EPA-454/R-99-054
December 1999
A Simplified Approach for Estimating Secondary Production
of Hazardous Air Pollutants (HAPs) Using the OZIPR Model
Replaces Section 2.4 in Dispersion Modeling of Toxic Pollutants in Urban
Areas, EPA-454/R-99-021
-------
Notice
This document has been reviewed by the Office of Air Quality Planning and Standards, U. S.
Environmental Protection Agency, and has been approved for publication. Any mention of trade
names or commercial products does not constitute endorsement or recommendation for use.
Acknowledgments
For their thoughtful comments on this document, the authors wish to thank Deborah J.
Luecken, Ned Meyer and Jawad S. Touma of the U. S. Environmental Protection Agency.
-------
Contents
Tables v
Figures vi
Executive Summary vii
Section 1 Introduction 1
Section 2 General Approach for Estimating Secondary Formation of Target HAPs 5
Section 3 Modifications to the Chemical Mechanism 8
Section 4 VOC Species—Reactivity Parameters 13
Section 5 Emissions Data and Boundary Conditions 18
Section 6 Meteorological Data 20
6.1 Approach to the Utilization of Meteorological Data with OZIPR 21
6.2 Typical Day Determination 22
Section 7 Running the OZIPR Model 33
7.1 Estimation of Secondary HAP Production 38
7.2 OZIPR Model Output 39
Section 8 Results and Discussion 40
8.1 Summary Tables of OZIPRModel Results 40
8.2 Discussion of OZIPR Output Tables 52
8.2.1 Variation by Pollutant 52
8.2.2 Variation by Hour of Day 52
8.2.3 Variation by Season 53
8.2.4 Urban/Rural Variation 53
in
-------
8.3 Annual Averages of Target Compounds from Secondary Formation ... 54
8.4 Comparison of Target Compounds with Other Models or
Measurements 60
8.5 Limitations of Current Study 63
8.6 Study Summary 64
Section 9 Practical Use of the Results from This Study 65
9.1 Use of the OZIPR Model to Estimate Secondary HAP Production 65
9.2 Use of Secondary Formation Values with Dispersion Models 66
9.3 Estimation of Secondary Formation without Using Tables 67
References 69
***The following Appendices are in separate files:***
Appendix A Running the OZIPR Model A-l
Appendix B Secondary Production by Hour B-l
Appendix C Total, Primary, and Secondary Production by Hour C-l
Appendix D Secondary Production by Season D-l
Appendix E Weighted Seasonal Secondary Production by City E-l
Appendix F Annual Average Secondary Production by City F-l
IV
-------
Tables
2-1. Counties Selected in Each Study Area 7
4-1. Reactivity Factors Used for the 10 Cities Under Consideration 17
6-1. NWS Stations Used for OZIPR Meteorological Input 20
6-2. Mixing Height Stations Used for OZIPR Meteorological Input 21
6-3. Typical Conditions for Each City and Season 23
6-4. Results of Cluster Analysis for Each City and Season 27
8-1. Secondary Production of Formaldehyde in |_ig m"3 as estimated by the OZIPR Model . . 41
8-2. Secondary Production of Acetaldehyde (|_ig m"3) Estimated by the OZIPR Model 45
8-3. Secondary Production of Acrolein (|_ig m"3) Estimated by the OZIPR Model 49
8-4. Annual Averages for Formaldehyde Concentrations 55
8-5. Annual Averages for Acetaldehyde Concentrations 55
8-6. Annual Averages for Acrolein Concentrations 56
8-7. Annual Secondary Production as a Percentage of Total 56
8-8. Comparison Values for Carbonyl Compounds (|_ig m"3) from the Photochemical
Assessment Monitoring Stations (PAMS) Collected During the Photochemical
Pollutant Seasons for 1995-1997 61
8-9. Comparison Values for Carbonyl Compounds (|_ig m"3) from the RADM Model Run for
July 11-15, 1995, for Cities in This Study 62
-------
Figures
3-1. Chemical reactions added to basic mechanism to account for formation
and destruction of acrolein 9
3-2. Modified SAPRC97 mechanism for determining secondary formation of
formaldehyde, acetaldehyde, and acrolein 10
7-1. Complete OZIPR input file for summertime urban Houston, day type #1, with
primary emissions of carbonyls 33
8-1. Annual averages for secondary formaldehyde concentrations 57
8-2. Annual averages for secondary acetaldehyde concentrations 58
8-3. Annual averages for secondary acrolein concentrations 59
VI
-------
Executive Summary
Title in of the 1990 Amendments to the Clean Air Act stipulates regulation of chemicals
classified as hazardous air pollutants (HAPs). These are substances that are either known to cause
or suspected of causing a threat of adverse human health effects. These compounds are found in the
atmosphere as a result of primary emissions or from the transformation of organic compounds
emitted into the atmosphere by stationary or area sources. Carbonyl compounds represent an
important class of organic compounds found on the list of HAPs. This study examines three of these
compounds, formaldehyde, acetaldehyde, and acrolein, to determine the relative importance of their
formation through primary and secondary processes.
