flagpole
receptors, and allows for a default flagpole height to be specified
Identifies whether to run model or process setup information only
Specifies whether to generate an input file for EVENT model (Applies to ISCST Only)
Option to store intermediate results for later restart of the model after user or system
interrupt (Applies to ISCST Only)
Option to initialize model from file of intermediate results generated by SAVEFILE option
(Appl ies to ISCST Only)
Option to process multiple years of meteorological data (one year per run) and accumulate
high short term values across years (Applies to ISCST Only)
Option to generate detailed error listing file (error file is mandatory for CO RUNORNOT NOT
case)
Identifies the end of CONTROL pathway inputs
Type: M - Mandatory N - Non-Repeatable
0 - Optional R - Repeatable
1) Either HALFLIFE or DCAYCOEF may be specified.
If both cards appear a warning
B-3
-------
message will be issued and the first value entered will be used in calculations.
Default assumes a half life of 4 hours for SO? modeled in urban mode.
2) The CO ELEVUNIT card is obsolescent with this version of the ISC models. The new
RE ELEVUNIT card should be used instead to specify elevation units for receptors.
3) The EVENTFIL keyword controls whether or not to generate an input file for the
ISCEV (EVENT) model. The primary difference between ISCST and ISCEV processing is
in the treatment of source group contributions. The ISCST model treats the source
groups independently, whereas the ISCEV model determines individual source
contributions to particular events, such as the design concentrations determined
from ISCST, or user-specified events. By specifying the EVENTFIL keyword, an input
runstream file will be generated that can be used directly with the ISCEV model.
The events included in the generated ISCEV model input file are defined by the
RECTABLE and MAXIFILE keywords on the OU pathway, and are placed in the EVent
pathway. If more than one output type (CONC, DEPOS, DDEP, and/or WDEP) is selected
for the ISCST model, only events associated with the first output type, in the
order stated above, are included in the EVENT model input file.
4) The SAVEFILE and INITFILE keywords work together to implement the model's re-start
capabilities. Since the MULTYEAR option utilizes the re-start features in a
special way to accumulate high short term values from year to year, it cannot be
used together with the SAVEFILE or INITFILE keyword in the same model run.
B-4
-------
TABLE B-2
DESCRIPTION OF CONTROL PATHWAY KEYWORDS AND PARAMETERS
Keyword
TITLEONE
where:
TITLETWO
where:
MODELOPT
Parameters
Titlel
Titlel
TitleZ
TitleZ
First line of title for output, character string of up to 68 characters
Optional second line of title for output, character string of up to 68
characters
DFAULT CONC DRYDPLT WETDPLT RURAL GRDRIS NOSTD NOBID NOCALM MSGPRO NOSMPL (ST)
DEPOS
DDEP
and/or
WDEP
DFAULT CONC DRYDPLT
DEPOS
or
DDEP
or or
URBAN NOCMPL
RURAL GRDRIS NOSTD NOBID (LT)
or
URBAN
B-5
-------
where:
AVERTIME
where :
DFAULT
CONC
DEPOS
DDEP
WDEP
DRYDPLT
WETDPLT
RURAL
URBAN
GRDRIS
NOSTD
NOBID
NOCALM
MSGPRO
NOSMPL
NOCMPL
Timel Time2 TimeS
Ti meN
MONTH
PERIOD
ANNUAL
Specifies use of regulatory default options (final
rise, stack tip downwash, BID, calms processing,
"upper bound" wake calcs, default exponents and
DTDZ), overrides presence of GRDRIS, NOSTD, NOBID,
NOCALM. and MSGPRO keywords
Specifies calculation of concentration values
Specifies calculation of total deposition flux (both dry and wet)
for Short Term, and dry deposition flux for Long Term
Specifies calculation of dry deposition flux only
Specifies calculation of wet deposition flux only (ST only)
Specifies inclusion of plume depletion due to dry removal
Specified inclusion of plume depletion due to wet removal (ST
only)
Specifies use of rural dispersion
Specifies use of urban dispersion
Option to use gradual plume rise
Option to use no stack-tip downwash
Option to use no buoyancy- induced dispersion
Option to bypass calms processing routine (ST only)
Option to use missing data processing routines (ST only)
Option to suppress simple terrain calculations, i.e., use
COMPLEXl algorithms only (ST only)
Option to suppress complex terrain calculations, i.e., use
ISCST algorithms only (ST only)
ime4 MONTH PERIOD (ISCST and ISCEV only)
or
ANNUAL
Nth optional averaging time (1, 2, 3, 4, 6, 8, 12,
Ł4_-hr; number of periods limited by NAVE parameter)
Option to calculate MONTHly averages (counts toward
NAVE limit)
Option to calculate averages for the entire data
PERIOD
Option to calculate ANNUAL averages for the entire data
B-6
-------
TABLE B-2 (CONT.)
DESCRIPTION OF CONTROL PATHWAY KEYWORDS AND PARAMETERS
AVERTIME
where :
POLLUTID
where :
HALFLIFE
where :
DCAYCOEF
where :
JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC (ISCLT model)
WINTER SPRING SUMMER FALL or QUART1 QUARTZ QUARTS QUART4
MONTH SEASON QUARTR ANNUAL PERIOD
JAN
FEB
DEC
WINTER
SPRING
SUMMER
FALL
QUART1
QUARTZ
QUARTS
QUART4
MONTH
SEASON
QUARTR
ANNUAL
PERIOD
Pollut
Pollut
Option to calculate JANuary averages from STAR data
Option to calculate FEBruary averages from STAR data
Option to calculate DECember averages from STAR data
Option to calculate WINTER averages from STAR data
Option to calculate SPRING averages from STAR data
Option to calculate SUMMER averages from STAR data
Option to calculate FALL averages from STAR data
Option to calculate QUART1 averages from STAR data
Option to calculate QUARTZ averages from STAR data
Option to calculate QUARTS averages from STAR data
Option to calculate QUART4 averages from STAR data
Option to calculate averages for all twelve MONTHs
Option to calculate averages for all four SEASONS
Option to calculate averages for all four QUARTeRs
Option to calculate annual values from an ANNUAL STAR
summary
Option to calculate averages for the entire data
PERIOD
Identifies type of pollutant being modeled. Any name
of up to eight characters may be used, e.g., SOZ,
NOX. CO, PM10. TSP or OTHER. Selection of
SOZ with the URBAN DFAULT options forces use of
a half life of 4 hours for exponential decay. Use
of PM10, PM-10 or OTHER allows for the use of the
MULTYEAR option.
Haflif
Haflif
Half life used for exponential decay (s)
Decay
Decay
Decay coefficient for exponential decay (s"1) = 0.693/HAFLIF
B-7
-------
TERRHGTS
where :
ELEVUNIT
where :
FLAGPOLE
where :
FLAT or ELEV
FLAT
ELEV
Specifies that flat terrain will be assumed for all
calculations (default)
Specifies that receptors may be located on elevated
terrain (chopped off at release height)
Note that if ELEVated receptors are allowed.
then receptor heights must be input on the RE
pathway, or they will be assumed to be 0.0.
METERS or FEET
METERS
FEET
Specifies input units for terrain (receptor) elevations of
meters
Specifies input units for terrain (receptor) elevations of feet
Note: This keyword applies to receptor elevations
only.
(Flagdf)
Flagdf
Default value for height of (flagpole) receptors
above local ground level, a default value of 0.0 m
is used if this optional parameter is omitted
Note: The CO ELEVUNIT card is obsolescent with this version of the ISC models. The new RE ELEVUNIT card should be used
instead to specify elevation units for receptors. If the CO ELEVUNIT card is present, it will be processed as it
was in the previous version of the ISC models, but it cannot be used when an ELEVUNIT card is present on either the
SO, RE or TG pathways.
Bi
-1
-------
TABLE B-2 (CONT.)
DESCRIPTION OF CONTROL PATHWAY KEYWORDS AND PARAMETERS
RUNORNOT
where:
EVENTFIL
where:
SAVEFILE
where:
INITFILE
where:
MULTYEAR
where:
RUN or NOT
RUN
NOT
Indicates to run full model calculations
Indicates to process setup data and report errors,
but to not run full model calculations
(Evfile) (Evopt)
Evfile
Evopt
Identifies the filename to be used to generate a file
for input to EVENT model (Defaul t=EVENTFIL. INP)
Optional parameter to specify the level of output
detail selected for the EVENT model: either
SOCONT or DETAIL (default is DETAIL if this para-
meter is omitted)
(Savfil) (Dayinc) (SavflZ)
Savfil
Dayi nc
SavflZ
Specifies name of disk file to be used for storing
intermediate results (default = SAVE.FIL) file is
overwritten after each dump)
Number of days between dumps (optional: default is 1)
Optional second disk filename to be used on alternate
dumps - eliminates risk of system crash during the
dump. If blank, file is overwritten each time.
(Inifil )
Inifil
Savfil (Inifil)
Savfil
Inifil
Specifies name of disk file of intermediate results
to be used for initializing run (default = SAVE.FIL)
Specifies name of disk file to be used for storing
results at end of the year
Optional name of disk file used for initializing the
results arrays from previous year(s). The Inifil
parameter is not used for the first year in the
multi-year run.
B-9
-------
ERRORFIL
where:
(Errfil) (DEBUG)
Errfil
DEBUG
Specifies name of detailed error listing file
(default = ERRORS. LSI)
Option to provide detailed output for debugging
purposes, e.g., plume heights, sigmas, etc.
Generates Very Large Files -- Use with CAUTION!!!
B-10
-------
TABLE B-3
DESCRIPTION OF SOURCE PATHWAY KEYWORDS
SO Keywords
STARTING
ELEVUNIT
LOCATION
SRCPARAM
BUILDHGT
BUILDWID
LOWBOUND
EMISFACT
EMISUNIT
CONCUNIT
DEPOUNIT
PARTDIAM
MASSFRAX
PARTDENS
PARTSLIQ
PARTSICE
GAS-SCAV
HOUREMIS
SRCGROUP1
FINISHED
Type
M -
0 -
M -
M -
0 -
0 -
0 -
0 -
0 -
0 -
0 -
0 -
0 -
0 -
0 -
0 -
0 -
0 -
M -
M -
N
N
R
R
R
R
R
R
N
N
N
R
R
R
R
R
R
R
R
N
Keyword Description
Identifies the start of SOURCE pathway inputs
Defines input units for source elevations (defaults to meters), must be
SO STARTING if used.
first keyword after
Identifies coordinates for particular source
Identifies source parameters for a particular source
Building height values for each wind sector
Building width values for each wind sector
Switch to use non-DFAULT option for "lower bound" wake calculations, controlled by sector
Optional input for variable emission rate factors
Optional conversion factors for emissions, concentrations, and depositi
ons
Optional conversion factors for emissions and concentrations
Optional conversion factors for emissions and depositions
Input variables for optional input of particle size (microns)
Optional input of mass fraction for each particle size category
Optional input of particle density (g/cm3) for each size category
Optional input of scavenging coefficient (s-mm/hr)'1 of particulates for
Optional input of scavenging coefficient (s-mm/hr)'1 of particulates for
Optional input of scavenging coefficient (s-mm/hr)'1 of gases for liquic
precipitation
liquid precipitation
frozen precipitation
or frozen
Option for specifying hourly emission rates in a separate file
Identification of source groups
Identifies the end of SOURCE pathway inputs
B-ll
-------
1) Source groups are treated independently for ISCST. The ISCEV (EVENT) model
provides the contribution from each source to the group total for each specified
event.
B-12
-------
TABLE B-4
DESCRIPTION OF SOURCE PATHWAY KEYWORDS AND PARAMETERS
Keyword
ELEVUNIT
where:
LOCATION
where:
SRCPARAM
Parameters
METERS or FEET
METERS
FEET
Specifies input units for source elevations of
meters
Specifies input units for source elevations of feet
Note: This keyword applies to source elevations
only.
Srcid Srctyp Xs Ys (Zs)
S r c i d
Srctyp
Xs
Ys
Zs
Source identification code (alphanumeric string
of up to eight characters)
Source type: POINT. VOLUME. AREA. OPENPIT
x-coord of source location, corner for AREA and OPENPIT (in m)
y-coord of source location, corner for AREA and OPENPIT (in m)
Optional z-coord of source location (elevation above
mean sea level, defaults to 0.0 if omitted)
Srcid Ptemis Stkhgt Stktmp Stkvel Stkdia
Vlemis Relhgt Syinit Szinit
Aremis Relhgt Xinit (Yinit) (Angle) (Szinit)
Pitemis Relhgt Xinit Yinit Pitvol (Angle)
B-13
-------
where:
BUILDHGT
where:
BUILDWID
where:
S r c i d
Emi s
Hgt
Stktmp
Stkvel
Stkdia
Sy i n i t
S z i n i t
Xinit
Yinit
Angl e
Pitvol
Source identification code
Source emission rate: in g/s for Ptemis or Vlemis,
g/(sm2) for Aremis or Pitemis for concentration or deposition
Source physical release height above ground (center
of height for VOLUME, height above base of pit for OPENPIT)
Stack gas exit temperature (K)
Stack gas exit velocity (m/s)
Stack inside diameter (m)
Initial lateral dimension of VOLUME source (m)
Initial vertical dimension of VOLUME or AREA source (m) (optional parameter
for AREA sources, assumed to be 0.0 if omitted)
Length of side of AREA or OPENPIT source in X-direction (m)
Length of side of AREA or OPENPIT source in Y-direction (m) (optional for
AREA sources, assumed to be equal to Xinit if omitted)
Orientation angle of AREA or OPENPIT source relative to North (degrees),
measured positive clockwise, rotated around the source location,
(Xs,Ys) (optional parameter, assumed to be 0.0 if omitted)
Volume of open pit (m3)
Srcid (or Srcrng) Dsbh(i), i=l,36 (16 for LT)
S r c i d
Srcrng
Dsbh
Source identification code
Range of sources (inclusive) for which building
dimensions apply, entered as two alphanumeric
strings separated by a '-'
Array of direction-specific building heights (m)
beginning with 10 degree flow vector and increment-
ing by 10 degrees clockwise
Srcid (or Srcrng) Dsbw(i), i=l,36 (16 for LT)
Srcid
Srcrng
Dsbw
Source identification code
Range of sources (inclusive) for which building
dimensions apply
Array of direction-specific building widths (m)
beginning with 10 degree flow vector and increment-
ing by 10 degrees clockwise
B-14
-------
TABLE B-4 (CONT.)
DESCRIPTION OF SOURCE PATHWAY KEYWORDS AND PARAMETERS
LOWBOUND
where:
EMISFACT
where:
Srcid (or Srcrng) Idswak(i), i=l,36 (16 for LT)
S r c i d
Srcrng
Idswak
Source identification code
Range of sources (inclusive) for which LOWBOUND
option applies
Array of direction-specific wake option switches
beginning with 10 degree flow vector and increment-
ing by 10 degrees clockwise
(0=upper bound, l=lower bound)
Srcid (or Srcrng) Qflag Qfact(i), i=l,n
Srcid
Srcrng
Qflag
Qfact
Source identification code
Range of sources (inclusive) for which emission rate
factors apply
Variable emission rate flag:
Short Term Model :
SEASON for seasonal; MONTH for monthly;
HROFDY for hour-of-day; STAR for speed-by-
stability; SEASHR for season -by- hour
Long Term Model :
SEASON for seasonal; MONTH for monthly;
SSTAB for season-by-stabi 1 i ty ; SSPEED for
season-by-speed ; STAR for speed-by-stabi 1 i ty ;
SSTAR for season-by-speed-and-stabi 1 i ty
Array of scalar emission rate factors, for:
SEASON, n=4; MONTH, n=12; HROFDY, n=24;
STAR, n=36; SSTAB, n=24; SSPEED, n=24;
SEASHR, n=96; SSTAR, n=144
B-15
-------
EMISUNIT
where:
CONCUNIT
where:
DEPOUNIT
where:
Emifac Emilbl Conlbl
or
Deplbl
Emi f ac
Emilbl
Conlbl
Deblbl
Emission rate factor used to adjust units of output
(default value is 1.0 E06 for CONC for grams to
micrograms; and 3600. for DEPOS, DDEP or WDEP for grams/sec
to grams/hour;
Note that ISCLT emission rates are automatically
adjusted for the number of hours in the STAR period
for deposition calculations)
Label to use for emission units (default is grams/sec)
Label to use for concentrations (default is mi crograms/m3)
Label to use for deposition (default is grams/m2)
Emifac Emilbl Conlbl (Applies to ISCST Only)
Emi f ac
Emilbl
Conlbl
Emission rate factor used to adjust units of output
for concentration (default value is 1.0 E06)
Label to use for emission units (default is grams/sec)
Label to use for concentrations (default is mi crograms/m3)
Emifac Emilbl Deplbl (Applies to ISCST Only)
Emi f ac
Emilbl
Deblbl
Emission rate factor used to adjust units of output
for deposition (default value is 3600.)
Label to use for emission units (default is grams/sec)
Label to use for deposition (default is grams/m2)
B-16
-------
TABLE B-4 (CONT.)
DESCRIPTION OF SOURCE PATHWAY KEYWORDS AND PARAMETERS
PARTDIAM
where:
MASSFRAX
where:
PARTDENS
where:
PARTSLIQ
where:
PARTSICE
where:
GAS-SCAV
Srcid (or Srcrng) Pdiam(i), i=l,Npd
S r c i d
Srcrng
Pdi am
Source identification code
Range of sources (inclusive) for which size categories apply
Array of particle diameters (microns)
Srcid (or Srcrng) Phi(i), i=l,Npd
Srcid
Srcrng
Phi
Source identification code
Range of sources (inclusive) for
Array of mass fractions for each
category
which mass fractions apply
particle size
Srcid (or Srcrng) Pdens(i), i=l,Npd
Srcid
Srcrng
Pdens
Source identification code
Range of sources (inclusive) for
Array of particle densities (g/cm
size category
which particle densities apply
3) for each
Srcid (or Srcrng) Scavcoef (i ) , i=l,Npd
Srcid
Srcrng
Scavcoef
Source identification code
Range of sources (inclusive) for
Scavenging coefficient (s-mm/hr)"
for each size category
which scavenging coefficients apply
for liquid precipitation
Srcid (or Srcrng) Scavcoef (i ), i=l,Npd
Srcid
Srcrng
Scavcoef
Source identification code
Range of sources (inclusive) for
Scavenging coefficient (s-mm/hr)"
for each size category
which scavenging coefficients apply
for frozen precipitation
Srcid (or Srcrng) LIQ or ICE Scavcoef
B-17
-------
where:
HOUREMIS
where:
SRCGROUP
where:
S r c i d
Srcrng
m
ICE
Scavcoef
Source identification code
Range of sources (inclusive) for which scavenging coefficent applies
Specifies that inputs are for liquid precipitation
Specifies that inputs are for frozen precipitation
Scavenging coefficient (s-mm/hr)'1 for liquid or frozen precipitation
for each size category
Emifil Srcid's Srcrng 's
Emifil
Srcid's
Srcrng 's
Specifies name of the hourly emission rate file
Discrete source IDs that are included in the hourly emission file
Source ID ranges that are included in the hourly emission file
Grpid Srcid's Srcrng's
G r p i d
Srcid's
Srcrng 's
Group ID (Grpid = ALL specifies group including all
sources), number of source groups limited by NGRP
parameter in the computer code
Discrete source IDs to be included in group
Source ID ranges to be included in group
Note: Card may be repeated with same Grpid if
more space is needed to specify sources
B-18
-------
TABLE B-5
DESCRIPTION OF RECEPTOR PATHWAY KEYWORDS
(APPLIES TO ISCST AND ISCLT)
RE Keywords
STARTING
ELEVUNIT
GRIDCART
GRIDPOLR
DISCCART
DISCPOLR
BOUNDARY
BOUNDELV
FINISHED
Type
M
0
0
0
0
0
0
0
M
- N
- N
- R1
- R1
- R1
- R1
- R1
- R
- N
Keyword Description
Identifies the start of RECEPTOR pathway inputs
Defines input units for receptor elevations (defaults to meters)
RE STARTING if used.
