&EPA
United States
Environmental Protection .
Agencv
Office of Air Quality
Planning and Standards
Research Triangle Park N'C 2771
EPA-450/4-89-012
July 1989
PROCEDURES FOR
APPLYING
CITY-SPECIFIC EKMA
-------
-------
EPA-450/4-89-012
PROCEDURES FOR
APPLYING
CITY-SPECIFIC EKMA
Office Of Air Quality Planning And Standards
Office Of Air And Radiation
U. S. Environmental Protection Agency
Research Triangle Park, NC 27711
July 1989
-------
This report has been reviewed by the Office Of Air Quality Planning And Standards, U. S. Environmental
Eroitection Agency, and has been approved for publication. Any mention of trade names or commercial
products is not intended to constitute endorsement or recommendation for use.
EPA-450/4-89-012
11
-------
PREFACE
This document is one of five related to application of EKMA and the use
of OZIPM-4 (Ozone Isopleth Plotting with Optional Mechanisms), the computer
program used by EKMA. Listed below are the titles of the five documents and a
brief description of each. -
"Procedures for Applying City-specific EKMA", EPA-450/4-89-012, July 1989
- Describes the procedures for using the Empirical Kinetic Modeling
Approach (EKMA).. The major focus is on how to develop needed inputs for
OZIPM-4. In addition this document describes how to determine a control
target once OZIPM-4 has been run.
"A PC Based System for Generating EKMA Input Files", EPA-450/4-88-016,
November 1988
- Describes a program that creates EKMA input files using a menu driven
program. This sofware is only available for an IBM-PC or compatible
machine. Files built using this system can be uploaded to a mainframe
computer.
"User's Manual for OZIPM-4 (Ozone Isopleth Plotting with Optional Mechanisms}-
Volume 1", EPA-450/4-89-009a, July 1989
- Describes the conceptual basis behind OZIPM-4. It describes the
chemical mechanism, Carbon Bond 4, and each of the options available in
OZIPM-4. Formats for each of the options are outlined so that a user
can create input files using any text editor.
"User's Manual for OZIPM-4 (Ozone Isopleth Plotting with Optional Mechanisms)-
Volume 2: Computer Code", EPA-450/4-89-009b, July 1989
- Describes modifications to the computer code that are necessary in
order to use OZIPM-4 on various machines. A complete listing of OZIPM-4
is also found in this publication.
"Consideration of Transported Ozone and Precursors and Their Use in EKMA",
EPA-450/4-89-010, July 1989
- Recommends procedures for considering transported ozone and
precursors in the design of State Implementation Plans to meet national
ambient air quality standards for ozone. A computerized (PC) system for
determining whether an ozone exceedance is due to overwhelming transport
is described. This document is necessary, only if an area is suspected
of experiencing overwhelming transport of ozone or ozone precursors.
-------
EKMA may be used in several ways: (1) as a means for helping to focus more
resource-intensive photochemical grid modeling analyses on strategies most
likely to be successful in demonstrating attainment; (2) as a procedure to
assist in making comparisons between VOC and NOx controls; (3) in non-SIP
applications, such as in helping to make national policy evaluations assessing
cost/benefits associated with various alternatives and (4) for preparation of
control estimates consistent with limitations/provisions identified in Clean
Air Act Amendments.
iv
-------
TABLE OF CONTENTS .
Page
List of Tables.... .........'. Vll-
List of Figures Vii1
Acknowledgements.! "~ iy
°*»*»»»**e«*»*»*«,»»^,^e-e^w# l j{
1.0 Introduction j
2.0 The CB-4 Mechanism. 4
2.1 Organic Reactivity 4
2.2 Use of CB-4 in OZIPM4..... !!.'.*!.*•.'!!.*!;!!.*.' 9
3.0 Procedures for Applying EKMA/CB-4 11
3.1 Selection of Model ing Cases 14
3.2 Development of Model Inputs ...!.*!'."!.".*.*!.*.".'' 15
3.2.1 Light Intensity 16
3.2.2 Dilution 18
3.2.3 Post-0800 Emissions '.'.['• 18
3.2.4 Initial N02/N0 .... ; .'.'!.'!..*."." 21
3.2.5 Ozone Transport 22
A. Present Transport of Ozone at the Surface 22
B. Present Transport of Ozone Aloft 25
C. Future Transport of Ozone 27
3.2.6 Precursor Transport 29
3.2.7 . Organic Reactivity...., ......'..".'!!.'.'.'.!.'! 31
A. Surface NMOC 31
B. NMOC Aloft ......*.'!.'!.'!.'!!!!! 35
3.2.8 Temperature ,. "..'... 36
3.2.9 Water Vapor ... ............. 37*
3.2.10 Biogenic Emission Estimates 37
3.3 Predicting Peak Ozone 38
3.3.1 Procedures for Making Ozone Predictions 38
3.3.2 Comparisons of Predictions With Observations.. 41
3.3.3 Review and Adjustment to Model Inputs 43
-------
tTABLE OF CONTENTS (CONTINUED)
Page
3.4 Computing VOC Emission Reductions........ .... 46
3.4.1 Derivation of Empirical Data '.. 46
3.4.2 Daily Ozone Design Value............. •. 47
•3.4.3 NMOC/NOX Ratios... .., 47
3.5 Selection of the VOC Emission Reduction Target :. 49
3.5.1 Without Overwhelming Transport. 49,
3.5.2 Selection of a VOC Reduction Target at Sites
Subject to Overwhelming Transport.. T 53
References • 55
Appendix A - Listing of CB-4 Mechanism '.; A-l
Appendix B - Estimation of Mixing Heights for Use in OZIPM4 B-l
Appendix C - Computation of Carbon Bond Fractions From GC Data C-l
-------
LIST OF TABLES
Number paqe
2-1 Carbon Numbers for CB-4 Organic Species... 8
3-1 OZIPM4/CB-4 Model Inputs. 17
3-2 Example Calculations of the Design NMQC/NOX Ratio.. 50
3-3 Example Illustrating Effect of Model Predictions on
Selection of Control Target „ 52
A-l CB-4 Mechanism „ A-2
B-l NWS Radiosonde Stations , B-4
B-2 Preferential Order of Data Selection.;......,. B-6
B-3 Procedures for Estimating Mixing Heights B-9
B-4 Worksheet for Computing Mixing Heights . B-ll
B-5 Surface and Sounding Data B-13
B-6 •'Morning Mixing Height Determintion ..B-17
B-6A Example (Hypothetical Data) B-19
B-7 Maximum Mixing Height Determination ; B-20
B-7A Example (Hypothetical Data) • B-24
C-l Species Profiles by Bond Groups for CB-4... C-2
C-2 Example Problem - Part 1 C-16
C-3 Example Problem - Part 2 C-17
vii
-------
LIST OF FIGURES
Number
2-1 Example of Carbon Bond Lumping Procedure . 7
3-1 OZIPM4 Example Isopleth Run ...:... 12
3-2 Example Determination of Hourly Emissions 20
3-3 Examples of Acceptable Monitoring Locations for
Estimating Transported Ozone 'f 28
3-4 Future Ozone Transport as a Function of Present
Transport ,
B-l Flow Chart for Tab! e B-3 g_
12
viii
-------
ACKNOWLEDGEMENTS
Contributions from authors of earlier versions of this document, served
as a basis for this report. These include Gerald Gipson, and Marcia Dodge,
Atmospheric Research and Exposure Assessment Laboratory, EPA; Bob Kelly,
Region II EPA; and Henry Hogo, Mike Gery, and Gary Whitten, Systems
Applications. In addition,, Ke.lth Baugues and Edwin Meyer, Jr. should be
acknowledged for the time spent in reviewing and-revising earlier drafts of
this report. Special recognition is due Mrs. Cynthia Baines for her splendid
clerical support in preparing and assembling this report.
ix
-------
1.0 INTRODUCTION
In March of 1981, the U.S. Environmental Protection Agency (EPA) issued
guidelines for applying the city-specific Empirical Kinetics Modeling Approach
(EKMA) (Gipson, et al, 1981). EKMA is a procedure that can be used to
estimate emission reductions that are needed to achieve the national ambient
air quality standard (NAAQS) for ozone. Application of city-specific EKMA
according to the March 1981 guidelines entails using the Ozone Isopleth
Plotting Package (OZIPP) to relate peak ozone concentrations to its
precursors--nonmethane organic compounds (NMOC) and oxides of nitrogen (NOY)
**
(Whitten and Hogo, 1978; and EPA, 1978). OZIPP is a computer program that
incorporates a simplified trajectory model and a chemical kinetics mechanism
(known as the DODGE mechanism) that mathematically simulate ozone formation.
After the issuance of the'March 1981 guidelines, the use of other chemical
.mechanisms with EKMA was suggested (Jeffries, et al, 1981; and Carter, et al,
1982). In response-, supplemental guidance on using other mechanisms was
circulated to EPA Regional Offices in December of 1981 (Rhoads, 1981).
Specific guidance regarding the use of one alternative mechanism-the Carbon
Bond III mechanism (CB-S)--was issued in February of 1984 (Gipson, 1984).
Since 1984, newer chemical mechanisms have been developed (Gery, et al,
1988; and Lurmann, et al, 1987). This document focuses on information
necessary to apply EKMA utilizing the Carbon Bond IV mechanism (CB-4) and •
provides details on all necessary input parameters. The discussions that
follow will focus exclusively on using the CB-4 mechanism with the OZIPM4
program. This program, Ozone Isopleth Plotting With Optional Mechanisms 4
(OZIPM4), is an updated version of OZIPP which contains the most recent
-------
chemical kinetics and stoichiometric information. Any application of EKMA
should be carried out with the OZIPM4 code described in EPA (1989).
The remainder of this document is divided into two chapters. Chapter 2
contains a discussion of the CB-4 mechanism and its relationship to the .OZIPM4
program. Chapter 3 describes the information necessary to develop input data
suitable for use of EKMA/CB-4.
In the OZIPM4 model, a column of air containing ozone and precursors is
transported along an assumed straightline trajectory. The trajectory is
defined so that the simulated column of air over the city being simulated
arrives at the site observing the daily maximum ozone concentration at the
! '
time of the observed maximum. As the column moves, it encounters gridded
emissions of fresh precursors that are mixed uniformly within the column. The
column is assumed to extend from the earth's surface through the mixed layer.
The assumed horizontal dimensions of this column are such that the
concentration gradients are small enough to make the horizontal exchange of
air between the column and its surroundings insignificant. The air within the
'column is assumed to be uniformly mixed at all times.
At the beginning of a simulation, the column is assumed to contain some
specified initial concentrations of NMOC, NOY, and CO. As the column moves
A i
along the assumed trajectory, the height of the column will change because of
variations in mixing height; it is assumed to change with time during a user-
selected period (for example, 8 a.m. - 3 p.m.), and to be constant before and
after that period. As the height of the column increases, its volume
increases, and air above from the inversion layer is mixed in. Pollutants
above the mixed layer are described as "transported above the surface layer"
-------
or "transported aloft." Any ozone or ozone precursors above the mixed layer
that are mixed into the column as it expands are assumed to be rapidly mixed
throughout the column.
Concentrations of NMOC species, NO, N02, CO, and 03, within the column
are physically decreased by dilution due to the inversion rise, and physically
increased both by entrainment of pollutants transported aloft and by fresh
emissions. All species react chemically according to the kinetic mechanism
selected. Photolysis rates within that mechanism are functions of the
intensity and spectral distribution of sunlight, and they vary diiirnally
according to time of year and location.
-------
2.0 THE CB-4 MECHANISM
As the name implies, CB-4 is the fourth in a series of evolving chemical
kinetics mechanisms. Each of the successive carbon bond mechanisms contains
revisions that reflect increased knowledge of the photochemistry leading to
ozone formation. The CB-4 mechanism is currently the most recent version of
that generic series. It has been designed to simulate laboratory smog chamber
experiments using detailed data bases, as well as atmospheric situations in
which much less information, is typically available. While a comprehensive
discussion of the scientific basis of the,CB-4 mechanism is beyond the scope
• i i i . , | ,
of this document, some introductory material on basic concepts is included
below for those unfamiliar with CB-4.
A distinguishing feature of any chemical mechanism is the manner in
which organic'reactivity is treated. Because the construction and use of a
mechanism that includes all atmospheric species is virtually impossible,
individual organic species must be combined, or lumped, into some sort.of
functional group or groups. Thus,, the discussion of any chemical mechanism
must necessarily address the manner in which organic chemistry is represented
in the mechanism. The concepts underlying the treatment of organic reactivity
in CB-4 are discussed in Section 2.1 below.
As noted in Section 1.0, use of the CB-4 mechanism in a city-specific
EKMA analysis is most easily accomplished with the OZIPM4 computer program.
2.1 Organic Reactivity
As described in Section 2.0, a characteristic that typically
distinguishes chemical mechanisms is the manner in which organic compounds are
represented in the mechanism. A number of approaches have been taken,
-------
but most have focused on lumping similar species into a single, identifiable
molecular species that represents the chemistry of that particular class of
compounds. For example, propylene might be used to represent the chemistry of
all alkenes. The CB-4 mechanism is somewhat different, in that the primary
functional organic groups are based, on various, types of structural units
(e.g., single-bonded carbon atoms) as opposed to molecular type (e.g.,
alkanes). As will be seen below, this kind of structuring results in some
organic species being partitioned among more than one functional group.
In CB-4, nine functional groups are used to represent organic species,
each based ion various, types of carbon bonds:
(1) single-bonded paraffinic carbon atoms, and represented by PAR;
(2) slowly reacting olefinic double bonds, almost exclusively
ethylene and represented by ETH;
(3) relatively reactive olefinic double bonds, and represented by OLE;
(4) less reactive aromatic compounds represented by TOL;
(5) more reactive aromatic compounds represented by XYL;
(6) formaldehyde represented by FORM;
(7) acetaldehyde and high aldehydes represented by ALD2;
(8) isoprene, represented by ISOP;
(9) nonreactive compounds represented by NR.
Just as important as the definition of the functional groups themselves
is the manner in which individual organic species are apportioned to those
groups. As noted above, a particular organic compound is assigned to a CB-4
group, or groups, on the basis of molecular structure. To illustrate the
procedure, consider the propene molecule which contains one single carbon-bond
-------
and one double carbon-carbon bond (see Figure 2-1). In the CB-4 mechanism,
the propene molecule is represented by one paraffin and by one olefin. In
essence, the molecule has been apportioned on the basis of the carbon-carbon'
bonds: the double bond represented by OLE, and the one single- bond by PAR.
Similar classifications haye been determined for hundreds of other compounds,
and they provide the basis for establishing the overall reactivity of an urban
mix.
In the propene example illustrated in Figure 2-1, note that the number
of carbon atoms associated with PAR is one, while the number for OLE is two.
, l\ general principle underlying use of the carbon bond mechanism is that the
number of carbon atoms associated with any individual carbon bond group is
fixed. (Table 2-1 shows these characteristic carbon numbers for the carbon
bond functional groups.) By making use of the carbon numbers, concentrations
• of each CB-4 group can be determined from concentrations of.individual organic
species. To illustrate, consider the propene example discussed above, and
further assume that the concentration.of propene is 3 ppmC. Since propene is
represented in CB-4 by one PAR and by one OLE, the 3 ppmC total propene
concentration must be apportioned to these two carbon bond groups. Of the
three carbon atoms in a propene molecule, one is PAR and two are OLE (see
Figure 2-1). Thus, one-third of the carbon atoms can be thought of as PAR, and
two-thirds as OLE. Since concentration is proportional to the number of
carbon atoms, the concentrations of PAR and OLE in the CB-4 mechanism would be
1 ppmC and 2 ppmC, respectively.* This same concept can be.extended to
mul.ticomponent mixtures as well. In such cases, concentrations of the
i.e., CpAR =1/3x3 ppmC and CQLE =2/3x3 ppmC
6
-------
H
I
C =
H H
II
C - C
A
,
- H
Propene
H H
C = C
H
H
- C - H
A
1 OLE
1 PAR
Figure 2-1. Example of Carbon Bond Lumping Procedure
7,
-------
TABLE 2-1
CARBON NUMBERS FOR CB-4 ORGANIC SPECIES
Carbon Bond Group Carbon Number (carbon atoms per
molecule)
PAR 1
ETH 2
OLE 2
TOL • 7
XYL 8 '
FORM 1
- ALD2 2
NR !
ISOP
8
-------
individual organic species are first apportioned to their respective CB-4
group. The total concentration of any particular CB-4 group is then obtained
by summing the contributions due to the individual organic species. This
procedure will be more fully discussed in Appendix C.
In using the CB-4 mechanism with the OZIPM4 program, absolute -.
concentrations of the individual CB-4 groups are not directly input to the
model. Rather, the total NMOC concentration is specified, and the fraction of
carbon attributable to each CB-4 group is input. For example, assume that the
total NMOC concentration is 2.0 ppmC, of which 1.4 ppmC is PAR (as determined
by the procedure described in the preceding paragraph). Then the apportioning
factor, or carbon fraction, for PAR would be 0.70,' indicating that 70 percent
of the total carbon is categorized as PAR. A special set of default values
for CB-4 fractions is normally used. These defaults were derived from 1984-86
NMOC species data collected in many cities (Jeffries, 1987).
2.2 Use of CB-4 in OZIPM4
The CB-4 mechanism that is contained in OZIPM4 is outlined in
Appendix A. A discussion of the development and testing of this mechanism is
contained in Gery, et al, 1988. More extensive information on the evolution
of the carbon bond mechanism in general can be found in Killus and Whitten,
1983; Killus and Whitten, 1982; Whitten and Hogo, 1977; Whitten, et al, 1980;
Whitten, et al, 1979.
The specific inputs necessary to use CB-4 in OZIPM4 are contained in
EPA, 1988 and EPA 1989. The discussions in Section 2.1 and above have
provided a general overview of the CB-4 mechanism and its relationship to the
OZIPM4 program. In most instances, consideration of the details of the
-------
mechanism will not be required in any particular model application. The major
concern in most applications is the determination of the total NMOC
concentration. In cases where defaults are not used,"specification of the
carbon bond fractions required to apportion the total carbon concentration to
the individual carbon bond groups (i.e., PAR, ETH, OLE, TOL, XYL, FORM, ALD2,
ISOP, and NR) is also of concern. The procedures for developing these and
other model inputs for an EKMA application are the subject of the next
chapter.
10
-------
3.0 PROCEDURES FOR APPLYING EKMA/CB-4
Although the March 1981 guidelines deal explicitly with OZIPP and the
DODGE mechanism, many of the concepts described in that document are relevant
to the use of EKMA with OZIPM4 and the CB-4 mechanism. For example, selecting
the cases to model is unaffected by choice of chemical mechanism. Neverthe-
less, use of CB-4 with OZIPM4 does require some special considerations. This
chapter will focus primarily upon these circumstances, but will also describe
all other facets of conducting an EKMA modeling analysis.
The ensuing discussion of using CB-4 with EKMA can perhaps be facili-
tated by a brief overview of the general modeling procedure. While the
following section describes EKMA in terms of ozone isopleth diagrams, it is no
longer necessary to develop these diagrams in.order to determine the VOC
control requirements. By using the EKMA option the control requirement is .
determined without drawing isopleth diagrams. The OZIPM4 program is used to
generate ozone isopleth diagrams that explicitly relate peak hourly ozone
concentrations to. initial (i.e., 8 a.m.) ambient levels of the ozone pre-
cursors NMOC and NOX (see Figure 3-1). The diagrams are used with a measured
peak ozone concentration and a city's NMOC/NOX ratio to compute, on a percent
basis, the VOC emission reduction needed to lower the observed peak to the
level of the standard. While isopleth diagrams are explicit functions of
initial NMOC and NOX, the positioning of the ozone isopleths on the diagram is
affected by model input variables that are related to meteorology, emissions
occurring throughout the day, and pollutants transported from areas upwind of
the city under review. Because these factors vary from day to day, the
11
-------
21*0 OlvQ QO'O 90-0
CO
01-0 •
90*0 9Q-0
(Wdd) XON
00-0°
-12-
-------
highest VOC emission reduction estimate will not necessarily correspond to the
highest, observed ozone peak (Killus and Whitten, 1983; and Killus and
Whitten, 1982). To account for this phenomenon, the modeling approach
recommended in the March 1981 guidelines consisted of:
(1) modeling a number of high, observed ozone peak concentrations;
'(2) computing the amount of VOC emission reduction needed tb lower each
peak to the level of the standard;
(3) selecting a final VOC emission reduction target that is consistent
with the statistical form of the ozone standard.
i • . • '•'.,.. i
Subsequent to the distribution of these recommendations, EPA issued '
supplemental guidance further recommending that predictions of peak 'ozone bei
1 i
compared to observed levels (Rhoads, 1981). If the agreement between
predictions and observations is found to be poor, review and possible
adjustment to key model inputs are suggested prior to computing VOC emission
reductions. While good agreement between predictions and observations does
not completely ensure accurate control estimates, successful prediction of
observed ozone peaks does provide some confidence that the chemical and
physical processes leading to ozone formation are being adequately simulated.
The modeling procedure described in the preceding paragraphs can be
divided into five basic steps which should be followed:
(1) selecting the observed ozone peaks to model;
(2) formulating the model inputs;
«
(3) predicting peak ozone;
(4) computing VOC emission reductions; and
(5) selecting the overall VOC emission reduction target.