Several very complex models exist that include both dispersion and atmospheric chemistry
to yield HAPs concentration estimates. However, these models are very expensive to execute, often
requiring the use of supercomputers. The goal of this study was to explore whether a simplified
approach could provide useful estimates of total HAPs concentrations. The approach taken here was
to estimate secondary HAPs production with a stand-alone model, run in a personal computing
environment, that incorporated only nondispersive processes, such as photochemistry. The results
from this model would then be coupled to those from a relatively simple dispersion model. The study
results have been quite encouraging, because in comparison with available monitoring data, the
approach used here seems to perform as well as more complex models such as RADM.
The photochemical modeling was done using the OZIPR model, a one-dimensional box
model with a time-varying box height. Emissions were added to the box by time of day; factors such
as temperature, relative humidity, atmospheric pressure, solar radiation, and deposition were used
to determine chemical reaction rates. OZIPR was originally designed to predict ozone concentrations,
but the concentrations of other stable intermediate compounds, such as aldehydes, are also calculated
during the course of the simulations. The model is generally run only for the daylight hours on a
single day. The reaction mechanism used in the current study is based on the widely used SAPRC97
mechanism. The model outputs chemical concentrations as a function of time. These estimates can
then be used in conjunction with output from other models that account for dispersion but not for
vii
-------
chemical transformations. The output data from the OZIPR model are presented in several ways
(e.g., annual and seasonal averages, time series profiles) to facilitate their use with dispersion
models.
Ten study areas were selected for this project: Atlanta, Boston, Chicago, Denver, Houston,
Los Angeles, Phoenix, Pittsburgh, Seattle, and Washington D.C. In each study area, urban and rural
counties were chosen. The urban counties are centered on the cities in question; the rural counties
are near enough to the urban areas to have similar meteorological patterns, but different emissions
based on their lower population and distinct land use patterns. As many as 48 model runs were
needed to adequately characterize each city.
The results show that secondary formation generally accounted for approximately 90% of the
ambient formaldehyde and acetaldehyde and approximately 85% of acrolein; these percentages
varied only slightly between cities. Annual averages for the urban secondary formation of
formaldehyde ranged from 3.0 |_ig m"3 for Seattle to 13.4 |_ig m"3 for Los Angeles. For acetaldehyde,
the corresponding numbers are 5.0 |_ig m"3 for Phoenix to 18.0 |_ig m"3 for Los Angeles; for acrolein,
the values are 0.2 |_ig m"3 (five cities) to 0.7 |_ig m"3 for Los Angeles. Generally, the rural values for
each of the three HAPs ranged between 30% and 50% less than the urban values. The ambient
formaldehyde concentrations were usually greater for southern cities.
Tables of secondary concentrations were also developed for incorporation with dispersion
models. These tables are designed to provide an area-wide additive adjustment to concentrations of
the compounds as a function of time and season or on an annual basis. Section 8 contains these
results.
The user can choose to use the information in this report in the following ways:
1. The most elaborate information can be obtained from running OZIPR using city-
specific parameters as described in Section 9 and Appendix A.
2. Tables 8-1 through 8-6 can be used in conjunction with a dispersion model to provide
seasonal and hourly adjustments (Tables 8-1 through 8-3) or annually (Tables 8-4
through 8-6) for secondary production.
3. When the location of interest is dissimilar from those studied here and the OZIPR
model cannot be run, Section 9 provides guidance on adjusting estimates of aldehyde
concentrations for secondary production.
Section 9 discusses each of these approaches.