, must be
first keyword
after
Defines a Cartesian grid receptor network
Defines a polar receptor network
Defines the discretely placed receptor locations referenced to a
Defines the discretely placed receptor locations referenced to a
Cartesian system
polar system
Defines discrete polar receptor locations corresponding to minimum plant
for each 10 degree sector
Defines terrain elevations for discrete receptors specified with
BOUNDARY
boundary distances
keyword
Identifies the end of RECEPTOR pathway inputs
1) At least one of the following must be present: GRIDCART, GRIDPOLR, DISCCART,
DISCPOLR, or BOUNDARY. Multiple receptor networks can be specified in a single
run, including both Cartesian and polar, up to an overall maximum controlled by
the NREC parameter.
B-19
-------
TABLE B-6
DESCRIPTION OF RECEPTOR PATHWAY KEYWORDS AND PARAMETERS
(APPLIES TO ISCST AND ISCLT)
Keyword
ELEVUNIT
where:
GRIDCART
Parameters
METERS
METERS
FEET
Netid
or
or
STA
XYINC
XPNTS
YPNTS
ELEV
FLAG
END
FEET
Specifies input units for
meters
Specifies input units for
receptor elevations
receptor elevations
of
of feet
Note: This keyword applies to receptor elevations
only.
Xinit Xnum Xdelta Yini
Gridxl GridxZ GridxS ....
Gridyl GridyZ GridyS ....
Row Zelevl ZelevZ ZelevS
Row Zflagl ZflagZ ZflagS
t Ynum Ydelta
GridxN, and
GridyN
. . . Zel evN
... ZflagN
B-20
-------
where:
Netid
STA
XYINC
Xinit
Xnum
Xdelta
Yinit
Ynum
Ydelta
XPNTS
Gridxl
GridxN
YPNTS
Gridyl
GridyN
ELEV
Row
Zel ev
FLAG
Row
Zflag
END
Receptor network identification code (up to eight
alphanumeric characters)
Indicates STArt of GRIDCART subpathway, repeat for
each new Netid
Keyword identifying grid network generated from
x and y increments
Starting x-axis grid location in meters
Number of x-axis receptors
Spacing in meters between x-axis receptors
Starting y-axis grid location in meters
Number of y-axis receptors
Spacing in meters between y-axis receptors
Keyword identifying grid network defined by a series
of x and y coordinates
Value of first x-coordinate for Cartesian grid
Value of 'nth' x-coordinate for Cartesian grid
Keyword identifying grid network defined by a series
of x and y coordinates
Value of first y-coordinate for Cartesian grid
Value of 'nth' y-coordinate for Cartesian grid
Keyword to specify that receptor elevations follow
Indicates which row (y-coordinate fixed) is being
input
An array of receptor terrain elevations for
a particular Row
Keyword to specify that flagpole receptor heights
fol1ow
Indicates which row (y-coordinate fixed) is being
input
An array of receptor heights above local terrain
elevation for a particular Row (flagpole receptors)
Indicates END of GRIDCART subpathway, repeat for each
new Netid
B-21
-------
TABLE B-6 (CONT.)
DESCRIPTION OF RECEPTOR PATHWAY KEYWORDS AND PARAMETERS
(APPLIES TO ISCST AND ISCLT)
B-22
-------
GRIDPOLR
Netid STA
ORIG
or ORIG
DIST
DDIR
or GDIR
ELEV
FLAG
END
X i n i t Y i n i t ,
S r c i d
Ringl RingZ RingS ...
Dirl Dir2 Dir3
Dirnum Dirini Dirinc
Dir Zelevl ZelevZ ZelevS
Dir Zflagl ZflagZ ZflagS
RingN
DirN,
Zel evN
... ZflagN
B-23
-------
where:
Netid
STA
PRIG
Xinit
Yinit
S r c i d
DIST
Ringl
RingN
DDIR
Dirl
DirN
GDIR
Di rnum
D i r i n i
D i r i n c
ELEV
Di r
Zel ev
FLAG
Di r
Zflag
END
Receptor network identification code (up to eight
alphanumeric characters)
Indicates STArt of GRIDPOLR subpathway, repeat for
each new Netid
Optional keyword to specify the origin of the polar
network (assumed to be at x=0, y=0 if omitted)
x-coordinate for origin of polar network
y-coordinate for origin of polar network
Source ID of source used as origin of polar network
Keyword to specify distances for the polar network
Distance to the first ring of polar coordinates
Distance to the 'nth' ring of polar coordinates
Keyword to specify discrete direction radials for the
polar network
First direction radial in degrees (1 to 360)
The 'nth' direction radial in degrees (1 to 360)
Keyword to specify generated direction radials for
the polar network
Number of directions used to define the polar system
Starting direction of the polar system
Increment (in degrees) for defining directions
Keyword to specify that receptor elevations follow
Indicates which direction is being input
An array of receptor terrain elevations for a
particular direction radial
Keyword to specify that flagpole receptor heights
fol1ow
Indicates which direction is being input
An array of receptor heights above local terrain
elevation for a particular direction (flagpole
receptors)
Indicates END of GRIDPOLR subpathway, repeat for each
new Netid
B-24
-------
TABLE B-6 (CONT.)
DESCRIPTION OF RECEPTOR PATHWAY KEYWORDS AND PARAMETERS
(APPLIES TO ISCST AND ISCLT)
DISCCART
where:
DISCPOLR
where:
BOUNDARY
where:
BOUNDELV
Xcoord Ycoord (Zelev) (Zflag)
Xcoord
Ycoord
Zel ev
Zflag
x-coordinate for discrete receptor location
y-coordi nate for discrete receptor location
Elevation above sea level for discrete receptor
location (optional), used only for ELEV terrain
Receptor height (flagpole) above local terrain
(optional), used only with FLAGPOLE keyword
Srcid Dist Direct (Zelev) (Zflag)
S r c i d
Dist
Di rect
Zel ev
Zflag
Specifies source identification for which discrete
polar receptor locations apply (used to define the
origin for the discrete polar receptor)
Downwind distance to receptor location
Direction to receptor location, in degrees clockwise
from North
Elevation above sea level for receptor location
(optional), used only for ELEV terrain
Receptor height (flagpole) above local terrain
(optional), used only with FLAGPOLE keyword
Srcid Dist(i), i=l,36
Srcid
Dist
Specifies source identification for which boundary
distances apply
Array of 36 values corresponding to minimum plant
boundary distances for every 10-degree sector,
beginning with the 10 degree flow vector
Note: Discrete receptor coordinates are generated
with an origin referenced to the location
of the source identified with Srcid
Srci d Zel ev(i ) , i=l ,36
B-25
-------
where:
S r c i d
Zel ev
Specifies source identification for which boundary
distances apply
Array of 36 values corresponding to terrain elevation
for plant boundary distances for 10-degree sectors,
beginning with the 10 degree flow vector
B-26
-------
TABLE B-7
DESCRIPTION OF METEOROLOGY PATHWAY KEYWORDS
ME Keywords
STARTING
INPUTFIL
ANEMHGHT
SURFDATA
UAIRDATA
STARTEND
DAYRANGE
WDROTATE
WINDPROF
DTHETADZ
WINDCATS
AVESPEED
AVETEMPS
AVEMIXHT
AVEROUGH
FINISHED
Type
M
M
M
M
M
0
0
0
0
0
0
0
M
M
0
M
- N
- N
- N
- N
- N
- N
- R
- N
- R
- R
- N
M
- R
- R
- R
- N
Keyword Description
Identifies the start of METEOROLOGY pathway inputs
Describes input meteorological data file
Input height of anemometer above stack base
Describes surface meteorological station
Describes upper air meteorological station
Specifies start and end dates to be read from input meteorological data file (default is to
read entire file). (Applies to ISCST Only)
Specifies days or ranges of days to process (default is to process all data read in).
(Aoolies to ISCST Only)
May be used to correct for alignment problems of wind direction measurements, or to convert
wind direction from to flow vector
Input optional wind profile exponents
Input optional vertical potential temperature gradients
Input upper bounds of wind speed categories, five values input - sixth category is assumed to
have no upper bound. (Applies to Short Term Only)
Average (median) wind speed for each speed category in the STAR summary. (Applies to ISCLT
Only)
Average ambient temperatures for each stability category and season. (Applies to ISCLT Only)
Average mixing heights for each wind speed, stability category and season. (Applies to ISCLT
Only)
Roughness length for each season (Applies to ISCLT Only)
Identifies the end of METEOROLOGY pathway inputs
B'
- .
-------
TABLE B-8
DESCRIPTION OF METEOROLOGY PATHWAY KEYWORDS AND PARAMETERS
Keyword
INPUTFIL
where:
ANEMHGHT
where:
SURFDATA
where:
UAIRDATA
where:
Parameters
Metfil (Format)
Metfil
Format
Zref (Zrunit)
Zref
Zrunit
Specify filename for meteorological input file
Specify format for input file: options are to provide
FORTRAN read format for ASCII file,
(YR,MN,DY,HR,AFV (or WD) , WS , TA , KST , ZIRUR , ZIURB ) ;
use default ASCII format (412 , 2F9.4 , F6 . 1 , 12 , 2F7. 1)
if blank;
use free format if FREE;
use default ASCII format with hourly WINDPROF and
DTHETADZ if CARD ; or
use unformatted PCRAMMET file if UNFORM
Reference (anemometer) height above ground for
wind speed measurement; also assumed to be height
above stack base
Units of Zref: METERS or FEET (default is METERS)
Stanum Year (Name) (Xcoord Ycoord)
Stanum
Year
Name
Xcoord
Ycoord
Station number, e.g. 5-digit WBAN number for NWS
surface station
Year of data being processed (four digits)
Station name (optional)
x-coordinate of station location (m) (optional)
y-coordinate of station location (m) (optional)
Stanum Year (Name) (Xcoord Ycoord)
Stanum
Year
Name
Xcoord
Ycoord
Station number, e.g. 5-digit WBAN number for NWS
upper air station
Year of data being processed (four digits)
Station name (optional)
x-coordinate of station location (m) (optional)
y-coordinate of station location (m) (optional)
B-28
-------
STARTEND
where:
Strtyr Strtmn Strtdy (Strthr) Endyr Endmn Enddy (Endhr) (Applies to ISCST Only)
Strtyr
Strtmn
Strtdy
Strthr
Endyr
Endmn
Enddy
Endhr
Year of first record to be read
Month of first record to be read
Day of first record to be read
Hour of first record to be read (optional)
Year of last record to be read
Month of last record to be read
Day of last record to be read
Hour of last record to be read (optional)
Note: File read begins with hour 1 of the start
date and ends with hour 24 of the end date
if Stahr and Endhr are omitted.
B-29
-------
TABLE B-8 (CONT.)
DESCRIPTION OF METEOROLOGY PATHWAY KEYWORDS AND PARAMETERS
DAYRANGE
where:
STARDATA
where:
WDROTATE
where:
Rangel RangeZ
Rangel
RangeN
JAN FEB MAR AF
WINTER SPRING $
MONTH SEASON Qi
JAN
FEB
DEC
WINTER
SPRING
SUMMER
FALL
QUART1
QUARTZ
QUARTS
QUART4
MONTH
SEASON
QUARTR
ANNUAL
PERIOD
Rotang
Rotang
?ange3 ... RangeN (Applies to ISCST Only)
First range of days to process, either as individual
day (XXX) or as range (XXX-YYY); days may be input
as Julian dates (XXX) or as month and day (XX/YY)
The 'nth' range of days to process
3R MAY JUN JUL AUG SEP OCT NOVDEC (ISCLT Model)
SUMMER FALL or QUART1 QUARTZ QUARTS QUART4
JARTR ANNUAL
Option to calculate JANuary averages from STAR data
Option to calculate FEBruary averages from STAR data
Option to calculate DECember averages from STAR data
Option to calculate WINTER averages from STAR data
Option to calculate SPRING averages from STAR data
Option to calculate SUMMER averages from STAR data
Option to calculate FALL averages from STAR data
Option to calculate QUART1 averages from STAR data
Option to calculate QUARTZ averages from STAR data
Option to calculate QUARTS averages from STAR data
Option to calculate QUART4 averages from STAR data
Option to calculate averages for all twelve MONTHs
Option to calculate averages for all four SEASONS
Option to calculate averages for all four QUARTeRs
Option to calculate annual values from an ANNUAL STAR
summary
Option to calculate averages for the entire data
PERIOD
Specifies angle (in degrees) to rotate wind direction
measurements to correct for alignment problems;
value of Rotang is subtracted from WD measurements,
i.e., rotation is counterclockwise; may also be
used to adjust input of wind direction from values
to flow vector values by setting Rotang = 180
B-30
-------
WINDPROF
where:
DTHETADZ
where:
Stab Profl ProfZ ProfS Prof4 ProfB Prof6
Stab
Profl
ProfZ
ProfS
Prof4
ProfB
Prof6
Specifies stability category (A through F) for the
following six values by wind speed class
Wind speed profile exponent for first speed class
Wind speed profile exponent for second speed class
Wind speed profile exponent for third speed class
Wind speed profile exponent for fourth speed class
Wind speed profile exponent for fifth speed class
Wind speed profile exponent for sixth speed class
Note: Card is repeated for each stability class
Stab Dtdzl DtdzZ DtdzS Dtdz4 DtdzB Dtdz6
Stab
Dtdzl
DtdzZ
DtdzS
Dtdz4
DtdzB
Dtdz6
Specifies stability category (A through F) for the
following six values by wind speed class
Vertical temperature gradient for first speed class
Vertical temperature gradient for second speed class
Vertical temperature gradient for third speed class
Vertical temperature gradient for fourth speed class
Vertical temperature gradient for fifth speed class
Vertical temperature gradient for sixth speed class
Note: Card is repeated for each stability class
B-31
-------
TABLE B-8 (CONT.)
DESCRIPTION OF METEOROLOGY PATHWAY KEYWORDS AND PARAMETERS
WINDCATS
where:
AVESPEED
where:
AVETEMPS
where:
AVEMIXHT
Wsl Ws2 Ws3 Ws4 Ws5 (Applies to Short Term Only)
Wsl
Ws2
Ws3
Ws4
Ws5
Upper bound of first wind speed category (m/s)
Upper bound of second wind speed category (m/s)
Upper bound of third wind speed category (m/s)
Upper bound of fourth wind speed category (m/s)
Upper bound of fifth wind speed category (m/s)
(sixth category is assumed to have no upper bound)
Wsl Ws2 Ws3 Ws4 Ws5 Ws6 (Applies to ISCLT Only)
Wsl
Ws2
Ws3
Ws4
Ws5
Ws6
Median speed of first wind speed category (m/s)
Median speed of second wind speed category (m/s)
Median speed of third wind speed category (m/s)
Median speed of fourth wind speed category (m/s)
Median speed of fifth wind speed category (m/s)
Median speed of sixth wind speed category (m/s)
Aveper Tal Ta2 Ta3 Ta4 Ta5 Ta6 (Applies to ISCLT Only)
Aveper
Tal
Ta2
Ta3
Ta4
Ta5
Ta6
Specifies averaging period (see AVERTIME keyword)
for the following temperatures (K)
Average temperature of stability category A
Average temperature of stability category B
Average temperature of stability category C
Average temperature of stability category D
Average temperature of stability category E
Average temperature of stability category F
Note: Card is repeated for each averaging period
Aveper Stab Mixhtl Mixht2 MixhtS Mixht4 MixhtB Mixht6
(Applies to ISCLT Only)
B-32
-------
where:
AVEROUGH
where:
Aveper
Stab
Mixhtl
MixhtZ
MixhtS
Mixht4
MixhtB
Mixht6
Specifies averaging period (see AVERTIME keyword)
for the following mixing heights (m)
Specifies stability category (A through F) for the
following six values by wind speed class
Average mixing height for first speed class
Average mixing height for second speed class
Average mixing height for third speed class
Average mixing height for fourth speed class
Average mixing height for fifth speed class
Average mixing height for sixth speed class
Note: Card is repeated for each stability class
and for each averaging period
Aveper ZO (Applies to ISCLT Only)
Aveper
ZO
Specifies averaging period (AVERTIME keyword)
for the roughness length (m)
Roughness Length
Note: Card is repeated for each averaging period
B-33
-------
TABLE B-9
DESCRIPTION OF TERRAIN GRID PATHWAY KEYWORDS
TG Keywords
STARTING
INPUTFIL
LOCATION
ELEVUNIT
FINISHED
Type
M - N
M - N
M - N
0 - N
M - N
Keyword Description
Identifies the start of TERRAIN GRID pathway inputs
Describes input terrain grid data file
Specifies the origin of the terrain grid
Defines input units for terrain grid elevations (defaults
to meters)
Identifies the end of TERRAIN GRID pathway inputs
Note: The Terrain Grid (TG) pathway is optional. The TG pathway is only used for calculating dry depletion in elevated
or complex terrain. If it is omitted, then the terrain profile is linearly interpolated along the plume path from
source to receptor for dry depletion calculations.
B-34
-------
TABLE B-10
DESCRIPTION OF TERRAIN GRID PATHWAY KEYWORDS AND PARAMETERS
INPUTFIL
where:
LOCATION
where:
ELEVUNIT
where:
Tgfile
Tgfile
Specifies filename for the terrain grid data file
Xorig Yorig (Units)
Xori g
Yori g
Units
UTM X-coordinate of origin for the source and receptor locations
UTM Y-coordinate of origin for the source and receptor locations
Units for Xoriq and Yoriq (FEET, KM, or METERS - default
is in METERS)
METERS or FEET
METERS
FEET
Specifies input units for terrain grid elevations of
meters
Specifies input units for terrain grid elevations of feet
Note: This keyword applies to terrain grid elevations
only.
B-35
-------
TABLE B-ll
DESCRIPTION OF EVENT PATHWAY KEYWORDS
(APPLIES TO ISCEV MODEL ONLY)
EV Keywords
STARTING
EVENTPER
EVENTLOC
FINISHED
Type
M - N
M - R
M - R
M - N
Keyword Description
Identifies the start of EVENT pathway inputs
Describes data and averaging period for an event
Describes receptor location for an event
Identifies the end of EVENT pathway inputs
B-36
-------
TABLE B-12
DESCRIPTION OF EVENT PATHWAY KEYWORDS AND PARAMETERS
(APPLIES TO ISCEV MODEL ONLY)
Keyword
EVENTPER
where:
EVENTLOC
where:
Parameters
Evname Aveper Grpid Date
Name
Grpid
Aveper
Date
Specify name of event to be processed (e.g. H2H24ALL),
(up to eight alphanumeric characters)
Specify source group ID for event
Specify averaging period for event
Specify data period for event (ending YYMMDDHH for
averaging period)
Evname XR= Xr YR= Yr (Zelev) (Zflag)
or
RNG= Rnq DIR= Dir (Zelev) (Zflaq)
Evname
XR=
YR=
RNG=
DIR=
Zel ev
Zflag
Specify name of event to be processed (e.g. H2H24ALL),
(up to eight alphanumeric characters)
X-coordinate for event (discrete Cartesian receptor)
Y-coordinate for event (discrete Cartesian receptor)
Distance range for event (discrete polar receptor)
Radial direction for event (discrete polar receptor)
Terrain elevation for event (optional)
Receptor height above ground for event (optional)
Note: EVENT locations can be input as either discrete Cartesian receptors (XR=. YR=)
or as discrete polar receptors (RNG=. DIR=). Events that are specified in the
file generated by the ISCST model (CO EVENTFIL card) are always given as
discrete Cartesian coordinates. Discrete polar receptors are assumed to be
relative to an origin of (0,0).