13
-------
3.1 Selection of Modeling Cases
As noted in Section 3.0, the highest VOC control estimate may not
correspond to the highest observed ozone concentration. Further, the
statistical form of the ozone NAAQS permits on average, one daily maximum 1-
hour .average ozone concentration above 0.12 ppm per calendar year at each
site. Consideration of these two. factors Jed to the recommendation that a
number of observed peaks above 0.12 ppm be modeled for each site. The VOC
emission reduction target is then selected from these results in a manner that
is consistent with the statistical form of the ozone NAAQS'i For an EKMA/CB-4
analysis, the same procedure is recommended. . . - , , ; ,
Recommended Procedure: The five (5) highest daily, maximum ozone '
concentrations at each site should be selected as candidates for modeling.
Only ozone peaks that occur within or downwind of the urban area under review
should be included. The five highest such values should generally be chosen
from the most recent 3 years during which measurements were made at a site.
A State may choose to include an additional year if data from another
ozone season become available between the time of the SIP call and the time
when the analysis is conducted. While an additional year may be added, a year
may not be replaced (i.e., 4 years of data must be used). If 4 years of data
are included, the six (6) highest daily, maximum ozone concentrations at each
site should be selected as candidates for modeling. If there is a tie for the
last daily maximum value, both days should be modeled. In the event that a
..significant amount of time (a few years) passes between the time of the SIP
call and the start of the modeling analysis, the appropriate EPA Regional
office should be contacted to determine the appropriate years to model.
In some cases, it may happen that on days initially selected as
candidates for modeling, daily maximum ozone (03) is most likely the result of
"overwhelming transport" from upwind areas. That is, it is unlikely that •
locally generated emissions have an appreciable effect on the observed daily
maximum. Procedures for determining whether an observed daily maximum results
from overwhelming transport are described in detail by Meyer and Baugues
(1989). In general, overwhelming transport is a strong possibility if the
14
-------
daily maximum occurs before 10 a.m. or if the timing of the observed maximum
is inconsistent with available ,10 a.m. - 4 p.m. surface wind data, and the,
orientation of the monitoring site with respect to the Metropolitan
Statistical Area (MSA) under review. Even if it is likely that an, observed
daily maximum 03 concentration results from overwhelming transport, if is
possible that the selected day should still be modeled. This would be
appropriate if the following occurred:, (a) concentrations greater than 0.12
ppm occurred at other times of the day; (b) surface wind data and monitor
i
orientation were consistent with impacts from the local MSA at these times;
,i , , ,
and (c) the 63 concentration judged'to result (from'local emissions was on$ of
the top five local peaks. Unless all of the preceding three conditions are
met, the day should be discarded and replaced by the previously unselected day
having the highest observed daily maximum. Overwhelming transport can also
affect.selection of which estimate for VOC and/or NOY controls is needed to
J\ ^ •
attain the NAAQS. This latter issue is addressed in Section 3.5.
3.2 Development of-Model Inputs . . . •
As just described, the five or six highest, daily maximum ozone peaks at
each site are selected for modeling. Two basic objectives of the. modeling
analysis are to predict the observed ozone peaks, and to compute the VOC
emission reductions needed to lower each observed peak to the level of the
ozone NAAQS. To best accomplish these objectives, the model inputs should be
based on the atmospheric conditions associated with each observed peak. Thus,
their derivation ought to be done on a case-by-case basis. In some instances,
however, insufficient or inadequate data preclude such a determination, and
appropriate approximations or defaults are needed. The major purpose of this
15
-------
I
section is to describe the methodologies recommended for deriving the model
input values under both sets of circumstances.
Table 3-1 summarizes the model input variables that require
consideration, regardless of the intended purpose of the model simulation.
Before discussing each of the model input variables, one additional point
should be added. The recommendations discussed below deal with model inputs
that correspond to conditions associated with the observed ozone peak.(i.e.,
so called base-case conditions). Some of these conditions might be 'expected
i
to change in future years subsequent to the implementation of VOC control
programs'. Factoring these potential changes into the. modeling analyses wi'll
be discussed in Section 3.4. Thus, the recommendations discussed below
concerning the derivation of model input values will, necessarily focus on data
i
corresponding to emissions and atmospheric conditions associated with a
particular ozone peak observed in the base case.
3.2.1 Light Intensity
The OZIPM4 program uses a city's latitude^ longitude, and time
zone, and the "day of the year being modeled to generate the appropriate1
diurnal pattern of photolytic reaction rates. While updates have been; made to
some of the photolytic rates, these have been incorporated in the OZIPM4
computer code. Thus, no changes need be made for this set of model inpu;ts.
The correct set of numerical time zones for the continental Unitedl '
States is as follows: • i •
Numerical Time Zone Common Name
4.0 Eastern Daylight Time .
5.0 Central Daylight Time
6.0 Mountain Daylight Time
7.0 Pacific Daylight Time
16
-------
Model Input Variables
Sunlight Intensity
Dilution
Post-0800 Emissions
Ozone Transport
Precursor Transport
Reactivity
Temperature
Water Vapor.
Biogenic Emissions
Table 3-1
OZIPM4/CB-4 MODEL INPUTS
Section
Addressed
3.2.1 ,
3.2.2
i
3.2.3
3.2.'5
3.2:6 .
3.2.4, 3.2.7
3.2.8
3.2.9
3.2.10 i
i i
17
-------
To produce standard time simulations, even though the output will show
daylight time units, a false time zone of one unit (hour) more can be used.
Thus, Pacific Standard Time photolysis constants would be generated if a 8.0
were entered instead of the correct 7.0 time zone.
Recommended Procedure: To properly simulate light intensity in OZIPM4,
input the city's latitude-, longitude, time zone, and the day of the year beinq
modeled. . ^' . - , • .
3.2.2 Dilution
\
In the OZIPM4 model, dilution occurs as a result of the rise in
atmospheric mixing height that typically1 occurs between early morning and mid-
afternoon. The mixing height can be viewed as the top of a surface based
; • I i • i .
layer of air which is well-mixed due to^mechanical and thermal turbulence.
Specific inputs to OZJPM4 include the early morning mixing height, the maximum
i
afternoon mixing height, the time that the mixing height rise begins, and the
time at-which the maximum mixing height is finally attained. Procedures for
!
estimating the early morning mixing height and maximum afternoon mixing height
from available radiosonde measurements!are outlined in Appendix B of this
document. The OZIPM4 program will internally calculate,the rate of rise in
mixing height based upon a characteristic curve developed by Schere and
Demerjian (EPA 1981; Schere, et al, 1977).
Recommended Procedure: It is recommended that city-specific estimates
of 0800 LCT mixing height and maximum afternoon mixing height be computed
using procedures outlined in Appendix B.| Minimum 0800 LCT mixing heiqtit used
should be 250 meters. I
3.2.3 Post-0800 Emissions
Post-0800 emissions refer to emissions occurring along the
trajectory subsequent to the start of the model simulation. The actual model
inputs are expressed as emission densities (kg/km2) of NMOC, NOX, and carbon
18
-------
monoxide (CO) concentrations that should be added each hour to represent the
effect of fresh precursor emissions. The requirements have changed from the
March 1981 guidelines in two respects: (a) large point sources of NOY are
A
differentiated from other sources and (b) CO emissions are considered.
The following example illustrates how to determine which county's
T. ^ ' *
emissions should be used for each hour. Figure 3-2 shows an example
trajectory. In this case the peak ozone occurs in Rockdale County between 3
1 i
and 4 p.m. The parcel starts at 8 a.m. in Fulton County. Between 8 a.m. and
i > '' , - ''
4 p.m. are 8 hours, indicated pq the straight-line trajectory in Figure 3-2.
Each value on'this'line represents the location of the parcel at 1 .through 8
hours.
During hour 1 (8-9 a.m.), ,the parcel is entirely in Fulton County.
Emissions for 8-9 a.m. from Fulton County should be used in OZIPM4. During
the second hour the parcel moves into Dekalb County, Emissions for 9-10 (hour
2) should' be averaged between Fulton and Dekalb Counties. During hours 3
i I • .
through 6, the parcel is entirely in Dekalb County and emissions should be
1 I :
based upon Dekalb County, buying hour. 7, the parcel crosses into Rockdale
County. Emissions for hour 7 (2-3 p.m.) should be averaged between Dekalb and
! 1 . • .
Rockdale counties. Emissions for hour-8 should be from Rockdale County.
In developing post-8 a.m. emission densities for EKMA, it is necessary
I !
to determine whether significant NOX point sources have effective stack
heights greater than the initial mixing height (usually 250 meters). If the
effective stack height is greater than the initial mixing height, the NOY
A
emissions from that source will not be contained within the mixed surface
19
-------
LU
CO
O
CO
CO
LJJ
5:
O
u_
O
O
I
s
tr
o
UJ
a.
CJ
UJ
cc
20
-------
•layer during the first few hours of the OZIPM4 simulation. At sometime later
in the day, as the mixing height rises, these NOX emissions will be entrained
into the mixed surface layer. Sources which should be reviewed to determine
effective stack height include large industrial process boilers and power
plant boilers. Only significant saurces of NOX (those greater than 5 percent
of the total NOX emission inventory) should be reviewed to estimate effective
stack height.
Each day is modeled to determine hourly mixing heights. The days must
then be rerun adding in the tall stack NOX emissions.. These emissions should
be added to the hour (and only to that hour) where the mixing height first
equals or exceeds the effective stack height.
The recommended procedure for estimating effective stack height is to
run PTPLU (EPA, 1982). This model requires stack height, stack gas
temperature, stack gas velocity, and inside stack diameter. The
meteorological conditions of interest are C stability and 5 m/s wind speed (at
•
10 meter height).
An alternative procedure is contained on the following pages for those
individuals who do not have access to PTPLU.
Recommended Procedure: Post-0800 emissions should be entered as
emission densities (kg/km*1). Large NOX sources may require a review of
effective stack height to determine if the NOX emissions are within the mixed
surface layer. Emission densities are required for NMOC, NOV, and CO.
A
3.2.4 Initial NDo/NO.. . '
The March 1981 guidelines recommend a default value of 0.25.
Review of recent data indicates that this ratio may vary over a wide range
(.1-.9) and that "median" ratios for individual cities may also vary
21
-------
significantly. Modeling analyses have indicated that EKMA-CB4 is not
sensitive to this ratio. Thus the default value of 0.25 should be used.
Recommended Procedure: A city-specific value for the nitrogen dioxide
to oxides of nitrogen ratio [(N02)/NOYJ need riot be estimated. The default
value of 0.25 should be used.
3.2.5 Ozone Transport
The two possible mechanisms by which .ozone is transported.into an
urban area are:
1. Advection of ozone along the earth's surface, and
2. Advection of ozone aloft, typically at night and during early
morning hours above the ground-based mixed layer, with downward mixing when
the mixing layer increases later in the day. .
Ozone transported at the surface is subject to surface reactions and
scavenging by other species [e.g., nitric oxide (NO)] emitted during the
night. As a result of nighttime atmospheric stability, ozone transported
aloft does not come into contact with scavengers emitted during the night.
Thus, overnight advection of ozone aloft is the more significant mechanism of
transport from one urban area to another (EPA, 1977; and Chan, et al, 1979).
Control strategies designed to attain and/or maintain the ozone standard
in individual urban areas must take into consideration the impact of
transported ozone on peak afternoon concentrations.*
A. Present Transport of Ozone at the Surface
The, chief impact of ozone transport near the surface is expected to
be the more rapid conversion of NO to N02. Several field studies have shown
Recall the discussion in Section 3.1, there are days when transport is
the predominant cause of observed 03. The recommendations in Section 3.2.6
apply for instances in which this is not the case.
22
-------
ALTERNATIVE PROCEDURE FOR ESTIMATING
EFFECTIVE STACK HEIGHT
1. Estimate wind speed at stack height
—
h -
u - 5.0
m/s
10 '
where: h is the physical stack height in meters
2. Estimate F (the flux parameter)
F - g V? d2 DT/4TS
where: g = 9.8 m/s2
Vs = stack gas velocity (m/s)
d - inner stack exit diameter (m)
€ DT = Ts - T ( stack ogas temperature- ambient air
temperature) K
Ts = ambient air temperature °K (assume 297 °K)
3. Estimate crossover temperature DTC
For F < 55 DTC = 0.0297 TS V^/Vd1/3
For F 2: 55 DTC = 0.00575 TS Vs2/3/d1/3
where: TS = ambient temperature (°K)
Vs « stack gas velocity (m/s)
d = stack exit diameter (m)
If DT < DTC Go to step 4
If DT £ DTC Go to step .5
23
-------
4. Estimate momentum plume rise
Dh = 3 d Vs/u
where: d = stack exit diameter (m)
Vs = stack gas velocity (m/s)
u = wind speed at stack height (m)[from Step 1]
Dh = plume rise (m)
Go to Step 6
5. Estimate buoyant plume rise
For F < 55 Dh = 21.425 F'75/u
For F > 55 Dh = 38.71 F'6/u
where: Dh = plume rise (m)
F = flux parameter [from Step 2]
u = wind speed at stack height (m)[from Step 1]
6. Estimate effective stack height
• H = hs + Dh
where: H = effective stack height (m)
hs = physical stack height (m)
Dh = plume rise (m) [from Step 4 or 5]
24
-------
that ozone transported along the surface tends to be minimal (Chan, et al,
1979; Decker, et al, 1977; and Ludwig, 1979).
Recommended Procedure: Based on the previous discussion, it is
. recommended that, for most situations, the value for present ozone transported
at the surface be set equal to zero for each day modeled.
Alternate Procedure: If ozone levels are measured downtown during
6-9 a.m., surface ozone.transport may be- considered. It is recommended that
the 6-9 a.m. LCT average ozone concentration at an- urban site(s) be used as
the estimate of the concentration of ozone transported into the urban area
along the surface for the given day.
B. Present Transport of Ozone Aloft
As noted above, it appears that unscavenged ozone, transported aloft
is likely to have a far greater impact than surface transport on maximum
afternoon ozone levels observed within or downwind from cities. Thus,
estimates of ozone aloft .are needed for control strategy development with
OZIPM4/EKMA. Techniques for estimating the level of ozone transported aloft
have been the subject of two studies (Chan, et al, 1979; and Eaton, et al,
1979). Five different techniques, which were considered to be feasible, were
field tested in Philadelphia during the summer of 1978 "(Chan, et al, 1979).
The five methods are: (1) use of fixed ground based stations; (2) use of
airborne measurements in a dedicated aircraft; (.3) use of airborne-
measurements with a portable instrument package; (4) use of free lift bal.loon
soundings; and (5) use of soundings by tethered balloon. Chan, et al (1979),
contains a detailed description of each of these techniques as well as a
discussion of the findings of the study. Of the five measurement techniques
evaluated, surface measurements at fixed sites, airborne measurements by
dedicated instrumented aircraft, and soundings by ozonesonde beneath a free
25
-------
balloon were judged to be practical means of providing information on ozone
transported aloft.
During the summers of 1985 and 1986, measurements of ozone aloft were
made over six cities using aircraft. Cities involved in this analysis were:
Dallas, Texas; Tulsa, Oklahoma; Atlanta, Georgia; Birmingham, Alabama;
Philadelphia, Pennsylvania; and New York, New York. Ninety percent of the
ozone aloft values from this study fall between approximately 25 to 60 ppb.
[Baugues, 1987]. Thus, measurements at other sites should be near this range.
For those cities located in the ROMNET domain, an alternative procedure is
being developed. Present and future aloft values for NMOC, NOV, CO and 0,
X o
will be based upon results from the ROMNET simulations. Exact procedures and
data bases will be avail-able in mid-FY-90.
Recommended Procedure: In selecting this recommendation, consideration
has been given to such factors as technical capability and available funding,
and the intended use of the data. Ozone measurements taken on the day being-
modeled are recommended as the best estimate of ozone aloft. These
measurements should be obtained at surface monitoring sites upwind of the city
during the first hour after breakup of the nocturnal inversion. An acoustic
radar (sodar^can be used to determine the time of inversion breakup for the
day. If the time of the breakup of the nocturnal radiative inversion is not
known, the 1000-1200 Local Civil Time average ozone concentration recorded at
the upwind monitor should be used as the transport estimate. A major
advantage of surface measurements is that it is the only method which provides
continuous measurements and, thus, assurance that measurements exist for days
or for times of day which are later determined to be of interest. The sitefs)
should be located in as rural a location as possible so as not to be
appreciably affected bv local sources of precursors. The distance such
upwind sites should be located from a city depends on the extent of urban
development. Because it is desirable not to measure pollutants which are
recirculated from the citv under review, a distance of 40 km or more upwind
from the urban core should be sufficient. This distance perhaps could be
reduced for smaller cities. Figure 3-3 depicts orientations for acceptable
upwind sites.
Alternate Procedure: Information on the vertical distribution of ozone
transported above the surface layer in the early morning may be used directly
if it is available. Such information might include aircraft or free-lift
ozonesonde measurements. The reader is referred to Chan, et al, 1979, for a
detailed discussion of these techniques. Use of an alternate procedure is
26
-------
subject to the approval of the EKMA contact person in the appropriate U. S
EPA Regional Office.
Missing Data:' In the event that an estimate of transport is not
available for a given day being modeled, the median transport value from the
remaining days being modeled should be used as a default value. This procedure
applies to all the data gathering techniques described above: fixed site,
aircraft, and ozonesonde observations.
C. " Future Transport of Ozone -
If control programs are implemented in upwind areas, ozone "
transported into the city may be reduced. However, in most cases, the source
area and the level of future controls are not likely to be known to any degree
of certainty.
Recommended Procedure: Because of the considerable uncertainty .in the
location and future control levels of the source area(s) for ozone transported
into the urban area, the relationship depicted in Figure 3-4 is recommended
for estimating the future ozone transport level given the level of present
transport. The solid curve in Figure 3-4 was derived on the basis of changes
in VOC emissions which are projected assuming a national mix of source
categories; national estimates of projected, growth in stationary source
emissions and vehicle miles traveled; anticipated impact of applying
reasonably available control technology to stationary sources and the impact
of the Federal Motor Vehicle Control Program on mobile sources; and
consideration of natural background levels. It was assumed that future ozone
levels would not exceed the NAAQS. The sol id curve is most appropriate for
use by cities subject to impacts from large upwind nonattainment areas. The
dashed curve is most appropriate for use when a city is isolated and not
impacted by large designated nonattainment areas.
Future ozone transport levels can be-computed by use of the following
equations:
For areas with large designated nonattainment area upwind:
03 (future) - 0.7 * (03 (present) - 0.04) + 0.04
For isolated areas:
03 (.future) - 0.9 * (03 (present) - 0.04) + 0.04
Where:
03 (future) = future ozone transport level (ppm)
03 (present) = present ozone transport level (ppm)
27
-------
(8)
' STAGNATION
Denotes upwind area
Figure 3^3 Examples of acceptable monitoring locations for estimating
transported ozone.
28
-------
The coefficients "0.7" and "0.9" in the preceding expressions were '
obtained by reviewing OZIPM4 runs with varying conditions. The preceding
expressions assume an irreducible background component of 0.04 ppm.
Without information to the contrary, future transport along the surface
should be-assumed equal to zero. If significant nonzero concentrations are
found for present ozone transport along the surface, then future ozone
transport levels should be obtained using the relationships shown in
Figure 3-4. " ' . ' . . -
For those cities located in the ROMNET domain, an alternative procedure
is being developed. Present and future aloft values for NMOC, N0y, CO and 0-,
will be based upon results from the ROMNET simulations. Exact procedures and
data bases will be available in mid-FY-90.
3.2.6 Precursor Transport
Just as for ozone, precursor pollutants could be transported in
both the surface layer and aloft. However, outside urban areas, the surface
layer is expected to be very shallow. Thus, long-range transport of
precursors in the surface layer may not be significant. Transported precursor
i
concentrations tend to be substantially less than concentrations within urban
areas (EPA, 1978). Recent measurements of NMOC aloft over six cities
indicates that most NMOC aloft values fall within a range'of 0-50 ppbC. The
overall median value for these data is 30 ppbC [Baugues, 1987]. Future NMOC
aloft should be reduced 20 percent from present levels. Present and future
levels of NOX aloft should be set to 2 ppb (.002 ppm). Surface levels of NMOC
and NOX for both present and future conditions should be set to zero.
Carbon monoxide levels at the surface should be set to. zero. Concen-
trations a-loft should be set to 0.5 ppm. Future levels of aloft CO may be
reduced 20 percent from present levels. For those cities located in the
ROMNET domain, an alternative procedure is being developed. Present and
future aloft values for NMOC, NOX, CO and 03 will be based upon results from
29
-------
04
6
to
T~
•
o
\
\
\
\
\
CD
t—
•
O
-
CO g
3 o.
o rsj
O
10 f-
- o
•
o
L q
o
CM
I- o
•
o
CM i-
o" d-
O)
O
CO
o
q
o"
to
o
m
o
o
•
o
q
CD
q
o
i- o
o.
LJ
tn
LU
o:
Q.
tL §
D. 5
en
s-
ID
Q- 5
tn i
2 "
< (O
2 |
O)
re
"o
O
a.
ui
c;
o.
O
o
ce:
=>
CD
D:
o
a.
oo
O
r-j
o
3NOZO
30 •
-------
the ROMNET simulations. Exact procedures and data-bases will be available in
mid-FY-90.
Recommended Procedure: Transported concentrations of NOY and NMOC in
the surface layer should be set to zero. The recommended default NMOC aloft
value is 30 ppbC based upon recent data. Present and future levels of NOX
aloft should be set to 2 ppb. Carbon monoxide levels aloft are recommended to
be set to 0.5 ppm. Future levels aloft CO may be reduced 20 percent from
present levels. The reactivity-of NMOC aloft will be discussed in-the .
following section.