Vlll
-------
Section 1
Introduction
Title in of the 1990 Amendments to the Clean Air Act stipulates regulation of chemicals
classified as hazardous air pollutants (HAPs). These are substances that either are known to cause
or are suspected of causing adverse human health effects. The list includes 189 compounds, many
of which are volatile or semivolatile organic compounds. These compounds are found in the
atmosphere as a result of primary emissions and the transformation of organic compounds (both
hazardous and nonhazardous) emitted into the atmosphere by stationary, area, or mobile sources.
Carbonyl compounds represent an important class of organic compounds (Carlier et al., 1986)
found on the list of HAPs. At present, four aldehydes—acetaldehyde, acrolein, formaldehyde, and
propionaldehyde—and three ketones—acetophenone, 2-butanone (i.e., methyl ethyl ketone), and
methyl-/'-butyl ketone—are represented. A unique feature of these compounds is that they are emitted
directly into the atmosphere (by both natural and anthropogenic means) and formed as products from
the oxidation of other emitted hydrocarbons.
Relatively little previous work has been undertaken to evaluate the fraction of ambient
carbonyl compounds due to primary versus secondary processes. Grosjean and coworkers (1983)
undertook a study of the relative production of carbonyl compounds in the metropolitan Los Angeles
area in the early 1980s. Ambient levels as high as 70, 56, and 37 ppbv were measured for
formaldehyde, acetaldehyde, and propionaldehyde, respectively. Changes in carbonyI/carbon
monoxide (CO) ratios determined for individual pollution events allowed the investigators to
estimate that from 44% to 98% of the ambient formaldehyde, acetaldehyde, and other carbonyl
compounds resulted from secondary formation. (CO was assumed to be of primary origin only.)
Modeling studies by these investigators tended to confirm their results.
In another study, Harley and Cass (1994) ran a photochemical air quality model to predict
concentrations for several HAPs, including formaldehyde, acetaldehyde, acrolein, and phenol.
(Modeling results for acetone, which is not a HAP, were also presented.) The study was designed
to follow the secondary production of the HAPs at various points in the Los Angeles air shed. For
the two-day period studied, secondary products were found to contribute significantly to the observed
concentrations of these HAPs, particularly in downwind locations where the ozone concentrations
1
-------
tend to be highest in the afternoon. The results of the model were compared to ambient data obtained
as part of the summer 1987 Southern California Air Quality Study and gave only fair agreement.
Both the Grosjean and Harley and Cass studies are subject to a number of limitations. For
example, in the Grosjean study CO concentrations did not include contributions from secondary
chemistry; thus, the secondary component might have been an underestimate. Second, both studies
were conducted in a single city (Los Angeles) under summertime conditions, which is when
secondary products are expected to be the greatest. The modeling results were compared to
concentrations obtained from individual pollution episodes. Thus, data for seasonal or annual
averages may be difficult to extrapolate, especially for other less polluted areas. Moreover, the
studies were performed during the early to mid-1980s when ambient volatile organic compound
(VOC) emissions were considerably higher than they are today. With the advent of reformulated
gasoline, the composition of fuels has changed in several areas of the country, particularly in severe
nonattainment areas. Finally, photochemical modeling parameters for formaldehyde and other
aldehydes have been improved since the early 1980s. These factors taken together might lead to a
different secondary contribution of carbonyl compounds than found in these earlier studies.
The most accurate means of predicting concentrations and sources of HAPs would be to use
a full three-dimensional dynamic atmospheric model such as the Urban Airshed Model (UAM),
which accounts for emissions, chemistry, and dispersion simultaneously. However, considerable
expertise, and in some cases high costs and computing requirements, make these models burdensome
to run.
The goal of this study is to determine whether a simplified method for modeling secondary
HAPs production can provide reasonable estimates of the levels of these chemicals. The approach
taken in this study is to model secondary HAPs production apart from dispersion. The chemistry is
modeled using the OZIPR program (the research version of the EPA's Ozone Isopleth Plotting
Program). In this study, the OZIPR model is used to estimate the ratio of secondary to primary
formation and the concentrations for each target HAP. The results can be applied in several ways,
as discussed in Section 9. The chemistry and dispersion are not properly coupled in this approach,
but it is considerably simpler to make the calculations in this manner, and the required inputs for
meteorology and emissions are less burdensome.