B-37
-------
TABLE B-13
DESCRIPTION OF OUTPUT PATHWAY KEYWORDS
OU Keywords
STARTING
RECTABLE
MAXTABLE
DAYTABLE
MAXIFILE
POSTFILE1
PLOTFILE1
TOXXFILE
EVENTOUT2
FINISHED
Type
M -
0 -
0 -
0 -
0 -
n
0 -
0 -
M -
N
R
R
N
R
R
R
R
N
N
Keyword Description
Identifies the start of OUTPUT pathway inputs
Option to specify value(s) by receptor for output
Option to summarize the overall maximum values
Option to print summaries for each averaging period for each day processed
ISCST Only)
Option to list events exceeding a threshold value to file (if CO EVENTFIL
these events are included in the input file generated for the EVENT model)
Only)
Option to write results to a mass storage file for postprocessing. (Appli
Option to write certain results to a storage file suitable for input to pi
(Appl ies to
option is used,
. (ADD! ies to ISCST
es to ISCST Only)
otting routines
Option to write results to a storage file suitable for input to the TOXX model component of
TOXST or the RISK model component of TOXLT
Specifies the level of output information provided by the EVENT model. (Applies to ISCEV
Only)
Identifies the end of OUTPUT pathway inputs
1) POSTFILE is used to output concurrent concentration values for particular source
groups and averaging times across the receptor network, suitable for
postprocessing, such as might be done for implementing the intermediate terrain
policy. PLOTFILE is used to output specific design values, such as second high
concentrations, across the receptor network, suitable for plotting concentration
contours.
2) EVENTOUT is the only keyword on the OU pathway for the Short Term EVENT model.
B-38
-------
TABLE B-14
DESCRIPTION OF OUTPUT PATHWAY KEYWORDS AND PARAMETERS
Keyword
RECTABLE
where:
Parameters
Aveper FIRST SECOND . . . SIXTH (Short Term Model) or
Aveper 1ST 2ND . . . 6TH (Short Term Model)
INDSRC and/or SRCGRP ( Long Term Model )
Aveper
FIRST
SECOND
SIXTH
1ST
2ND
6TH
INDSRC
SRCGRP
Averaging period to summarize with high values
(keyword ALLAVE specifies all averaging periods)
Select summaries of FIRST highest values by receptor
Select summaries of SECOND highest values by receptor
Select summaries of SIXTH highest values by receptor
Select summaries of 1ST highest values by receptor
Select summaries of 2ND highest values by receptor
Select summaries of 6TH highest values by receptor
Note: If two keywords are input separated by a
dash (e.g. FIRST-THIRD), then summaries of
all high values in that range are provided.
The number of high values allowed is con-
trolled by the NVAL parameter in the computer
code (initially set at 3). Also, if the
CO EVENTFIL keyword is exercised, then the
events generated by the RECTABLE keyword are
included in the input file for EVENT model.
Specifies that summaries of individual source values
for each receptor point will be provided
Specifies that summaries of source group values for
each receptor point will be provided
Note: Either INDSRC or SRCGRP or both may be
specified
B-39
-------
MAXTABLE
Aveper
Maxnum
Maxnum
INDSRC
and/or SRCGRP and/or SOCONT
(Short
(Long
Term Model )
Term Model )
where:
Aveper
Maxnum
INDSRC
SRCGRP
SOCONT
Averaging period to summarize with maximum values
(keyword ALLAVE specifies all averaging periods)
Specifies number of overall maximum values to
summarize (number of maximum values permitted is
limited by the NMAX parameter in the computer code,
initially set at 50 for Short Term and 10 for Long
Term)
Specifies that summaries of maximum values for
individual sources will be provided (independent of
source group maxima)
Specifies that summaries of maximum values by source
group will be provided
Specifies that summaries of individual source contri-
butions for locations of maximum source group
values will be provided
Note: Any combination of Long Term parameters
is acceptable
B-40
-------
TABLE B-14 (CONT.)
DESCRIPTION OF OUTPUT PATHWAY KEYWORDS AND PARAMETERS
DAYTABLE
where:
MAXIFILE
where:
POSTFILE
where:
Avperl AvperZ AvperS . . . (Applies to ISCST Only)
Avperl
Averaging period to summarize with values by receptor
for each day of data processed (keyword ALLAVE for
first parameter specifies all averaging periods)
Aveper Grpid Thresh Filnam (Funit) (Applies to ISCST Only)
Aveper
Grpid
Thresh
Fi 1 nam
Funi t
Specifies averaging period for list of values equal to
or exceeding a threshold value
Specifies source group to be output to file
Threshold value (e.g. NAAQS) for list of exceedances
Name of disk file to store maximum values
Optional parameter to specify the file unit
Note: If the CO EVENTFIL keyword is exercised,
then the events generated by the MAXIFILE
keyword are included in the input file for
the EVENT model .
Aveper Grpid Format Filnam (Funit) (Applies to ISCST Only)
Aveper
Grpid
Format
Fi 1 nam
Funi t
Specifies averaging period to be output to file,
e.g., 24 for 24-hr averages, PERIOD for period
averages
Specifies source group to be output to file
Specifies format of file, either UNFORM for
unformatted files or PLOT for formatted files for
pi otti ng
Specifies filename for output file
Optional parameter to specify the file unit
B-41
-------
PLOTFILE
where:
TOXXFILE
where:
EVENTOUT
where:
Aveper Grpid Hivalu Filnam (Funit) (ISCST short term values)
Aveper Grpid Filnam (Funit) (ISCLT model and ISCST
PERIOD averages)
Aveper
Grpid
Hivalu
Fi 1 nam
Funi t
Specifies averaging period to be output to file,
e.g., 24 for 24-hr averages, PERIOD for period
averages, WINTER for winter averages, etc.
Specifies source group to be output to file
Specifies high value summary (e.g. FIRST . SECOND . 1ST .
2ND, etc.) to be output to file (must be selected on
a RECTABLE card)
Specifies filename for output file
Optional parameter to specify the file unit
Aveper Cutoff Filnam (Funit) (ISCST short term values)
Aveper Grpid Filnam (Funit) (ISCLT model)
Aveper
Cutoff
Grpid
Fi 1 nam
Funi t
Specifies averaging period to be output to file,
e.g., 1 for 1-hr averages, PERIOD for period
averages ( LT only), WINTER for winter averages, etc.
Specifies cutoff (threshold) value in g/m3 for outputting
results for ISCST model
Specifies source group to be output to file (LT only)
Specifies filename for output file
Optional parameter to specify the file unit
SOCONT or DETAIL (Applies to ISCEV Only)
SOCONT
DETAIL
Specifies the option to provide source contribution
information only in the event output
Specifies the option to include hourly concentrations
for each source and hourly meteorological data in
the event output
B-42
-------
APPENDIX C. UTILITY PROGRAMS
C.I CONVERTING INPUT RUNSTREAM FILES - STOLDNEW
The STOLDNEW.EXE program is a file conversion utility that may be used to convert
original ISCST model (EPA, 1987a) input files to the proper format for the ISCST2 model
(EPA, 1992). With the exception of the source inputs for the dry deposition algorithm,
the ISCST2 model inputs generated by STOLDNEW will be compatible with the ISCST3 model.
To run the file conversion utility, type STOLDNEW at the DOS prompt. The program
will prompt the user for the name of the original ISCST input file being converted and
for the name of the new file to be generated in the ISCST2 format. The program will
also generate a file called SUMMARY.OLD that contains a summary of model inputs in the
same format as would appear at the beginning of an original ISCST model run.
Even though the STOLDNEW utility should convert most ISCST input files without any
difficulty, users are strongly encouraged to check the results of STOLDNEW carefully
before using the input file with the ISCST3 model. The purpose of this is primarily to
check for rounding of the inputs in the conversion process. Some inputs that may vary
over a considerable range, such as the emission rate, are converted using an Fortran G
format with a full seven significant digits. However, most inputs are converted using a
Fortran F format specifier that uses a fixed number of decimal places. Some rounding is
possible on some of these fixed format inputs, depending on how many decimal places were
used for inputting the data in the original format.
C-l
-------
The STOLDNEW utility program will prompt the user to input additional filenames
where appropriate. Specifically, the program prompts for the name of the meteorological
data file (including a DOS path if desired), which is inserted into the appropriate
field on the ME INPUTFIL keyword. If the option for using unformatted preprocessed data
was specified for the original ISCST input, then the meteorology data filename should be
the name of the file containing the preprocessed data. If the "card image"
meteorological data option was specified for the original ISCST model input, then the
hourly "card image" meteorological data are included as part of the original runstream
option file. In this case, the STOLDNEW program prompts for the name of the file that
it uses for writing out the card image data in the ASCII format used by the ISCST3
model. The format field on the ME INPUTFIL card will include the default ASCII format
used by the ISCST3 model (which would have the same effect as leaving the field blank),
unless the card image data includes hourly wind profile exponents or hourly vertical
potential temperature gradients. In the latter case, STOLDNEW will insert the CARD
keyword for the meteorological data format on the ME INPUTFIL card.
Another case where the STOLDNEW program will prompt for a filename is when the
option for generating a separate file of concurrent concentration values is selected in
the original runstream file (ISW(5)=1). In this case, the program will request the name
to use for the concentration file, and will insert that name in the appropriate field
for the OU POSTFILE keyword inputs. A separate POSTFILE card will be generated for each
combination of averaging period and source group, with all of the concentration results
being written to a single file on file unit 20. This will result in a concentration
file that is nearly identical to the file generated by the original ISCST model.
C-2
-------
It should be noted that the ISCST3 model does not support the use of hourly decay
coefficients, which were allowed for the original ISCST model when "card image"
meteorological data were used. If hourly decay coefficients are detected in the
original ISCST runstream file, then STOLDNEW will write a warning message to the screen
and within the new runstream file indicating that the hourly values of decay
coefficients will be ignored. The only other option available in the original ISCST
model that is not available with ISCST3 is the option to list the meteorological data
for each day processed as part of the main printed output file. In lieu of this option,
a separate utility program, called METLIST, is available with the ISC2 package that
produces a listing of meteorological data for the period of interest. The METLIST
program is described in more detail in Section C.3.
C.2 CONVERTING UNFORMATTED PCRAMMET FILES TO ASCII FORMATTED FILES - BINTOASC
The BINTOASC.EXE program is a utility program that converts unformatted (binary)
meteorological data files generated by the PCRAMMET or MPRM preprocessor programs to the
default ASCII format used by the ISCST3 Model. The ASCII data file consists of
sequential hourly records.
To run this program, type BINTOASC at the DOS prompt. The program will prompt for
the name of the unformatted data input file and the name of the ASCII formatted output
file. The BINTOASC program will convert unformatted data files generated by a
Microsoft-compiled version of PCRAMMET, as well as files generated by versions of
PCRAMMET or MPRM compiled with either the Lahey or the Ryan-MeFarland FORTRAN compilers.
The program will write a message to the screen indicating which of the three types of
files has been identified. If the program encounters an error reading the data file,
C-3
-------
then a message will be written to the screen indicating which compilers are supported.
The program may also have encountered a read error due to the use of "short integers"
(INTEGER*2) in the storing of some of the data in the unformatted file. The program
assumes that all integer variables occupy four bytes of storage.
Once the type of unformatted file has been determined the program will prompt the
user as follows:
Do You Want to Convert the Entire Data File? (Y or N)
If the user responds with either a 'Y' or a 'y', then the program will convert the
entire data file (up to 366 days for a leap year). If the user responds with either an
'N' or an 'n', then the program will prompt the user as follows:
Enter the Start Date and End Date (e.g. 1,365):
The user can select a single day or a range of (Julian) days within the year to convert
to the ASCII file.
If the BINTOASC program encounters a calm hour in the unformatted data file, which
is identified by a wind speed of 1.0 m/s and a flow vector equal to the flow vector for
the previous hour, then it writes out a wind speed of 0.0 for that hour, which is
interpreted by the ISC2 Short Term models as a calm hour. The flow vector variable
written to the ASCII file corresponds to the randomized flow vector in the unformatted
data file. The structure of the PCRAMMET-generated unformatted data file and the
default ASCII file are described in detail in Appendix F.
C-4
-------
C.3 LISTING HOURLY METEOROLOGICAL DATA - METLIST
The METLIST.EXE program is a utility program that creates a listing file of
meteorological data for a specified day or range of days, which can be sent to a
printer. The program lists one day of data per page, with appropriate column headers
for the meteorological variables. The original version of the ISCST model included an
option to print the hourly meteorological data within the main output file. This option
has not been included in the ISCST3 model. The user can use the METLIST program instead
to create a listing for the data period of interest, and refer to that listing as needed
to examine the meteorological data. Since the ISCST3 model also uses ASCII sequential
hourly files (see Sections 3.5.1 and C.I), the meteorological data file can be examined
directly through an editor or listing program, or the ASCII file itself can be printed.
Therefore, the need for an option to list meteorological data within the program has
been reduced. Also, the ISCEV2 model contains the option to list the hourly
meteorological data for specific events that are of interest to the user.
To use this program, type METLIST from the command line prompt. The program will
prompt the user for the following information:
Enter Meteorology File Name: (Enter the name of the file containing the
meteorological data)
Options for File Formats are:
ASCII
UNFORM
FREE
CARD
Fortran format specifier
C-5
-------
Enter File Format: (Select the format of the meteorological file by entering one of
the four keywords above or by entering a Fortran format specifier, e.g.
(4I2,2F9.4,F6.1,I2,2F7.1) )
Enter Output File Name: (Enter the name of the file to which the meteorological
data listing will be stored)
Enter Day Range: (Enter the Julian start day and Julian end day, e.g. 1,10)
The ASCII data format option for the METLIST program corresponds with the default
ASCII format used by the ISCST3 and ISCEV3 models. The Fortran specifier for this
format is '(412,2F9.4,F6.1,12,2F7.1)'. The other format options are described in
Section 3.5.1.1. The METLIST program was compiled using the Microsoft FORTRAN Compiler,
and therefore only supports unformatted data files generated by Microsoft versions of
PCRAMMET or MPRM. To use unformatted data files generated by either the Lahey or the
Ryan-MeFarland compiler, the user should first convert the unformatted data file to the
default ASCII format using the BINTOASC utility program (described in Section C.2), and
then use the METLIST program and select the ASCII format option.
C-6
-------
APPENDIX D. BATCH FILE DESCRIPTIONS FOR
COMPILING THE MODELS ON A PC
D.I MICROSOFT/DOS VERSIONS
The ISC models were developed on an IBM-compatible PC using the Microsoft
Optimizing FORTRAN Compiler (Version 5.1). The models are provided on the Support Center
for Regulatory Air Models (SCRAM) Bulletin Board System (BBS) as executable files
designed to run on DOS PCs. These DOS versions were compiled with the Microsoft
emulator library option that allows the models to utilize a math coprocessor if
available, but also run in the absence of one. The batch file provided for compiling
the ISCST model with the Microsoft compiler (FLMSISCS.BAT) includes the following
commands:
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
/c
/c
/c
/c
/c
/c
/c
/c
/c
/c
/c
/c
/c
/c
/c
/c
/c
/c
/c
/c
/FPi
/FPi
/FPi
/FPi
/FPi
/FPi
/FPi
/FPi
/FPi
/FPi
/FPi
/FPi
/FPi
/FPi
/FPi
/FPi
/FPi
/FPi
/FPi
/FPi
/AH ISCST3.FOR
/AH /DMICRO PCCODE.FOR
/AH SETUP.FOR
/AH COSET.FOR
/AH SOSET.FOR
/AH RESET.FOR
/AH MESET.FOR
/AH TGSET.FOR
/AH OUSET.FOR
/AH INPSUM.FOR
/AH METEXT.FOR
/AH CALC1.FOR
/AH CALC2.FOR
/AH DEPFLUX.FOR
/AH PRISE.FOR
/AH SIGMAS.FOR
/AH CALC3.FOR
/AH CALC4.FOR
/AH PITAREA.FOR
/AH OUTPUT.FOR
D-l
-------
LINK iFLMSISCS.LRF
where /c instructs the compiler to compile without linking; the /FPi option instructs
the compiler to use in-line instructions for floating point operations and link with an
emulator library (uses 80x87 coprocessor if present); and the /AH option that the huge
memory model be used, allowing arrays or common blocks to exceed 64K. The /DMICRO
option for the PCCODE.FOR source file instructs the compiler to use the conditional
compilation blocks defined for the Microsoft compiler. These enable the PC-specific
features, such as writing the date and time on each page of the output file and writing
an update to the screen on the status of processing. Each of the source files (*.FOR)
for the ISCST model are listed separately in this batch file, which assumes that all of
the source code modules and the include files are in a single directory, or that the
compiler has been setup to search for the include files in the appropriate directory.
The command line options for the compiler make full use of the compiler's optimization
routines to speed up the code. To disable optimization, the /Od option would be added.
Disabling optimization will increase the model's execution time by about 10 percent, and
will also increase the size of the code.
Once the source files have been compiled successfully, and object (.OBJ) files have
been generated for each source file, the model is ready to be linked and an executable
file created. The executable file on the SCRAM BBS was linked using a memory overlay
manager so that only certain portions of the code are resident in memory at any given
time. This allows for a more efficient use of available memory by the model, and
therefore allows for larger runs to be performed than would be possible without using
overlays. This is accomplished with the following command line for the linker provided
with the Microsoft compiler, which is included in the link response file, FLMSISCS.LRF:
D-2
-------
/E /SE:256 ISCST3+PCCODE+SETUP+(COSET)+(SOSET)+(RESET)+(MESET)+(TGSET)+(OUSET)+(INPSUM)+(METEXT+
CALCl+CALC2+CALC3+PRISE+SIGMAS+CALC4+DEPFLUX+PITAREA)+(OUTPUT)
The /E option instructs the linker to produce a packed executable file that occupies
less disk space. The /SE:256 option increases the number of segments allowed to 256.
With this memory overlay structure, the ISCST3, PCCODE and SETUP modules are always
memory resident, and any module or group of modules within parentheses are overlayed
into the same area of memory only when needed. Linking without the overlay manager will
increase the minimum load size for the executable file by about 200K for the ISCST
model.
Similar batch files are available for compiling and linking the ISCLT and ISCEV
models. The batch file for the ISCLT model, FLMSISCL.BAT, includes the following
commands:
FL /c /FPi /AH ISCLT3.FOR
FL /c /FPi /AH /DMICRO PCCODELT.FOR
FL /c /FPi /AH SETUPLT.FOR
FL /c /FPi /AH COSETLT.FOR
FL /c /FPi /AH SOSETLT.FOR
FL /c /FPi /AH RESETLT.FOR
FL /c /FPi /AH MESETLT.FOR
FL /c /FPi /AH TGSETLT.FOR
FL /c /FPi /AH OUSETLT.FOR
FL /c /FPi /AH INPSUMLT.FOR
FL /c /FPi /AH METEXTLT.FOR
FL /c /FPi /AH CALC1LT.FOR
FL /c /FPi /AH CALC2LT.FOR
FL /c /FPi /AH CALC3LT.FOR
FL /c /FPi /AH PRISELT.FOR
FL /c /FPi /AH SIGMASLT.FOR
FL /c /FPi /AH PITAREAL.FOR
FL /c /FPi /AH DEPFLUX.FOR
FL /c /FPi /AH OUTPUTLT.FOR
D-3
-------
LINK iFLMSISCL.LRF
The only difference between this and the file for the ISCST model is the source file
names. This file invokes the following command line from the FLMSISCL.LRF link response
file:
/E /SE:256 ISCLT3+PCCODELT+SETUPLT+(COSETLT)+(SOSETLT)+(RESETLT)+(MESETLT)+(TGSETLT)+(OUSETLT)+
(INPSUMLT)+(METEXTLT+CALC1LT+CALC2LT+CALC3LT+PRISELT+SIGMASLT+PITAREAL+DEPFLUX)+(OUTPUTLT)
The batch file for the ISCEV model, FLMSISCE.BAT, includes the following commands:
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
/c
/c
/c
/c
/c
/c
/c
/c
/c
/c
/c
/c
/c
/c
/c
/c
/c
/c
LINK
/FPi
/FPi
/FPi
/FPi
/FPi
/FPi
/FPi
/FPi
/FPi
/FPi
/FPi
/FPi
/FPi
/FPi
/FPi
/FPi
/FPi
/FPi
/AH
/AH
/AH
/AH
/AH
/AH
/AH
/AH
/AH
/AH
/AH
/AH
/AH
/AH
/AH
/AH
/AH
/AH
iFLMSISCE.
EVISCST3
/DMICRO
EVSETUP.
EVCOSET.
EVSOSET.