3.2.7 Organic Reactivity
A. Surface NMOC
The fundamental concepts underlying the treatment of organic
reactivity in the CB-4 mechanism were described in Section 2.1. As noted in
that section, the organic reactivity input that is required by OZIPM4 consists
of specifying a set of apportioning factors, or as they are more commonly
termed, carbon-fractions. Specification of these fractions permits the OZIPM4
program to apportion total NMOC concentration" into the individual carbon
groups—PAR, ETH, OLE, ALD2, TOL, XYL, FORM, ISOP, and NR. (The apportioning
procedure is carried out within the model for the NMOC concentrations that
occur both initially and as a result of subsequent post-0800 emissions.)
Two basic approaches are possible for estimating the carbon-fractions.
The recommended approach consists of using a set of default fractions that
have been derived through analyses of available ambient organic species data,
and review of pertinent, scientific experimental results. The second, an
alternative approach,' requires the analysis by gas chromatography (GC) of
individual organic species' concentrations in ambient air within the city
under review. Typically, this latter approach requires a special field study.
31
f
-------
The recommended approach of using a default value rather than making a
city-specific determination arises primarily as a consequence of two factors.
First, the default values are estimated to be representative of typical urban
reactivity based on an analysis of ambient sampling results conducted in a
number of locales (Jeffries, 1987). While some city-to-city variations in
organic composition are to be expected, the default recommendations should
adequately represent most -United States cities (Jeffries, 1987). The second
factor relates to the resource requirements associated with the alternative
approach. The cost of conducting a special ambient sampling program can be
substantial.
Instead of using default values, carbon-fractions can be computed from
GC analysis of ambient samples. Monitoring considerations in performing GC
sampling/analysis are discussed by Singh (1980) and EPA (1980), and will not
be repeated here. However, it should be noted that GC analysis is not an
automated technique, and is most often done on a special study basis. Thus, a
monitoring program of limited duration is the most pragmatic approach for
developing the information needed to compute carbon-fractions. While it is
difficult to prescribe exactly the number of samples needed, enough should be
analyzed to ensure that representative, average carbon-fractions can be
computed. Since the carbon-fractions will be used to apportion initial
concentrations and concentrations due to fresh emissions* the most appropriate
sampling period is one prior to the onset of significant photochemical
reaction. As with NMOC monitoring, sampling during the 6-9 a.m. time period
within the area of maximum emission density (i.e., usually the center city) is
32
-------
generally recommended. Ambient samples for GC analysis should be collected by
Integration over a period of 3 hours.
The GC analysis must identify all species up to and including C-12
'(compounds containing 12 carbons). Identification of any peak over 0.5 parts
per billion (ppbC) is required. Where an individual peak cannot be identified
as a specific compound, it must be analyzed to determine the carbon number and
the class (paraffin, olefin,- or aromatic)..
Regardless of the technique employed in their derivation, the carbon-
fractions are used to apportion total concentrations of organic compounds
which are based upon ambient measurements. Of the two organic compound
monitoring techniques [i.e., PDFID (preconcentration direct flame ionization
detection) and GC], both rely on a flame ionization detector,that is
relatively inefficient in responding to many oxygenated compounds such as
aldehydes and ketones (i.e., these techniques measure hydrocarbons only). SAI
"has estimated that, initially, total carbonyl compounds (i.e., those including
aldehyde and ketones, as well as-some surrogate carbonyls) are about 5 percent
of total nonmethane hydrocarbon concentrations (Killus and Whitteri, 1983).
Only about 1 percent of the total carbon that is measured can be classified in
the carbonyl group (i.e., surrogate carbonyls). The remainder of the
carbonyls (i.e., 5 percent of the nonmethane hydrocarbons that are measured)
I :
is attributable to oxygenates that are not detected. The carbon-fractions
which would sum to 1.05 (or 105 percent) are then adjusted so that, they total
only 1.00 or 100 percent. If ambient measurements of aldehydes are available,
a city-specific determination of the carbonyl fraction can be made. However,
these measurements tend to be complex, using techniques that are mostly
33
-------
conducted by research groups. As a consequence, carrying out a special
aldehyde monitoring program cannot be routinely recommended.
Whenever city-specific estimates are made by the techniques discussed
above, special care must be taken to ensure that the computed carbon-fractions
represent a realistic distribution of NMOC species.
Reactivities of NMOC in numerous cities have been.computed based upon
KOH values- KOH values are rate constants which give a measure of the
reactivity of a class of compounds with OH radicals. The weighted sum of
these KOH values give an estimate of the overall reactivity of the NMOC mix.
The KQH value for an NMOC mix can be determined using the following
equation:
KQH = PAR * 1203 + ETH * 5824 + OLE * 20422 + ALD2 *'11833'+ TOL * 1284 +
XYL * 4497 + FORM * 15000
Where:. KQH is the average KQH value for the NMOC mix, PAR is the
fraction of the mix considered paraffin (based upon the CB-4
splits).
A typical city is expected to have an average KQH value that falls
between 2700 and 3600 min"1. If the computed KQH value, based upon a city-
specific NMOC distribution, does not fall within this range, the process of
determining the city-specific distribution should first be redone to ,check for
errors. If no errors are found, the default reactivity should be utilized.
Requests to use reactivities other than the default must be reviewed and
approved by the appropriate Regional Office in cooperation with the Model
Clearinghouse.
34
-------
Recommended Procedure. The carbon-fractions recommended for use in an
EKMA/CB-4 analysis are listed below:
PAR = .564
ETH = .037
OLE = .035
ALD2 = .052
FORM - .021
TOL = .089
XYL = .117 . - -,
ISOP - 0 '-
NR = .085
They should normally be used unless sufficient information is available to
derive city-specific information by the method discussed below.
Alternate Approach. If analyses of ambient air samples by gas
chromatography are available for a particular city, the results can be used to
derive carbon-fractions. The ambient samples should be taken in the high
emission density area (normally the urban core) within the 6-9 a.iiu Local
Daylight Time (LDT-) period during the ozone season. Integrated samples are
required. It is desirable that enough samples be analyzed to provide a
representative average. For supplemental-information regarding monitoring
aspects, the reader is referred to Singh (1,980) and Rhoads (1987), and for
details on how carbon-fractions are computed from the sampling results, the
reader is referred to Appendix C of this document. Those considering this
approach should discuss it with the EKMA contact in the appropriate Regional
'Office.
Caveat> If the alternative approach is^sed, the resultant reactivity
must fall within the range of 2700-3600 min"1. If it does not, it is strongly
recommended that the data and computations be thoroughly checked to ensure
that no errors have been introduced. If the problem cannot be resolved, use
of the default carbon-fractions listed in the recommended procedure above is
preferable.
B. NMOC Aloft' ' '
OZIPM4 also requires carbon-fractions for the NMOC aloft. The
recommended approach is to' use the -default value provided. These values are
based upon GC speciation of aircraft samples taken over six cities during the
summers of 1985 and 1986 [Baugues, 1987]. In order to develop a city-specific
distribution, a special field study would be required. Such an analysis is .
not recommended.
35
-------
For those cities located in the ROMNET domain, an alternative procedure
is being developed. Present and future aloft values for NMOC, NCL, CO and 0,
x o
will be based upon results from the ROMNET simulations. Exact procedures and
data bases will be available in mid-FY-90.
Recommended Procedure. The carbon-fractions recommended for use in
EKMA/CB-4 for NMOC aloft are:
PAR
ETH
OLE
ALD2
FORM
TOL
XYL
I SOP
NR.
_
=
=:
=
=
=
=
=
.498
.034
.020
.037
.070
.042
.026
0
.273
3.2.8 Temperature
Hourly temperature data must be utilized in OZIPM4. Use of hourly
temperatures allows reaction rates to be increased or decreased according to
the hourly temperature. If not specified, OZIPM4 uses a default temperature
of 303°K. The hourly surface.temperatures to be utilized, in OZIPM4 should be
from an urban meteorological station. Tapes which provide complete hourly
data are available from the National Oceanic and Atmospheric Administration
(NOAA) in Asheville, NC.
Recommended Procedure. Hourly surface temperatures from an urban
meteorological station are recommended for use in OZIPM4.
36
-------
3.2.9 Water Vapor
Recent work has shown that ozone predictions are sensitive to the
amount of atmospheric moisture content. A new option has been included in
OZIPM4 which will estimate the atmospheric moisture content given relative
humidity values and an ambient pressure level. Hourly values of relative
humidity can be found on meteorological tapes available from NOAA (in
Asheville, NC).
Recommended Procedure. Hourly relative humidity values are recommended
for use in OZIPM4. .
3.2.10 Biogenic Emission Estimates
OZIPM4 has recently been modified to contain an option to allow
inclusion of biogenic emission rates. The inputs to OZIPM4 are emission
estimates of the biogenics, typically broken out as: isoprene, a-pinene,
monoterpenes and unknowns. The units for these values are kilograms per
square kilometer per hour (kg/km2/hr).
The biogenic emission estimates, are sensitive to several meteorological
parameters: air temperature, wind speed, relative humidity and cloud cover.
Therefore, biogenic emission estimates must be developed for each day modeled
and the values based upon day specific meteorological parameters.
EPA will provide a computer program that can be run on an IBM-PC (or
compatible machine) which will estimate biogenic emissions rates on a county
basis. The user would need, to provide day specific meteorological parameters.
This program should be available by mid-1990.
As with man-made emissions (Section 3.2.3), emission rates should be
included for the county in which the straight line trajectory is over for each
hour.
37
-------
OZIPM4 also requires initial values for biogenic species. All initial
values should be set to 0.0001 ppm in the absence of measured concentrations.
They should not be left.at zero, as this may cause the program to "hang up".
3.3 Predicting Peak Ozone
In one study, it was found that EKMA could yield a lower control
estimate for a case when peak ozone is underpredicted as compared to one in
which peak ozone is more accurately predicted (Jeffries, et ,al, 1981). In
extreme cases of underprediction, a solution may not be possible with EKMA.
In a similar fashion, a large overprediction could lead to a control estimate
that is higher than that obtained when good agreement is found. As part of
the supplemental EKMA guidance issued in December of 1981, EPA addressed this
problem by recommending that predictions of peak ozone be made, and
appropriate adjustments or compensations be made if poor agreement is found
(Rhoads, 1981). In this section, the procedures for making the predictions,
comparing them with observations, and making appropriate adjustments are
•
described.
3.3.1 Procedures for Making Ozone Predictions
In Section 3.2, most of the OZIPM model inputs that are- needed
either to predict peak ozone or to estimate VOC emission reductions were
discussed. In order to make predictions of peak ozone, one additional set of
model input variables is needed: the concentrations of NMOC, N0vy and CO that
X
are representative of the initial (i.e., 8 a.m.), urban core levels. These
model" inputs are the most critical for making predictions, and should be
estimated on a case-by-case basis. Because of the model sensitivity to these
inputs, use of mean or median values compiled from measurements taken across a
38
-------
number of days may lead to erroneous results. Thus, day-specific measurements
should normally be used to make these estimates.
As for estimating the initial concentrations, the recommended procedure
is to make use of ambient NMOC, NOX, and CO measurements routinely taken in
the urban core,.but which represent neighborhood scale levels. The initial
concentrations are intended to represent the NMOC, NOY, and CO that is
A
initially present within the mixed layer at the start of the model simulation
(i.e., 8 a.m.). While several approaches could be taken, the recommended
method is to use the 6-9 a.m. average concentration measured by collocated
NMOC, NOX, and CO monitors within the urban core which represent neighborhood
scale values. If more than one set" of measurements are available from several
such monitors, then the 6-9 a.m. average concentration at each monitor should
be averaged to obtain an overall, urban average NMOC, NOV, and CO
A
concentration. Algebraically, the above procedure can be expressed as
follows:
'(NMOC)o
(NMOC)6-9
1 (3-4a)
N
and
'(NOx)o
n
2
1-1
and
'(C0)o
1=1
'(NOx)6-9
N
C(CO)6-9
N
(3-4b)
(3-4c)
39
-------
where
.,„._ .,ln (CNMOC)o, (cNOv)o, (CCO)o -initial concentrations of
NMOC, NOX, and CO (in units of ppmCT, ppm, and ppm, respectively) input to
OZIPM4 simulation
, . C(^NMQC)6-9]i, [(cNOx)6-9]i -the 6-9 a.m. average
[rCO)6-9]i concentrations of NMOC, NO , and CO
(in units of ppmC, ppm, and ppm,
respectively) taken in the urban
core (or high emission density area)
at site i
N = total number of collocated
monitors for which day-specific NMOC and NOY measurements are available.
' '• < . **
As noted above, the initial NMOC and NOY concentrations are derived from
i i f^
day-specific measurements of NMOC1 and NOX. In some instances, an NMOC
measurement may not be available for the day being modeled. In such a case,
the initial NMOC concentration can be approximated by making use of the
median NMOC/NOX ratio (see Section 3.4.3) and a day-specific measurement of
NOX alone, provided it -is available. The initial NMOC concentration for use
with the OZIPM4 simulation can be computed as the product of the median
NMOC/NOX ratio and initial NOX concentration, or
r • ' ' • r
(LNMOC)o = (LNOx)o (NMOC/NOX) (3-5)
where
(CNMOC)o = the initial NMOC concentration for
the OZIPM4 simulation, ppmC
C
( N0x)o = the initial NOX concentration
. calculated by equation 3-4b, ppmC
(NMOC/NOX) = the median NMOC/NOX ratio as derived
according to the procedures outlined
in Section 3.4
It should be emphasized that this approach is an approximation, and the
one described in the preceding paragraph is preferable.
40
-------
With the estimates of initial NMOC, NOX, and CO, and the corresponding
day-specific inputs listed in Table 3-1, the CALCULATE option of OZIPM4 may be
used" to perform a single model simulation. An example simulation and
additional information are contained in EPA, 1989. Thus, no additional
discussion will be Included here.
. Recommended Procedure. The CALCULATE option of the OZIPM4 program
should be used to predict peak ozone for comparison with the observed peak.
The model inputs discussed in Section 3.2 should be used, with initial
concentrations that have been derived according to equations 3-4a, 3-4b, and
3-4c, using data that are specific to the .day being modeled. In the event -
that day-specific NMOC measurements are unavailable, the initial NMOC
concentration can be approximated by means of equation 3-5, with the
recognition that some uncertainty may be introduced in the analysis. If day-
specific measurements pf NMOC, NOX, and CO are not available, predictions of
peak ozone cannot be made. In this case, computation of VOC control estimates
are recommended, but without the requirement of reasonable agreement between
prediction and'observation.
3.3.2 Comparisons of Predictions With Observations
The principal output of concern obtained with a CALCULATE
simulation is the predicted ozone. By numerically integrating the
differential equations describing ozone formation processes (i.e., chemical
reaction, emissions, dilution, etc.), instantaneous concentrations of ozone
are computed throughout the simulation period. From this computed profile of
instantaneous ozone concentrations, the OZIPM4 program.calculates the hourly
average concentrations occurring during the model simulation. The predicted
ozone concentration that occurs, at the time of the observed peak is used in
the performance measure that is recommended to evaluate model performance.
This performance measure is the relative deviation of the prediction from the
observation, or
DEV = C - C0
_£ » x 10Q (3-6)
Co
41
-------
where
DEV = deviation...of the model prediction from the
observation, percent
Cp = maximum 1-hour average predicted peak ozone, ppm
CQ, - observed peak ozone, ppm
If the relative deviation is found to be no more than ± 30 percent, then
agreement between the prediction and the observed peak is judged to be
sufficient to proceed with control estimate calculations. If the deviation is
outside the ± 30 percent range, a comparison between the measured peak and
predicted peaks 1 hour before (or after) the time of the observed peak should
be1made. Due to the uncertainty in trajectories, it is possible for the time
of the predicted peak to be off by an hour.
If the model underpredicts by more than 30 percent (i.e., DEV < - 30
percent) or overpredicts by more than 30 percent (i.e., DEV > + 30 percent),
the review of, and possible adjustment to, key model inputs according to the
discussion of Section 3:3.3 below is warranted. R should be noted that the
observed ozone peak (not the predicted) is recommended for subsequent control
calculations. . '
Recommended Procedure. The relative deviation of the model prediction
from the observed peak should be computed according to equation 3-6 above.
The model predicted peak to be used in this computation is the hourly average
ozone concentration calculated by the OZIPM4 program at the time of the
observed peak. If the computed "deviation is within ± 30 percent, then the
model results are sufficiently accurate for control estimate calculations. If
the deviation is outside the ± 30 percent envelope comparisons between the
measured peak and the predicted values 1 hour before (or after) the time of
the measured peak should be made. If the ± 30 percent test is not met, then
the procedures discussed in Section 3.3.3 should be applied, in an attempt to
improve the simulation results.
42
-------
3.3.3 Review and Adjustment to Model Inputs
If inadequate agreement between a model prediction and an observed
peak is found, review of the model inputs should be conducted. The objective
of this review is to investigate whether some modifications to key model
inputs can be justified on some physical basis in order to improve the model
predictions. This review should focus on those model inputs that most
critically affect predictions of peak ozone. Of most importance are the
initial NMOC, NOX, and CO concentrations; dilution; and post-0800 emissions.
Adjustment of these inputs, within the uncertainties associated with their
development, is warranted if improvements in model predictions can be made.
Obviously, any errors that may have been made in their derivation should be
corrected as we!1.
While specific recommendations in trouble-shooting poor model perform-
ance are difficult to make, some general guidelines can be made depending on
the nature of the problem, be it an underprediction or an overprediction. If
« i
some uncertainty exists with regard to the data from which they were derived,
then the inputs can be adjusted within that range of uncertainty. In general,
initial NMOC,'NOX, and CO levels may be adjusted by ± 15 percent and maximum
afternoon mixing height by ± 200 meters (Seila, 1986 and Rhodes and Evans,
1986). Outliers in the data may be deleted if adequate justification is
available. For example, assume that an initial NMOC concentration had been
derived from 6-9 a.m. measurements taken at three monitoring sites. However,
the 6-9 a.m. concentration at one site (say site A) deviated substantially
from the concentrations measured at the other two sites (say sites B and C).
Then, improved agreement between"model predicted and observed ozone might be
43
-------
found if the initial NMOC concentration for the day in question was derived
solely from the measurements taken at the two sites in agreement (i.e., sites
A and B). Consider the case of underproduction first. The following steps
should be taken:
1. Check inputs for errors (especially morning and afternoon mixing
heights and ozone aloft). ~-
2. Increase the initial NMOC, NOX and CO concentrations by 15 percent.
3. Reduce afternoon mixing height by 200 meters.
4. Increase original .afternoon mixing height by 200 meters (may
improve situations where ozone aloft is high).
5. Increase morning mixing height by 50 meters.
All steps are cumulative, except for 3/4, where the step that improves
the situation should be included with Step 5. Steps are to be followed in the
order above, and carried out only until the deviation is within the ± 30
percent range. Further adjustments should not be carried out to reduce the
deviation.
1
1 ' When changes are made to the morning mixing height, make sure that
changes are reflected in all options. The following options use the morning
mixing height: DILU (Dilution), MASS (Emissions), CRED (CO) and BIOG
(Biogenics).
Guidelines for correcting a problem of overprediction are similar in
concept to those for underprediction. The following steps should be taken:
1. Check inputs for errors (especially morning and afternoon mixing
heights and ozone aloft). Also make sure that all emission rates are being
2
read in as kg/knr/hr and not in fractions of the initial concentration.
44
-------
2. Reduce initial NMOC, NOX and CO concentrations by 15 percent.
3. Increase afternoon mixing height by 200 meters.
4. Reduce original afternoon mixing height by 200 meters (may improve
cases where ozone aloft is high).
5. Reduce morning mixing height by 50 .meters.
The same rules apply that were described earlier for the underprediction
case.
As noted-above, the model inputs that substantially affect model
predictions of peak ozone include the initial NMOC, NO , and CO
A
concentrations, and initial, mixing height. The possibility exists that mass
balance techniques could be used to evaluate the appropriatenessiof a
particular set of initial concentrations and an initial mixing height. For.
example, one could test by means of a simplified box model whether or not a
city's emissions are sufficient to generate the measured initial
concentrations within a mixed layer corresponding to the postulated initial
mixing height. While such an approach is intuitively appealing, such
calculations may not be able to account properly for ventilation, and for
advection of pollutants from source areas nearby the precursor monitors.
Nevertheless, it does provide one means of assessing the reasonableness of the
postulated model inputs.
mnft Recommended Procedure. To improve agreement with observed 0-3 levels
so canTcited"1 b^npv^d'^th in reasonable ranges if justification for doing
l»Arn f sPeciflcally, the key model inputs are initial NMOC NO
be aSiUSpf^hati0+nhS' and miXlng heights' Fina11*' model inputs shouldlnly
c.th i£fjed W]thl" the ra?9e of reas°nable uncertainty, and not just selected
obtained ^Ll^T* bet??? the mode1 Prediction and observed peak ts
should stirS IV! ac"ptable Demerit cannot be found, control estimates
should still be made and the procedures discussed in Section 3.5 applied.
45
-------
3.4 Computing VOC Emission Reductions
The recommended procedure for. computing VOC emission reductions is to
use the EKMA option in OZIPM4. Use of this option eliminates the need to
generate isopleth diagrams. OZIPM4 performs the necessary calculations and
determines the VOC emission reduction. Several.variables are needed to
perform this calculation. These include: "initial NMOC, initial NOY, initial
A
CO, NMOC/NOX ratio, maximum observed ozone, present and future levels of NMOC
aloft, NOX aloft, 03 aloft, and assumptions regarding future levels of NOX and
CO. Use of this option is discussed further in EPA, 1989.