In this study a simplified approach was developed for modeling the concentrations for a
subset of aldehydes (i.e., formaldehyde, acetaldehyde, and acrolein) in selected study areas to allow
comparison of the primary and secondary formation. While emission inventories exist for these
chemicals, the estimation of atmospheric concentration is complicated by the fact that they are highly
reactive chemically; that is, all three aldehydes are produced and destroyed in the atmosphere on
short time scales. Chemical lifetimes for each of the compounds range between 3 and 6 hours in
daylight urban atmospheres. On balance, the production of these HAPs from other organic species
-------
is expected to outweigh their chemical removal, yielding higher atmospheric concentrations than if
no chemical reactions occurred.
This report provides justification for the approach taken and discusses the modifications to
the base chemical mechanism and other input parameters required to estimate secondary formation
for the compounds of interest. The SAPRC chemical mechanism is modified to include additional
reactions for modeling acrolein. The overall approach to the study is described in Section 2.
Modifications to the chemical mechanism are summarized in Section 3. The distribution of VOCs
used in the OZIPR runs is discussed in Section 4. The methods of preparing emission rates and
meteorological data for input to the model are discussed in Sections 5 and 6, respectively. Section
7 details the method used for distinguishing primary from secondary formation and describes the
preparation of OZIPR input files; a more detailed explanation of the preparation of input files is
given in Appendix A. Section 8 contains the results of the study and discusses their use in adjusting
estimates of primary production for the three aldehydes. Section 9 explains the practical use of the
results obtained from this study.
Finally, a word of caution about this study. EPA is interested in obtaining comparative
annual-average estimates of ground level concentrations of the 188 HAPs that are considered to be
important air toxics. The estimates-to be comparable-must include both concentrations associated
with HAPs directly emitted (primary production) and indirectly produced via atmospheric processes
(secondary production). The Agency also is interested in obtaining these estimates for relatively
large geographical domains, such as urban or metropolitan areas. Obtaining accurate annual average
estimates for pollutants that undergo chemical transformations in the ambient air is very difficult.
This report, then, addresses topics in which there remains a number of significant uncertainties.
Some examples of these uncertainties are: (1) ozone photochemistry in the winter, which usually is
not addressed; and (2) using chemical mechanisms which accurately predict ozone, but with
relatively less attention paid to the ability to predict the carbonyl species. Other difficulties include
temporal differences inherent in short-term modeling of ozone episodes, which is usually done,
versus the need for long-term averages associated with the modeling of point and area sources to
evaluate their chronic cancer and non-cancer potential.
In practice, there is a great incompatibility between the extensive amount of resources needed
to exercise photochemical models, applied to episodes of limited duration, and the problem at hand:
the estimation of annual average concentrations for certain photochemically active chemicals using
relatively unsophisticated emission inventories. Added to this is the context within which results
using the described methodology will be used. That is, results will likely be used in concert with unit
risk factors to estimate the risk of getting cancer if subjected to long term exposure to the
concentrations calculated with the approach. Thus, it is important that the report results are used in
the context within which the described estimates are being made. It is easy to identify features in
-------
the methodology which fall short of the most scientifically rigorous treatment; however, based on
the results of this study, EPA staff believes that it is unlikely that the most rigorous treatment of
secondary-production processes would substantially improve our ability to predict ground-level
HAPs concentration given the current state of scientific knowledge and available data. Thus, it will
take a significant amount of scientific and technical research to substantially reduce uncertainty in
these estimates. Until that work is accomplished, EPA staff believes that it is important to address
the issue using the best information available, and this study utilizes the most current knowledge of
the subject.
-------
Section 2
General Approach for Estimating Secondary Formation of Target HAPs
The approach taken in this study was to separate the modeling of chemistry and dispersion,
focusing on the photochemical formation of secondary HAPs. Chemical transformations of confined
urban and rural atmospheres were modeled using the OZIPR program. In general, any comprehensive
model describing reactive transformation in a large area should couple chemistry and dispersion to
achieve proper spatial and temporal chemical distributions. However, in addition to the computing
power required, chemical and meteorological inputs to such three-dimensional models are quite
demanding, especially in terms of the mass emissions of VOCs as a function of time and location.
While a three-dimensional model might be suitable for a research study of a limited number of
episodes, the simplified approach described here, because of its limited input requirements, is far
more amenable to examining a broad range of cities, seasons, and meteorological conditions.