EVMESET.
EVTGSET.
EVEVSET.
EVOUSET.
EVINPSUM
EVMETEXT
EVCALC1.
EVCALC2.
EVPRISE.
EVSIGMAS
EVPITARE
DEPFLUX.
EVOUTPUT
,LRF
.FOR
EVPCCODE.FOR
FOR
FOR
FOR
FOR
FOR
FOR
FOR
.FOR
.FOR
FOR
FOR
FOR
.FOR
.FOR
FOR
.FOR
which invokes the following command from the ISCEV.LRF link response file:
/E /SE:256 EVISCST3+EVPCCODE+EVSETUP+CEVCOSET)+(EVSOSET)+(EVMESET)+(EVTGSET)+(EVEVSET)+(EVOUSET)+
(EVINPSUM)+(EVMETEXT+EVCALCl+EVCALC2+EVPRISE+EVSIGMAS)+(EVOUTPUT)
D-4
-------
D.2 LAHEY/EXTENDED MEMORY VERSIONS
While the ISC models were developed on an IBM-compatible PC using the Microsoft
Optimizing FORTRAN Compiler (Version 5.1), the models have also been compiled using the
Lahey F77L-EM/32 Fortran Compiler (Version 5.2) to generate PC-executable files capable
of utilizing extended memory on 80386 and 80486 PCs with at least 8 MB of RAM for the
Short Term model and at least 4 MB of RAM for the Long Term model. The extended memory
(EM) versions of the models are also provided on the SCRAM BBS. The batch file provided
for compiling the ISCST model (ISCST3EM.EXE) with the Lahey compiler (F77LISCS.BAT)
includes the following commands:
F77L3 ISCST3.FOR /NO /NW
F77L3 PCCODE.FOR /NO /NW /D1LAHEY
F77L3 SETUP.FOR /NO /NW
F77L3 COSET.FOR /NO /NW
F77L3 SOSET.FOR /NO /NW
F77L3 RESET.FOR /NO /NW
F77L3 MESET.FOR /NO /NW
F77L3 TGSET.FOR /NO /NW
F77L3 OUSET.FOR /NO /NW
F77L3 INPSUM.FOR /NO /NW
F77L3 METEXT.FOR /NO /NW
F77L3 CALC1.FOR /NO /NW
F77L3 CALC2.FOR /NO /NW
F77L3 PRISE.FOR /NO /NW
F77L3 SIGMAS.FOR /NO /NW
F77L3 CALC3.FOR /NO /NW
F77L3 CALC4.FOR /NO /NW
F77L3 DEPFLUX.FOR /NO /NW
F77L3 PITAREA.FOR /NO /NW
F77L3 OUTPUT.FOR /NO /NW
3861 ink §F77LISCS.LRF
cfig386 ISCST3EM.EXE -nosignon
where /NO option instructs the compiler not to list the compiler options to the screen,
the /NW option suppresses a certain level of warning messages, and the /D1LAHEY option
D-5
-------
for the PCCODE.FOR source file instructs the compiler to use the conditional compilation
blocks defined for the Lahey compiler. These conditional blocks of code enable the
PC-specific features, such as writing the date and time on each page of the output file
and writing an update to the screen on the status of processing. Each of the source
files (*.FOR) for the ISCST model are listed separately in this batch file,
which assumes that all of the source code modules and the include files are in a single
directory, or that the compiler has been setup to search for the include files in the
appropriate directory. The '3861ink @F77LISCS.LRF' links the model using the
F77LISCS.LRF link response file, which includes the following command:
ISCST3,PCCODE,SETUP,COSET,SOSET,RESET,MESET,TGSET,OUSET,INPSUM,METEXT,CALC1,CALC2,
CALC3,CALC4,PRISE,SIGMAS,DEPFLUX,PITAREA,OUTPUT -STUB RUNB -EXE ISCST3EM.EXE - PACK
There are no memory overlays used for the Lahey versions, since they make use of
extended memory.
Similar batch files are available for the ISCLT (F77LISCL.BAT) and the ISCEV
(F77LISCE.BAT) models, except for the specification of the appropriate source file names
provided in the previous section. The executable filenames for these models are
ISCLT3EM.EXE and ISCEVEM.EXE.
D-6
-------
APPENDIX E. EXPLANATION OF ERROR MESSAGE CODES
E.I INTRODUCTION
One of the significant operational improvements of the ISC models is an improved
error handling procedure. The input runstream is checked to identify parameters that
are missing or potentially in error, and the input source and meteorological data are
checked and flagged for possible erroneous values.
The ISC models use a "defensive programming" approach to eliminate as much as
possible of the user's work in debugging the input runstream file. Also, a great deal
of effort has been made to eliminate the possibility of run time errors, such as "divide
by zero," and to point out questionable input data. Error messages are reported to the
user in two ways. A summary of messages is provided in the main output result file, and
the user can also request a detailed message listing file.
Message Summary: Whether the user selects a detailed error listing file or not,
the ISC models output a summary of messages within the output result file. This message
table gives the number of messages of each type, together with a detailed list of all
the fatal errors and warning messages. During setup processing, if no errors or warnings
are generated, then the model simply reports to the user that "SETUP Finishes
Successfully."
Detailed Message Listing File: The ISC models provide the option of saving a
detailed list of all messages generated by the model in a separate output file. The
user can select this option by specifying the keyword "ERRORFIL" followed by a filename
E-l
-------
inside the COntrol pathway. For example, the following statements will save all the
error messages to an ASCII text file named "errormsg.out":
CO STARTING
ERRORFIL errormsg.out
CO FINISHED
E.2 THE OUTPUT MESSAGE SUMMARY
There are two message summaries provided in the standard output file of the ISC
models. The first one is located after the echo of input runstream file images and
before the input data summary. This summary will take one of two forms, depending on
whether any fatal error or non-fatal warning messages were generated, and also depending
on whether the option to RUN or NOT to run was selected on the CO RUNORNOT card. If
there are no errors or warnings generated during the setup processing, and the RUN
option was selected, then the model simply reports that "SETUP Finishes Successfully."
If any fatal errors or warning messages were generated during the setup processing, or
if the option NOT to run was selected, then a more detailed summary is provided. This
summary provides a message count for each type of message, and a detailed listing of
each fatal error and warning message generated. The second message summary table is
located at the very end of the standard output result file, and it sums up the messages
generated by the complete model run - both setup processing and run-time processing.
E-2
-------
An example of a setup processing message summary is shown in Figure E-l
E-3
-------
*** Message Summary For The ISC3 Model Setup ***
Summary of Total Messages
A Total of 0 Fatal Error Message(s)
A Total of 0 Warning Message(s)
A Total of 0 Information Message(s)
******** FATAL ERROR MESSAGES ********
*** NONE ***
******** WARNING MESSAGES ********
*** NONE ***
***********************************
*** SETUP Finishes Successfully ***
***********************************
E-4
-------
FIGURE E-l. EXAMPLE OF AN ISC MESSAGE SUMMARY
E.3 DESCRIPTION OF THE DETAILED MESSAGE LAYOUT
Three types of messages can be produced by the models during the processing of
input runstream images and during model calculations. These are described briefly
below:
• Errors that will halt any further processing, except to identify additional
error conditions (type E);
• Warnings that do not halt processing but indicate possible errors or suspect
conditions (type W); and
• Informational messages that may be of interest to the user but have no direct
bearing on the validity of the results (type I).
The messages have a consistent structure which contains the pathway ID, indicating
which pathway the messages are generated from; the message type followed by a
three-digit message number; the line number of the input runstream image file for setup
messages (or the meteorology hour number for runtime messages); the name of the module
(e.g. the subroutine name) from which the message is generated; a detailed message
corresponding to the message code; and an 8-character simple hint to help the user spot
the possible source of the problem.
E-5
-------
The following is an example of a detailed message generated from the CO pathway:
CO E100 8 EXPATH: Invalid Pathway Specified. The Troubled Pathway is FF
The message syntax is explained in more detail below (values in parentheses give the
column numbers within the message line for each element):
E-6
-------
644444444444444444444444444444444444447
5 PW Txxx LLLL mmmmmm : MESSAGE Hints 5
944444444444444444444444444444444444448
* * * * * * * +)))))))))))))))))))))))))))))))))))))),
* * * * * * * *Hints to help you determine the nature*
* * * * * * .)>*of errors (keyword, pathway where the *
* * * * * * *error occurs ,.. .etc. ) (73:80) *
* * * * * * .))))))))))))))))))))))))))))))))))))))-
* * * * * * +)))))))))))))))))))))))))))))))))))))))))))),
* * * * * .)))>*Detailed message for this code (22:71) *
* * * * * .))))))))))))))))))))))))))))))))))))))))))))-
* * * * * +)))))))))))))))))))))))))))))))))))))))))))),
* * * * * *Name of the code module from which the *
* * * * .)))))))))))>*message is generated (14:19) *
* * * * .))))))))))))))))))))))))))))))))))))))))))))-
* * * * +)))))))))))))))))))))))))))))))))))))))))))),
* * * * *The line number of the input runstream image*
* * * * *file where the message occurs; If message *
* * * . )))))))))))))))))>*occurs in runtime operation, the hour number*
* * * *of the meteorology file is given (9:12) *
* * * .))))))))))))))))))))))))))))))))))))))))))))-
* * * +)))))))))))))))))))))))))))))))))))))))))))),
* * -)))))))))))))))))))))>*Numeric message code (a 3)digit number) (5 : 7) *
* * .))))))))))))))))))))))))))))))))))))))))))))-
* * +)))))))))))))))))))))))))))))))))))))))))))),
* .)))))))))))))))))))))))>*Message type (E, W, I) (4:4)
*Pathway ID (CO, SO, RE, ME, EV, or OU) (1:2)
•))))))))))))))))))))))))))>*or MX for met data extraction,
*or CN for calculation messages
The three message types are identified with the letters E (for errors) , W (for
warnings) , and I (for informational messages) . The 3 -digit message codes are grouped
into general categories corresponding to the different stages of the processing. Theses
categories are:
E-7
-------
100 - 199 Input Runstream Image Structure Processing
200 - 299 Parameter Setup Processing
300 - 399 Data and Quality Assurance Processing
400 - 499 Run Time Message Processing
500 - 599 Input/Output Message Processing
A detailed description of each of the message codes currently used in the models is
provided in the next section.
E.4 DETAILED DESCRIPTION OF THE ERROR/MESSAGE CODES
INPUT RUNSTREAM IMAGE STRUCTURE PROCESSING, 100-199
This type of message indicates problems with the basic syntax and/or structure of
the input runstream image. Typical messages include errors like "Missing mandatory
keyword", "Illegal Keyword", ..., etc. If a fatal error of this kind is detected in a
runstream image, a fatal error message is written to the message file and any attempt to
process data is prohibited, although the remainder of the runstream file is examined for
other possible errors. If a warning occurs, data may still be processed, although the
inputs should be checked carefully to be sure that the condition causing the warning
does not indicate an error.
100 Invalid Pathway Specified. The pathway ID should be a 2 character string. It
should be one of the following: CO for control pathway, SO for source pathway, RE
for receptor pathway (or EV for event pathway for ISCEV model), ME for meteorology
data setting pathway, and OU for output format pathway. Its position is normally
confined to columns 1 and 2 (1:2) of the input runstream file. However, the model
Eo
- o
-------
does allow for a shift of the entire input runstream file of up to 3 columns. If
the inputs are shifted, then all input records must be shifted by the same amount.
The invalid pathway is repeated at the end of the message.
105 Invalid Keyword Specified. The keyword ID should be an 8-character string. Its
position is normally confined to columns 4 to 11 (4:11) of the input runstream
file. However, the model does allow for a shift of the entire input runstream file
of up to 3 columns. If the inputs are shifted, then all input records must be
shifted by the same amount. There should be a space between keyword ID and any
other data fields. For a list of valid keywords, refer to Appendix A or Appendix
B. The invalid keyword is repeated at the end of the message.
110 Keyword is Not Valid for This Pathway. The input keyword is a valid 8-character
string, but it is not valid for the particular pathway. Refer to Appendix A,
Appendix B or Section 3 for the correct usage of the keyword. The invalid keyword
is repeated at the end of the message.
115 Starting and Finishing Statements do not match. Only One STARTING and one FINISHED
statement,respectively, is allowed at the very beginning and the very end of each
pathway block. Check the position and frequency to make sure the input runstream
file meets the format requirement. The pathway during which the error occurs is
included at the end of the message.
120 Pathway is Out of Sequence. The pathways are not input in the correct order. The
correct order is CO, SO, RE, ME, and OU for the ISCST and ISCLT models, and CO, SO,
ME, EV, and OU for the ISCEV model. The offending pathway is given as a hint.
125 Missing FINISHED Statement - Runstream file is incomplete. One or more FINISHED
statements are missing. A 5-digit status variable is given as a hint. Each digit
corresponds to a pathway in the appropriate order, and is a '!' if the pathway is
complete and a '0' if the FINISHED is missing. For example, a status of '10111'
indicates that the SO pathway was missing a FINISHED statement. Normally such an
error will generate additional messages as well.
130 Missing Mandatory Keyword. To run the model, certain mandatory keywords must
present in the input runstream file. For a list of mandatory keywords, see
Appendix A or Appendix B. For more detailed information on keyword setup, see the
description of message code 105. The missing keyword is included with the message.
E-9
-------
135 Duplicate Non-repeatable Keyword Encountered. More than one instance of a
non-repeatable keyword is encountered. For a list of non-repeatable keywords, see
Appendix A or Appendix B. The repeated keyword is included with the message.
140 Invalid Order of Keyword. A keyword has been placed out of the acceptable order.
The order for most keywords is not critical, but the relative order of a few
keywords is important for the proper interpretation of the input data. The keyword
reference in Section 3 identifies any requirements for the order of keywords. The
keyword that was out of order is included with the message.
143 Conflicting Options: UNFORM with Dry or Wet Deposition. The dry and wet deposition
algorithms of the Short Term model require additional meteorological variables that
are not included in the unformatted data file generated by the PCRAMMET or MPRM
meteorological processors. The user must use PCRAMMET or MPRM to generate an ASCII
meteorological data file with the necessary variables.
144 Conflicting Options: NOSMPL with FLAT Terrain. The NOSMPL option specifies that
only the COMPLEXl algorithms will be used, whereas the FLAT option specifies that
flat terrain will be used (i.e., all receptor elevations are at stack base
elevation). Since the COMPLEXl algorithms apply only to receptor elevations that
are above the release height, these two options are in conflict.
145 Conflicting Options: MULTYEAR and Re-Start Option. The multiple year option for
processing PM-10 values makes use of the re-start routines in the model with some
slight changes to handle the period averages from year to year. As a result, the
MULTYEAR keyword cannot be specified with either the SAVEFILE or INITFILE keywords.
150 Conflicting Options: MULTYEAR for Wrong Pollutant. The multiple year option is
provided specifically for the processing of PM-10 values to obtain the
"high-sixth-high in five years" design value. Its treatment of the high short term
values for multiple year periods is not consistent with existing air quality
standards for other pollutants. To use the MULTYEAR option, the user must specify
a pollutant type (on the CO POLLUTID card) of PM-10, PM10, or OTHER.
151 CO ELEVUNIT card is obsolescent: use RE ELEVUNIT card. With the release of the
ISC3 models, the CO ELEVUNIT card has been designated as obsolescent - it will
still be processed as before by the model, but the user is encouraged to use the
E-10
-------
new RE ELEVUNIT card instead. The RE ELEVUNIT card has the same effect as the
original CO ELEVUNIT card.
152 ELEVUNIT card must be first for this pathway. The ELEVUNIT card must be the first
non-commented card after STARTING when used on the SO or RE pathway. This
requirement is made in order to simplify reviewing runstream files to determine the
elevation units used for sources and receptors.
153 Cannot use CO ELEVUNIT card with ELEVUNIT card for the SO, RE or TG pathway. With
the release of the ISC3 models, the CO ELEVUNIT card has been designated as
obsolescent - it will still be processed as before by the model if it is the only
CO ELEVUNIT card encountered in the runstream. This is to allow for compatibility
of the model with old input files. However, if any of the new ELEVUNIT cards (on
the SO, RE or TG pathways) are used, then the CO ELEVUNIT card must be removed.
155 Conflicting Decay Keyword. The ISC models allow for the user to specify the rate
of exponential decay either in terms of the half-life (HALFLIFE keyword) or the
decay coefficient (DCAYCOEF keyword). If both keywords are specified, then only
the first one will be used, and inputs for the second one will be ignored.
157 EMISUNIT keyword used with more than one output type. If both concentration and
deposition are being output for the ISCST model, then the EMISUNIT keyword cannot
be used. To specify emission or output units, the CONCUNIT and/or DEPOUNIT keyword
should be used.
158 EMISUNIT keyword used with CONCUNIT or DEPOUNIT keyword. The EMISUNIT keyword may
be used if a single output type (CONG, DEPOS, DDEP or WDEP) is being generated,
whereas the CONCUNIT or DEPOUNIT keywords must be used if more than one output type
is generated.
160 Duplicate ORIG Secondary Keyword for GRIDPOLR. Only one origin card may be
specified for each grid of polar receptors. The network ID for the effected grid
is included with the message.
170 Invalid Secondary Key for Receptor GRID. The network ID for the effected grid is
included with this message. Refer to Appendix B for the correct syntax of secondary
keywords.
E-ll
-------
175 Missing Secondary Keyword END for Receptor Grid. The END secondary keyword is
required for each grid of receptors input by the user (keywords GRIDCART and
GRIDPOLR). It signals the end of inputs and triggers the processing of data for
that particular network.
180 Conflicting Secondary Keyword for Receptor Grid. Two incompatible secondary
keywords have been input for the same grid of receptors, e.g. GDIR and DDIR for the
keyword GRIDPOLR, where GDIR specifies to generate directions with uniform spacing,
and DDIR specifies that discrete, non-uniform directions are being specified.
185 Missing Receptor Keywords. No Receptors Specified. Since none of the RE pathway
keywords are mandatory, a separate error check is made to determine if any of the
RE keywords are specified. At least one of the following keywords must be present:
GRIDCART, GRIDPOLR, DISCCART, DISCPOLR, or BOUNDARY.
190 No Keywords for OU Pathway and No PERIOD or ANNUAL Averages. All of the OU pathway
keywords are optional, and in fact the model will run if no keywords are specified
on the OU pathway as long as PERIOD or ANNUAL averages are being calculated.
However, if there are no OU keywords and no PERIOD or ANNUAL averages, then there
will be no output generated by the model, and this fatal error message will be
generated.
195 Incompatible Option Used With SAVEFILE or INITFILE. Either a non-fatal message to
warn the user that DAYTABLE results will be overwritten if the model run is
re-started, or a fatal error message generated if the TOXXFILE option is selected
with either the SAVEFILE or INITFILE options.
PARAMETER SETUP PROCESSING, 200-299
This type of message indicates problems with processing of the parameter fields for
the runstream images. Some messages are specific to certain keywords, while others
indicate general problems, such as an invalid numeric data field. If a fatal error of
this kind is detected in a runstream image, a fatal error message is written to the
message file and any attempt to process data is prohibited, although the remainder of
the runstream file is examined for other possible errors. If a warning occurs, data
E-12
-------
may still be processed, although the inputs should be checked carefully to be sure that
the condition causing the warning does not indicate an error.
200 Missing Parameter(s). No options were selected for the indicated keyword. Check
Appendix B for the list of parameters for the keyword in question.
201 Not Enough Parameters Specified For The Keyword. Check if there are any missing
parameters following the indicated keyword. See Appendix B for the required
keyword parameters.
202 Too Many Parameters Specified For The Keyword. Refer to Appendix B or Section 3
for the list of acceptable parameters.
203 Invalid Parameter Specified. The inputs for a particular parameter are not valid
for some reason. Refer to Appendix B or Section 3. The invalid parameter is
included with the message.
204 Option Parameters Conflict. Forced by Default to: Some parameters under the
indicated keyword conflict with the other model parameters setting. Refer to
Appendix B or Section 3 for the correct parameter usage. The default setting is
specified with the message.