3.4.1 Derivation of Empirical Data
Two pieces of empirical data are needed for calculating control
requirements. The first is the maximum 1-hour average ozone concentration
observed at the site of interest. The degree of emission control necessary to
reduce this "peak" to 0.12 ppm is to be calculated; hence, the peak level will
be termed the daily site-specific ozone maxima.
i
The second piece of information needed is the NMOC/NOV ratio. This
J\
ratio is derived from the 6-9 a.m. concentrations of NMOC and NOV within the
X
urban area. The ratio will be termed the- design ratio. The procedures for
deriving both the daily site-specific ozone control values and the design
ratios are described below.
3.4.2 Daily Site-Specific Ozone Control Value
The daily site-specific ozone maxima is used in conjunction with
the NMOC/NOX ratio for calculating control estimates needed to reduce the day-
specific and site-specific observed peak ozone to 0.12 ppm. The daily ozone
maxima should be expressed in ppm units rounded to two decima-1 places.
46
-------
Recommended Procedure: A daily site-specific ozone maxima is obtained
for each site which is downwind of the city, and/or within the city in the
case of light and variable winds on the day for which the control strategy is
to be developed. Surface wind data should be examined to assure that the site
is not "upwind" of the city. Based on the'results of field studies and
reviews in which ozone gradients dov/nwind from urban areas were examined, peak
ozone concentrations should generally be observed within 15-45 km downwind of
the central business district (EPA, 1978; Martinez and Meyer, 1976; and EPA,
1976).
3.4.3 NMOC/NOV Ratios
A
The prevailing 6-9 a.m. LCT NMOC/NOX ratio measured in the urban
core of the city is the second piece of empirical data required. The design
ratio is viewed as characteristic of the.city which would prevail during the
remainder of the morning and early afternoon in the, absence of chemical
reactions. OZIPM4 expresses peak ozone concentrations as a function of the
initial concentration of NMOC and NOX. Thus, the 6-9 a.m. LCT NMOC/NOy ratio
is considered to be the appropriate ratio for use in OZIPM4 since this ratio.
is consistent with the conceptual basis of the model (Dimitriades, 1977). To
ensure that representative ratios are obtained, the NMOC and NOY instruments
J\
should be collocated in the central core of the urban area. The site(s)
should be located in an area of relatively uniform emission density and not
significantly influenced by any individual source. More detailed guidance on
siting NMOC instruments is contained in EPA, 1980. Guidance on the operation
of NMOC instruments is available in EPA, 1985.
Significant discrepancies have been found between NMOC/NOX ratios
calculated on the basis of ambient measurements and those obtained from
emission inventory data (Drivas, 1978). Reasons for the lack of correlation
between the two ratio calculation procedures have not been resolved. As a
result, only ambient NMOC/NOX ratios should be used with EKMA since these
47
-------
ratios are consistent with the conceptual basis of the model and the. emission
ratios have been shown to be poor, surrogates for these ambient ratios.
NMOC data analyzed only with the PDFID (preconcentration direct flame '
ionization detection) or.GC (gas chromatograph) should be used with OZIPM4
(Rhoads, 1985).. Due 'to the large uncertainty in low NOX values, any day with
a 6-9 a.m. NOX concentration at or below 0.020 ppm should be excluded from' the
process of estimating NMOC/NO • ratios.
1 A ' '
The NMOC data are to be collected during the season of peak ozone
concentrations (summer). Because NMOC concentrations are apt .to be relatively
high in central urban locations at those times of the day (early morning) when',
these measurements are required for use in EKMA more confidence can be placed
in the estimate. Because of the variability .in individual NMOC readings, the
NMOC/NOX ratio calculated for a single day is not recommended for use in city-
specific EKMA. Considering instrument reliability and model sensitivity, the
following procedure is recommended for calculating NMOC/NO ratios.
. .. 'A
Recommended
A
avail fhio F PH:cursor measurements from more than one urban site are
««"» -
3. The peak ozone level for all days with NMOC/NO ratine -u
48
-------
4. Once-the local ozone peaks have been determined for all days with
NMOC/NOX ratios, the highest ozone days are selected. This consists of either
all days with ozone peaks above the NAAQS or the top 10 days if less than 10
days exceed the NAAQS.
5. The days selected in Step 4 are then ranked and a median NMOC/NtX,
ratio determined. " , x
It is the median NMOC/NOX ratio that is to be utilized in OZIPM4. Two
examples of calculating a median NMOC/NOV ratios are shown in Table 3-2.
A
3.5 Selection of the VOC Emission Reduction Target
3.5.1 Without "Overwhelming" Transport • ' '
• After all site/day combinations have been modeled, the final step '
of the modeling analysis involves the selection of the overall VOC emission
.' ' • ' ' ' ' i
reduction target. In essence, this procedure is dictated by the form of the ,
ozone NAAQS, and is identical to the method recommended in the March 1981
guidelines document (Gipson, 1981). In summary, a control target is selected
for each site that permits, on average, one hourly-average concentration above
0.12 ppm per year. This corresponds to selecting the fourth highest control
level if 3 valid years of data are available, the third highest control for
2 years of data, and the second highest control estimate for only 1 year.*
The overall control target is then chosen as the h.ighest of the site-specific
control estimates to ensure t.hat the ozone standard is attained at all sites.
As noted in the March 1981 guidelines, an additional factor that could
affect the procedure just described is the consideration of model predictions
A site is considered to have a valid year of data if valid daily maxima
exist for at least 75 percent of the days during the ozone season and there is
no obvious pattern of missing data during periods when maximum ozone is most
nkeiy. A valid daily maxima exists if 75 percent of the hours In a day
report data and there is no systematic lack of data during times of day when
high ozone is most likely. The reader is referred to EPA 1979b and FR (March
19, 1981) for further information.
49
-------
Example 1
TABLE 3-2. EXAMPLE CALCULATION OF THE DESIGN NMOC/NOY RATIO
Date
7/1
6/2
7/3
7/4
8/5
8/8'
7/9
6/10
7/11
9/12
7/15
Max
Oo fPDITl)
.15
.15
.14,
.14
.14
.14
.13' '
.13
.13
.13
.13
NMOC/NOY ratio
Site 1
6.9,
7.5
. 11.3
14.0
5.3
8.7
9.2
6.7
8.4
9.5
12.1
Rank NMOC/NO.. ratio
A
1 ' 5.3
•2 , 6.7
3 ; 6.9
4 7.5 '
5 8.4 V :
6' , , ; 8.7<---Median
7 '912
8 9.5
9 11.3
10 12.1
11 14.0
Example 2
Date
8/1
7/2
6/3
7/5
8/8
9/9
6/10
7/11
6/12
9/15
Max
Oo fppm)
.16
.16
.15
.14
.13
.13
.13
.12
.12
.12
NMOC/NOY ratio
Site 1
5.8
8.4
12.3
8.8
7.8
7.8
8.3
10.1
9.0
8.8
Rank NMOC/NOV ratio
A
1 5.8
2 7,8
3 7.8
4 8.3
5 8.4
6 8.6<---Median
7 .8.8
8 9.0
9 10.1
10 12.3
50
-------
versus observations. Recall from Section 3.3.2 that a VOC emission reduction
estimate should not be used when the model predicted peak ozone disagrees with
the observed peak by more than ± 30 percent. This does not apply to days
which do not have day-specific measurement of NMOC and NOV. However, it has
i , A
been observed that substantial underpredictions of base case, peak ozone may
lead to control estimates which are too low (Jeffries, et al, 1981).
'Conversely, significant .overpredictions of base case, peak ozone may yield
control estimates which are too high. Under some circumstances, this finding
enables one to use control estimates for those days in which base case, peak
ozone is poorly predicted. To illustrate, consider an example 'in. which model
" * •
predictions and control estimates have been made for a site with 3 years of
ozone data (see Table 3-3). Note that for Day 1, peak ozone is substantially
underpredicted, and the control estimate is the highest,of all days. If any
improvements were made to predicted peak ozone, the control estimate for this
day would likely be increased even more. Since the control estimate for Day 1
is already higher than the control target (i.e., 45 percent), any improvements
in model predictions would not affect the selection of the final control
target. Thus, the results from Day 1 can be used, even though the model
significantly underpredicted peak ozone. The converse situation occurs for
Day 5. In this case, any improvements in model predictions would likely
reduce the control estimate for that day, again having no bearing on the
choice of the final control target. If the model prediction is poor, but
neither of the situations described above occur (i.e., overpredictipn and high
control estimate, or underprediction and low control estimate), then it is
51
-------
TABLE 3-3. EXAMPLE ILLUSTRATING EFFECT OF MODEL PREDICTIONS
ON SELECTION OF CONTROL TARGET
Observed Predicted Relative^
Day Ozone, ppm Ozone, ppm Deviation.
Control
Deviation
observed x 100
Rank of
Estimate. % Control Estimates
1
2
3
4
5
0.27
0.22
0.20 '
0.18
. 0.15
0.18
0.20
0.22
0.18
.0.21
-33
- 9 .
+10
0
+40
predicted - observed
55 : ' '
47
51 '•
45**
42
tO \*\Jlt 1st \J
1
1
3
2
4
5
I i
Control Target = fourth highest control estimate (for 3 years of data)
1 I
52
-------
recommended that the site/day be discarded, and replaced by the day with the
next lowest peak ozone concentration.
Recommended Procedure. To obtain the final VOC emission reduction
target, site-specific control requirements must first be determined. In
general, a candidate control estimate is chosen for each site Abased on the
number of years of data and the statistical form of the ozone standard (i.e.,
fourth highest control for 3 years, third highest for 2 years, and the second
highest for 1 year). -Of the candidate site-specific control estimates, the
highest one is selected as the overall VOC emission reduction target.
However, all cases in which predictions and observations disagree by more than
30 percent, should be discarded,.unless: , ,
(1) peak ozone is underpredicted and the VOC reduction estimate is
greater than the candidate site-specific estimate; ' '
I i •
(2) peak ozone is overpredicted and the VOC reduction estimate is lower
than the candidate site-specific estimate. I '
In the event that a day is eliminated, the next lowest peak at the site
in question should be added for modeling. ',
i
3.5.2 Selection of a VOC Reduction Target at Sites Sub.lect to
Overwhelming Transport
I.
In Section 3.1, .we noted that some days initially selected may be
discarded (for modeling purposes), if shown to be subject to "overwhelming
i i I
transport." Nevertheless, a demonstration that the local MSA will attain1 the
I , i' • I ' '
NAAQS still needs to be made. Once a day has been discarded for local
modeling analysis, a- determination needs to be made concerning what MSA/CMSA
is most likely responsible for the excluded observed daily maximum.
i
Procedures for doing this are suggested by Meyer and Baugues ,(1989). These
procedures require a review of all surface National Weather ^ervice (NWS) wind
data within 100 miles of the monitoring site plus any special study surface
wind data collected at properly exposed sites (EPA, 1986). In general, if the
wind data suggest an air parcel located at the monitor at the time of the
observed maxima may have been within an upwind MSA/CMSA between 8 a.m. to
53
-------
noon, that upwind CMSA/MSA may be instrumental in causing the observed daily
maximum. If the discarded day for the locaj MSA has a daily maximum ozone
level higher than that for the fifth highest modeled day for the upwind
CMSA/MSA, the discarded day should be included in the upwind CMSA/MSA's
t
modeling analysis. If a discarded daily maxima is included in an upwind
MSA/CMSA's analysis, it may be ignored in the local attainment demonstration. '
In some cases, however, it may not be possible to identify the upwind
CMSA/MSA most likely responsible for an observed daily maximum ozone
i i1 '
concentration. If this happens, the event is referred to as an "irreducible
exceedance." Presence .of'one or more irred^cibl^ exceedances at a .monitoring ,
• site has. the effect of raising the local VOC control target needed to meet the
NAAQS at that site. Fbr example, the site-specific control requirement at a
site with three valid years of data would become the third (rather than the
fourth) highest control estimate if there were1 ope "irreducible exceedance" at
the site. If the particular site-specific control requirement were the
highest amongst all the' sites assigned io the local MSA, the local MSA's
' I " '
overall VOC reduction target would be slimtlarly affected.
54
-------
• • REFERENCES
Baugues, K. A. (1987), "Support Document for Selection of Default Upper
Air Parameters for EKMA."
'Benkley, C. W. and L. L. Schulman (1979), "Estimating Hourly Mixing Depths
From Historical Meteorological Data," Journal of Applied Meteorology.
• V -IS, pp. 772-780. ' .. . .
Carter, W. P. L., A. M. Winer and J. N. Pitts, Jr. (1982), "Effects of
Kinetic Mechanisms and Hydrocarbon Composition on Oxidant-Precursor
Relationships Predicted by the EKMA Isopleth Technique," Atmospheric
Environment. Volume 16, No. 1, January 1982.
i . i >' . •
Chan, M. W., D. W. Allard and I|. Jonibach (1979), "Ozone and Precursor
Transport Intp an Urban Area-Evaluation of Approaches," EPA-450/4-79-
039, U.S. Environmental Protection Agency, Research Triangle'Park, North
Carolina. •
Decker, C. E., et al (1977), "Ambient Monitoring Aloft of Ozone and
Precursors Near and Downwind of St. Louis," EPA-450/3-77-009,. U.S.
Environmental Protection Agency, Research Triangle Park, North Carolina.
Dimitriades, B. (1977), "An Alternative to the Appendix J Method for
. Calculating Oxidant and NOo Related Control Requirements," International
Conference on Photochemical Oxidant Pollutant and Its Control:
Proceedings. Volume II.. EPA-600/3-77-01~6, U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina.
Drivas, P. J. (1978), "Comparison of Ambient NMHC/NOX Ratios Calculated
From Emission Inventories," EPA-450/3r78-026, U.S. Environmental
Protection Agency, Research Triangle Park, North Carolina.
Eaton, W. C., M. L. Saegar, Ui D. Bach, Jj. E. Sickles, II and C. E Decker
(1979), "Study of the Nature of Ozone, Oxides of Nitrogen, and
Nonmethane Hydrocarbons in Tulsa, Oklahoma - Volume III: Data Analysis
and Interpretation," EPA-450/4,-79-008c, U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina.
EPA (1977), "Uses, Limitations and Technical Basis of Procedures for
Quantifying Relationships Between Photochemical Oxidants and
Precursors," EPA-450/2-77-021a, U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina.
55
-------
EPA (1978), "Ozone Isopleth Plotting Package (OZIPP)," EPA-600/8-78-014b,
U.S. Environmental Protect ion. Agency, Research Triangle Park, North
Carolina.
EPA (1979), "Procedures for the Preparation of Emission Inventories for
Volatile Organic Compounds, Volume II: Emission Inventory Requirements
for Photochemical Air Quality Models," EPA-450/4-79-018, U.S.
Environmental Protection Agency, Research Jri angle Park, North Carolina.
EPA (l979b)y "Guideline for the Interpretation of Ozone Air Quality
Standards," EPA-450/4-79-003, U.S. Environmental Protection Agency
Research Triangle Park, North Carolina. ,
EPA (1980), "Guidance for Collection of Ambient NonMethane Organic Compound
(NMOC) Data for Use in 1982 Ozone SIP Development, and Network Design and
Siting Criteria for the NMOC and NO Monitors," EPA-450/4-80-011, U. S.
Environmental Protection Agency, Research Triangle Park, North Carolina.
EPA (1981), "Addendum 1 to the User's Manual for the Kinetics Model and
Ozone Isopleth Plotting Package (OZIPP)," U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina.
EPA (1982), "PTPLU - A Single Source Gaussian Dispersion Algorithm,"
EPA-600/8-82-014, U.S. Environmental Protection Agency, Research Trianqle
Park, North Carolina.
EPA (1985), "A Cryogenic Preconcentration Direct FID (PDFID) Method for
Measurement of NMOC in Ambient Air," EPA-600/4-85-063, U.S. Environmental
Protection Agency, Research Triangle Park, North Carolina.
.EPA (1986), "On Site Meteorological Program Guidance for Regulatory
Modeling Applications," EPA-450/4-87-013, U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina.
EPA (1988), "A PC Based System for Generating EKMA Input Files",
. EPA-450/4-88-016, U. S. Environmental Protection Agency, Research
Triangle Park, North Carolina.
EPA (1989), "User's Manual for OZIPM-4 - Ozone Isopleth Plotting
With Optional Mechanisms,". EPA-450/4-89-009a, U. S. Environmental
Protection Agency, Research Triangle Park, North Carolina.
Federal Register. March 19, 1986.
Gery, M. W., J. P. Killus, and G. Z. Whitten (1988), EPA Report,
"Development and Testing of the Carbon Bond IV Mechanism for Urban and
Regional Modeling," to be published (1988).
56
-------
Gipson, 6. L., W. P. Freas, R. K. Kelly and E. L. Meyer (1981), "Guideline
for Use of City-Specific EKMA in Preparing Ozone SIPs," EPA-450/4-80-027,
U.S. Environmental Protection Agency, Research Triangle Park, North
Carolina.
Gipson, G. L. (1984), "Guideline for Using the Carbon Bond Mechanism in
City-Specific EKMA," EPA-450/4-84-005, U.S. Environmental Protection
Agency, Research Triangle Park ,^ North Carolina.
Godowitch, J. M.','j. K. Ching and J. F. Clarke (1979), "Dissipation
of the Nocturnal Inversion Layer at an Urban and Rural Site in St.
Louis," Fourth Symposium on Turbulence,- Diffusion, and Air Pollution,
Reno, Nevada. • , ,
Hewson, E. W. (1976), "Meteorological Measurements," Air Pollution.
Volume I. 3rd edition, A. C. Stern, ed., Academic Press, pp.~591-597.
Holzworth, G. C (1972), "Mixing Heights, Wind Speeds, and Potential for
Urban Air Pollution Throughout the Contiguous United States," AP-101,
U.S. Environmental Protection Agency, Research Triangle Park, North
Carolina.
Jeffries, H, E., K. G. Sexton and C. N. Salmt (1981), "Effects of Chemistry
and Meteorology on Ozone Control Calculations Using Simple Trajectory
Models and the EKMA Procedure," EPA-450/4-81-034, U.S. Environmental
Protection Agency, Research Triangle Park, North Carolina.
Jeffries, H. E., K. G. Sexton and J. R. Arnold (1987), "Analysis of
Hydrocarbon Composition from Ground-Level and Aloft Measurements for the .
Carbon Bond and Carter, Atkinson and Lurman Photochemical Mechanisms,"
Cooperative Agreement CR-813107, U. S. Environmental Protection Agency-,
Research Triangle Park, North Carolina.
Killus, J. P. and G. Z. Whitten (1982), "A New Carbon Bond Mechanism for
Air Quality Simulation Modeling," EPA-600/3-82-041, U.S. Environmental
Protection Agency, Research Triangle Park, North Carolina,
Killus, J. P.' and G. Z. Whitten (1983), "Technical Discussions Relating to
the Use of the Carbon Bond Mechanism in OZIPM/EKMA," EPA-450/4-84-009,
U.S. Environmental Protection Agency, Research Triangle Park, North
Carolina.
Ludwig, F. L. (1979), "Assessment of Vertical Distributions of
Photochemical Pollutants and Meteorological Variables in .the Vicinity of
Urban Areas," EPA-450/4-79-017, U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina.
Lurmann, F. W., W. P. L. Cartre and L. A. Coyner (1987), "A surrogate
Species Chemical Reaction Mechanism for Urban-Scale Air Quality
Simulation Models, (Vols. I and II)."
57
-------
Meyer, E. L. and K. A. Baugues (1989), "Consideration of Transported
Ozone and Precursors and Their Use in EKMA," EPA-450/4-89-010, U.S.
Environmental Protection Agency, Research Triangle Park, North Carolina.
Rhoads, R. G. (1981), memorandum to Director, Air and Hazardous Materials
Division, Regions I-X, "Effects of Chemijstry and Meteorology on Ozone
Control Calculations Using Simple Trajectory Models and the EKMA
Procedure," December 3, 1981.
Rhoads,^R. G. (f985), memorandum to Director, Air Management Division,
Regions I-X, "Partictpation in the Summary 1986 NMOC Sampling Program,"
December 23, 1985.
Rhoads, R. G. (1987), memorandum to Darryl Tyler, "Proposed Draft
Federal Register Notice on Use of EKMA in Post-1987 Ozone SIP's, April
30, 1987. .
Rhodes, R. C. and E. G. Evans (1986), "Precision and Accuracy Assessments
for State and Local Air Monitoring Network, 1983," EPA-600/4-86-012, U.S.
Environmental Protection Agency, Research Triangle Park, 'North Carolina.
Schere, Ki L. and K. L. Demerjian (1977), "A Photochemical Box Model for
Urban Air Quality Simulation," Proceedings. 4th Joint Conference on
Sensing of Environmental Pollutants. American Chemical Society.
Seila, R. L. (1986), "GC - Personal Computer System for Determination of
Ambient Air C-2 to C-12 Hydrocarbon Species," Presented at 1986 EPA/APCA
Symposium on Measurement of Toxic Air Pollutants, Raleigh, North
Carolina.
Singh, H. (1980), "Guidance for the Collection and Use of ambient
Hydrocarbon Species Data in Development of Ozone Control Strategies,"
EPA-450/4-80-008, U.S. Environmental Protection Agency, Research Triangle
Park, North Carolina.
Slade, D. H. (1968), "Meteorology and Atomic Energy," NTIS No. TID-24190,
NTIS U.S. Department of Commerce, Springfield, Virginia 22161, ppg.33-39.
Wanta, R. C. and W. P. Lowry, "The Meteorological Setting for Dispersal
. of Air Pollutants," Air Pollution. .Volume I. 3rd Editiion, A. C. Stern,
ed., Academic Press, pp. 337-352.
Whitten, G. Z. and M. W. Gery (1986), "Development of CBM-X Mechanisms for
Urban and Regional AQSM's, "EPA-600/3-86-012, U.S. Environmental
Protection Agency, Research Triangle .Park, North Carolina.