The OZIPR model is a one-dimensional box model with a time-varying box height (i.e., the
height of the mixing layer). This model was originally designed to predict ozone concentrations;
however, the concentrations of other relatively stable intermediate compounds, especially aldehydes,
are necessarily determined during the course of the calculations. Moreover, since carbonyl
compounds and ozone result mainly from photochemical secondary processes, the model may be
expected to provide reasonable estimates for carbonyl compounds.
The OZIPR model, which is generally run for daylight hours on a single day, calculates solar
radiation from the zenith angle of the sun based on date, location, and time of day. Inputs include
initial concentrations and hourly emission rates for the relevant chemical species and the
meteorological parameters for temperature, humidity, mixing heights, and atmospheric pressure. To
allow flexibility in using the model, the chemical species and reactions are not fixed in the program
code, but are provided as inputs to the model when it is run. The modeling runs take advantage of
this feature by allowing modification to a standard chemical mechanism that includes reactions
affecting the target HAPs, such as acrolein.
-------
The hourly emissions input data are typically obtained from emissions inventories.
Temperature, barometric pressure, and relative humidity are provided by the National Weather
Service. For this study, mixing height data were supplied by EC/R, Inc.
The OZIPR model was designed to be quite flexible, with the user controlling the specific
details of the model run through an input file. All of the model runs done for this study used the same
compiled program code with the only difference between runs found in the input file; the changing
input file reflected, for example, different cities and meteorology. Thus, individual changes to the
scenarios were easily implemented. The OZIPR user's guide (Gery and Grouse, 1990) provides a
more general explanation on how to run the model, and information provided in Section 7 and
Appendix A explains the purpose and structure of the input files and other information used to
estimate the primary and secondary HAP production.
To obtain an adequate representation of conditions across the United States, 10 study areas
were selected: Atlanta, Boston, Chicago, Denver, Houston, Los Angeles, Phoenix, Pittsburgh,
Seattle, and Washington, DC. In each study area, urban and rural counties were selected to allow
reasonably defined emissions data to be used. When multiple counties were selected in one study
area, they were treated as effectively one larger county by calculating average emissions over the
total area. The urban counties were centered on the selected cities; the rural counties were close
enough to the urban areas to have similar weather but different emissions based on their lower
population and distinct land-use patterns. The counties were selected to be representative and did
not include all possible counties in each study area. The land areas of each county were used in
calculating emission factors (mass per unit area per hour). Table 2-1 lists the study areas and
provides geographic information for each.
Up to 48 runs were made for each study area. There were four seasons: winter (December-
February), spring (March-May), summer (June-August); and autumn (September-November). For
each season, two or three typical days, differing in meteorology, were used. To determine primary
and secondary formation, two runs were made for each urban and rural area; one run included
primary emissions of the target HAPs and one did not.
-------
Table 2-1. Counties Selected in Each Study Area
Study Area
Atlanta
Atlanta
Atlanta
Atlanta
Boston
Boston
Boston
Boston
Chicago
Chicago
Denver
Denver
Denver
Houston
Houston
Los Angeles
Los Angeles
Los Angeles
Phoenix
Phoenix
Pittsburgh
Pittsburgh
Seattle
Seattle
Washington, DC
Washington, DC
Washington, DC
Washington, DC
County
Fulton
DeKalb
Cobb
Walton
Norfolk
Suffolk
Plymouth
Rockingham
Cook
McHenry
Denver
Clear Creek
Elbert
Harris
Colorado
Los Angeles
Orange
Ventura
Maricopa
Final
Allegheny
Westmoreland
King
Snohomish
DC
Arlington
Alexandria
Loudoun
State
GA
GA
GA
GA
MA
MA
MA
NH
IL
IL
CO
CO
CO
TX
TX
CA
CA
CA
AZ
AZ
PA
PA
WA
WA
DC
VA
VA
VA
FIPS Code
13121
13089
13067
13297
25021
25025
25023
33015
17031
17111
08031
08019
08039
48201
48089
06037
06059
06111
04013
04021
42003
42129
53033
53061
11001
51013
51510
51107
Type
Urban
Urban
Urban
Rural
Urban
Urban
Rural
Rural
Urban
Rural
Urban
Rural
Rural
Urban
Rural
Urban
Urban
Rural
Urban
Rural
Urban
Rural
Urban
Rural
Urban
Urban
Urban
Rural
Area (km2)
1369.3
694.9
881.2
852.8
1035.0
151.6
1720.9
1800.7
2449.3
1564.7
397.0
1024.2
4793.8
4478.1
2494.3
10515.3
2045.3
4781.0
23838.5
13908.3
1891.3
2648.5
5506.6
5413.6
159.1
67.0
39.6
1346.6
-------
Section 3
Modifications to the Chemical Mechanism
To determine the secondary formation of the aldehydes under consideration, a chemical
mechanism is required to quantitatively describe the transformation of VOCs and NOX to
photochemical products. OZIPR has been structured to permit the use of arbitrary chemical
mechanisms. There are three major chemical mechanisms used in a wide range of oxidant formation
applications in the United States: Carbon Bond IV (CBIV), RADM H, and SAPRC97. In CBIV,
VOCs are grouped together in quasi-chemical units, whereby individual VOCs lose their chemical
identity. Since this application requires individual species, such as acetaldehyde and acrolein, CBIV
would not be appropriate here. SAPRC97 was selected for this study since it is the most up to date
mechanism available; it has been placed in the public domain and allows detailed conversion of
VOCs to products including formaldehyde (HCHO) and acetaldehyde (ALD2). Input parameters to
SAPRC97 are straightforwardly generated from ambient measurements of 6-9 a.m. nonmethane
hydrocarbons. However, modifications to the mechanism are required to include the formation and
loss processes for acrolein.