205 No Option Parameter Setting. Forced by Default to: No setting was specified for a
particular parameter. Refer to Appendix B or Section 3 for the correct parameter
usage. The default setting is specified with the message.
206 Regulatory DFAULT Specified With Non-default Option. The DFAULT option on the CO
MODELOPT card always overrides the specified non-default option, and a warning
message is generated.
207 No Parameters Specified. Default Values Used For. The keyword for which no
parameters are specified is included with the message. Refer to Appendix B or
Section 3 for a discussion of the default condition.
208 Illegal Numerical Field Encountered. The model may have encountered a
non-numerical character for a numerical input, or the numerical value may exceed
E-13
-------
the limit on the size of the exponent, which could potentially cause an underflow
or an overflow error.
209 Negative Value Appears For A Non-negative Variable. The effected variable name is
provided with the message.
210 Number of Short Term Averages Exceeds Maximum. The user has specified more short
term averages on the CO AVERTIME card than the model array limits allow. This
array limit is controlled by the NAVE PARAMETER specified in the MAIN1.INC file.
The value of NAVE is provided with the message.
211 Duplicate Parameter(s) Specified for Keyword. A duplicate parameter or set of
parameters has been specified for the indicated keyword. For example, if more than
one POSTFILE keyword is included for the same averaging period and source group,
then this error message will be generated.
212 END Encountered Without (X,Y) Points Properly Set. This error occurs during
setting up the grid of receptors for a Cartesian Network. This message may occur
for example if X-coordinate points have been specified without any Y-coordinate
points for a particular network ID.
213 ELEV Inputs Inconsistent With Option: Input Ignored. This happens when the user
inputs elevated terrain heights for receptors when the TERRHGTS option is FLAT.
The input terrain heights are ignored and the model proceeds with FLAT terrain
modeling.
214 ELEV Inputs Inconsistent With Option: Defaults Used. This happens when the user
does not input elevated terrain heights for receptors when the TERRHGTS option is
ELEV. The model assumes that the missing terrain heights are at 0.0 meters for
those receptors and proceeds with ELEV terrain modeling.
215 FLAG Inputs Inconsistent With Option: Input Ignored. This happens when the user
inputs receptor heights above ground for flagpole receptors when the FLAGPOLE
keyword option has not been specified. The input flagpole heights are ignored in
the model calculations.
216 FLAG Inputs Inconsistent With Option: Defaults Used. This happens when the user
does not input receptor heights above ground for flagpole receptors when the
E-14
-------
FLAGPOLE keyword option has been specified. The model assumes that the missing
flagpole heights are equal to the default value specified on the CO FLAGPOLE card.
If no default height is specified on the FLAGPOLE card, then a default of 0.0
meters is assumed.
217 More Than One Delimiter In A Field. For example, 12//34 is an illegal input data
item for the DAYRANGE card, and STACK1--STACK-20 is an illegal specification for a
range of sources.
218 Number of (X,Y) Points Not Match With Number Of ELEV Or FLAG. Check the number of
elevated terrain heights or flagpole receptor heights for the gridded network
associated with the indicated line number in the runstream file.
219 Number Of Receptors Specified Exceeds Maximum. The user has specified more
receptors on the RE pathway than the model array limits allow. This array limit is
controlled by the NREC PARAMETER specified in the MAIN1.INC file. The value of NREC
is provided with the message.
220 Missing Origin (Use Default = 0,0) In GRIDPOLR. This is a non-fatal warning
message to indicate that the ORIG secondary keyword has not been specified for a
particular grid of polar receptors. The model will assume a default origin of
(X=0, Y=0).
221 Missing Distance Setting In Polar Network. No distances have been provided
(secondary keyword DIST) for the specified grid of polar receptors.
222 Missing Degree Or Distance Setting In Polar Network. Missing a secondary keyword
for the specified grid of polar receptors.
223 Missing Distance or Degree Field. No data fields have been specified for the
indicated secondary keyword.
224 Number of Receptor Networks Exceeds Maximum. The user has specified more receptor
networks of gridded receptors on the RE pathway than the model array limits allow.
This array limit is controlled by the NNET PARAMETER specified in the MAIN1.INC
file. The value of NNET is provided with the message.
E-15
-------
225 Number of X-Coords Specified Exceeds Maximum. The user has specified more
X-coordinate values for a particular grid of receptors than the model array limits
allow. This array limit is controlled by the IXM PARAMETER specified in the
MAIN1.INC file. The value of IXM is provided with the message.
226 Number of Y-Coords Specified Exceeds Maximum. The user has specified more
Y-coordinate values for a particular grid of receptors than the model array limits
allow. This array limit is controlled by the IYM PARAMETER specified in the
MAIN1.INC file. The value of IYM is provided with the message.
227 No Receptors Were Defined on the RE Pathway. Either through lack of inputs or
through errors on the inputs, no receptors have been defined.
228 Default(s) Used for Missing Parameters on Keyword. Either an elevated terrain
height or a flagpole receptor height or both are missing for a discrete receptor
location. Default value(s) will be used for the missing parameter(s).
229 Too Many Parameters - Inputs Ignored on Keyword. Either an elevated terrain height
or a flagpole receptor height or both are provided when the corresponding option
has not been specified. The unneeded inputs are ignored.
230 Not Enough Numerical Values Specified. For example, less than 36 distance fields
may have been specified for a particular group of BOUNDARY receptors.
231 Too Many Numerical Values Specified. For example, more than 36 distance fields may
have been specified for a particular group of BOUNDARY receptors.
232 Number Of Specified Sources Exceeds Maximum. The user has specified more sources
than the model array limits allow. This array limit is controlled by the NSRC
PARAMETER specified in the MAIN1.INC file. The value of NSRC is provided with the
message.
233 Building Dimensions Specified for a Non-POINT Source. Building dimensions can only
be specified for a POINT source, since the VOLUME, AREA and OPENPIT source
algorithms do not include building downwash.
234 Too Many Sectors Input. For example, the user may have input too many building
heights or widths for a particular source.
E-16
-------
235 Number of Source Groups Specified Exceeds Maximum. The user has specified more
source groups than the model array limits allow. This array limit is controlled by
the NGRP PARAMETER specified in the MAIN1.INC file. The value of NGRP is provided
with the message.
236 Not Enough BUILDHGTs Specified for a Source ID. There should be 36 building
heights for Short Term and 16 for Long Term.
237 Not Enough BUILDWIDs Specified for a Source ID. There should be 36 building widths
for Short Term and 16 for Long Term.
238 Not Enough LOWBOUNDs Specified for a Source ID. There should be 36 lower bound
flags specified for Short Term and 16 for Long Term.
239 Not Enough QFACTs Specified for a Source ID. The number of variable emission rate
factors specified for a particular source is less than the model expects based on
the variable emission rate flag. Check the EMISFACT keyword on the SO pathway in
Appendix B of Section 3 for the appropriate number.
240 Inconsistent Number of Particle Size Categories for a particular source. The
number of parameters must be the same for the PARTDIAM, MASSFRAX and PARTDENS
keywords for a particular source.
242 No Particle Size Categories Specified for Source ID. There were no settling/removal
categories specified for the indicated source. When modeling for total deposition,
the user must include the PARTDIAM, MASSFRAX and PARTDENS keywords for each source.
243 No Scavenging Coefficients Specified for Source ID. There were no scavenging
coefficients specified for the indicated source. When modeling for total
deposition, wet deposition, or wet depletion, the user must include the PARTSLIQ
and PARTSICE keywords for particulate sources or the GAS-SCAV keyword for gaseous
sources.
244 Too Many Settling and Removal Parameters specified for a particular source. The
limit is controlled by the NPDMAX PARAMETER in the computer code, set initially to
20.
E-17
-------
245 Number of Particle Size Categories Exceeds Maximum. The user has specified more
settling/removal categories than the model array limits allow. This array limit is
controlled by the NPDMAX PARAMETER specified in the MAIN1.INC file. The value of
NPDMAX is provided with the message.
248 No Sources Were Defined on the SO Pathway. There must be at least one LOCATION
card and one SRCPARAM card to define at least one source on the SO pathway. Either
no cards were input or there were errors on the inputs.
250 Duplicate XPNT/DIST or YPNT/DIR Specified for GRID. One of the grid inputs, either
an X-coordinate, Y-coordinate, polar distance range or polar direction, has been
specified more than once for the same grid of receptors. This generates a non-fatal
warning message.
252 Duplicate Receptor Network ID Specified. A network ID for a grid of receptors
(GRIDCART or GRIDPOLR keyword) has been used for more that one network.
255 Boundary Receptor Distances Not Defined Yet. The user has input the BOUNDELV
keyword for a particular source before any BOUNDARY keyword has been specified for
that source.
260 Number of Emission Factors Exceeds Maximum. The user has selected an option for
variable emission rate factors that exceeds the array storage limit for emission
rate factors. The array limit is controlled by the NQF PARAMETER specified in the
MAIN1.INC file. The value of NQF is provided with the message.
270 Number of High Values Specified Exceeds Maximum. The user has selected a high
short term value on the OU RECTABLE card that exceeds the array storage limit for
high values by receptor. The array limit is controlled by the NVAL PARAMETER
specified in the MAIN1.INC file. The value of NVAL is provided with the message.
280 Number of Maximum Values Specified Exceeds Maximum. The user has selected a value
for the number of overall maximum values on the OU MAXTABLE card that exceeds the
array storage limit for overall maximum values. The array limit is controlled by
the NMAX PARAMETER specified in the MAIN1.INC file. The value of NMAX is provided
with the message.
E-18
-------
285 Number of Output Types Specified Exceeds Maximum (for Short Term only). The user
has specified more than the maximum number of output types allowed (CONG, DEPOS,
DDEP, and/or WDEP). The number of output types is controlled by the NTYP PARAMETER
specified in the MAIN1.INC file. The value of NTYP is provided with the message.
290 Number of Events Specified Exceeds Maximum. The user has specified more events
than the ISCEV model array limits allow. The array limit is controlled by the NEVE
PARAMETER specified in the EVMAIN1.INC file. The value of NEVE is provided with
the message.
SETUP DATA AND QUALITY ASSURANCE PROCESSING, 300-399
This type of message indicates problems with the actual values of the parameter
data on the input runstream image. The basic structure and syntax of the input card is
correct, but
one or more of the inputs is invalid or suspicious. These messages include quality
assurance checks on various model inputs. Typical messages will tell the consistency of
parameters and data for the setup and run of the model. If a fatal error of this kind
is detected in a runstream image, a fatal error message is written to the message file
and any attempt to process data is prohibited. If a warning occurs, data may or may not
be processed, depending on the processing requirements specified within the run stream
input data.
300 Specified Source ID Has Not Been Defined Yet. The message indicates that the user
attempts to use a source ID on a keyword before defining this source ID on a SO
LOCATION card. It could indicate an error in specifying the source ID, an omission
of a LOCATION card, or an error in the order of inputs.
310 Attempt to Define Duplicate LOCATION Card for Source. There can be only one
LOCATION card for each source ID specified. The source ID is included with the
message.
E-19
-------
315 Attempt to Define Duplicate SRCPARAM Card for Source. There can be only one
SRCPARAM card for each source ID specified. The source ID is included with the
message.
320 Source Parameter May Be Out-of-Range for Parameter. The value of one of the source
parameters may be either too large or too small. The name of the parameter is
provided with the message. Use the line number provided to locate the card in
question.
322 Release Height Exceeds the Effective Depth for an OPENPIT Source. The release
height for an OPENPIT source is measured from the base (bottom) of the pit. If the
release height exceeds the effective depth of the pit, calculated from the lateral
dimensions and volume of the pit, a fatal error message is generated.
323 No Particle Categories Specified for an OPENPIT Source. Since the OPENPIT
algorithm is applicable for particulate emissions, particle category data must be
specified for open pit sources using the PARTDIAM, MASSFRAX, and PARTDENS keywords.
This fatal error message will be generated if no particle information is specified
for an open pit source.
325 Negative Exit Velocity (Set=1.0E-5) for Source ID. The exit velocity for the
specified source ID was input as a negative value. Since the model currently
cannot handle sources with downward momentum, the exit velocity is set to a very
small value (l.OE-5 m/s) and modeling proceeds. This non-fatal message is generated
to warn the user that the input may be in error.
330 Mass Fraction Parameters Do Not Sum to 1. (within +/- 2 percent) for a particular
source.
332 Mass Fraction Parameter Out-of-Range for a particular source. Must be between 0.0
and 1.0, inclusive.
334 Particle Density Out-of-Range for a particular source. Must be greater than 0.0.
340 Possible Error in the Anemometer Height. The value of the anemometer height may be
either too large or too small
E-20
-------
350 Julian Day Out Of Range. This error occurs if the Julian Day selected is less than
zero or greater than 366. Check ME setup to ensure the Julian Day selection.
355 Specified Averaging Period Not Being Calculated. This is a non-fatal warning
message for the ISCLT model generated when average temperatures or mixing heights
are specified for a STAR averaging period that was not specified on the CO AVERTIME
card. The inputs will be ignored, and processing will continue.
360 2-digit Year Specified. Valid for the range 1901-2099. Four-digit years are valid
for the entire range of Gregorian dates, but two digit years are accepted.
362 Averaging Time Conflict: PERIOD with ANNUAL Data. The PERIOD average is not
compatible with the specification of an ANNUAL STAR summary on the CO AVERTIME card
or the ME STARDATA card.
364 Averaging Time Conflict: PERIOD with MONTH and SEASON or QUARTR. The PERIOD
average is not compatible with the presence of monthly STAR summaries and seasonal
or quarterly summaries in the same data file.
366 Possible Averaging Time Conflict: PERIOD Average Only. The CO AVERTIME card has
specified the PERIOD average only. There could be a conflict unless the ME
STARDATA card is used to specify the STAR summaries in the data file.
368 Averaging Time Conflict: PERIOD Average with No STARDATA. The ISCLT model cannot
process the PERIOD average unless the STAR summaries in the data file are
identified, either through the CO AVERTIME card or the ME STARDATA card.
369 Averaging Time Conflict: Both SEASON and QUARTR. The ISCLT model cannot process
both seasonal and quarterly STAR summaries in the same model run, since they occupy
the same areas in the data storage.
370 Invalid Date: 2/29 In a Non-leap Year. The year has been identified as a leap
year, and a date of 2/29 (February 29) has been specified on the DAYRANGE card.
Check the year and/or the date specification.
380 This Input Variable is Out-of-Range. The indicated value may be too large or too
small. Use the line number to locate the card in question, and check the variable
for a possible error.
E-21
-------
385 Averaging period does not equal 1-hour averages for the TOXXFILE option for the
ISCST model. The ISCST model will generate TOXXFILE outputs for other averaging
periods, but the TOXX model component of TOXST currently supports only the 1-hour
averages. This is a non-fatal warning message.
390 Invalid Averaging Period Specified for the Event. An invalid averaging period has
been specified for the event name indicated for the ISCEV model. This may be an
averaging period that was not selected on the CO AVERTIME card, or it may be an
averaging period of greater than 24 hours, which cannot be handled by ISCEV.
391 Aspect ratio (length/width) of an area source is greater than 10. The new area
source algorithm in the ISC3 model allows for specifying area sources as elongated
rectangles, however, if the aspect ratio exceeds 10 a warning message will be
printed out. The user should subdivide the area so that each subarea has an aspect
ratio of less than 10.
392 Aspect ratio (length/width) of an open pit source is greater than 10. The new open
pit algorithm in the ISC3 model allows for specifying open pit sources as elongated
rectangles, however, if the aspect ratio exceeds 10 a warning message will be
printed out. Due to the way open pit sources are treated by the model, an open pit
source should not be subdivided. The user should therefore use extra caution when
interpreting results of the open pit algorithm for sources that exceed an aspect
ration of 10.
393 Terrain grid value differs by more than 50 percent from the source elevation for
the specified source. The ISC model will compare source elevations to an
interpolated elevation from a terrain grid file (from the TG pathway) if one is
used. A warning message is generated if the elevations differ by more than 50
percent. Several warning messages could indicate an error in specifying the
elevation units for either source elevations or terrain elevations. Elevation
units are in meters by default, but may be specified as feet by using the ELEVUNIT
keyword.
394 Terrain grid value differs by more than 50 percent from the receptor elevation for
the specified receptor. The ISC model will compare source elevations to an
interpolated elevation from a terrain grid file (from the TG pathway) if one is
used. A warning message is generated if the elevations differ by more than 50
percent. Several warning messages could indicate an error in specifying the
E-22
-------
elevation units for either receptor elevations or terrain elevations. Elevation
units are in meters by default, but may be specified as feet by using the ELEVUNIT
keyword.
395 Monthly QFACT Specified With No Monthly Averages. The monthly variable emission
rate option for the ISCLT model can only be used with monthly STAR summaries.
398 STAR Data Not Available for the Specified Average. The STAR summaries identified
on the ME STARDATA card do not include one of the averaging periods selected on the
CO AVERTIME card for the ISCLT model.
RUNTIME MESSAGE PROCESSING, 400-499
This type of message is generated during the model run. Setup processing has been
completed successfully, and the message is generated during the performance of model
calculations. Typical messages will tell the information and error during the model
run. If a fatal error of this kind is detected during model execution, a fatal error
message is written to the message file and any further processing of the data is
prohibited. The rest of the meteorological data file will be read and quality assurance
checked to identify additional errors. If a warning occurs, data may or may not be
processed, depending on the processing requirements specified within the run stream
input data.
400 No Convergence Reached in SUB. CUBIC. The CUBIC module is used to solve a cubic
equation for the Schulman-Scire BLP plume rise and for the vertical virtual
distance for URBAN mode. The routine uses Newton's method, which is an iterative
approach to determining the solution to the cubic equation. This message is
generated if the routine does not converge within 24 iterations. The message is
provided for informational purposes and processing will continue. The date of
occurrence is provided with the message.
E-23
-------
410 Flow Vector Out-of-Range. The flow vector must be between 0 and 360 degrees,
inclusive. The date of occurrence is provided with the message (in the form of
year, month, day, hour as YYMMDDHH)
420 Wind Speed Out-of-Range. The wind speed value may be either too large or too
small. An error is generated if the speed is less than 0.0, and a warning is
generated if the speed is greater than 30.0 m/s. The date of occurrence is
provided with the message (in the form of year, month, day, hour as YYMMDDHH).
430 Ambient Temperature Data Out-of-Range. The ambient temperature value may be either
too large or too small. A warning is generated if the temperature is less than
250.0 K or greater than 320 K. The date of occurrence is provided with the message
(in the form of year, month, day, hour as YYMMDDHH).
435 Surface Roughness Length Out-of-Range. The surface roughness value may be too
small or missing. A warning is generated if the surface roughness length is less
than l.OE-05 meters. The value is set to l.OE-05 to avoid possible division by
zero. The date of occurrence is provided with the message (in the form of year,
month, day, hour as YYMMDDHH).
440 Calm Hour Identified in Meteorology Data File. This message is generated if a calm
hour is identified, and provides the date of occurrence (in the form of year,
month, day, hour as YYMMDDHH). The message will be generated whether or not the
calms processing option is used.
450 Error in Meteorology File - Record Out of Sequence. There is an error in the
sequence of the hourly meteorological data file. The message also provides the
date of occurrence (in the form of year, month, day, hour as YYMMDDHH).
455 Date/Time Mismatch on Hourly Emission File. There is mismatch in the date/time
field between the meteorological data file and the hourly emission file. The
message also provides the date of the occurrence from the surface/scalar file (in
the form of year, month, day, hour as YYMMDDHH).
460 Missing Hour Identified in Meteorology Data File. At least one of the
meteorological variables is missing or invalid for the hour specified (in the form
of year, month, day, hour as YYMMDDHH). If the missing data processing option is
not used, then this message will be generated and any further calculations with
E-24
-------
the data will be aborted. The model will continue to read through the
meteorological data file and check the data.
470 Mixing Height Value is Less Than or Equal to 0.0. This is an informational message
that may indicate an error in the meteorological data file. Since the plume will
always be above a mixing of 0.0 or less, no calculations are performed for the hour
specified (in the form of year, month, day, hour as YYMMDDHH).