Whitten, G. Z. and H. Hogo (1977), "Mathematical Modeling of Simulated
Photochemical Smog," EPA-600/3-77-011, U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina.
58
-------
Whitten, G. Z. and H. Hogo (1978), "User's Manual for Kinetics Model and
Ozone Isopleth Plotting Package," EPA-600/8-78-014a, U.S. Environmental
Protection Agency, Research Triangle Park, North -Carolina.
Whitten, 6. Z., H. Hogo, M. J. Meldgin, J. P. Killus and
P. J. Bekowies (1979), "Modeljng of Simulated Photochemical Smog With
Kinetic Mechanisms," EPA-600/3-79-011a, U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina.
Whitten, G. Z-., H. Hogo and J, P. Killus (1980), "The Carbon Bond
Mechanism:- A Condensed Kinetic Mechanism for Photochemical Smog,"
Environmental Science and Technology. Volume 14, No. 6.
59
-------
APPENDIX A
LISTING OF CB-4 MECHANISM
A-l
-------
TABLE A-l. CB-4 MECHANISM
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
mmm
N02
0 --
03 -f
Reaction
--> NO + 0
•>o3
• NO --> N02
0 + N02 — > NO
0 •*
0 -t
N02
°3-
03-
01D
01D
°3 +
o3 +
N03
N03
N03
N03
N205
N205
NO +
NO +
• N02 --> N03
• NO — > N02
+ 03 -> N03
-> 0 '
-> 01D
--> 0 .
+ H20 — > 20H
OH --> H02
H02 --> OH
--> 0.89 N02 +'0.89 0 + 0.11 NO
+ NO --> 2 N02
+ N02 --> N02 + NO
+ N02 --> N205
+ H20 --> 2 HN03
--> N03 + N02
NO --> 2 N02
N02 + H20 -> 2 HN02
A- 2
Rate Constant
at 298°K
Cppm"1 min"1)
1.0
4.323 x 10s
26.64
1.375 x 104
2309
2438
0.04731
0.053
1.0
4.246 x 105
3.26
100
3
33.9
4.416 x 104
0.5901
' " 1853
1.9 x 10'6
2.776
1.539 x 10"4
1.6 x 10"11
Activation
Energy
(10
0
- 1175
1370
0
- 687
-. 602
2450
• o "
0
•- 390 .
0
940
580
0
•- 250
1230
- 256
0
1.09 x 104
- 530
0
-------
TABLE A-l. CB-4 MECHANISM (CONTINUED)
„ ,. .
Reaction
22. NO + OH -> HN02
23. HN02 --> NO + OH
24. "- OH + HN02 --> N02
25. HN02 + HN02 --> NO + N02
26: N02 +OH ->HN03
27. OH + HN03 -> N03
28. H02 + NO -> OH +
29. H02 + N02 --> PNA
30. PNA --> H02 -f N02
31. OH + PNA --> N02
32. H02 + H02 -> H202
33. H02 + H02 + H20 -
34. H202 -> 2 OH
35. OH + H202 --> H02
36. OH + CO --> H02
37. FORM + OH --> H02 + CO
38. FORM --> 2 H02'+ CO
39. FORM --> CO
40. FORM + 0 --> OH + H02
41. FORM + N03 --> HN03 +
42. ALD2 + 0 --> C203 + OH
43. ALD2 + OH -> C203
"
) + N02
N02
> «202
h CO
)
i02 + CO
I + H20 + CO
• OH
Rate Constant
at 298°K
(pom min"}
9799
.1975
9770
1.5 x 10"5
1.682 x 104
217.9
1..227 x 104
2025
5.115
6833
4144
.2181
.189
2520
322
1.5 x 104
1.0
1.0 •
237
0.93
636
2.4 x 104
Activation
Energy
f'n
-. 806
0
0
0
- 713
- 1000
- 240
- 749
1.012 x 104
- 380
-. .1150
- 5800
0
187
0
0
0
0
1550
0
986
- 250
A-3
-------
TABLE A-l. CB-4 MECHANISM (CONTINUED)
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
Reaction
ALD2 + N03 — > C203 + HN03
ALD2 --> FORM + 2H02 + CO + X02
C203 + NO --> FORM + N02 + H02 + X02
C203 + N02 — > PAN
PAN --> C203 + N02
C203 + C203 --> 2 FORM + 2X02 + 2H02
C203 + H20 --> 0.79 FORM + 0.79 X0? +
0.79 H02 + 0.79 OH
OH --> FORM + X02 + H02
PAR + OH --> 0.87 X02 + 0.13 OX?N +
0.11 HOp + 0.11 ALD2 -
0.11 PAR + 0.76 ROR
ROR --> 0^96 X02 + 1.1 ALD2 + 0.94 H0? -
2.1 PAR + 0.04 X02N + 0.02 ROR
ROR --> H02
ROR + N02 -->
0 + OLE --> 0.63 ALD2 + 0.38 H0? +
0.28 X02 + 0.3 CO +
0.2 FORM + 0.02 XO?N +
0.22 PAR + 0.2 OH
OH -f OLE --> FORM + ALD2 - PAR + X0?
+ H02
03 + OLE --> 0.5 ALD2 + 0.74 FORM +
Rate Constant
at 298°K
7ppm min )
3^7
1.0
. 1.831 x 104
1.223 x 104
.0222
.3700
9600
21
1203
1.371 x 105
9.544 x 104
2.2 x 104
5920
4.2 x 104
.018
Activation
Energy
f°K)
0
• -0
- 250
- 5500
1.4 x 104
0
0
1710
0
8000
0
0
' 324
• - 504
2105
0.22 X02 + 0.1 OH +
0.33 CO + 0.44 H02 - PAR
59. N03 + OLE --> 0.91 X0? + FORM +
0.09 XOoN + ALD2 +
N02 - PAR
11.35
A-4
-------
TABLE A-1. CB-4 MECHANISM (CONTINUED)
,*
Rate Constant Activation
04..'' at ?98°K i Energy
Reaction fopm"1 min"1) f°K)
60. 0-+ ETH --> FORM + 1.7 HO? + CO + 1080 79?
0.7 X02 + 0.30 H '
61. OH + ETH --> X0? + 1.56 FORM + 1.192 x 104 - 411
0.22 ALD2 + H02
62. 03 + ETH --> FORM + 0.42 CO + 0.13 H02 2.702 x 10'3 2633
63. TOL + OH --> 0.44 HOo + 0.8 X0? + 9150 - 322
0.36 CRES + 0.56 T02
64. T02 + NO --> 0.9 NOo .+ 0.9 HOo + 1 2 x 104 0
0.9 OPEN ' . •*
65. T02 --> CRES + H02 250 0
66. OH + CRES --> 0.4 CRO + 0.6 X0? + 6.1 x 104 0
0.6 H02 + 0.3 OPEN
67. CRES + N03 --> CRO + HN03 3.25 x 104 0
68. -CRO + N02 --> ' - 2 x.104 0
69. OPEN --> C203 + H02 + CO 8.4 0
70. OPEN + OH -> X02 + 2 CO + 2 HOo H- 4.4 x 104 ' 0
C203 + FORM
71. OPEN + 03 --> 0.3 ALD2 + 0.62 C?0, + 0.015 500
0.7 FORM + 0.3 XO^ +
0.69 CO + 0.8 OH +
0.76 H02 + 0.2 MGLY
72. OH + XYL --> 0.7 H09 + 0.5 XOo + 0.2 CRES 3.62 x 104 - 116
+• 0.8 MGCY + 1.1 PAR +•
0.2 T02
73. OH + MGLY --> X02 + C203 ' 2.6 x 104 0
74. MGLY --> C203 + H02 +. CO 8.96 0
A-5
-------
TABLE A-l. CB-4 MECHANISM (CONTINUED)
Reaction
75. 0 + ISOP --> 0.6 H0? + 0.8 ALD2
+ 0.55 OCE +.0.5 X0?
+ 0.5 CO + 0.45 ETH
+ 0.9 PAR
76. OH + ISOP --> X0? + FORM + 0.67 H0? .
+ 0.13 XO?N + ETH +
0.4 MGLY + 0.2 C?0, +
0.2 ALD2
77. ' 03 + ISOP --> FORM + 0.4 ALD2 +
0,55 ETH + 0.2 MGLY +
0.1 PAR + 0.06 CO +
0.44 H02 + 0.1 OH
78. N03 + ISOP --> X02 N
79. X02 + NO --> N02
80. X02N + NO -->
81. X0
X02 --
Rate Constant Activation
at 298°K , Energy
(ppm"1 min"*)
2.7 -x 104
1.42 x 105
,018
82. NR — > NR
470
1.2 x 104
1000
2000
1
( 10
0
0
0
- 1300
0
A-6
-------
APPENDIX B
ESTIMATION OF MIXING HEIGHTS FOR USE IN 02IPM4
B-l
-------
In OZIPM4, the rate of dilution of atmospheric pollutants is governed by
the diurnal change in mixing height. The mixing height is the top of a
surface-based layer of air which is well-mixed due to mechanical and thermal
turbulence-. As described in Section 3.2.2, the input variables required for
OZIPM4 include: the mixing height at 0800 LOT, the maximum mixing height, the
time at which the mixing height begins to rise if it starts to rise after 0800
LCT, and the time at which the mixing height reaches its maximum. The rate of
rise is computed internally by OZIPM4.
Three different procedures exist for determining daily morning and
afternoon mixing heights; The recommended procedure entails the use of
temperature soundings taken routinely by the National Weather Service at
various locations throughout the United States. If more direct measurements
are available (e.g., radiosondes taken in the urban area or sodar data), they
may be used instead of NWS data. If neither of the above two sets of
measurements can be used, then the use of 250 m for the 0800 LCT mixing height
and the climatological mean value for the maximum mixing height is
recommended. The procedures to be followed for each approach are described
below:
B.I RECOMMENDED PROCEDURE USING NWS RADIOSONDES
Temperature soundings are taken by the NWS at sites throughout the
United States. , Soundings are usually taken every 12 hours at 1200 and 0000
Greenwich Mean Time (GMT), corresponding to 0800 and 2000 Eastern Daylight
Time, respectively. Therefore, to estimate daily mixing heights (1), a NWS
site must be selected which is representative of the city of interest,
B-2
/
-------
(2) appropriate sounding data and urban surface data must be obtained, and (3)
these data must be used to compute the morning and
maximum mixing heights. Each of these steps is discussed below.
B.I.I Site Selertinn
In selecting a NWS site as the basis for mixing height estimation,
care should be taken to ensure that the site is meteorologically
representative of the city of interest. .Table B-l contains recommended
sites for a number of cities. Backup sites are listed for those cases in
which radiosonde data may not be available for a given day, or if the
site has significantly different meteorological conditions. Examples of
the latter are the case in which a surface front lies between the sounding
site and the city or the city .is clear but cloudiness or, precipitation
occurs at the sounding site.
B.I.2 Selection of Day.Specific Data
The daily morning mixing height for the model is normally estimated
using the 1200 GMT (0800 EOT) sounding, while the maximum mixing height
is estimated using the 0000 GMT (2000 EOT) sounding. In some cases,
these soundings may not be available or appropriate and alternate approaches
will be necessary. Table B-2 summarizes the order of preference'in
selecting the radiosondes for estimating the daily mixing heights. The
actual data may be obtained from the National Climatic Center (NCC).* '
Please allow about 4 weeks'for NCC to fill an order.
B-3
-------
TABLE B-l. NWS RADIOSONDE STATIONS
CitV
Allentpwn, PA
Baltimore, MD
Boston, MA
Bridgeport, CT
Chicago, IL/IN
Cincinnati, OH/KY
Cleveland, OH
Dayton, OH
Denver, CO
Detroit, MI
:resno, CA
iartford, CT
jouston, tX
Indianapolis, IN
.os Angeles, CA
.ouisville, KY/IN
lilwaukee, WI
iahsville, TN
lew Haven, CT
ew York, NY/NJ
shladelphia, PA/NJ
hoenix, AZ
ittsburg, PA
ortland, OR
rovidence, RO
ichmond, VA
acramento, CA
t. Louis, MO/IL
alt Lake City, UT
an Bernardino, CA
an Diego, CA
an Francisco, CA
cranton, PA
aattle, WA
Dringfield, MA
"enton, NJ
Primary
NYC, NY; At! City, NJ
" Dulles AP, VA
Portland, ME
NYC, NY; At! City, NJ
Peoria, IL
Dayton, OH
Dayton, OH
*Dayton, OH
*Denver, CO
Flint, Mi-
Oakland, CA
Albany, NY
Victoria, TX
Dayton, OH
Vandenberg AFB, CA
Dayton, OH
Green Bay, WI
*Nashville, TN
NYC, NY; Atl Cityj'NJ
*NYC, NY; Atl City, NJ
NYC, NY; Atl City, NJ
Tucson, AZ
* Pittsburgh, PA -
Salem, OR
New York, NY
Dulles AP, VA
Oakland, CA
Salem, IL
*Salt Lake City, NT
San Diego, CA
*San Diego, CA
Oakland, CA
NYC, NY; Atl City, NJ
Quilayute, WA
Albany, NJ
NYC, NY; Atl City, NJ
Backups)
Climatological
Mixing Heights
(m AGL)
MAX
Albany, NY; Dull as AP,, VA
Wallops Is., VA; Atl City,
Albany, NY; Chatham, MA
Albany, NY
Green Bay, WI
Huntington, WV
Buffalo, NY
Huntington, WV
Grand Junction, CO
Dayton, OH
Vandenberg AFB, CA
NYC, NY; Atl City, NJ
Lake Charles, LA
Peoria, IL; Salem, IL .
San Diego, CA
Nashville, TN
Peoria, IL
Jackson, AL
Albany, NY
Albany, NY
Dulles AP, VA
Winslow, AZ
Dayton, OH; Dulles AP, A
Medford, OR; Quilayute, WA
Albany, NY; Chatham, MA;
Atl City, NJ
Greensboro, NC;
Wallops Is., VA
Vandenberg AFB, CA
Peoria, IL; Monette, MO
Grand Junction, CO
Vandenberg AFB, CA .
Vandenberg AFB, CA
Vandenberg AFB, CA
Albany, NY; Atl City, NJ
Dulles AP, VA
Salem, OR
NYC, NY; Atl City, NJ
Dulles AP, VA |
1825
NJ 1825
1375
1500
1575
1650
1650
1661
3358
1700
2000
1500
1525
1600
603
1700
1575
1845
1450
1512
1700
3250
1794
1575
1350
1725
1600
1625
3673
1200
564-
625
1850
1398
1600
1700
B-4
-------
TABLE B-l (CONTINUED)
Climatological
Mixing Heights
(m A6L)
^ Erimary Backup^ MAy
«KT^3xsrwp>« »?SC\A ..' ;««
il.ington.-DE Dulles A^VA; Kallops Is.'; VA; - , 1884
At I City, NJ New York NY ,-,**'
)rcester, MA Albany, NY Portland; JJE; 17°°
>ungstown, OH Pittsburg, PA Bu^fa™' K; 15°°
. Dayton, OH • ' 1700
sr s a«s
TE: The NYC, NY radiosonde station was replaced by AtUntic City, NJ on September 2, 1980,
B-5
-------
TABLE B-2. PREFERENTIAL ORDER OF DATA SELECTION
Horning Mixing Height Estimate
1. 1200 GMT Sounding at Primary Site
2. 0600 GMT Sounding at Primary Site* - ,
3. 1200 GMT Sounding at Backup Site
4. 0600 GMT Sounding at Backup Site*
Maximum Mixing Height Estimate
1. 0000 GMT Sounding at Primary Site
2. 1800 GMT Sounding at Primary Site*
3. 1200 GMT Sounding at Primary Site
4. 0000 GMT Sounding at Backup Site
5. 1800 GMT Sounding at Backup. Site*
6. 1200 GMT Sounding at Backup Site
*Soundings are not normally taken at these times, but may be available in
some instances.
B-6
-------
In addition to the sounding data, surface temperature and pressure data
are also needed for each day modeled. The urban surface temperature at 0800
LCT (or the average temperature between 0800-0900 LOT) and the maximum
temperature occurring prior to 1800 LCT are needed to estimate the morning and
maximum-mixing height, respectively.. The surface temperature data should be,
measured to the nearest 0.1° C-at a well ventilated site (EPA, 1986). The
site should be located near the center of the urban area. Surface atmospheric
pressure measurements are needed at the same time and location of the urban
surface temperature measurements, if at all possible. If these measurement?
are not available, a local NWS or-Federal Aviation Administration weather
reporting station's barometer reading may be used.
If the elevation of the pressure reading and the urban temperature site
are different, an adjustment should be made to the pressure measurement using
equation (3) ' .
Psfc = pobs'+ [.ilmb/m x (Zobs - Zsfc)] (3)
where
' zobs = the elevation, in meters above sea level (mASL),
of the pressure measurement
zsfc = tne elevation (mASL) of the urban temperature
measurement
pobs = the Pressure, in millibars, at ZQbs
Psfc = the pressure, in millibars, at the urban
temperature site
NOTE: Zobs w111 be eclual to-zero meters ASL when a pressure reduced at
sea level is used.
B-7
-------
The value of Pgfc from equation (3) is an approximate value and can be rounded
to the nearest whole millibar.
B.I.3 Mixing Height Estimation
The procedures for estimating the 0800 LCT mixing height and the maximum
mixing height are outlined in Table Br3. The procedures in Table B-3 are
designed for use with the worksheet displayed in Table B-4. Figure B-l
contains a flow diagram of the process. The procedures use the mandatory and
significant pressure levels reported for each sounding (Table B-5). The steps
lead to determination of the height at which the.adiabatic lapse rate
(extended from the surface temperature and pressure) intersects the vertical
temperature profile). -(For background information, the reader is referred to
Wanta and Lowry, 1976; Hewson,•1976; and Slade, 1968). An example problem is
presented in Section B.4. •
In some instances, the mixing heights estimated by this procedure may
not be representative. If the 0800 LCT morning mixing height is estimated to
be less than 250 meters, then a value of 250 meters should be used. This
assumed minimum value for the 0800 LCT mixing height accounts for the effects
of mixing due to mechanical turbulence caused by increased surface roughness
in the urban area (Godowitch, et al, 1979; and Bentley and Schulman, 1979).
Similarly, if the city's maximum mixing height is greater than twice the
climatological maximum value (e.g., see Table B-l), the surface temperature
and pressure used and the choice of sounding site should be checked for
representativeness using the guidelines in B.I.I and B.I.2 above. If no
backup data are available, twice the climatological value should be used as
•the maximum. Also, a maximum mixing height less than or equal to the morning
B-8
-------
TABLE B-3. PROCEDURES FOR ESTIMATING MIXING HEIGHTS
Step 1 -- For reference, the information at the top of Table B-4 should
rfri,1,? J Ve;i!''n2Se,V?1ty'*etc'}- If the morn™9 mixing height is to be
calculated the 0800 LCT surface data are used. If the maximum mixing height
is to be calculated, the data corresponding to the time of maximum temperature
(i.e., between 800-1800 LCT.) are used. In the row Tabled URBAN SURFACE DATA
enter the following information: (1) the elevation of the urban temperature'
•site in meters above sea level; (2) the surface pressure in millibars (this
value is Psfc); and (3) the surface temperature in degrees Celsius (°C).
Convert the surface temperature in column four to degrees Kelvin (°K) bv
adding 273.2, and enter the result in column five. This value is T$fc( K).
Use Equation 1 below and the values just entered to calculate the
potential temperature at the surface.(Of in Y, to the nearest O.l'K) and
enter this value under column six "( K)?* .
. ' -0.286
sfc (in K) = Tsfc (in °K) £sfc...(1n mb) (1)
~™0 mb
i iSt!E 2 Iu Using the temperature sounding data, find the highest pressure
level other than the founding's surface value that is less than the. pressure
at the urban surface. From this pressure level on the sounding, enter the
5?J?« (1£ i,lstndi' Pressure> and temperature (in °C) into the row marked
(Z) on Table B-4.
**
For example, if the urban surface pressure is 985 mb, and the soundinq
pressures are: 1005, 1000, 963, 850 mb, etc., 963 mb is the "highest
pressure level that is less than the pressure at the urban surface." 850
mb is the "next lowest pressure level" needed in Step 4.
B-9
-------
TABLE B-3 (CONTINUED)
Step 3 -- Convert the temperature at this level to the Kelvin scale and
enter in column 5. Compute the potential temperature (D) to the nearest 0.1°K
using the pressure (P, in mb) and temperature (TD in °K|) at this level in
Equation 2 below: v
-0.286
„ (in K) = T_ (in °K) P Mn mb) (2)
v , H 1000 mb
Enter the value of n found from Equation (2) into the same row under the
column labeled ir(K)7"
Step 4 -- If the potential temperature of the last row that
was entered is greater than the potential temperature sfc, and this is the
first level above the surface, then 250 meters should 5e used as the mixing
height (if given), pressure and temperature of the next lowest pressure level
found on the sounding into the next row of Table B-4 and return to Step 3.
Step 5 -- The mixing height is between the last two levels entered into
Table B-4. If height values are given for both of these levels, the elevation
of the mixing height can be found using Step 6. If one of the levels does not
have a height value, use linear interpolation to find the pressure value for
the potential temperature value sfc + 0.1°K. Enter this pressure into the row
marked "MIXING HEIGHT" at the bottom of Table B-4 under the column "PRESSURE
in mb." Proceed to Step 7.