SAPRC97, developed by Dr. W.P.L. Carter at the University of California-Riverside, is
based largely on the SAPRC93 mechanism and earlier versions, with updates for more accurate rate
constants (Carter, 1990; Carter and Lurmann, 1991; detailed information on the mechanism can be
found at the following Web address: www.cert.ucr.edu/~carter/saprc97.htm.). The SAPRC
mechanism consists of 134 reactions. Of these reactions, the first 38 reactions represent standard
reactions of inorganic species applicable in all tropospheric systems. The balance of reactions
represent reactions of organic radicals and stable organic compounds. Many of the organic
compounds are combined according to the monikers found in Section 4. However, the compounds
of interest in this study retain their identity. In the SAPRC97 mechanism, 22 reactions lead to
formation and 5 reactions lead to removal of formaldehyde (HCHO). Similarly for acetaldehyde
(ALD2), 18 reactions lead to formation and 3 reactions lead to removal. The most important removal
processes for formaldehyde and acetaldehyde are reactive loss with the hydroxyl radical (OH) and
photolytic loss. Removal of acetaldehyde from the atmosphere by reaction with OH is approximately
60% faster than for formaldehyde. By contrast, photolysis is somewhat more significant for
formaldehyde than for acetaldehyde due to formaldehyde's greater absorption and more efficient
-------
quantum yield in the actinic region of the solar spectrum. Photolysis is a function of the sun's zenith
angle and thus is highly dependent on the time of day and the season. However, the mechanism
contains no formation or removal reactions for acrolein.
The mechanism was modified to allow formation and removal of acrolein. Only one
hydrocarbon readily found in the atmosphere, 1,3-butadiene (BDIE in the mechanism), is known to
lead to acrolein formation. The five reactions summarized in Figure 3-1 were added to the
mechanism along with associated rate constants (units: cm3 molecule"1 s"1) in Arrhenius form (A e"
Ea/RT), where A is the pre-exponential factor, Ea is the activation energy for the reaction, R is the gas
constant, and T is the temperature. Rate constants for the OH reaction were based on values taken
from Atkinson (1989).
Figure 3-1. Chemical reactions added to basic mechanism to account for formation and
destruction of acrolein.
{135}
{136}
{137}
{138}
{139}
ACRO -
ACRO -
ACRO -
BDIE -
BDIE -
h OH
h 03
h N03
h OH = ACRO
h 03 = ACRO
#1
#2
#1
#1
#3
.9900E
.8000E
.2000E
.4800E
.3000E
-11;
-18;
-15;
-11 @ 448;
-14 @ -2500;
The extremely rapid reaction of OH with 1,3-butadiene causes reaction 138 to dominate the
removal of 1,3 -butadiene during daylight hours. (Reaction numbers refer to reactions found in Figure
3-2.) The reaction of 1,3-butadiene with the nitrate radical is negligible and is not included in the
mechanism. The stoichiometry for the formation of acrolein is shown to be 1:1 on a molar basis.
While relatively few product studies have been performed to verify this stoichiometry, it would not
be possible to form more than one molecule of acrolein per molecule of 1,3-butadiene reacted.