480 Sum of STAR Frequencies Does Not Total to 1.0. The ISCLT model accepts STAR data
files with either normalized frequencies or with a frequency count. For normalized
frequencies, the sum of the STAR frequencies should total 1.0. If the sum is less
than 0.98 or greater than 1.02, this non-fatal warning message is generated. The
actual sum of the frequencies for each STAR summary is included in the printed
output file at the end of the listing for the STAR frequency input. The frequency
array is not automatically normalized to 1.0 as was done by the original ISCLT
model.
INPUT/OUTPUT MESSAGE PROCESSING, 500-599
This type of message is generated during the model input and output. Typical
messages will tell the type of I/O operation (e.g., opening, reading or writing to a
file), and the type of file. If a fatal error of this kind is detected in a runstream
image, a fatal error message is written to the message file and any attempt to process
data is prohibited. If a warning occurs, data may or may not be processed, depending on
the processing requirements specified within the run stream input data.
500 Fatal Error Occurs During Opening of the Data File. The file specified can not be
opened properly. This may be the runstream file itself, the meteorological data
file, or one of the special purpose output files. This may happen when the file
called is not in the specified path, or an illegal filename is specified. If no
errors are found in the filename specification, then this message may also indicate
that there is not enough memory available to run the program, since opening a file
causes a buffer to be opened which takes up additional memory in RAM. For the
E-25
-------
special purpose output files, the hint field includes character string identifying
the type of file and the file unit number, e.g., 'PLTFL312'.
510 Fatal Error Occurs During Reading of the File. File is missing, incorrect file
type, or illegal data field encountered. Check the indicated file for possible
problems. If the file is identified as 'DEP-MET', then the problem may be that the
additional surface variables needed for the new deposition algorithms are missing.
As with error number 500, this message may also indicate that there is not enough
memory available to run the program if no other source of the problem can be
identified.
520 Fatal Error Occurs During Writing to the File. Similar to message 510, except that
it occurs during a write operation.
530 Error Occurs Reading Met Station or Year: File Says. This error occurs only with
the ST models. The surface and upper air station numbers and years specified on
the ME pathway do not agree with the values on the first record of the
meteorological data file. The value from the file is printed out to help resolve
the problem.
540 No RECTABLE/MAXTABLE/DAYTABLE for Averaging Period. No printed output options
selected for a particular averaging period. This is a non-fatal warning condition
for the ISCST model.
550 File Unit/Name Conflict for the Output Option. This error indicates that a problem
exists with the filename and file unit specification for one of the special purpose
output files. The associated keyword is provided as a hint. The same filename may
have been used for more than one file unit, or vice versa.
560 User Specified File Unit < 20 for OU Keyword. A file unit of less than 20 has been
specified for the indicated special purpose output files. This is a fatal error
condition. File units of less than 20 are reserved for system files. Specify a
unit number in the range of 20 to 100.
565 Possible conflict With Dynamically Allocated FUNIT. A file unit specified for the
indicated special purpose output files is in the range > 100, and may therefore
conflict with file units dynamically allocated for special purpose files by the
model. This is typically a non-fatal warning condition.
E-26
-------
570 Problem Reading Temporary Event File for Event. The ISCST model stores high value
events in a temporary file that is used to create the input file for the ISCEV
model, if requested, and also to store the high values for the summary tables at
the end of the printed output file. A problem has been encountered reading this
file, possibly because the concentration or deposition value was too large and
overflowed the fixed format field of F14.5.
575 End-of-File Reached Trying to Read STAR Data. The ISCLT model has encountered an
end-of-file for the STAR meteorological data trying the read the indicated
averaging period. Check the data file for the correct number of STAR summaries or
modify the CO AVERTIME and/or ME STARDATA cards.
580 End-of-File Reached Trying to Read a Data File. The ISCST model has encountered an
end-of-file trying the read the indicated file. This may appear when trying to
"re-start" a model run with the CO INITFILE card if there is an error with the
initialization file. Check the data file for the correct filename.
E'
-.
-------
APPENDIX F. DESCRIPTION OF FILE FORMATS
F.I ASCII METEOROLOGICAL DATA
The ISCST and ISCEV models are designed to accept a wide range of ASCII
meteorological data file formats. The use of ASCII files for meteorological data has
two distinct advantages over the use of unformatted data files, such as are generated by
the PCRAMMET and MPRM preprocessors (see the next section). The first advantage is the
portability of the data files to different compilers and computer systems used for
running the models. The second advantage is that the data file can be examined easily
to determine its contents, and listed to the computer screen or to a printer for later
reference. The user may specify the use of the default ASCII format by leaving the
formet field blank on the ME INPUTFIL card. The user may also specify FREE-formatted
reads for the meteorological data, may specify the Fortran read format explicitly, or
may select the CARD option, which allows for the input of hourly wind profile exponents
and vertical potential temperature gradients.
The first record of the meteorological data input file contains the station number
and year for both the surface station and the upper air (mixing height) station. For
the formatted ASCII files, these four integer variables are read using a free-format
READ, i.e., the variables must be separated by either a comma or by one or more blank
spaces. The order of these variables is as follows:
Surface Station Number, e.g., WBAN Number for NWS data
Year for Surface Data (2 or 4 digits)
Upper Air Station Number (for Mixing Height Data)
Year for Upper Air Data (2 or 4 digits)
F-l
-------
The model checks these variables against the values input by the user on the ME SURFDATA
and ME UAIRDATA cards (see Section 3.5.3).
The rest of the records in the file include the sequential meteorological data.
The order of the meteorological variables for the formatted ASCII files and the default
ASCII format are as follows:
Variable
Year (last 2 digits)
Month
Day
Hour
Flow Vector (deg.)
Wind Speed (m/s)
Ambient Temperature (K)
Stability Class
(A=l, B=2, . . . F=6)
Rural Mixing Height (m)
Urban Mixing Height (m)
Wind Profile Exponent
(CARD only)
Vertical Potential
Temperature Gradient (K/m)
(CARD only)
Friction velocity (m/s)
(Dry Deposition Only)
Monin-Obukhov Length (m)
(Dry Deposition Only)
Fortran Format
12
12
12
12
F9.4
F9.4
F6.1
12
F7.1
F7.1
F8.4
F8.4
F9.4
F10.1
Columns
1-2
3-4
5-6
7-8
9-17
18-26
27-32
33-34
35-41
42-48
49-56
57-65
49-57
(66-74
for CARD)
58-67
(75-84
for CARD)
F-2
-------
Surface Roughness Length (m)
(Dry Deposition Only)
Precipitation Code (00-45)
(Wet Deposition Only)
Precipitation Rate (mm/hr)
(Wet Deposition Only)
F8.4
14
F7.2
68-75
(85-92
for CARD)
76-79
(93-96
for CARD)
80-86
(97-103
for CARD)
Calm hours are identified in the ASCII meteorological data files by a wind speed of
0.0 m/s. For unformatted PCRAMMET files that are converted to the ASCII format by
BINTOASC (see Section C.2), the conversion program checks for calm hours based on the
PCRAMMET convention of a wind speed equal to 1.0 m/s and a flow vector equal to the flow
vector for the previous hour, and sets the wind speed to 0.0 in the ASCII file.
F.2 PCRAMMET METEOROLOGICAL DATA
The PCRAMMET preprocessor generates an unformatted file of meteorological data from
National Weather Service observations suitable for use by several dispersion models,
including the ISCST model. The file contains two types of records, the first is a
header record and the second is the meteorological data. The second contains the data
for one 24-hour period (midnight to midnight) and is repeated until all data are listed.
The data are written unformatted to the file. This type of file may also be generated
by the MPRM processor designed for processing on-site meteorological data.
The format of the header record is:
F-3
-------
READ(u)
ID1
5
5
5
5
5
5
5
5
5
5
94
,IYEAR1,ID2,IYEAR2
5 55
5 5 94 Last 2 digits
5 5 heightdata.
5 5
5 94 5-digit station i
5 heightdata.
5
94 Last 2 digits of beginni
surface data.
5-digit station identificati
surface data.
of beginni
denti f i cati
ng year of
ng year of mixing
on of mixing
hourly
on of hourly
The format of the meteorological records are:
READ(u)
IYEAR,
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
94
MONTH
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
94
Last
,IDAY,
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
94
Month
2 digi
PGSTAB, SPEED
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
94
Day
5
5
5
5
5
5
5
5
5
5
5
5
5
5
94
Array
,TEMP,FLWVEC
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
94
94 Array
Kelvi
Array of
of P a s q u i
wi
11
,RANFLW,MIXHGT
5 5
5 94 Array of mixing
5 heights (m)
5
94 Array of randomized
flow vectors (to
nearest degree)
Array of flow vectors (to
nearest 10 degrees)
of temperatures (degrees
n)
nd speeds (m/s)
stability categories
of month (1-31)
of year ( 1-
ts of
year
12)
F-4
-------
The DIMENSION statements used to define the arrays are:
DIMENSION IKSK24), AWSC24), ATA(24), AFV(24), AFVRC24), AZI(2,24)
The first index in the AZI (mixing height) array controls which of the two mixing
height values is referenced. AZI(1,1) refers to the rural mixing height values, where i
equals from 1 to 24 and refers to hour of day in local standard time. AZI(2,1) refers to
the urban mixing height values.
The following preset values are used to indicate missing data:
IKST 0
AWS -9
ATA -99
AFV - 9 9
AFVR - 9 9
AZI -999
F.3 STAR SUMMARY JOINT FREQUENCY DISTRIBUTIONS
For the ISC Long Term dispersion model, the input file describing the
meteorological conditions is a joint frequency distribution. These frequency
distributions are called STAR summaries for STability ARray. The frequency distribution
is constructed using 16 wind direction sectors, with the first 22.5 sector centered on
winds from the North (increasing clockwise), six wind speed classes and six stability
classes. The wind speed classes are 0-3, 3-6, 6-10, 10-16, 16-21 and >21 kts. The
F-5
-------
Pasquill stability categories for the ISCLT dispersion model are grouped into classes
as,
Pasquill
Class category Remarks
1 A Very unstable conditions
2 B Moderately unstable conditions
3 C Slightly unstable conditions
4 D Neutral conditions
5 E Slightly stable conditions
6 F Very stable conditions
A separate STAR summary may be used for each averaging period, such as a month or a
season, or for the entire annual data period.
The format of the meteorological file is:
LOOP ON 1=1,6
LOOP ON K=l,16
READ(u.f) FREQC (I,J,K),J=l,6 )
5 555
5 5 5 94 Index associated with wind speed class
5 55
5 5 94 Index associated with wind direction sector
5 5
5 94 Index associated with stability class
5
94 Frequency of occurrence (decimal), of stability class I, with
wind speed class J, for wind from wind sector K
F-6
-------
FORMATC6F10.0)
Hence the meteorological file consists of 96 records for each STAR summary, the first
16 are for stability class 1, the next 16 are for stability class 2, and so forth.
F.4 THRESHOLD VIOLATION FILES (MAXIFILE OPTION)
The OU MAXIFILE card for the ISCST model allows the user the option to generate a
file or files of threshold violations for specific source group and averaging period
combintations. The file consists of several header records, each identified with an
asterisk (*) in column one. The header information includes the model name and version
number, the first line of the title information for the run, the list of modeling option
keywords applicable to the results, the averaging period and source group included in
the file, and the threshold value. Any value equal to or exceeding the threshold value
will be included in the file. The header also includes the format used for writing the
data records, and column headers for the variables included in the file. The variables
provided on each data record include the averaging period, the source group ID, the date
(YYMMDDHH) for the end of averaging period, the X and Y coordinates of the receptor
location, the receptor terrain elevation and flagpole receptor height, and the
F-7
-------
concentration or deposition value that violated the threshold.
from a threshold file identifies the contents of the MAXIFILE:
The following example
* ISCST3 (95250): A Simple Example Problem for the ISCST Model
* MODELING OPTIONS USED:
* CONC RURAL FLAT DFAULT
* MAXI-FILE FOR 3-HR VALUES >= A THRESHOLD OF 30.00
* FOR SOURCE GROUP: ALL
* FORMAT: (IX 13
*AVE
*
3
3
3
3
3
3
3
3
3
3
3
GRP
ALL
ALL
ALL
ALL
ALL
ALL
ALL
ALL
ALL
ALL
ALL
DATE
64010206
64010218
64010424
64010506
64010506
64010512
64010515
64010518
64010521
64010524
64010524
IX, A8
76
76
76
76
153
86
86
76
128
0
-0
,1X,I8,2(1X.F13.5) .2QX.F7.2)
X
.60445
.60445
.60445
.60445
.20889
.60254
.60254
.60445
.55753
.00000
.00001
64
64
64
64
128
50
50
64
153
100
200
Y
.27876
.27876
.27876
.27876
.55753
.00000
.00000
.27876
.20889
.00000
.00000
ELEV
0
0
0
0
0
0
0
0
0
0
0
00
00
00
00
00
00
00
00
00
00
00
1X.F13.
FLAG
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
5)
AVERAGE
30.
42.
34.
38.
33.
36.
33.
44.
34.
58.
38.
CONC
24433
91793
63943
86485
00018
78835
48914
44987
85760
49796
87197
F.5 POSTPROCESSOR FILES (POSTFILE OPTION)
The OU POSTFILE card for the ISCST model allows the user the option of creating
output files of concurrent concentration or deposition values suitable for
postprocessing. The model offers two options for the type of file generated - one is an
unformatted file similar to the concentration file generated by the previous version of
ISCST, and the other is a formatted file of X, Y, CONC (or DEPO) values suitable for
inputting to plotting programs.
F-!
-------
The unformatted POSTFILE option generates a separate unformatted data record of
concurrent values for each averaging period and source group specified. The averaging
period and source group combinations may be written to separate files, or combined into
a single file. Each record begins with the date variable for the end of the averaging
period (an integer variable of the form YYMMDDHH), the averaging period (e.g., an
interger value of 3 for 3-hour averages), and the source group ID (eight characters).
Following these three header variables, the record includes the concentration or
deposition values for each receptor location, in the order in which the receptors are
defined on the RE pathway. If more than one output type (CONG, DEPOS, DDEP, and/or
WDEP) is calculated, then all of the output values for a particular averaging period and
source group are included on a single record, in the order listed here. The results are
output to the unformatted file or files as they are calculated by the model.
The formatted plot file option for the POSTFILE keyword includes several lines of
header information, each identified with an asterisk (*) in column one. The header
information includes the model name and version number, the first line of the title
information for the run, the list of modeling option keywords applicable to the results,
the averaging period and source group included in the file, and the number of receptors
included. The header also includes the format used for writing the data records, and
column headers for the variables included in the file. The variables provided on each
data record include the X and Y coordinates of the receptor location, the concentration
or deposition value for that location, the receptor terrain elevation, the averaging
period, the source group ID, and either the date variable for the end of the averaging
period (in the form of YYMMDDHH) for short term averages or the number of hours in the
period for PERIOD averages. The last column provides the eight-character receptor
network ID for receptors that are defined as part of a gridded network. For discrete
F-9
-------
receptors, the network ID field includes the character string ' NA '. When more
than one output type is selected among the list of CONG, DEPOS, DDEP, and/or WDEP, the
PLOT formatted post-processing output file will include all of the output types
selected, in the order listed here. The results for each output type will be printed in
separate columns, one record per receptor. The following example from a formatted
postprocessor file for PERIOD averages identifies the contents of the POSTFILE:
* ISCST3 (95250): A Simple Example Problem for the ISCST Model
* MODELING OPTIONS USED:
* CONC RURAL FLAT DFAULT
* POST/PLOT FILE OF PERIOD VALUES FOR SOURCE GROUP: ALL
* FOR A TOTAL OF 180 RECEPTORS.
* FORMAT: (3 ( IX . F13 . 5) . IX . F8 . 2 . 2X . A6 . 2X . A8 . 2X . 18 . 2X . A8)
*
*
17
34
52
86
173
34
68
102
171
X
36482
72964
09445
82409
64818
20201
40403
60604
01007
98
196
295
492
984
93
187
281
469
Y AVERAGE CONC
48077
96155
44232
40387
80774
96926
93852
90778
84631
0
0
0
0
0
0
0
0
0
09078
04353
02323
00646
00389
00053
22839
14398
06481
ZELEV
0
0
0
0
0
0
0
0
0
00
00
00
00
00
00
00
00
00
AVE
PERIOD
PERIOD
PERIOD
PERIOD
PERIOD
PERIOD
PERIOD
PERIOD
PERIOD
GRP
ALL
ALL
ALL
ALL
ALL
ALL
ALL
ALL
ALL
NUM MRS
240
240
240
240
240
240
240
240
240
NET ID
POL1
POL1
POL1
POL1
POL1
POL1
POL1
POL1
POL1
F.6 HIGH VALUE RESULTS FOR PLOTTING (PLOTFILE OPTION)
The OU PLOTFILE card for the ISCST model allows the user the option of creating
output files of highest concentration or deposition values suitable for importing into
graphics software to generate contour plots. The formatted plot files generated by the
PLOTFILE include several lines of header information, each identified with an asterisk
F-10
-------
(*) in column one. The header information includes the model name and version number,
the first line of the title information for the run, the list of modeling option
keywords applicable to the results, the averaging period and source group included in
the file, the high value (e.g. 2ND highest) included for plotting, and the number of
receptors included. The header also includes the format used for writing the data
records, and column headers for the variables included in the file. The variables
provided on each data record include the X and Y coordinates of the receptor location,
the concentration or deposition value for that location, the receptor terrain elevation,
the averaging period, the source group ID, and either the high value included for short
term averages or the number of hours in the period for PERIOD averages. The last column
provides the eight-character receptor network ID for receptors that are defined as part
of a gridded network. For discrete receptors, the network ID field includes the
character string ' NA '. When more than one output type is selected among the list
of CONG, DEPOS, DDEP, and/or WDEP, the PLOTFILE output file will include all of the
output types selected, in the order listed here. The results for each output type will
be printed in separate columns, one record per receptor. The following example from a
F-ll
-------
formatted postprocessor file for high second highest 24-hour averages identifies the
contents of the PLOTFILE:
* ISCST3 (95250): A Simple Example Problem for the ISCST Model
* MODELING OPTIONS USED:
* CONC RURAL FLAT DFAULT
* PLOT FILE OF HIGH 2ND HIGH 24-HR VALUES FOR SOURCE GROUP: ALL
* FOR A TOTAL OF 180 RECEPTORS.
* FORMAT: (3 ( IX . F13 . 5) . IX . F8 . 2 .3X . A5 . 2X . A8 . 2X . A4 . 6X . A8)
*
*
17
34
52
86
173
34
68
102
171
X
36482
72964
09445
82409
64818
20201
40403
60604
01007
98
196
295
492
984
93
187
281
469
Y AVERAGE CONC
48077
96155
44232
40387
80774
96926
93852
90778
84631
0
0
0
0
0
0
0
0
0
00038
00759
00223
00058
00012
00032
73597
46271
22714
ZELEV
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
AVE
24
24
24
24
24
24
24
24
24
-HR
-HR
-HR
-HR
-HR
-HR
-HR
-HR
-HR
GRP
ALL
ALL
ALL
ALL
ALL
ALL
ALL
ALL
ALL
HIVAL NET ID
2ND
2ND
2ND
2ND
2ND
2ND
2ND
2ND
2ND
POL1
POL1
POL1
POL1
POL1
POL1
POL1
POL1
POL1
The PERIOD and ANNUAL average PLOTFILE uses the same format for the data records as
the PERIOD and ANNUAL average formatted POSTFILE shown in the previous section.
F.7 TOXX MODEL INPUT FILES (TOXXFILE OPTION)
The OU TOXXFILE card for the ISCST model allows the user the option to generate an
unformatted file or files of threshold violations for a specific averaging period for
use with the TOXX model component of TOXST. The file consists of three header records,
including the first line of the title information for the run, the number of source
groups, receptors and averaging periods, information on the type of receptor network,
F-12
-------
and the threshold cutoff value. Following the header records are pairs of records
identifying the specific averaging period, source group and receptor location and
corresponding concentration value for the values exceeding the user-specified threshold.