Step 6 — From the two levels where height is given on the sounding •
surrounding the mixing height level, use linear interpolation to find the
height On meters ASL) at the value _fc + 0.1°K (i.e., the potential
temperature at the mixing height). Enter the value found by linear
interpolation into the row labeled "MIXING HEIGHT" under the column "HEIGHT
(mASL)" and proceed to'Step 8.,
Step 7 -- Use linear interpolation to find the height above sea level of
the mixing height using the pressure at the mixing height (found in Step 5)
and the pressure .levels on the sounding above and below the mixing height
pressure that have both pressure and height values. Enter the height value
found into the row "MIXING HEIGHT" under the column marked "HEIGHT, (mASL)"
and proceed to Step 8.
Step 8 -- Subtract the elevation of the urban site (mASL) from the height
(mASL) of the mixing height, the result is the height of the mixing height in
meters above the surface of the city (mAGL). Enter this value into Table B-4.
NOTE: Despite the fact that pressure and height, and potential
temperature and height, are not linearly related, linear
interpolation does not produce significant errors over the limited
ranges used above.
B-10
-------
Date:
TABLE B-4. WORKSHEET FOR COMPUTING MIXING HEIGHTS
Time of Mixing Height for Input Into Model:
Sounding Method:
Time of Sounding: LOT, Surface Elevation:
Location of Sounding: , ' ' • -
LOCATION OF URBAN SURFACE AREA (IF DIFFERENT THAN ABOVE)
1 ' . 2 3 * 56
mASL
LEVEL ' r .
Urban (
Surface r
Data ' (1)
(2)
HEIGHT
(mASL)
i
•
PRESSURE
' (mb)
i i .
TEMP.
Co
TEMP.
(*K)
•'
•• .
(°K)
REMARKS
<"sfc
MIXING
HEIGHT
(Lfr + 0.1 K
C(°K)
PRESSURE
HEIGHT
(mASL)
Height
(mAGL)
HEIGHT USED IN
MODEL (mAGL)
B-ll
-------
- ENTER URBAN SURFACE DATA
INTO TABLE A-«f,
- CONVERT SURFACE TEMPERATURE
TO •*,
-FIND 6 , USING EQUATION (1).
(5)- on TABLEAU LIST DATA TROM
W SOUNDING FROM THE FIRST PRESSURE
LEVEL ABOVE THE URBAN SURFACE LEVEL.
; - CONVERT TEMPERATURE TO *K AND
USE EQUATION .(2) TO COMPUTE PO-
TENTIAL TEMPERATURE (6p) FOR
THE LEVEL JUST ENTERED.
- ENTER NEXT PRESSURE
LEVEL INTO TABLE A -M-.
- fll I.ISG HV.I(.HT IS
AT e
sfe
- ARE THE
"HEIGHT VALUES GIVEN
KIR THE LAST TWO ROWS
ENTERED INTO
JTABLE A-4"
NO
TES
- USE LTREAK INTERPOLATION
TO FIND THE PRESSURE AT
- FIND THE HEIGHT (aASL) OF
OF THE MTITHG HEIGHT BY
LINEAR INTERPOLATION FROM
TEE PRESSURE AT THE MIXING
HEIGHT.
D- USE LUilEAS. INTER-.
POLATIOH TO FIND THE
HEIGHT' (BIASD AT
a
ymixing height.
V
fi)- FIND TEE HEIGHT- ABOVE THE
URBAN AREA (mAGL) OF THE MIS-
ING HEIGHT -(THIS GIVES THE
ANSWER.} .
Figure,B-l. Flow Chart for TableB-3. Numoers in circles are step numbers in Table pi-3.
B-12
-------
TABLE B-5. SURFACE AND SOUNDING DATA
Hour Starting at. LCT Temperature °C . Pressure, mb
1
,
-
1200
Pressure
(mb)
S 1015
M 1000
S 967
M 850
'S 827
S 817
M 700
S 680
S 661
S 608
M 500
S 491
S 453
S 438
M 400
S 388
S 349
S 324
M 300
S 267
M 250
M 200
M 150
S 148
8
9
10
11
12
13
14 ,
15
16
17
18
GMT Sound ina
Height
(m,ASL)
1 . 8* i
139
—
1550
_'__
___
3168
—
—
—
5860
> —
—
—
7560
—
—
—
"9640
—
10890
12370
14190
—
Temp.
( °Q)
. 23.0
23.0
24.4
16.2
14.2
13.6
4.6
5.6
5.6
0.4
- 8.3
- 9.3
-12.7
-13.9
-18. 7
-20.1
-26.3
-29.7
-33.7
-39.5
-47.7
-51.7
-60.9
-61.5
23.2
23.9
25.8
27.3
28.7
29.3
3Q.1 ,
30.4
30.8
31.4
31.2
Sounding Data
1010.3
1010.7
1Q10.8
1010.6
1010.3
1010.0
1009.
1009.
1008.8
1008.6
1008.5
0000 GMT Sounding
Pressure
(mb)
S 1012
M 1000
M 850
S 831
S 791
S 778
S 760
M 700
S 628
S 560
M 500
M 400
S 371 -
M 300
S 265
M 250
S 205
M 200
M 150
S 127
S 120
M 100
M 70
M 50
M 30
M 20
S 15
Height
(m ASL)
8*
114
1537
—
—
—
___
3164
• , —
—
5860
7560
—
9650
--.-
10900 "
—
12370
14190
—
—
16690
18900
21040
24350
27030
—
Temp
( °Q)
31.0
30.6
16.4
15.4
13.2
11.8
11.2
7.0
1.6
- 1.5
- 7.3
-18.9
-21.7
.-33.1
-39.9
-42.9
-52.9
-53.3
-61.1
-64.9
-61.7
-63.3
-58.5
-54.5
-49.9
-44.7
-42.1
NOTE: M = Mandatory Levels and S = Significant Levels
If NWS data are used, both the mandatory and significant levels
are needed.
The 0000 GMT sounding is the following day in GMT.
*The lowest level of the sounding should not be used in the mixing
height calculations.
B-13
-------
mixing height, or less than one-third the climatological maximum mixing height
value is suspect. Using data from a backup site may provide a more realistic
value. However, if the low afternoon mixing height is due to the existence of
a surfacebased stable layer, an adjustment to. the procedures putlined in.Table
B-3 can be employed. Replace the "Urban Surface Data" with the following data
from the sounding site: (1) the maximum temperature, (2) the estimated or
observed surface pressure at the time of maximum temperature, and (3) the
height of the sounding surface level. Then compute the mixing height
according to'the procedure in Table B-3. If this problem occurs on a majority
of modeling days, then an alternative, more representative site should be used
for all the modeling days.
B.2 USE OF ALTERNATE DATA
Other, more direct measurements of mixing height may be used to increase
the representativeness of the estimated values. These methods include direct
urban temperature sounding and sodar data. The measurements should be taken
over the urban area near the center of the city at 0800 LCT, and close to the
time of the climatological maximum surface temperature. It is not recommended
that these measurements be taken specifically for the OZIPM4/EKMA techniques;
however, they may be employed if available. Examples are discussed below.
1. Local Urban Radiosonde --'The methods described in Section,8-1.3 can
be used to find the mixing height from radiosondes taken within the urban area
as opposed to NWS sites, the radiosonde surface temperature and pressure
should be used in place of the URBAN SURFACE DATA.
B-14
-------
2- Urban Helicopter Soundings -- Similarly, vertical temperature
profiles obtained from helicopter soundings can be used in place of the NWS
soundings. The urban site surface temperature and pressure should be used as
the URBAN SURFACE DATA. • .
3. Sodar -- (also known as Acoustic Radar) the mixing height found by
sodar (in mAGL) can be used directly in the model.
NOTE: Regardless of the procedure applied, the limitations concerning the
morning and maximum mixing heights that were described in Section B.I.3 should
be observed.
B.3 USE OF CLIMATOLOGICAL MEANS
If radiosonde data are not available, 250 m should be used for the 0800
LCT mixing height and the city-specific climatological mean value may be used
for the maximum,mixing height. Table B-l lists representative values for
several cities, and Holzworth, 1972, contains information for the contiguous
United States. If Holzworth is used, values for summer, nonprecipitation days
should be used. The appropriate starting and ending tiems of the mixing
height rise in the model are 0800 LCT and the time of the maximum temperature.
If the latter is unknown, 1400 1ST (1500 LOT) may be assumed.
B.4 EXAMPLE PROBLEM
To illustrate the procedure described in Section B.I.3, an example
problem is included for reference. Table B-5 shows relevant data typically
available. Note that both the 1200 GMT and the 0000 GMT soundings are used in
B-1.5
-------
the calculations, the former for the morning mixing height and the latter for
the maximum mixing height. Table B-6 shows the individual computational steps
for the morning mixing height calculation, while Table B-7 shows the same for
the maximum mixing height.
B-16
-------
TABLE B-6. MORNING MIXING HEIGHT DETERMINATION
Example from Table B-5:
08 LCT temperature = 23.2°C
Maximum temperature after 08 LCT = 3.4°C at 17 LCT
08 LCT pressure = 1010.3 mb - - -
Pressure at time of maximum temperature (1700 LCT)
morning mixing height = 0800 LCT
Time of maximum mixing height = 1700 LCT
1008.6 mb%T1nfe of
Problem:
Find the 0800 LCT mixing height using data from the sounding shown in
Table B-5 (i.e., the 1200 GMT sounding). A worksheet is shown as Table
B-6A. The elevation of the urban surface site is 62 mASL.
Solution:
STEP 1
Enter 62., 1010.3, and 23.2 into row (1) of Table B-6A
(URBAN SURFACE DATA)
Temp ( C) = 23.2
Converted to °K = 23.2 + 273.2 - 296.4°K
Enter 296.4 into row (1) of Table B-6A under "TEMP(°K)" Using
Equation (l).on the Urban Surface Data:
-0.286
1010.3 mb
1000 mb
Osfc = 296.4°K
Osfc = 295.5'K
STEP 2 - Enter 139., 1000., and 23.0 into row (2) of Table B-6A
B-17
-------
STEP 3 - 23.0 + 273.2 =296.2 K
Using Equation (2):
Op = 296.2 K
-0.286
1000 mb
1000 mb
Op = 296.2°K (enter this value into Table B-6A)
i
STEP 4 - Op (29612eK) is greater than Osfc (295.5°K). ,
Since 0_ is from the first level above the surface, the 250 m
defaulrvalue should be used for the 0800 LCT mixing height.
B-18
-------
TABLE B-6A. EXAMPLE (HYPOTHETICAL DATA)
Date: Date of Modeling Time of Mixing Height for Input Into Model: 0800 EOT
City: City to be Modeled Sounding Method: NWS, Urban Radiosonde or Helicopter
Time of Sounding: 0800 LCT. Surface Elevation: (of sounding) mASL
Location of Sounding: Name'of Sounding. Site "
LOCATION OF URBAN SURFACE DATE (IF DIFFERENT THAN ABOVE) - Street Address,
Building or Park, etc.
1 2 3 4 R' « • 7
LEVEL ,
Urban
Surface
Data (1)
(2)
HEIGHT
(mASL)
62
139
PRESSURE
(mb)
1010.3
1000.
TEMP.
(•c)
23.2
23.0
TEMP.
(°K)
296.4
296.2
(°K)
295.5
296.2
REMARKS
<"sfc
level is
higher than
sfc +0-1 °K
MIXING
HEIGHT
°sfc,t 0.1'K
( K)
295.6
PRESSURE
(mb)
. HEIGHT
(mASL)
0
Height
(mAGL)
0
HEIGHT USED IN
MODEL (mAGL)
250
B-19
-------
TABLE B-7. MAXIMUM MIXING HEIGHT DETERMINATION
Example from Table B-5:
08 LCT temperature = 23.2°C
Maximum temperature = 31.4 C at 17 LCT
08 LCT pressure = 1010.3 mb
17 LCT pressure = 10Q8..6 mb • " , -
Time of morning mixing height = 0800 LCT
Time of maximum mixing height = 1700 LCT
Problem:
Find the maximum afternoon mixing height using data from the sounding
shown in Table B-5 (i.e., the 0000 GMT sounding). A worksheet is shown
as Table B-7A. The elevation of the urban site is 62 mASL.
Solution:
STEP 1 .
Enter 62, 1008.6, and 31.4 into row (1) of Table B-7A (Urban
Surface Data)
Temp (°C) = 31.4°C
Converted to °K - 31.4 + 273.2 - 304.6°K
Enter 304.6°K into row (1) of Table B-7A under "Temp ( K)"
Using Equation (1) on the Urban Surface Data
-0.286
sfc
= 304.6 K
1000 mb
1000 mb
sfc
303.9
STEP 2
Enter 114., 1000., and 30.6 into Table B-7A.
B-20
-------
TABLE'B-7 (CONTINUED)
STEP 3
30.6°C + 273.2 = 303.8°K
Using Equation (2):
n = 303".8°K 1000. mb
1000 mb
-0.286
p = 303.4°K
STEP 4 - 303.8°K is less than 303.9°K
Therefore, enter 1537., 850., and 16.4 in Table B-7A and
return to STEP 3.
STEP 3 - 16.4'C =273.2 = 289.6°K
Using Equation (2)
289.6°K
= 303.4°K
-0.286
850 mb
1000 mb
STEP 4 - 303.4°K is less than 303.9°K
Therefore, enter 831. and 15.4 into Table B-7A (note
that there is no height value for this pressure level)
and return to STEP 3.
STEP 3 - 15.4°C = 273.2 = 288.6°K
Using Equation (2):
-0.286
= 296.2°K
304.3CK
831 mb
1000 mb
B-21
-------
TABLE B-7 (CONTINUED)
STEP 4 - 304.3°K is greater than 303.9°K
STEP 5 ...
sfc
+ 0.1 K = 303.9 K = 0.1 K = 304.0 K
Using linear interpolation from temperature (0) to
pressure since a highest value is not xjiven for the 831
mb pressure level - -" . .
0 ( K)
303.4
304.0
304.3
Pressure (mb)
850
P mixing height
831
pmixing height = 831 mb - (850 mb - 831 mbU304.0°K - 304.3°K)
304.3 K - 303.4 K
- 831 - (19 mbU-0.3'tO
0.9 K
= 837.3 mb
The pressure at the mixing height (rounded to the nearest
•whole millibar) is 837 mb.
STEP 7
Use linear interpolation to find the height above sea
level of the mixing height. Enter 3164. and 700. into
Table B-7A.
Pressure (mb)Height (mASL)
8501537.
837Z mixing height
7003164.
Z mixing height = 1537 m + (3164 m - 1537 mH837 mb - 850 mb)
700 mb - 850 mb
= 1537 m + (1627 mU-13 mb)
-150 mb
Z mixing height = 1678 m
B-22
-------
TABLE B-7 (CONTINUED)
STEP 8
1678 mASL - height of mixing height
- 62 mASL - elevation of urban surface site
1616 mAGL = mixing height in meters above the urban
area
1616 m is the height of the maximum mixing height to
be used in the model with the time of 1700 LCT
B-23
-------
TABLE B-7A EXAMPLE (HYPOTHETICAL DATA)
'Date:
City:
Time of Sounding:
Time of Mixing Height for Input Into Model: 1700 EOT
Sounding Method: NWS
LCT. Surface Elevation: 8. mASL
Location of Sounding: Name of Sounding Site
LOCATION OF URBAN SURFACE AREA (IF DIFFERENT THAN ABOVE)
1 2 3 45
LEVEL
Urban
Surface
Data (1)
(2)
HEIGHT
(mASL)
62
114
1537
—
3164
PRESSURE
(mb)
1008.6
1000.
850.
831.
700
TEMP.
(°C)
31.4 .
30.6
16.4
15.4
—
TEMP.
(°K)
304.6
303.8
289.6
288.6
—
(°K)
303.9.
303.8
303.4
304.3
—
1
REMARKS,
<"-sfc
Mixing height
is between
- these two
levels at
304 0°K
Needed to
•provide upper
height value
•for interpola-
tion
MIXING
HEIGHT
O-f. + OU'K-
SfC(°K)
304.0
PRESSURE
(mb)
837.
HEIGHT
(mASL)
1678.
Height
(mAGL)
1616.
HEIGHT USED IN
MODEL (mAGL)
•
1616.
B-24
-------
APPENDIX C
COMPUTATION OF CARBON BOND FRACTIONS FROM GC DATA
In this appendix, the computation of carbon-fractions from results of
gas chromatographic (GC) analysis is discussed. As noted earlier, GC analysis
actually measures the concentrations of individual organic species, which must
then be grouped according to the CB-4 organic reactivity classes. In order to
keep the computations relatively simple for illustrative purposes,
hypothetical examples are discussed. For more detailed discussion, the reader
is referred to EPA, 1989.
At the heart of the computational procedure is the definition of how
individual species should be categorized according to carbon bond type.
Definitions for numerous individual species and are listed in Table C-l.
These definitions, or species profiles, give the number of bond types found in
each CB-4 category. Using this information, along with the carbon numbers
shown in Table 2-1, it is possible to compute concentrations of individual
carbon bond classes, and then determine percentages of carbon in each class.
These computations will be illustrated by the examples in Tables C-2 and C-3,
respectively.
i
Table C-2 presents example calculations for a hypothetical example. The
individual species that might be detected by GC analysis are shown in the left
hand column, and their associated concentrations, in units of ppbC and ppb, •
are shown in the next two columns. The remaining columns are associated with
the carbon bond computations.
C-l
-------
TABLE C-l SPECIES PROFILES-BY BOND GROUPS FOR CB-4
COMPOUND HAKE PAR' OLE ETH TOL XYL FORH ALD2 I50P NR
1,1,1-TRICHLORQETHAHE 2.0
1,1,2-IRICKLQROETHftHE 2,0
1,2,3,4-TETRAHETHYLBEHZENE 2.0 . 1.0
1,2,3,5-TETRAHEIHYLBEHZEHE 2.0 1.0
1,2,3-TRIMETHYtBEHZENE 1.0 1.0
1,2,4,5-TETRAHETHYLBEHZENE Z.fl 1.0
1,2,4-TRIHETHYLBEHZENE 1.0 '. 1.0
1,2-DIETHYLBEHZENE 2.0 1.0
V-HIKETHYL-3-ETHYIBEHZEHE 2.0- ' ' ' 1.0
1,2-BIHETHYM-ETHYLBENZENE 2.0 ' 1.0
1,3,5-TRIHETHYLBEHZENE 1.0 . 1.0
1,3-BOTAHEHE 2.0
1,3-BIETHYlBENZENE 2.0 , 1.0
1,4-BUTAHEBIOL 4.0
1,4-BIETHYLBEHZENE 2.0 1.0
1-BUTENE 2.0 1.0
1-BBTYHE 3.0 1.0
1-CHlflROBOTANE 4.0
HECEHE 8.0 ' 1.0
l-ETHOXY-2-PROPAHOL 3.0 1.0
1-HEPIEHE , 5.0 1.0
1-HEXEHE ' 4.0 1.0
1-KETHYtCYCLOHEXENE 5.0 1.0
HETHYL-2-ETHYLBEHZEHE 1.0 1.0
l-HETHYL-3-EIHYLBENZENE 1.0 1.0
l-KETHYL-3-ISOPROPYLBESJZEHE 2.0 1.0
HETHYL-3-H-PROPYLBENZEHE 2.0 ' 1.0
l-KETHYL-4-ISOPROPYLBENZENE 2.0 ' 1.0
1-HOHEHE 7.0 1.0
1-OCTENE 6.0 1.0
1-PEHTEHE 3.0 1.0
HHBECEHE ?.0 1.0
2,2,3-TRIHETHYLBOTANE ' 7.0
2,2,3-TRIKETHYLPE«TAHE 8.0
2,2,4-TRIIiETHYLPEHTftNE 8.0
2,2,5-TRIHETHYLHEXMlE ?.0
2,2-BICHLOROKITROANILINE 1.0 5.0
2,2-BIHETHYLBOTAHE 6.0
2,2-JIHEIHYLHEXAHE 8.0
2,2-IIHETHYLPROPANE 5.0
2,3,3-TRIKETliYLPEHTAHE 8.0
2,3j3-TRI«EIHYl-l-BUTEKE 6.0 1.0
2,3,4-TRIHET«YLPEHTftKE 8.0
2,3,5-IRIIiETHYLHEXANE 9.0
2,3-BI«ETHYLBUTAW£ ' 6.0
2,3-DIHETHYLHEPTftNE 9.0
2,3-BIHETHYLHEXANE 8.0
2,3-!I«EIHYLOCTANE lO.fl
2,3-H«ETHYLPEKIANE 7.0
2,3-BIHETHYL-l-BDTENE 5.0 ' 1.0
2,4,4-TRIHETHYL-l-PEHTEHE 7.0 1.0
2,4,5-TRIHETHYLHEPTAHE 10.0
2,4-DIMETHYLHEPTAHE 9.0
2,4-JIKETKTLHEXANE fi.O
C-2
-------
iwmm m GLE ETH TDL m FORM AIM ISOP HR
2,4-JIUEIHYLflCTANE 10.0
2,4-BltSETHYlPENTANE 7.0
2,5-BIHETHriHEPTAHE 9.0
2,5-DIHETHYLHEXANE 8.0 • •
2,6-JIKETHYLflCTANE 10,0
2,H>MTHYLSTYRENE 1,0 1.0
2-BUIYLTETRAHYIROFORftN 6.0 1.0
2-BUTYNE 3.0 1.0
2-ETHYLHEXANOL 8.0 -
2-ETHYL-HUTENE 5.0 1.0
2-Eim-l-HEXAHflL '8.0
2-FBRFHRAL 1.0 1,0 1.0
2-HEXEHE • 2.0 2.0
2-HETHYOEMNE 11.0
2-HETHYLHEPT(\NE 8.0
2-HETHYLHEXftWE 7.0
2-HETHYL8CT/SHE 9.0
2-lfETHYLPEHIAHE 4,0
2-HETHYLPROPftHE 4.0
2-KETHYLPROFENE 2.0 1.0
2-liETHYLPROPENE 2.0 1.0
2-«ETKYL-l,3-B«TASIENE 1.5
2-RETHYL-l-BUIEHE 4.0 ' 1.0
2-BETHYL-l-PEHTENE ' 5.0 1.0
2-HETHYL-2-BOTENE 3.0 1.8
2-KETHYL-2-PEHTEHE 4.0 1.0
2-8ETHYL-3-HEXMONE 7.0
2-(2-BOTOXYETHQXYHTMffil 4.0 2,0
3,3-IIHETHYLPEKTftHE 7.0
35H!lfiE!(lYLOCTAHE 10.0
3,5,5-TRIHETHYLHEXftHE 9.0
3,5-5IHETHYiHEPT(\HE 9.0
3-HEPTENE . 3.0 2.0
3-NETHYlHEPTftNE 8.0
3-«EIHYLHEXftNE - 7.0
3-»ETHTlO£TflKE 9.0
3-KETHYLPEHTftNE ' 6.0
3-BETHYL-i-IOTEffi 3.0 1.0
3-!!EIHYL-l-PEHIENE 4.0 1.0
3-(fETHYL-CIS-2-PEHTENE ' 4.0 1.0
3-HEIHYL-TRAN8-2-PENTEHE 4.0 1.0
3-(CHLORQHETHYL)-S)EPTftHE 8,0
4,HETHYLENE SlftNILINE 1.0 u 5 0
4-lfETHYLANILINE i.o
4-HETHYLHEPTANE 8.0
4-HETHYLHDHME JO.O
4-HETHYLOCTAHE 9.0
4-«ETHYL-l-PEHTESE 4.0 1.0
4-BETHYL-CIS-2-PENTENE 2.0 z.O
4-HETHYL-TRflHS-2-PENTEHE 2.0 2*0
4-PHENYL-l-BUTEHE 1.0 1.0 1.0
ACEHftPHTHENE 1.0 i.o 3 fl
ACEHAPHTHYLEKE l.fl }.o 2'0
ACETALIEHYSE " 1 0 *"
C-3 i
-------
COMPOUND HADE PAR OLE ETH TOL XYL FORK ALD2 ISOP NR
ACETIC ACID 1.0 1.0
ACETIC ANHYDRIDE 2.0 2.0
ACETONE 3.0
ACETYLENE 1.0 1.0
ACROLEIH (PROPEHAL) 1.0 1.0
ACRYLIC ACID 1.0 , 1.0
ACRYLONITRILE 1.0 1.0
ADIPICACID 4.0 . 2.0'
ALIPHATICS (per carbon) • 1.0
ALKENE KETONE 2.0 1.0
AHINOAHTHRA8UINQNE 2.0' 12.0
ANILINE 1.0 5.0
AHTHANTHRENE 1.0 1.0 1.0 7.0
ANTHRACENE 1.0 1.0 ' 5.0
AHTHRAfiUIHOHE 2.0 12.0
ft-PINEHE 8.0 0.5 1.5
BEHZALIEHYBE 1.0 5.0
BENZENE 1.0 5.0
1EHZOIC ACIB 1,0 4.0
BENZOPYRENES 1.0 1.0 1.0 3.0
1EHZOTHIAZOLE 2.0 5.9
BEHZ8(a)AKTHRACEH£ . 1.0 1.0 . 3.0
lEHZBUJPYRENE 1.0 1.0 1.0 3.0
BENZO(b)FLllflRAHTHEK 1.0 1.0 5.0
lENZfl(c)PHENA«THRENE 1.0 1.0. 3.0
BEHZOUJPYRENE 1.0 1.0 1.0 3.0
BEHZO(g,h,i)FLUORAHTHENE 1.0 1.0 3.0
UHZO(9,l,i)PERYLENE 1.0 1.0 1.0 . 4.0
BEHZO(k)FLUORANTHEHE 1.0 I.fl ' . 5.0.