Products are not assigned for secondary reactions of acrolein (reactions 135-137), since relatively
little reliable information is available from the reaction products from its ozone and radical reactions.
Given the extremely low concentration of acrolein in the system, it is expected that inclusion of
acrolein products would not increase the reactivity of the system. The complete reaction scheme for
the generation and removal of formaldehyde, acetaldehyde, and acrolein (i.e., SAPRC97 with
modifications) is presented in Figure 3-2. An explanation of the terminology used in Figure 3-2 can
be found in Appendix A and the OZIPR user's guide (Gery and Grouse, 1990).
-------
Figure 3-2. Modified SAPRC97 mechanism for determining secondary formation of
formaldehyde, acetaldehyde, and acrolein. This listing represents the contents of the file,
CAL97.MEC, named in the OZIPR input file.
The SAPRC97 mechanism
3CH [CM] >
CNUM=
ALK4 =4.5,
ALK7 = 7.0,
ETHE = 2.0,
PRPE =3.0,
TBUT =4.0,
TOLU = 7.0,
XYLE = 8.0,
TMBZ = 9.0,
HCHO = 1.0,
ALD2 = 2.0,
RCHO = 3.0,
ACRO =3.0,
BDIE =4.0,
NRHC = 1.0;
REACTIONS =
{1} NO2
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
O
0 + N02
0 + N02
NO + O3
NO2 + O3
NO + N03 = 2
NO + NO =2
NO2 + NO3
N2O5
N205 + H20 = 2
N02 + N03
NO3
NO3
03
03
O1D + H2O = 2
O1D
NO + OH
HONO
NO2 + H2O =
NO2 + OH =
HN03 + OH
CO + OH
OH
O3 + OH
NO + H02
N02 + H02
HNO4
HNO4 + OH
03 + H02
H02 + H02
HO2 + HO2 + H2O
NO3 + HO2
N03 + H02 + H20
H202
H2O2 + OH
37a}HO2 + OH
38
39
40
41
42
43
44
45
46
47
R02 + NO
RC03 + NO
RCO3 + NO2
RO2 + HO2
RC03 + H02
R02 + R02
RO2 + RCO3
RCO3 + RCO3
R02R + NO
R02R + H02
, with two added species: 1
NO + O
O3
NO
N03
NO2
NO 3
*N02
*N02
N2O5
NO2 + NO3
*HN03
NO + N02
NO
NO2 + O
0
01D
*OH
O
HONO
NO + OH
HONO + -1* NO2 + HNO3
HNO3
N03
H02
HCHO + RO2R + RO2
HO2
N02 + OH
HN04
NO2 + HO2
NO2
OH
H202
H2O2
HNO3
HN03
= 2*OH
HO2
NO
NO
NO2
HO2
H02
=
=
=
N02 + H02
ROOH
,3-butadiene (BDIE) and acrolein (ACRO).
#0.016667 /LI;
#1.0500E+03 @ -1282.0;
#6.5000E-12 @ -120;
#1.1100E-13 @ -894.0;
#2.0000E-12 @ 1400.0;
#1.4000E-13 @ 2500.0;
#1.7000E-11 @ -150.0;
#8.1500E-20 @ -528.0;
#4.6200E-13 @ -273.0;
#1.3300E+15 @ 11379;
#1.0000E-21;
#2.5000E-14 @ 1228.0;
#0.016667 /Rl;
#0.016667 /R2;
#0.016667E-03 /R3 ;
#0.016667E-03 /R4 ;
#2 . 2000E-10;
#7.2000E+08;
#4.0300E-13 @ -833.0;
#0.016667 /R5;
#4 .OOOOE-24;
#9.5700E-13 @ -737.0;
#6.4500E-15 @ -818;
#2.400E-13;
#28.3^2 @ 1280.0;
#1.6002E-12 @ 941.0;
#3.7000E-12 @ -242.0;