If any source group exceeds the threshold for a given averaging period and receptor
location, then the concentrations for all source groups are output for that period and
receptor. The structure of the unformatted file for the ISCST model TOXXFILE option is
described below:
Record
# Description
1 Title (80 characters)
2 IYEAR, NUMGRP, NUMREC, NUMPER, ITAB, NXTOX, NYTOX, IDUM1, IDUM2, IDUM3
3 CUTOFF, RDUM1, ..., RDUM9
where: TITLE = First line of title (80 characters)
IYEAR = Year of simulation
NUMGRP = No. of source groups
NUMREC = Total no. of receptors
NUMPER = No. of averaging periods (e.g., number of hours in the year)
ITAB = 1 for polar grid; 2 for Cartesian grid; 0 for discrete receptors or
mixed grids
NXTOX = No. of x-cooordinates (or distances) in receptor network
NYTOX = No. of y-coordinates (or directions) in receptor network
IDUM1, IDUM2, IDUM3 = dummy integer variables, arbitrarily set equal to zero
CUTOFF = User-specified threshold for outputting results (g/m3)
RDUM1, ..., RDUM9 = Dummy real variables (nine) arbitrarily set equal to zero
Following the header records, the file consists of pairs of records including an ID
variable identifying the data period, source group number and receptor number, and the
corresponding concentration values. The number of values included in each record is
F-13
-------
controlled by the NPAIR PARAMETER, which is initially set at 100 in the MAIN1.INC file.
The identification variable is determined as follows:
IDCONC = IPER*100000 + IGRP*1000 + IREC
where: IPER = the hour number for the year corresponding to the concentration
value
IGRP = the source group number (the order in which the group was defined on
the SO pathway)
IREC = the receptor number (the order in which the receptor was defined on
Llie RE
* ISCLT3 (95250): TEST RUN FOR NEW ISCLT MODEL - BASED ON SCRAM BBS TEST CASE
* MODELING OPTIONS USED:
works somewhat differently from the ISCST
notiel optiQSnA deTs\er3fbed3%MSveTPRS.The format of the TOXXFILE output file for ISCLT is the
* ITAB = 1; NXTOX = 9; NYTOX = 4
same formaOtRM/ws wsedFifioao ,tiiei4PEJ,QTFFlLf;2»pi&fi2®p\8in ISCLT, except for some slight differences
in* some of the header records, and cne fact that the TOXXFILE output file includes the
d *SIfrcA group. The following is an example of
tesultl:
an
Seach
in t-
800.00000
2000.00000
4000.00000
8000.00000
16000.00000
20000.00000
.00000
.00001
.00001
.00002
.00005
.00010
.00019
.00039
.00049
-.00003
-.00009
-.00017
-.00035
-.00070
-.00087
-125.00000
-250.00000
-400.00000
-800.00000
-2000.00000
-4000.00000
-8000.00000
-16000.00000
-20000.00000
-2^eCi
1.680273
11.524000
8.915471
5.361694
3.010265
1.210022
.918835
.000001
.004480
1.500647
10.346320
9.384181
6.173569
3.782269
1.583979
1.202485
$
7.62
10.36
10.67
10.97
15.24
30.48
30.48
1.52
3.05
7.62
10.36
10.67
10.97
15.24
30.48
30.48
WINTER
WINTER
WINTER
WINTER
WINTER
WINTER
WINTER
WINTER
WINTER
WINTER
WINTER
WINTER
WINTER
WINTER
WINTER
WINTER
F-14
-------
The ITAB, NXTOX, and NYTOX variables included in the header records for the ISCLT
TOXXFILE output are the same as defined above for the ISCST model option.
F-15
-------
APPENDIX G. QUICK REFERENCE FOR ISCST AND ISCLT MODELS
CO Keywords
TITLEONE
TITLETWO
MODELOPT
AVERTIME
POLLUTID
HALFLIFE
DCAYCOEF
TERRHGTS
ELEVUNIT
FLAGPOLE
RUNORNOT
EVENTFIL
SAVEFILE
INITFILE
Type
M-N
0-N
M-N
M-N
M-N
0-N
0-N
0-N
0-N
0-N
M-N
0-N
0-N
0-N
Parameters
Titlel
TitleZ
DFAULT CONC DRYDPLT WETDPLT RURAL GRDRIS NOSTD NOBID NOCALM MSGPRO NOSMPL (ST)
DEPOS or or
DDEP URBAN NOCMPL
and/or
WDEP
DFAULT CONC DRYDPLT RURAL GRDRIS NOSTD NOBID (LT)
DEPOS or
or URBAN
DDEP
1 2 3 4 6 8 12 24 MONTH PERIOD (ST Model)
or
ANNUAL
JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC ( LT Model)
WINTER SPRING SUMMER FALL or QUART1 QUART2 QUARTS QUART4
MONTH SEASON QUARTR ANNUAL PERIOD
Pollut
Haflif
Decay
FLAT or ELEV
METERS or FEET
(Flagdf)
RUN or NOT
(Evfile) (Evopt) (ST model only)
(Savfil) (Dayinc) (Savfl2) (ST model only)
(Inifil) (ST model only)
Sec.
3.2.1
3.2.1
3.2.2
3.2.3
3.2.4
3.2.5
3.2.5
3.2.6
3.2.6
3.2.7
3.2.8
3.2.9
3.2.10
3.2.10
G-l
-------
IMULTYEAR
ERRORFIL
0-N
0-N
Savfil
(Errfil
(Inifil)
) (DEBUG)
(ST model
only)
13.2.11 1
3.2.12
Type: M - Mandatory
0 - Optional
N - Non-repeatable
R - Repeatable
G-2
-------
SO Keywords
ELEVUNIT
LOCATION
SRCPARAM
BUILDHGT
BUILDWID
LOWBOUND
EMISFACT
EMISUNIT
CONCUNIT
DEPOUNIT
PARTDIAM
MASSFRAX
PARTDENS
PARTSLIQ
PARTSICE
GAS-SCAV
HOUREMIS
SRCGROUP
Type
0-N
M-R
M-R
0-R
0-R
0-R
0-R
0-N
0-N
0-N
0-R
0-R
0-R
0-R
0-R
0-R
0-R
M-R
Parameters
METERS or FEET
Srcid Srct.yp Xs Ys (Zs) (Srct.yp = POINT, VOLUME, AREA, or OPENPIT)
Srcid Ptemis Stkhgt Stktmp Stkvel Stkdia (POINT Source)
Vlemis Rel hgt Syinit Szinit (VOLUME Source)
Aremis Relhgt Xinit (Yinit) (Angle) (Szinit) (AREA Source)
Opemis Relhgt Xinit Yinit Pitvol (Angle) (OPENPIT Source)
Srcid (or Srcrng) Dsbh( i ) , i=l ,Nsec
Srcid (or Srcrng) Dsbw( i ) , i=l ,Nsec
Srcid (or Srcrng) Idswak( i ) , i=l ,Nsec
Srcid (or Srcrng) Qfl ag Qfact (i ) , i=l , Nqf
Emifac Emilbl Conlbl (or Deplbl)
Emifac Emilbl Conlbl
Emifac Emilbl Deplbl
Srcid (or Srcrng) Pdi am(i ) , i = l , Npd
Srcid (or Srcrng) Phi(i),i=l,Npd
Srcid (or Srcrng) Pdens (i ) , i = l , Npd
Srcid (or Srcrng) Scavcoef ( i ) , i=l ,Npd (ST model only)
Srcid (or Srcrng) Scavcoef ( i ), i=l , Npd (ST model only)
Srcid (or Srcrng) LIQ or ICE Scavcoef (ST model only)
Emifil Srcid's Srcrng's
Grpid Srcid's Srcrng's
Section
3.3
3.3.1
3.3.2
3.3.3
3.3.3
3.3.3
3.3.4
3.3.5
3.3.5
3.3.5
3.3.6
3.3.6
3.3.6
3.3.7
3.3.7
3.3.7
3.3.8
3.3.9
RE Keywords
ELEVUNIT
Type
0-N
Parameters
METERS or FEET
ISecti on
3.4
G-3
-------
GRIDCART
GRIDPOLR
DISCCART
DISCPOLR
BOUNDARY
BOUNDELV
0-R
0-R
0-R
0-R
0-R
0-R
Netid STA
XYINC Xinit Xnum Xdelta Yinit Ynum Ydelta
or XPNTS Gridxl GridxZ GridxS ... GridxN, and
YPNTS Gridyl GridyZ GridyS ... GridyN
ELEV Row Zelevl ZelevZ ZelevS ... ZelevN
FLAG Row Zflagl ZflagZ ZflagS ... ZflagN
END
Netid STA
PRIG Xinit Yinit,
or PRIG Srcid
DIST Ringl RingZ RingS ... RingN
DDIR Dirl Dir2 Dir3 ... DirN
or GDIR Dim urn Dirini Dirinc
ELEV Rad Zelevl ZelevZ ZelevS ... ZelevN
FLAG Rad Zflagl ZflagZ ZflagS ... ZflagN
END
Xcoord Ycoord (Zelev) (Zflag)
Srcid Range Direct (Zelev) (Zflag)
Srcid Dist(I) ,1=1,36
Srcid Zelev(I),I=l,36
3.4.1
3.4.1
3.4.3
3.4.3
3.4.4
3.4.4
Note: While all RE keywords are optional, at least one receptor must be defined for each run.
G-4
-------
ME Keywords
INPUTFIL
ANEMHGHT
SURFDATA
UAIRDATA
STARTEND
DAYRANGE
STARDATA
WDROTATE
WINDPROF
DTHETADZ
WINDCATS
AVESPEED
AVETEMPS
AVEMIXHT
AVEROUGH
Type
M-N
M-N
M-N
M-N
0-N
0-R
0-N
0-N
0-R
0-R
0-N
0-N
M-R
M-R
0-R
Parameters
Metfil (Format)
Zref (Zrunit)
Stanum Year (Name) (Xcoord Ycoord)
Stanum Year (Name) (Xcoord Ycoord)
Strtyr Strtmn Strtdy (Strthr) Endyr Endmn Enddy (Endhr) (ST only)
Rangel RangeZ RangeS ... RangeN (ST model only)
JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC (LT model only)
WINTER SPRING SUMMER FALL or QUART1 QUARTZ QUARTS QUART4
MONTH SEASON QUARTR ANNUAL PERIOD
Rotang
Stab Profl ProfZ ProfS Prof4 ProfB Prof6
Stab Dtdzl DtdzZ DtdzS Dtdz4 DtdzB Dtdz6
Wsl Ws2 Ws3 Ws4 Ws5
Wsl Ws2 Ws3 Ws4 Ws5 Ws6 ( LT model only)
Aveper Tal Ta2 Ta3 Ta4 Ta5 Ta6 ( LT model only)
Aveper Stab Mixhtl MixhtZ MixhtS Mixht4 MixhtB Mixht6 ( LT model only)
Aveper ZO ( LT model only)
Secti on
3.5.1
3.5.2
3.5.3
3.5.3
3.5.5
3.5.5
3.5.4
3.5.6
3.5.8
3.5.9
3.5.7
3.5.10
3.5.11
3.5.12
3.5.13
TG Keywords
INPUTFIL
LOCATION
ELEVUNIT
Type
M-N
M-N
0-N
Parameters
Tgfile
Xorig Yorig (Units)
METERS or FEET
Secti on
3.6
3.6
3.6
G-5
-------
OU Keywords
RECTABLE
MAXTABLE
DAYTABLE
MAXIFILE
PLOTFILE
POSTFILE
TOXXFILE
Type
0-R
0-R
0-N
0-R
0-R
0-R
0-R
Parameters
Aveper FIRST SECOND ... SIXTH or 1ST 2ND ... 6TH (ST Model)
INDSRC and/or SRCGRP (LT Model)
Aveper Maxnum (ST Model )
Maxnum INDSRC and/or SRCGRP and/or SOCONT ( LT Model)
Avperl AvperZ AvperS Avper4 (ST model only)
Aveper Grpid Thresh Filnam (Funit) (ST model only)
Aveper Grpid Hivalu Filnam (Funit) (ST model)
Aveper Grpid Filnam (Funit) ( LT model & ST period ave)
Aveper Grpid Format Filnam (Funit) (ST model only)
Aveper Cutoff Filnam (Funit) (ST model)
Aveper Grpid Filnam (Funit) ( LT model)
Section
3.8.1
3.8.3
3.8.1
3.8.3
3.8.1
3.8.1
3.8.1
3.8.3
3.8.1
3.8.1
3.8.3
G-6
-------
APPENDIX H. QUICK REFERENCE FOR ISCEV (EVENT) MODEL
(USED FOR SHORT TERM EVENT/SOURCE CONTRIBUTION ANALYSES)
CO Keywords
TITLEONE
TITLETWO
MODELOPT
AVERTIME
POLLUTID
HALFLIFE
DCAYCOEF
TERRHGTS
FLAGPOLE
RUNORNOT
ERRORFIL
Type
M-N
0-N
M-N
M-N
M-N
0-N
0-N
0-N
0-N
M-N
0-N
Parameters
Titlel
TitleZ
DFAULT CONC DRYDPLT WETDPLT RURAL GRDRIS NOSTD NOBID NOCALM MSGPRO NOSMPL
DEPOS or or
DDEP URBAN NOCMPL
and/or
WDEP
1 2 3 4 6 8 12 24 MONTH PERIOD
or
ANNUAL
Poll ut
Haflif
Decay
FLAT or ELEV
(Flagdf)
RUN or NOT
(Errfil) (DEBUG)
Sec.
3.2.1
3.2.1
3.2.2
3.2.3
3.2.4
3.2.5
3.2.5
3.2.6
3.2.7
3.2.8
3.2.12
Note: MONTH. PERIOD. and ANNUAL averages are ignored by the EVENT model, which can only handle short term averages of up to 24
hours. Also, only the first output type, in the order of CONC, DEPOS, DDEP and WDEP, is used.
SO Keywords
ELEVUNIT
LOCATION
SRCPARAM
Type
0-N
M-R
M-R
Parameters
METERS or FEET
Srcid Srct.yp Xs Ys (Zs) (Srct.yp = POINT, VOLUME, AREA, or OPENPIT)
Srcid Ptemis Stkhgt Stktmp Stkvel Stkdia (POINT Source)
Vlemis Rel hgt Syinit Szinit (VOLUME Source)
Aremis Relhgt Xinit (Yinit) (Angle) (Szinit) (AREA Source)
Opemis Relhgt Xinit Yinit Pitvol (Angle) (OPENPIT Source)
Section
3.3
3.3.1
3.3.2
H-l
-------
BUILDHGT
BUILDWID
LOWBOUND
EMISFACT
EMISUNIT
CONCUNIT
DEPOUNIT
PARTDIAM
MASSFRAX
PARTDENS
PARTSLIQ
PARTSICE
GAS-SCAV
HOUREMIS
SRCGROUP
0-R
0-R
0-R
0-R
0-N
0-N
0-N
0-R
0-R
0-R
0-R
0-R
0-R
0-R
M-R
Srcid (or Srcrng) Dsbh( i ) , i=l ,Nsec
Srcid (or Srcrng) Dsbw( i ) , i=l ,Nsec
Srcid (or Srcrng) Idswak( i ) , i=l ,Nsec
Srcid (or Srcrng) Qfl ag Qfact (i ) , i=l , Nqf
Emifac Emilbl Conlbl (or Deplbl)
Emifac Emilbl Conlbl
Emifac Emilbl Deplbl
Srcid (or Srcrng) Pdi am(i ) , i=l , Npd
Srcid (or Srcrng) Phi(i),i=l,Npd
Srcid (or Srcrng) Pdens (i ) , i=l , Npd
Srcid (or Srcrng) Scavcoef ( i ) , i=l ,Npd (ST model only)
Srcid (or Srcrng) Scavcoef ( i ), i=l , Npd (ST model only)
Srcid (or Srcrng) LIQ or ICE Scavcoef (ST model only)
Emifil Srcid's Srcrng's
Grpid Srcid's Srcrng's
3.3.3
3.3.3
3.3.3
3.3.4
3.3.5
3.3.5
3.3.5
3.3.6
3.3.6
3.3.6
3.3.7
3.3.7
3.3.7
3.3.8
3.3.9
Type: M - Mandatory
0 - Optional
N - Non-repeatable
R - Repeatable
ME Keywords
INPUTFIL
ANEMHGHT
SURFDATA
UAIRDATA
WDROTATE
WINDCATS
WINDPROF
DTHETADZ
Type
M-N
M-N
M-N
M-N
0-N
0-N
0-R
0-R
Parameters
Metfil (Format)
Zref (Zrunit)
Stanum Year (Name) (Xcoord Ycoord)
Stanum Year (Name) (Xcoord Ycoord)
Rotang
Wsl Ws2 Ws3 Ws4 Ws5
Stab Profl ProfZ ProfS Prof4 Prof5 Prof6
Stab Dtdzl DtdzZ DtdzS Dtdz4 Dtdz5 Dtdz6
Secti on
3.5.1
3.5.2
3.5.3
3.5.3
3.5.6
3.5.7
3.5.8
3.5.9
H-2
-------
TG Keywords
INPUTFIL
LOCATION
ELEVUNIT
Type
M-N
M-N
0-N
Parameters
Tgfile
Xorig Yorig (Units)
METERS or FEET
Secti on
3.6
3.6
3.6
EV Keywords
EVENTPER
EVENTLOC
Type
M-R
M-R
Parameters
Evname Aveper Grpid Date
Evname XR= Xr YR= Yr (Zelev) (Zflag)
or
RNG= Rng DIR= Dir (Zelev) (Zflag)
Secti on
3.7.1
3.7.2
OU Keywords
EVENTOUT
Type
M-N
Parameters
SOCONT or DETAIL
1 Secti on
3.8.2
Note: RE Pathway is not used for the ISCEV (EVENT) model. Receptor locations for specific events are identified on the EVent
Pathway in combination with particular data periods.
H-3
-------
GLOSSARY
ASCII -- American Standard Code for Information Interchange, a standard set of codes used by
computers and communication devices. Sometimes used to refer to files containing only
such standard codes, without any application-specific codes such as might be present in
a document file from a word processor program.
CD-144 Format -- Card Deck-144 data format available from NCDC for National Weather Service
surface observations commonly used for dispersion models. Each record represents an
80-column "card image".
CO -- COntrol, the 2-character pathway ID for input runstream images used to specify overall
job control options.
CO Pathway -- Collective term for the group of input runstream images used to specify the
overall job control options, including titles, dispersion options, terrain options,
etc.
Directory -- A logical subdivision of a disk used to organize files stored on a disk.
Dispersion Model -- A group of related mathematical algorithms used to estimate (model) the
dispersion of pollutants in the atmosphere due to transport by the mean (average) wind
and small scale turbulence.
DOS -- Disk Operating System. Software that manages applications software and provides an
interface between applications and the system hardware components, such as the disk
drive, terminal, and keyboard.
EBCDIC -- Extended Binary Coded Decimal Interchange Code, the collating sequence used on IBM
mainframe computers.
Echo of inputs -- By default, the ISC models will echo the input runstream images, character
by character, into the main printed output file. This serves as a record of the inputs
as originally entered by the user, without any rounding of the numerical values. The
echoing can be suppressed with the NO ECHO option.
EOF -- End-of-File.
GLOSSARY-1
-------
EPA -- U. S. Environmental Protection Agency.
Error message -- A message written by the model to the error/message file whenever an error
is encountered that will inhibit data processing.
Error/Message File -- A file used for storage of messages written by the model.
EV -- EVent, the 2-character pathway ID for input runstream images used to specify event
inputs for the Short Term EVENT model.
EV Pathway -- Collective term for the group of input runstream images used to specify the
event periods and location for the Short Term EVENT model.
EVENT Model -- A new ISC Short Term model (ISCEV) developed with Version 2 of ISCST,
specifically designed to provide source contribution (culpability) information for
specific events of interest, e.g., design values or threshold violations.
Extended Memory -- Additional memory on 80386 and 80486 PCs that allows programs to address
memory beyond the 640 KB limit of DOS. Special software is required to utilize this
extra memory.