BENZYLCHLORIEE 1.0'
BIPHENYL i.O . 5.0
BIPHENYLOL 1.0 5.0
BROMOD1HITROANILINE 1.0 5.0
BROHODIHITROBENZEKE 1.0 ' 5.0
BUTEHE ' . 2.0 1.0
BUTOXYBUTEHE ' 4.0 1.0 1.0
BUTOXYETHOXYETHANOL 4.0 2.0
BUTOXYETHOXYETHAHOL ACETATE 5.0 2.0 1.0
BUTYL CARBITOL 4.0 2.0
BUTYL CELLOSOLVE 4.0 1.0
BOIYLACRYLATE 4.0 1.0 1.0
BliTYLBEHZENE 3.0 1.0
ByTYLBENZOATE 5.3 4.0"
BOTYLBENZYLPHTHALATE 5.0 1.0 7.0
BUTYLCYCLOHEXANE 10.0
BBHLISOPROPYLPHTHALATE 8.0 7.0
BUTYRALIEHYBE 2.0 1.0
B-PHELLANDRENE 4.0 2.0
B-PINEHE , 8.0 1.0
Cl COHPQUNBS (DIESEL EXHAUST) 0.01 0.79
CIO AROMATIC 3.0 1.0
CIO COMPOUNDS (DIESEL EXHAUST) 5.48 0.189 0.289 0.245
CIO GLEFIHS 8.0 1.0
CIO PARAFFIHS 10.0
C-4
-------
CONPOOW NAtiE ' PAR OLE ETH TOL XYL FORH AID2 ISOP fffi
C10H12' 2.0 U
mm 10.0
C10H168 8,0 1.0
Cll CQHPQUNBS (DIESEL EXHAUST! 10.7 0.15 t
Cll GLEFINS ?.0 1.0
Cll PARAFFIN 11.0
C11B10 3.0 1.9
C11H140 2.0 . 1,0 1.0
C12 COMPOUNDS (DIESEL EXHAUST) 5.0 1.0
C12 OLEFINS 10.0 1.0
.£12 PARAFFIN , 12,0
C12H22 12.0
C13 CQHPQUNDS (DIESEL EXHAUST) 4.0 1.0
C13 PARAFFIN , 13.0
C14 COmmBS (BIESEL EXHAUST) 7.0 1.0
CIS COHPOUNDS (DIESEL EXHAUST) 8.0 1.0
C16 BRANCHES ALKANE 16.0
Mi COHPUONSS (SIESEL EXHAUST] 9.0 1.0
C17 COKFOUNDS (SIESEL EXHAUST) 10.0 1.0
CIS WOUNDS (DIESEL EXHAUST) 11.0 , 1.0
C19 COHPflONSS.(DIESEL EXHAUST) 12.0 1.0 '
C2 ALKYLANTHRACENES 3.0 1.0 5.0 .
C2 ALMENZANTHRACENE - 2.0 1.0 1.0 3.0
C2 ALKYLBEHZOPHEHAHTHREHE 2.0 1.0 1.0 3.0
C2 ALKYLCHRYSEHES 2.0 1.0 1.0 3.0
C2 ALHLCYCLOHEXANE 8.0
C2 ALCYLINSAN 3.0 1.0
C2 ALKYLNAPTHALEHE 4.0 1.0
C2 ALXYLPHEHAHTHRENES 3.0- 1.0 5,0
C2 COHPOOHSS (SIESEL EXHAUST) 0.77 0.115 ' . 1.0
C20 COfiPOUHSS (IIESEL EXHAUST) 13.S i.O
C21 CBMPflUMSS (DIESEL EXHAUST) 14.0 i.O
C22 CQ8POUH8S (DIESEL EXHAUST) 15.0 1.0
C23 WOUNDS (SIESEL EXHAUST) 16.0 1.0
C24 COMPOUNDS (DIESEL EXHAUST) 17.0 1.0
C25 COMPOUNDS (DIESEL EXHAUST) 18.0 1.0
C-26 COMPOUNDS (DIESEL EXHAUST) 19.0 1.0
C27 COaPQUNBS (DIESEL EXHAUST) 20.0 1.0
C28 COMPOUNDS (DIESEL EXHAUST) 21.0 1.0
C2? COMPOUNDS (DIESEL EXHAUST) 22.0 1.0
C3 ALKYLCYCLOHEXANE 9.0
C3 ALmSTYRENE 2.0 1.0 1.0
C3 CGKPOUHSS (DIESEL EXHAUST) 1.07 0.904 0.122
C3 PARAFFIN 3.0
C30 COBPOUNDS (DIESEL EXHAUST! 23.0 1.0
C31 WOUNDS (DIESEL EXHAUST) 24.0 1.0
C32 COfiPflUNDS (DIESEL EXHAUST) 25.0 1.0
C33 COfiPflUNDS (DIESEL EXHAUST) 26.0 1.0
C34 COHPOUNDS (DIESEL EXHAUST) 27.0 1.0
C35 COHPOUNDS (DIESEL EXHAUST) 28.0 1.0
C36 CORPOUNDS (DIESEL EXHAUST) 29.0 1.0
C37 CORPOUNDS (DIESEL EXHAUST) 30.0 1.0
C38 COHPflUHDS (DIESEL EXHAUST) 31.0 1.0
C39 COHPOUNDS (DIESEL EXHAUST) 32.0 1.0
C-5
-------
COMPOUND HAKE
PAR OLE ETH IOL XYL FORK ALD2 ISOP KR
C3/C4/C5 ALCTLBENZENES 3.8 1.0
C4 ALKYLPHEHOLS 3.0 1.0
C4 ALKYLSTYRENES 3.0 1.0 1.0
C4 COMPOUNDS (DIESEL EXHAUST) 3.7 0.03?
C4 OLEFIH 2.0 1.0
C4 PARAFFIN 4.0
M'SUBSTITUTED CYCLOHEXANE 10.0
C4 SUBSTITUTED CYCLOHEXANONE .10.0
C40 COMPOUNDS (DIESEL EXHAUST) 33.0 1.0
C41 COMPOUNDS (DIESEL EXHAUST) 34.0 1.0
C42 COMPOUNDS (DIESEL EXHAUST). 35.0 1.0 . .
C43 COMPOUNDS (DIESEL EXHAUST)" 36.0 1.0
C5 ALIYL CYCLOHEXANE 11.0
C5 ALKYLBENZENES 4.0 1.0
C5 AUYLBENZEHES (UNSATURATED) 2.0 1.0 1.0
C5 ALKYLPHENOLS 4.0 1.0
C5 COMPOUNDS (DIESEL EXHAUST) 4.6 0.045
C5 ESTER 6.0
C5 OLEFIH 3.0 1.0
C5 PARAFFIN 5.0
C5 PARAFFIN/OLEFIN 4.0 0.5 . •
C5 SUBSTITUTED CYCLOHEXAKE 11.0
C5H100 - 5.0
C6 ALKYLBENZENE 5.0 • 1.0
C6 COMPOUNDS (DIESEL EXHAUST) 4.5 0.218
C6 OLEFINS 4.0 1.0
C6 PARAFFIN 6.0
C6 SUBSTITUTED CYCLOHEXANE 12.0
C6H1803SI3 • 6.0
C7.ALKYLBENZENE • 6.0 1.0
C7 COMPOUNDS (DIESEL EXHAUST) 1.4 0.8
C7 CYCLOPARAFFINS 7.0
C7 OLEFINS 5.0 1.0
C7 PARAFFINS 7.0 r
C7H12 5.0 1.0
C7H120 5.0 1.0
C7-C16 11.0
C8 COMPOUNDS (DIESEL EXHAUST) 4.? 0.21 0.335
CB CYCLOPARAFFINS 8.0
C8 OLEFINS 6.0 1.0
C8 PARAFFIN 8.0
C8 PHENOLS 1.0 1.0
C8H14 6.0 1.0
C8H2404SI4 8.0.
C? COMPOUNDS (DIESEL EXHAUST) 3.62 0.056 0.608
C? CYCLOPARAFFINS 9.0
C? OLEFINS 7.0 1.0
C9 PARAFFIN 9.0
C? PHENOLS 2.0. 1.0
CAMPHENE 8.0 1.0
CAPROLACTAM 5.0
CARBITOL 2.0
CARBON BISULFIDE 1.0
CARBOH TETRACHLORIDE
0.111
0.155
1.0
0.065 0.?34
0.062
2.0
1.0
1.0
C-6
-------
COMPOUND MAKE PAS OLE ETH Tfll XTL FORH .ALD2 !SOP MR
SARBOHYL SULFIDE ' * t £l
CARYOFHYLLENE 9.0 3.0
CELLOSOLVE 2.0 i.o
CELLOSOLVE ACETATE 3.6 ,, jj 1 0
CHLflROBENZENE j.Q - ' 5'0
CHLORflDIFLOQROHETHANE J'0
CHLflRQFflRH ^
CHLOROPEHTAFLUOROETHftNE z'ft
CHLOROPREHE • 2.0
CHLOROTRIFLIIQROHETHANE I 0
CHRYSENE ' U u 3i'e
CIS-I,4-BIBETHYLCYCLOH£XAHE 8.0 '''
CIS-2-BOTENE 2 0
CIS-2-HEPIENE 3.0 2.'fl
CIS-2-HEXENE 2.0 2.0
CIS-2-OCTEHE 4.0 2*0
CIS-2-PEHTENE 1.0 2!'0
CIS-3-HEXEHE 2.0 \ 9
CflRONEE LC i.o ' ?.0
' CREOSOTE , u .1.0 20
C8ESQL • u
CROTOMBEHYBE 1.0 i.o
COHEffi USOPROPYL B£«ZEE) 2.0 1.0
CYCL8HEPTAKE 7.0
CYCLOHEXiiHE 6.0
CYCLOHEXftKBL 6.0
EYCLOHEXAHONE , 6.0
CYCL08EXEIC 2.0 20
CYCLOPEHTftfiHTHRftCEHES • 3.0 1.0 ' 50
CYCLOPENTAHE 5.0
CYCLDPEKTftPflESAHTHRENES 3.0 l o 55
CYCLOPEHTft(c,d!PYl!EE 2.0 1.0 1.0 To
CYCIQPENTENE 1.0 • 20
' CYCLOPEHTYLCYCLOPEHTftHE 10.0
BECALIHS 10.«
BEHATURAHT 1.0
HACETflKE ALCOHOL 4.0 : i o
BIBE8ZAHTHRSCEHES 1.0 }.0 1.5 ' 6 0
HBENZOPY8EHES 1.0' i.o 1.0 8*0
JIBESZOfijiJAKTHRACESE 1.0 1.0 l.'o <'o i
SIBEHZPHEWKREffiS l.fl i.o i o 6°0 '
JIBUTYL ETHER 6.0 u
HBIITYLPHTHALATE ?.o 7 0 !
5ICHLOROBEHZEHES 1.0 5"0 i
HCHLOROBIFLUOROIIETHAHE ' j'j '
BICHLflRflHETHAE j'0 |
BICHOLROTETRAFLflflROETHANE 2V '
BIETHYLCYCLOHEXAHE 10.0 ' ;
BIETHYLEHE fiLYCOL 2.0 1 0 '
BIETHYLHETHYLCYCLDHEXANE 11.0
BIHYBROHAPTHALENE 2.0 1.0
BIHYBRQXYHAPTHALEUEBIONE 2.0 H
BIISOPROPYLBEHZEKE 4.0 } o
\BIHETHYL flLKYL ADINES 3.0
C-7
-------
COMPOUND MANE PAR OLE ETH TOL XT', FQRH AU2 ISOP NR
SIKETHYLBENZriALCOHOL 1.0 1.0
HHETHYLBUTANE 4.0
BIKETHYL8UTAHEBIOATE 4.8 2.0
BIHETHYLBUTENE 4.0 1.0 . •
BIHETHYLBUTYLCYCLOHEXANE 12.0
BIHETHYLCYCLOBUTANOHE f.O
BIRETHYLCYCLOHEXAHE 8.0
BIHETHYLCYCLOPENTANE 7.0
BIKETHYLCYCL0PENTEHES 5.0 1.0
BIHETHYLBECAHE 12.0
DIHETHYLETHER ' . 2.0 - . •
BIHETHYLETHYLBEHZOIC ACIB 2.0 1.0 1.0
SIKETHYLETHYLCYCLOHEXAKE 10.0
BIHETHYLFORHAKIBE 2.0 1.9
DIKETHYtHEPTftNES 9.0 ,
DieHYLHEPTANOL 9.0
BIHETHYtHEXABIENE 2.0 1.0 2.0
IIHeHYLHEXAHEHOATE i.O 2.0
SIKETHYLHEXA«ES 8.0
DIKETHYLHEXEME 4.0 1.0
DIKETHYLINIAHS 3.0 1.0
BIKETHYLINDENE 1.0 1.0 1.0
BIKEIHYLNAPHTHYRIBINE 3.0 1.0
51KETHYLKAPIHALEHE U 1.0
BIHEIHYLHflHAHES 11.0
BIliETHYLOCTANES 10.0
BIKEIHYLOCTAKflL 10.0
BIKETHYLOCTENES 8.0 1.0
BIHETHYLOCHNE ?.0 • 1.0
WHETHYLPEHTAHE 7.0
8IKEIHYIPEHTAHEBIOATE 5.0 2.0
BIETHYLPEHTAHQL 7.0
BIKETHYLPEKTEE ' ' 5.0 1.0
BIKETHYIPHTHALATE 3.0 - 7.0
1IHETHYLTEREPHTHALATE 3.0 - . 7.0
BINETHYLUHBECAE . 13.0
BIPHEKYLETHAHE ' 2.0
B1PROPYLEHE 6LYCOL 4.0 1.0
BIPROPYLPHTHALATE 7.0 7.0
BIVIHYUEHZEHE 1.0 1.0
BKETHYLPHEHYUETHAHE 4.0 2.0
B1-C8 ALWL PHTHALATE 17.0 7.0
BOBECEHE 10.0 1.0
D-LIHOHEE 4.0 1.0 2.0
EICUSAHE 20.0
EPICHLOROHYBRIH 3.0
ETHAE 0.4 1.6
ETHAHOLAHIKE 0.4 ' U
ETHYL ACETATE 3.0 1.0
ETHYL ACRYIATE 2.0 1.0 1.0
ETHYL ALCOHOL 0.4 1.6
ETHYL CHLORIBE 2.0
ETHYL ETHER 2.0 1.0
ETHYLAMIHE 0.4 ' 1.6
C-8
-------
CONPOUM MADE FftR OLE ETH TOL XTL FQRtl M.D2 ISOP NR
ETHYLBEHZE8E i.O f.o
ETHYLBICYCLGHEPTAHE 14. 0
ETHYLCYCLBHEXANE 8.0
ETHYLCYCLflPENTftNE. 7.0
ETHYLCYCLflPEHTENE 5.0 1.0 '
ETHYLDIIiETHYLBENZENE 2.0 1.0
ETHYLBMTHYLCYCLOHEXANE 10,0
ETHYLBIHETHYLOCTANE • 12.0
ETHYLDIIIETHYLPEMTAH ?.0
ETHYLBIfiETHYLPHENOL 2.0 i 0
ETHYLEHE - j.o
ETHYLEHE H1ROHI8E ' ' zo
ETBYLEIff 8ICHLORHE . 2'o
ETHYLENE GLYCflL 6,4 ^
ETHYLEHE OXIIE U ' j'o
ETHYLENEAKIHES (J.4 . ^
ETHYLFURAS ' 2.0 2.0
ETHYLHEPTAKE ?.0
ETHYLHEPTEHE 7.0 1.0
EIHYLHEXANE 8.0 '
ETHYLHEXAHflftTE 7.0 ' t 0
EIHYLIMAN 3.0 -u
ETHYLISOPROPYL ETHER 3.0 I.O
ETHYLHERCAPHK 2.8
ETHYLKETHYLCYCLOHEXAHE 9.0
ETHYLHETHYLCYCLOPEHTANE 8.0
EWIHETHYIHEXAHE ?.0
ETHYLHETHYLflCTfiffi u.o
ETHYLOCTAHE 10.0.
ETHYlflCTENE g.o ik«
ETHYLPEHTEHE 5.0 i.«
ETHYLPHEHYlPaEKYLETHAE 1.0 1.0 1.0
ETHYLPRSPYLCTCLflHEXAHE 11.0
ETHYLSTYREffi l.o l.fl }.{
ETHYIT.QLOEHE 1.0 " It0
ETHYL-T-BOTYL ETHER 4.0 ' i {
FLUORAHTHEHE 1.0 1 0 ' 10
FLLIOREHE u ,*,
FORHALIEHYSE . j 5
FflRHIC ACIB ' j 0
FORFURYL ALCOHOL 1.0 2.0
6LYCEROL 1.5 j 5
BLYCOL o.l i I
6LYCOL ETHER 0.8
6LYOXAL l.o lfl " ' ;
HEHEICOSAfff 21.0 !
HEPTABIENftL 1.0 l.fl 2 0 '
HEPTANE 7.0
HEPTAHOHE 7.0 . i
HEPTENE 5.0 1.0 '
HEXABECAHE 16.0
HEXA8ECAHOIC ACIP 15.0 i 0 i
HEXABIENftL 1.0 2 0 * I
HEXAFLUOROETHAHE '
C-9
-------
COMPOUND NAHE
pftR GLE ETH TOL XYL FORM ALS2 ISBP
HEXAHETHYIEHEIIMKIHE
HEXAHAL
HEXANE
HEXENE
KEXYLENE GLYCOL
HEXYHE
INSANE
mm .