Fatal Error -- Any error which inhibits further processing of data by the model. Model
continues to read input images to check for errors during setup, and will continue to
read input meteorological data during calculation phase.
Flow Vector -- The direction towards which the wind is blowing.
GMT -- Greenwich Mean Time, the time at the DO meridian.
Informational Message -- Any message written to the error/message file that may be of
interest to the user, but which have no direct bearing on the validity of the results,
and do not affect processing.
Input Image -- User supplied input, read through the default input device, controlling the
model options and data input. A single card or record from the input runstream file.
Each input image consists of a pathway ID (may be blank indicating a continuation of
the previous pathway), a keyword (may also be blank for continuation of a keyword), and
possibly one or more parameter fields.
GLOSSARY-2
-------
Input Runstream File -- The basic input file to the ISC models controlling the modeling
options, source data, receptor locations, meteorological data file specifications, and
output options. Consists of a series of input images grouped into functional pathways.
ISCEV -- Industrial Source Complex - Short Term EVENT Dispersion Model.
ISCST -- Industrial Source Complex - Short Term Dispersion Model.
ISCLT -- Industrial Source Complex - Long Term Dispersion Model.
JCL -- Job Control Language, an IBM mainframe's operating system control language for batch
jobs.
Joint Frequency Distribution -- The joint frequency of wind direction sector, wind speed
class and stability category (see also STAR).
Julian Day -- The number of the day in the year, i.e., Julian Day = 1 for January 1 and 365
(or 366 for leap years) for December 31.
KB -- Kilobyte, 1000 bytes, a unit of storage on a disk
Keyword -- The 8-character codes that follow immediately after the pathway ID in the input
run stream data.
LST -- Local Standard Time.
Math Co-processor -- A computer chip used to speed up floating point arithmetic in a
personal computer.
MB -- Megabyte, one million bytes, a unit of storage on a disk
ME -- MEteorology, the 2-character pathway ID for input runstream images used to specify
meteorological data options
ME Pathway -- Collective term for the group of input runstream images used to specify the
input meteorological data file and other meteorological variables, including the period
to process from the meteorological file for the ISCST model.
GLOSSARY-3
-------
Meteorological Data File -- Any file containing meteorological data, whether it be mixing
heights, surface observations or on-site data.
Missing Value -- Alphanumeric character(s) that represent breaks in the temporal or spatial
record of an atmospheric variable.
Mixing Height -- The depth through which atmospheric pollutants are typically mixed by
dispersive processes.
MPRM -- Meteorological Processor for Regulatory Models, a program designed for the purpose
of processing on-site meteorological data to prepare them for input to the regulatory
models, such as ISC. Produces a file comparable to the PCRAMMET pre-processor output,
and also capable of producing STAR summaries.
NCDC -- National Climatic Data Center, the federal agency responsible for distribution of
the National Weather Service upper air, mixing height and surface observation data.
NO ECHO -- Option to suppress echoing of the runstream input images to the main printed
output file.
NWS -- National Weather Service.
On-site Data -- Data collected from a meteorological measurement program operated in the
vicinity of the site to be modeled in the dispersion analysis.
OU -- Output, the 2-character pathway ID for input runstream images used to specify output
options.
OU Pathway -- Collective term for the group of input runstream images used to specify the
output options for a particular run.
Overlay -- One or more subprograms that reside on disk and are loaded into memory only when
needed.
Pasquill Stability Categories -- A classification of the dispersive capacity of the
atmosphere, originally defined using surface wind speed, solar insolation (daytime) and
cloudiness (nighttime). They have since been reinterpreted using various other
meteorological variables.
GLOSSARY-4
-------
Pathway -- One of the six major functional divisions in the input runstream file for the ISC
models. These are COntrol, SOurce, REceptor, MEteorology, EVent, and Output (see these
entries in this section for a description).
PC -- Personal Computer, a wide ranging class of computers designed for personal use,
typically small enough to fit on a desktop.
PCRAMMET -- Meteorological processor program used for regulatory applications capable of
processing twice-daily mixing heights (TD-9689 format) and hourly surface weather
observations (CD-144 format) for use in dispersion models such as ISCST, CRSTER, MPTER
and RAM.
Quality Assessment -- Judgment of the quality of the data.
Quality Assessment Check -- Determining if the reported value of a variable is reasonable
(see also Range Check).
Quality Assessment Message -- Message written to the error/message file when a data value
is determined to be suspect.
Quality Assessment Violation -- Occurrences when data values are determined to be suspect
(see also Range Check Violation).
RAM -- Random Access Memory on a personal computer.
Range Check -- Determining if a variable falls within predefined upper and lower bounds.
Range Check Violation -- Determination that the value of a variable is outside range defined
by upper and lower bound values (see also Quality Assessment Violation).
RE -- REceptor, the 2-character pathway ID for input runstream images used to specify
receptor locations.
RE Pathway -- Collective term for the group of input runstream images used to specify the
receptor locations for a particular run.
Regulatory Applications -- Dispersion modeling involving regulatory decision-making as
described in the Guideline on Air Quality Models (Revised), (EPA, 1987b).
GLOSSARY-5
-------
Regulatory Model -- A dispersion model that has been approved for use by the regulatory
offices of the EPA, specifically one that is included in Appendix A of the Guideline on
Air Quality Models (Revised), (EPA, 1987b), such as the ISC model.
Runstream File -- Collectively, all input images required to process input options and input
data for the ISC models.
SCRAM BBS -- Support Center for Regulatory Air Models - Bulletin Board System, an electronic
bulletin board system used by EPA for disseminating air quality dispersion models,
modeling guidance, and related information.
Secondary Keyword -- A descriptive alphabetical keyword used as a parameter for one of the
main runstream keywords to specify a particular option.
SO -- SOurce, the 2-character pathway ID for input runstream images used to specify input
source parameters and source groups.
SO Pathway -- Collective term for the group of input runstream images used to specify the
source input parameters and source group information.
STAR -- STability ARray, a joint frequency distribution summary of stability category, wind
speed and wind direction. The STAR data are used as input for the ISC Long Term
dispersion model.
Station Identification -- An integer or character string used to uniquely identify a station
or site as provided in the upper air (TD-5600 and TD-6201), mixing height (TD-9689),
and surface weather (CD-144 and TD-3280) data formats available from NCDC. There are
no standard station numbers for on-site data or card image/screening data, and the user
may include any integer string
Subdirectory -- A directory below the root, or highest level, directory or another
subdirectory, used for organization of files on a storage medium such as a PC hard
disk.
Surface Weather Observations -- A collection of atmospheric data on the state of the
atmosphere as observed from the earth's surface. In the U.S. the National Weather
Service collect these data on a regular basis at selected locations.
GLOSSARY-6
-------
Surface Roughness Length -- Height at which the wind speed extrapolated from a near-surface
wind speed profile becomes zero.
Syntax -- The order, structure and arrangement of the inputs that make of the input
runstream file, specifically, the rules governing the placement of the various input
elements including pathway IDs, keywords, and parameters.
TD-1440 Format -- A format available from NCDC for summarizing NWS surface observations in
an 80-column format; the CD-144 format is a subset of this format. This format has
been superseded by the TD-3280 format.
TD-3280 Format -- The current format available from NCDC for summarizing NWS surface weather
observations in an elemental structure, i.e., observations of a single atmospheric
variable are grouped together for a designated period of time.
TD-5600 Format -- A format available from NCDC for reporting NWS upper air sounding data.
This format has been superseded by the TD-6201 format.
TD-6201 Format -- The current format available from NCDC for reporting NWS upper air data.
The file structure is essentially the same as the TD-5600 format except that there is
more quality assurance information.
TD-9689 Format -- The format available from NCDC for mixing heights estimated from morning
upper air temperature and pressure data and hourly surface observations of temperature.
UNAMAP -- User's Network for Applied Modeling of Air Pollution, a collection of dispersion
models and closely related support utilities, used for disseminating models prior to
the SCRAM BBS.
Unformatted File -- A file written without the use of a FORTRAN FORMAT statement, sometimes
referred to as a binary file.
Upper Air Data (or soundings) -- Meteorological data obtained from balloon- borne
instrumentation that provides information on pressure, temperature, humidity, and wind
away from the surface of the earth.
Vertical Potential Temperature Gradient -- The change of potential temperature with height,
used in modeling the plume rise through a stable layer, and indicates the strength of
GLOSSARY-7
-------
the stable temperature inversion. A positive value means that potential temperature
increases with height above ground and indicates a stable atmosphere.
Warning Message -- A message written by the model to the error/message file whenever a
problem arises that may reflect an erroneous condition, but does not inhibit further
processing.
Wind Profile Exponent -- The value of the exponent used to specify the profile of wind speed
with height according to the power law (see Section 1.1.3 of Volume II).
GLOSSARY-8
-------
INDEX
Anemometer height specification 3-74
Area sources
emission rate parameter 3-28, 3-33
input parameters 3-27,B-8
irregularly-shaped areas 3-23
specification of location 3-23
specification of source type 3-23
ASCII meteorological data files 1-11, F-l
converting from binary C-3
default format for ISCST 3-67, F-2
Averaging periods
options for Long Term model 3-10
options for Short Term model 3-8
specifying options for 3-8
Binary meteorological data 2-22
Building downwash
BUILDHGT keyword 3-35, B-8
BUILDWID keyword 3-35, 3-38, B-8
example of building inputs 2-16
LOWBOUND keyword 3-35, 3-39, B-9
modeling options 1-8, 1-9, 2-1, 2-7, 3-6, 3-21
specification of building dimensions 3-25, 3-35
specifying "lower bound" option 3-39
Buoyancy-induced dispersion
and the regulatory default option 2-7,3-6
NOBID parameter 3-5
specifying not to use on MODELOPT card 2-8, 3-5, B-4
Calm and missing data flags 3-8
Calm flag in output file 2-36
Calms processing 3-6
specifying NOCALM option 3-5
Card image meteorological data
specification of CARD format for 3-66, 3-70
Cartesian grid receptors 3-53
INDEX-1
-------
specifying a receptor network 3-53
specifying discrete receptors 3-62
CO pathway 2-2
brief tutorial 2-12
example of inputs for 2-15
keyword reference 3-2,B-3
modeling options 2-7,2-12
order of keywords within 2-5
Command line for running ISCST 2-33, 3-126
Compiling options 4-3
Lahey D-4
Microsoft D-l
Complex terrain algorithms 1-16,3-6,3-14
Concentration
adjusting emission rate units for 3-44, B-9
specifying calculation of 2-13, 2-41, 3-4, B-4
Concentration file
converting options with STOLDNEW C-2
description of files generated by ISCST F-7
POSTFILE option for generating 3-103
Daily table option 3-100
Data period
specifying period to process for ISCST 3-78
Decay coefficient 2-5, 3-12
DCAYCOEF keyword 3-13, A-2, B-3, B-5
DECAY parameter 3-13
default for urban S02 3-12
relationship to half life 3-13
specifying 3-13
Depletion options 3-7
Deposition 2-41
specifying calculation of 2-41
Deposition algorithms
additional meteorology variables E-22
meteorology inputs 3-69, 3-87
Discrete receptors 3-61
with Cartesian coordinates 3-62
INDEX-2
-------
with polar coordinates 3-63
DOS
limits for DOS versions of models 2-9, 4-6
DOS redirection 2-33, 3-126
Dry deposition
adjusting emission rate units for 3-44, B-9
DEPOS keyword on MODELOPT card 3-4
MASSFRAX keyword 3-46
number of particle size categories 3-46
number of settling categories 3-48
PARTDENS keyword 3-46
PARTDIAM keyword 3-46
specifying calculation of 2-13, 2-41, 3-4, B-4
specifying emission rates for 3-25, 3-26, 3-28, 3-33
specifying input parameters for 3-46, B-10
Echoing of the runstream file
suppressing with NO ECHO 2-35
Elevated terrain
example of inputs for Cartesian grid 3-55
example of inputs for polar network 3-59
modeling options 1-10, 2-15, 2-42, 3-13
specifying boundary receptor elevations 3-64
specifying receptor elevations 3-53, 3-54, 3-58, 3-62, 3-63,
B-12, B-13, B-14
specifying units with ELEVUNIT 3-14
TERRHGTS keyword 2-42, 3-13
truncation above stack height 1-10
Elevation units
ELEVUNIT keyword 3-14
specifying for receptors 3-53, 3-54, 3-58
specifying for sources 3-23
specifying for terrain grids 3-92
Error handling capabilities 2-27
detailed message descriptions E-6
example message summary table 2-31
general description E-l
message summary table E-2
INDEX-3
-------
message types 2-27
syntax of messages E-3
Error message 2-27, E-3
example of syntax 2-28
Error/message file 3-119
EV pathway
keyword reference 3-92, B-21
EVENT model (ISCEV)
naming convention used for events 3-95
specifying event inputs 3-92
user defined events 3-95
using events defined by ISCST 3-94
Extended memory 3-119, GLOSSARY-2
limits for extended memory versions 2-9,4-6
Flagpole receptor heights
default receptor height, FLAGDF 3-15, B-5
example of inputs for Cartesian grid 3-55
example of inputs for polar network 3-59
FLAGDF parameter 3-15
FLAGPOLE keyword 3-15, B-3, B-5
modeling options 1-10, 2-16, 3-15
specifying boundary flagpole receptors 3-65
specifying flagpole receptors 3-53, 3-54, 3-58, 3-62, 3-63,
B-12, B-13, B-14
Flat terrain modeling 2-15, 3-14
Gradual plume rise
and the regulatory default option 2-7,3-6
GRDRIS parameter 3-4
specification of on the MODELOPT card 3-4, B-4
specifying the non-regulatory option 3-4
Half life
default value for urban S02 3-12
HAFLIF parameter 3-13
HALFLIFE keyword 3-13, A-4, B-3, B-5
relationship to decay coefficient 3-13
INDEX-4
-------
High value options for ST 3-97
Hourly emission rate file 3-49
Initial lateral dimension
for volume sources 3-27
Initial vertical dimension
for volume sources 3-27
Input meteorological data files 3-117
Input runstream file GLOSSARY-2
definition GLOSSARY-2
Intermediate terrain processing 1-16, 3-6
ISCEV model output options 3-110
Julian day
definition GLOSSARY-3
selecting specific days for processing 3-79
Keyword
definition GLOSSARY-3
detailed reference 3-1, B-l
Keyword/parameter approach
advantages explained 2-5
description of 2-1
Line sources, modeled as volumes 3-27
Linking the models 4-5, D-2
using memory overlays 4-5, D-2
Locations
specifying receptor location inputs 3-52
specifying source location inputs 3-22
Long Term model output options 3-111
Maximum value options
for the Long Term model 3-112
for the Short Term model 3-99
ME pathway 2-2
brief tutorial 2-21
example of inputs for 2-22
INDEX-5
-------
keyword reference 3-65, B-15, B-19
Message summary table
example for sample problem 2-31
example showing error condition 2-32
Meteorological data
ASCII format 1-11
card image format 1-11
options for Long Term 3-72
options for Short Term 3-66
unformatted or binary files 1-11
Missing data processing option 3-5, 3-7
Mixing heights
specifying averages for ISCLT 3-86, 3-87
Multiple year analyses for PM-10 3-19
Open pit sources 3-23
input parameters 3-32
OU pathway 2-2
brief tutorial 2-24
example of inputs for 2-24
keyword reference 3-96,B-23
Output file
organization of main print file 2-34
Output options
for ISCEV model 3-110
for Long Term model 3-111
overview 1-12
Pathways
input runstream pathways explained 2-2
order of 2-2
PCRAMMET preprocessed data files F-3
converting to ASCII format C-3
Plotting files 3-106, 3-123, F-9
Plume depletion 3-7
Point sources
and building downwash 3-35
input parameters 3-24, B-8
INDEX-6
-------
specification of location 3-23
specification of source type 3-23
Polar receptors 3-57
specifying a receptor network 3-57
specifying discrete receptors 3-57, 3-63
Postprocessing files 3-103, 3-122
estimating the size 3-105
Postprocessor files F-7
Precipitation scavenging
specifying input parameters 3-47
Printed output file 3-118
RE pathway 2-2
brief tutorial 2-20
example of inputs for 2-20
keyword reference 3-52, B-ll
Re-start capability 3-17
file descriptions 3-117, 3-120
INITFILE keyword 3-17
SAVEFILE keyword 3-17
Receptor networks
Cartesian grid 3-53
defining receptor grids 3-53
example of defining polar 2-20
modifying inputs for 2-43
polar 3-57
using multiple 3-60
Receptor options 1-10
Receptors
limits on number of 2-9
Regulatory default option 1-8,2-7
description 3-6
DFAULT parameter 3-4
specifying on the MODELOPT card 3-4
Repeat value
using repeat values for numeric input 3-38,3-39
Runstream file 1-2, 2-1
converting old inputs to new format C-l
INDEX-7
-------
debugging a 2-26
definition GLOSSARY-5
description of 3-116
example file for sample problem 2-26
Fortran unit number 3-116
functional keyword reference B-l
generated for ISCEV 3-17
modifying existing 2-41
numeric inputs 2-18
records or input images 2-3
rules for structuring 2-3
setting up an example 2-10
structure 2-2
use of DOS redirection with 3-117, 3-126
using the RUNORNOT option with complex 2-14
Rural dispersion option 1-8, 2-13, 3-4
potential temperature gradients 3-6
selection of on MODELOPT card 3-4
wind profile exponents 3-6
Secondary keywords
use of for certain input parameters 2-2, 2-7
Settling and removal
MASSFRAX keyword 3-46
PARTDENS keyword 3-46
PARTDIAM keyword 3-46
specifying input parameters for 3-46
SO pathway 2-2
brief tutorial 2-16
example of inputs for 2-17
keyword reference 3-21, B-7
Source code
portability to other systems 3-127
Source contribution analyses 1-13
use of the EVENT model for 1-14
use of the SOCONT option for ISCLT 3-112
Source groups 3-51
limits on number of 2-9
INDEX-8
-------
specifying a group of ALL sources 3-51
SRCGROUP keyword 3-51, A-5, B-7
Source IDs
specifying alphanumeric 3-23
Source ranges
specifying and interpreting 3-36
Sources 3-21
limits on number of 2-9
specifying source location inputs 3-22
specifying source parameter inputs 3-21, 3-24
Stack parameters
see Point sources 3-24
Stack-tip downwash
and the regulatory default option 2-7,3-6
NOSTD parameter 3-5
specifying not to use on MODELOPT card 3-5, B-4
STAR frequency files F-5
specifying contents of the STAR file 3-12, 3-76
Storage limits 2-8
modifying the storage limits 4-6
Surface roughness length 3-87
Temperatures
specifying averages for ISCLT 3-85
Terrain 1-10
Terrain grid data 3-90
Threshold violation files 3-101, 3-108, 3-121, 3-124, F-6, F-10
Unformatted meteorological data
description of file structure F-3
Unformatted meteorological data files
converting to default ASCII format C-3
specifying as input to ISCST 3-66, 3-70
Units
input units for numeric data 2-4
Upper case vs lower case inputs 2-4
Urban dispersion option 1-8, 2-13, 3-4
and decay for S02 2-5, 3-12, 3-13
INDEX-9
-------
potential temperature gradients 3-6
selection of on MODELOPT card 3-4
wind profile exponents 3-6
Variable emission rates 3-40, A-3, B-7
EMISFACT keyword 3-40, 3-42, B-7, B-9
factors for the Long Term model 3-42
factors for the Short Term model 3-40
hourly emission file option 3-49
Vertical potential temperature gradients
regulatory default values for 3-6
specifying inputs for 3-83
Volume source 3-27
Volume sources
input parameters 3-25,B-8
specification of location 3-23
specification of source type 3-23
Warning message 2-27, E-3
example of syntax 2-28
Wet deposition
GAS-SCAV keyword 3-48
PARTSLIQ and PARTSICE keywords 3-47
specifying input parameters 3-47
Wind profile exponents
regulatory default values for 3-6
specifying inputs for 3-82
INDEX-10
-------
INDEX-11
------- |