IHKHOaI2I3-cd)PrREffi
•ISOAKYL ALCOHOL
ISOAHYLBENZEHE
ISOBUTANE
ISOiUTYL ALCOHOL
ISOBUTYLACETATE
ISOBOTYLACRYLATE
ISOBOTYLBEHZEHE
ISOBBTYLENE
ISOBUTYLISOBUTYRATE
ISOBHTYRAUiEHYDE
ISOREBS OF 8UTEKE
ISOHERS OF BliTYLBEHZEHE
ISOHERS OF C10H10
ISORERS OF C10H18
ISOHERS OF C11H20
ISOHERS OF C5H16
ISOHERS OF DECAHE
ISOHERS OF DIETHYLEEHZEKE
ISOHERS OF 10IECAHE
ISORERS OF ETHYLTOLUEHE
ISOHERS OF HEPTADECAHE
ISOHERS OF HEPTAE
ISOHERS OF HEXANE
ISOKERS OF NQHftHE
ISOHERS OF OCTADECAffi
ISOKERS OF OCTANE
ISOHERS OF PEHTADECAHE
ISOKERS OF PEKTAHE
ISOHERS OF PEHTENE
ISOKERS OF PROPYLBENZENE .
ISOHERS OF TETRWECANE
ISOHERS OF TRIJECAME
ISOHERS OF KCAKE
ISOHERS OF XYLEHE
ISOOCTANE
ISOPEHTANE
ISOPREKE
ISOPROFYL ALCOHOL
ISOPROPYLACETATE
ISOPRQPYLBEHZENE
ISOPROPYLCYCLOHEXANE
ISOPROPYLCYCLOPEHTANE
ISOPROPYLHETHYLCYCLOHEXAHE
ISQVALERAL1EHYBE
LACTOL SPIRITS
6.0
4.0
6.0
4.0 1.0
6.0
5.0
1.0
0.5
1.0
5.0
4.0
4.0
4.0
$.0
4.0 1.0
3.0
' 2.0 1.0
7.0
' 2.0
2.0 1.0 ,
3.0
. 1.0
8.0 1.0
11.0
7.0 1.0
10.0
2.0
12.0
1.0
17.0
7.0
6.0
9.0
18.0
8.0.
15..0
5.0
3.0 1.0
2.0
14.0
13.0
11.0
8.0
5.0
1.5
4.0
2.0
9.0
8.0
10.0
3.0
8.0
1
• %
1.0
1.0
a$s i.o i.o
a
1.0
1.0
1.0
1.0
1.0
1.0.
1.0
1.0
1.0
1.0
1.0
4.0
1,0
1.0
1.0
1.0
1.0
1.5
1.0
1.0
C-10
-------
CflMPOiJNJ NAHE
PAR OLE £TH TOL XTL FORK M.I2 ISO?
LIMflNEHE 4.0 l.fl
HALEIC ANHY8RISE 2.0
HETHANE
HETHOXYETHOXYBUTAHOHE 5.0
HETHQXYETHOXYETHANQL 3.0
HETHOXYNAPHTHALENE 3.6
HETHYL ALCOHOL 1.9
METHYL Cll ESTER !2.0
HETHYL C12 ESTER 13.0
METHYL C13 ESTER 14.0
HETHYL C14-ESTER 15.9
HETHYL CIS ESTER 14.0
METHYL C19 ESTER 20.0
METHYL C20 ESTER 21.0
HETHYLACETATE 2.9
HETHYLACETOPHEHO.NE 1.9
METHYLACETYLEHE (PRQPYNE) 2.9
HETHYLACRYLATE 1.9 i.o
HETHYLAL 3.0
HETHYLALLENE 1.9 1.5
HETHYLAHYL JEETONE 7.6
BETHYLAHTHRACEHES 2.0
HETHYLBE8ZANTHRACENES ' 1.9
HETHYLBE8ZPHENANTHREE l.fl
HETHYLBIPHEMYL
HETHYLBUTABIEffi i.o 2.9
HETHYLBUTEHE 3.0 l.fl
METHYLBOTYL IETONE U
HETHYLCARBITQL 3.9
HETHYLCELLOSOLVE 1,5
«ETHYLCKLflRIi!E
METHYLCHRYSEHES 1.8
HETHYLCYCLOHEXASIEffi 1.5 i.o
KETHYLCYCLOHEXANE ' 7.0
METHYLCYCLflHEXEHE ' 5.0 1.0
HETHYLCYCLeflCTAHE 9.0
METHYLCYCLOPEHTABIEE i.o
HETHYLCYCLOPEHTANE U
HETHYLCYCLOPEHTEHE 4.0 l.fl
HETHYLDECALIHS n.fl
METHYL5ECAHES H.5
HETHYL1ECENE ?.o 1.0
HETHYLBIHYBRQNAPHT8ALE 3.6
SETHYLDOSECANE- 13.0
HETHYL508ECAHQATE 12.0
HETHYLEHE SROHIIE
HETflYLEHE CHLflRISE
HETHYLEHEB1S(C4H4NCO) 1.0
HETHYLEHEfbM-PHEKYLISOCYAHATE
METHYLETHYL KETflNE 4.fl
HETHYLETHYLHEPTANE lO.fl
METHYLETHYLPENTAHOATE 7.0
METHYLFLUQRANTHEHES 1.0
HETHYLFflRHATE i.o
2.0
i.O
1.0
1.0
l.fl
1.0
1.9
1.6
l.S
1.0
1,5
1.0 1.0
1.0-
1.0
2.0
2.0
l.fl
l.fl
1.0
1.0 LO
1.0
1.0
1.0
l.fl
1.0
i.O
1,0
1.0
1.0
1.0
5.0
3.0
3.0
5.0
1.3
3.0
1.0
1.0
1.0
7.0
8.0
LO
1.0
1.0
C-ll
-------
COKPOUHD KftBE PAR OLE ETH ICL XYL FORK ALD2 1SOP KR
IIETHYLGLYOXAL ' - 1.0 1.0
HETHYIHEPTAHE 8.0
HETHYLHEPTANflL 8.0
liETHYLHEPTEME 6.0 1.0
HETHYLHEPTYNE 7.0 1.0
KETHYLHEXAOIENE 1.0 1.0 2.0
KETHYLHEXAHAL 5.0 1.0
KETHYLHEXANE 7.0
HETHYLHEXEHES 5.0 1.0
HEIHYLIHIAHS 2.0 " 1.0
HETHYUHBENE ' , 1.0 1.0
KETHYLISOBUTYL KEIOHE 6.0
NETHYUSGPROPYLCYCLOHEXAHE 10.0
HETHYUIETHACRYLATE ' 2.0 1,0 1.0
HETHYLHETHYLPROPENOATE 2.0 1.0 1.0,
HETHYLKYRISTATE 14.0 ' 1.0
KETHYtHAPHTHALEHES 3.0 1.0
KETHYLHQHAKE 10.0
HETHYLKKE 8.0 1.0
liETHYLOCTANES 9.0
HETHYIPALIIITATE . 16.0 1.0
HETBYLPENTAHE 6.0 .
HETHYLPENTENES 4.0 1.0
HEIHYLPHEMIHREKES . 2.0 1.0 5.0
HETHYLPROPYLCYCLOHEXANE 10.0
•eHYLPROPYLdaKAHE 13.0
HEIHYLSTEARATE 18.fi . 1.0
KTHYLSTYREKE , 1.0 1.0
HETHYLUH8ECAHE .12.0
«ETHYL-T-BUTYL ETHER 3.0 1.0
HIKERAL SPIRITS 6.0 1.0
HYRCESE - 4.0 3.0
K-BICHIOROBEHZEKE 1.0 5.0
H-IIETHYLBEHZEHE 2.0 i.O
K-ETHYLTflLUEKE 1.0 . 1.0
H-XYLENE ' .1.0
H-XYLEffi AHB P-XYLEKE 1.0
HAPHTHA 8.0
HAPTHAIEHE 2.0 1.0
NITROBEHZEKE . • 1.0 5.0
KGHASEGAHE 17.0
KOKA6IEKE 3.0 1.0 2.0
!«KE 9.0
mm 7.0 i.o
NOHEKOE 7.0 1.0
•NOHYIPHEHOL 8.0 1.0
H-AHYLJENZEKE 4.0 1.0
H-BUTAE 4.0
H-BUTYL ALCOHOL 4.0
H-BUTYLACETATE 5.0 i.O
K-iECAKE 10.0
H-EQSECAHE 12.0
H-HEPTAJECAffi 17.0
K-HEXYLBEHZEHE 5.0 1.0
C-12
-------
COHPOONB h'AHE PAR OLE ETH TCL XTL FORK AL52 ISOP MR
N-PENTft&ECANE
M-PEHTftHE
H-fENTENE
H-PEHTYLCYCLOHEXANE
N-PHEHYLAMILINE
N-PROPYL ALCOHOL
M-PRQPYLACETATE
N-PROPYLBENZENE
N-TETRftBECANE
N-TRIBECANE
N-UNBECAHE
QWHYBROINDENES
OCTAHETHYLCYCLQTETRASILOXAHE
OCTANE
OCTAKOL
OCTATRIENE
flCTENE
OXY6EHATES
0-JICKLOROBENZENE
0-ETHYLTOLUEH,E
8-XYLENE
PALMITIC ACIB
PARAFFINS (C1H34J
PARAFFINS (C2-C7)
PARAFFINS/OLEFISS {C12-C16}
PENTABIEHE
FENTAKOL
PENTEHYHE
PEKTYLBEKZEHE
PENTYLCYCLOHEXA8E
POJTYLIDEffiCYCLOHEXfiHE
PENTYHE
PERCHL8ROETHYLENE
prnvi ryr
f LA I UCHC
PHENANTHREKE
PHENOL
PfiENYLISOCYAHAIE
PHEHYLNAPHTHALENES
FHTHALIC AHHY5RIBE
PIPERYLEKE
FflLYETHYLEHE 8LYCOL
PROPABIENE
PROPANE
PROPEHE
PROPENYLCYCLOHEXftNE
PROPIOHALBEHYBE
PROFIONIC ACID
PROPYLBENZEHE
FROPYLCYCLOHEXANE
PROPYLENE 5ICHLORIBE
PftOPYLENE fiLYCOL
PROPYLEKE OXIBE
PROPYLHEPTENES
15.0
5.0
3.0
11.0
1.5
4.0
2.0
14.0
13.0
.11.0-
9.0
8.0
8.0
8.0
6.0
4.0
1.0
1.0
15.0
25.0
4.5
13.0
1.0
5.0
2.0
4.0
11.0
10.0
4.0
1.0-
1.0
1.0
1.0
1.0
1.0
2.0
1.5
1.0
7.0
f.O
2.0
2.0
9.0
1.5
1.5
2.0
8.0
1.0
2.0
1.0
0.5
2.0
1.0
1,0
2.0
1.5
1.8
1.0
1.0
1.0 5.0
1.5
1,0
1.0
2.0
5.0
l.o
1.0
1.5
0.0
1.0
2 0
l.o 1.0 5.'o
1.0 5 0
.j.'o
" 4.0
1.0 1.0
7.0
1.5
1.0
1.0
1.5
1.5
1.0
1.0 1.0 1.0
C-13
-------
COMPOUND IMISE PAR OLE ETH ICL XYL FORK, ftLDZ ISOP NR
P-BICHLQROeENZEk'E 1.0 • 5.0
P-ETHYLTOLIIEHE 1.0 1.0
P-TOLIIALBEHYSE 2.0 4.0
P-XYLENE 1.0
SEC:BUTYL ALCOHOL 4.0
SEC-8UTYLBENZEHE 3.0 1.0
SILOXANE
STYRENE 0.5 1.0
SUBSTITUTED C9 ESTER (C12) 12.0 1.0
TEREPHTHALIC ACID i.O 7.0
TERPEHES • 8.0 1.0
TETRACHLOROBEHZEKES 1.0 -. 5.0
TETRAFLUOROHETHAHE 1.0
TETRAHETHYLiEHZEKE 2.0 1.0
TETRAKETHYLCYCLOBDTENE 6.0 1.0
TEIRAHETHYLCYCLOPEKTAHE 9.0
TETRAMETHYLHEXANE 10.0
TETRAKETHYLPENTANONE • 9.0
TETRAHETHYLSIIAKE 4.0
IETRAHETHYLTHIOUREA 4.0 ' 1.0
TOLUEKE 1.0
TOLUEffi IIISOCYANATE -1.0 '2.0
TOLUENE ISOCYAKATE • 1.0 1.0
TOTAL AROKATIC AlilKES 1.0 5.0
TOTAL C2-C5 ALDEHYDES 1.5 1.0
TRAHS-1-PHEHYLBOTENE 1.0 1.0. 1.0
TRAHS-2-BOTENE 2.0
TRftHS-2-HEPTEHE 3.0 2.0
TRA«S-2-HEXEH£ 2.0 2.0
TRftHS-2-KQHEN£ 5.0 2.0
TRANS-2-PEHTEffi 1.0 2.0
TRAKS-3-HEXENE 2.0 2.0
TRICHLOROBEHZEKES 1.5 5.0
TRICHLOROFLUQROhETHANE • _ . 1.0
TRICHLDROTRIFLUOROETHftHE ' 2.0
TRICHOLROETHYLEXE 1.0
TRIETHYLEHE 6LYCOL 2.0 . 2'.0
TRIFLUORONETHA«£ 1.0
TRIKETHYLAHIHE 3.0
TRIHETHYLBEIJZE)£ 1.0 1.0
TRMETHYLCYCLOHEXANES 9.0
TRIKETHYLCYCLflHEXAOOL 9.0
TRIKETHYLCYCLOPEHTAHE 8.0
TRIXETHYLCYCLOPEHTAHONE 8.0
TRIHEIHYLDECANE • 13.0
TRIHETHYLBECEK 11.0 1.0
TRIIiETHYLFLUQROSILAHE 2.0 1.0
TRHETHYLHEPTANES 10.0
TRIKETHYLHEXANES 9.0
TRIMETHYLHEXEHE 7.0 1.0
TRIHETHYIINDAH 4.0 1.0
TR1HETHYLKONEKE 10.0 1.0
TRIHETHYLQCTAKES 11.0
TRIHETHYLPEHTADIEKE 4.0 2.0
C-14
-------
PAR OLE ETH TOL XTL FORK ALD2 ISOP NR
TRIETMLPEHTANE . 8.0
T-EUIYL ALCOHOL 4.0
T-BUTYLBENZENE 2.0 1.0 1.0
UNIDENTIFIED 6.0
VINYL ACETATE 1.0 1.0 . 1.0
VIHYL CHLORIBE . 1.0
XYLENE BASE ACIDS ' 1.0
C-15 -'
-------
TABLE C-2. EXAMPLE PROBLEM - PART I
COMPUTATION OF CARBON BOND CONCENTRATIONS
leasured Compound
Carbon Bond Concentration
Species
Ethyl ene
Propene
n-Butane
T-2-Butane
2, 3-Dimethyl butane
Toluene
M-xylene
Benzene
TOTAL
ppbC
20
30
;170
10
100
70
40
60
500
PPb ,
10
10
42.5
i
2.5
16.7
10
5
10
106.7
OLE
10
10
PAR
i
10
170
100
10
290
TOL
10
10
XYL
5
5
FORM
0
ALD2
5
5 |
ETH
10
:
10
NR
•
50
50
I SOP
0
This is a hypothetical problem, and is not necessarily intended to be indicative of the
MOC composition of ambient air.
C-16
-------
TABLE C-3. EXAMPLE PROBLEM - PART 2
COMPUTATION OF CARBON-FRACTIONS
5-4 Class
OLE
PAR
TOL
XYL
FORM
ALD2
ETH
I SOP
NR
FAL
Concentration (ppb)1
10
290 ,
10
i
5
,0
5
10 .
0
50
Concentration (ppbr.)2
20
290
70
40'
0
10
20
0
50
500
Initial Carbon
Fraction**
"6.04
0,58
0.14
0.08
o.oo :
0.02
0.04
0.00
0.10
1.00
Final 'Carbo
Fraction4
0.04'
0.55
, 0.13
'
1
• 0.08
< i
. i 0.02
0.05
0.04
0.00
0.09
1.00
Table C-2
2
Computed by taking concentration in ppb times carbon numbers from Table 2-1.
Computed by dividing concentrations in ppbC by the total NMOC in ppbC (e.e., 500 ppbC)
Unmeasured aldehydes added (.02 to FORM AND .03. to ALD2) and total readjusted to 1.00.
C-17
-------
The individual concentrations in the nine entries in the right hand
column are obtained by multiplying the concentration of each species times the
number of bonds for that species found in Table C-l. For example, propene has
a concentration of 10 ppb in the example problem. Table C-l shows that
propene has one olefin bond and one paraffin bond. Thus, 10 ppb are put in
each category (OLE and PAR).
After all of'the species concentrations have been apportioned to the
carbon bond groups, then each column is totalled. These concentrations are in
ppb. To convert to ppbC, the assumed carbon number for each carbon bond class
are utilized. These are found in Table 2-1. When the concentrations in ppb
i
are multiplied by these carbon numbers, concentrations in ppbC are obtained.
Each total for a carbon bond class is then divided by the total NMOC (in
ppbC) to obtain initial carbon-fractions. The final step is to add (3.02 to
the FORM fraction and 0.03 to the ALD2 fraction. These adjustments are to
account for aldehydes which are not detected by the sampling/analytical
procedure. The fractions are then adjusted so that they total up to 100
percent.
C-18
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
*•
Procedures for Applying City-Specific EKMA
a. f-tiRFORMING ORGANIZATION NAME AND ADDRESS
12. SPONSORING AGENCY NAME AND ADDRESS
U. S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
3. RECIPIENT'S ACCESSION-NO.
5. REPORT DATE
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO.
EPA-450/4-89-012
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
... |
EPA Contact: Keith Baugues ,
FvnrM,1? Vodel :wll1ch allows the user to estimate the volatile organf'c compound
qua°?lt"sSndardqfore:zSLf0r ' ""* 1n °rder *° ich1av8 the national •?*'-«« ^
S^nSSp^vJ^Mf! thS 1flPUtS rSqUired t0 rUn °ZIPM-4 and 6XPla1nS h°W *> de-V^0P
the needed variables.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
ozone
VOC control strategies
photochemical modeling
13. DISTRIBUTION STATEMENT
Unlimited
L
EPA Form 2220-1 (3-73)
b.lDENTIFIERS/OPEN ENDED TERMS )c. COSATI Field/Group
- SECURITY CLASS (ThisReport)
Unclassified
10. SECURITY CLASS (Thispage)
Unclassified
21. NO. OF PAGES
118
22. PRICE
-------
INSTRUCTIONS
1. REPORT NUMBER
Insert the EPA report number as it appears on the cover of the publication.
2. LEAVE BLANK
3. RECIPIENTS ACCESSION NUMBER
Reserved for use by each report recipient.
4. TITLE AND SUBTITLE
Title should indicate clearly and briefly the subject coverage of the report, and be displayed prominently. Set subtitle, if used, in smaller
type or otherwise subordinate it to main title. When a report is prepared in more than one volume,, repeat the primary title, add volume
number and include subtitle for the specific title. . '
5. REPORT DATE
Each report shall carry a date indicating at least month and year. Indicate the basis on which it was selected (e.g., date of issue date of
approval, date of preparation, etc.). ...'.-.' ' .
6. PERFORMING ORGANIZATION CODE
Leave blank.
7. AUTHOR(S)
Give name(s) in conventional order (John R. Doe, J. Robert Doe, etc.). List author's affiliation if it differs from the performing organi-
zation. • , .
8. PERFORMING ORGANIZATION REPORT NUMBER i I (l
Insert if performing organization wishes to assign this number.
I '
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Give name, street, citjf, state, and ZIP code.' List no more than two levels of an organizational hirearchy. : ; '
10. PROGRAM ELEMENT NUMBER
Use the program element number under which the report.was prepared. Subordinate numbers may be included in parentheses.
11. CONTRACT/GRANT NUMBER
Insert contract or grant number under which report was prepared. ' ' (
12. SPONSORING AGENCY NAME AND ADDRESS '
Include ZIP code.
13; TYPE OF REPORT AND PERIOD COVERED
Indicate interim final, etc., and if applicable, dates covered.
14. SPONSORING AGENCY CODE
Leave blank. .
15. SUPPLEMENTARY NOTES I
Enter information not included elsewhere but useful, such as: Prepared in cooperation with, Translation of. Presented at conference of
To be published in, Supersedes, Supplements, etc.
16. ABSTRACT ' |
Include a brief (200 words or less} factual summary of the most significant information contained in the report. If the report contains a
significant bibliography or literature survey, mention it here. I
• i
17. KEY WORDS AND DOCUMENT ANALYSIS
(a) DESCRIPTORS - Select from the Thesaurus of Engineering and Scientific Terms the proper authorized terms that identify the major
concept of the research and are sufficiently specific and precise to be used as index entries for cataloging. • i
(b) IDENTIFIERS AND'OPEN-ENDED TERMS - Use identifiers for project names, code names, equipment designators, etc. Use open-
ended terms written in descriptor form for those subjects for which no descriptor exists.
(c) COSATI FIELD GROUP - Field and group assignments are to be taken from the 1965 COS ATI Subject Category List. Since the ma-
jority of documents are multidisciplinary in nature, the Primary Field/Group assignment(s) will be specific discipline, area of human
endeavor, or type of physical object. The application(s) will be cross-referenced with secondary Field/Group assignments that will fol ow
the primary posting(s). . , ]
18. DISTRIBUTION STATEMENT
Denote releasability to the public or limitation for reasons other than security for example "Release Unlimited." Cite any availability to
the public, with address and price. /
19.&20. SECURITY CLASSIFICATION !
DO NOT submit classified reports to the National Technical Information service.
21. NUMBER OF PAGES
Insert the total number of pages, including this one and unnumbered pages, but exclude distribution list, if any.
22. PRICE
Insert the price set by the National Technical Information Service or the Government Printing Office, if known.
EPA Form 2220-1 (9-73) (Reverse)
-------
-------
I }
------- |