#1.0200E-13 @ -773.0;
#4.3500E+13 @ 10103;
#1.3000E-12 @ -380;
#1.1000E-14 @ 500.0;
# 2.200E-13 @ -619.0;
# 3.100E-34 @ -2818.0;
# 2.200E-13 @ -619.0;
# 3.100E-34 @ -2818.0;
# 0.016667E-03 /R6 ;
# 3. 300 OE- 12 @ 200.0;
# 4.6000E-11 @ -230;
# 4.2000E-12 @ -181.0;
# 4.2000E-12 @ -180.0;
# 2.8000E-12 @ -180.0;
# 3.4000E-13 @ -800;
# 3.4000E-13 @ -800;
# l.OOOOE-15;
# 1.8600E-12 @ -530;
# 2.8000E-12 @ -530;
# 4.2000E-12 @ -181.0;
# 3.4000E-13 @ -800;
10
-------
{48} R02R H
{49} RO2R H
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
ROOH
HCHO
HCHO
HCHO H
HCHO H
HCHO H
ALD2
ALD2 H
ALD2 H
MC03 H
MC03 H
MCO3 H
MCO3 H
MC03 H
PAN
RCHO
RCHO H
RCHO H
PC03 H
PCO3 H
PCO3 H
PC03 H
PC03 H
PPN
ACET
ACET H
MEK
MEK +
78
79
80
GLYX
GLYX H
GLYX H
81
82
83
84
85
86
87
88
89
90
GC03 H
GC03 H
GPAN
GCO3 H
GC03 H
GC03 H
MGLY
MGLY H
MGLY H
ALK4 H
h R02
h RCO3
h OH
h NO3
h H02
=
h OH
h NO3
h NO
h N02
h HO2
h RO2 = 0
h RC03 =
=
=
h OH
h N03
h NO
h NO2
h HO2
h R02 = 0
h RC03 =
=
=
h OH
=
OH = 1
=
h OH
h NO3
h N02
h NO
=
h HO2
h R02 = 0
h RC03 =
=
h OH
h N03
h OH = 0
= 0.5*H02 + R02 # l.OOOOE-15;
= 0.5*HO2 + RCO3 # 1.8600E-12 @ -530;
HO2 + OH # 0.016667E-03 /R7 ;
2*H02 + CO # 0.016667E-03 /R8 ;
CO # 0.016667E-03 /R9 ;
HO2 + CO # 1.125000E-12*2.0 @ -648;
HNO3 + HO2 + CO #2.8000E-12 @ 2516;
R02R + R02 #1.0000E-14;
CO + HCHO + H02 + R02R + RO2 #0 . 016667E-03 /RIO;
MCO3 + RCO3 #5.55000E-12 @ -311;
HNO3 + MCO3 + RCO3 #1.40000E-12 @ 1860;
N02 + HCHO + R02R + RO2 #4.2000E-12 @ -180;
PAN #2.8000E-12 @ -180;
ROOH + HCHO #3.4000E-13 @ -800;
.5*HO2 + HCHO + RO2 #1.8600E-12 @ -530;
H02 + HCHO + RC03 #2.8000E-12 @ -530;
MC03 + N02 + RC03 #2.0000E+16 @ 13542;
ALD2 + HO2 + CO + RO2R + RO2 #0 . 016667E-03 fR.11;
RCO3 + PCO3 #8.5000E-12 @ -252;
HN03 + PC03 + RC03 #1.4000E-12 @ 1860;
N02 + ALD2 + R02R + RO2 #4.2000E-12 @ -180;
PPN #8.4000E-12;
ROOH + ALD2 #3.4000E-13 @ -800;
.5*H02 + ALD2 + RO2 #1.8600E-12 @ -530;
H02 + ALD2 + RC03 #2.8000E-12 @ -530;
PCO3 + NO2 + RCO3 #1.6000E+17 @ 14073;
MCO3 + HCHO + RCO3 + RO2R + RO2 #0 . 016667E-03 /R12 ;
R02R + R02 + MC03 +RCO3 + HCHO #4.81E-13 * 2.0 @ 230
MC03 + ALD2 + RCO3 + RO2R + RO2 #0 . 016667E-03 /R13;
.5*R2O2 + 1.5*RO2 + 0.5*MCO3 + 0.50*ALD2
+ 0.5*HCHO + 0.5*PCO3 + RCO3 #2 . 9200E-13^2 . 0@ -414
0.13*HCHO + 1.87*CO #0 . 016667E-03 /R14 ;
0.6*H02 + 1.2*CO + 0.4*GC03 + 0.4*RCO3 #1.1400E-11;
HNO3 + 0.6*HO2 + 1.2*CO + 0.4*GCO3
+ RCO3 #1.4000E-12 @ 1860;
GPAN #2.8000E-12 @ -180.0;
N02 + H02 + CO #4.2000E-12 @ -180.0;
GCO3 + NO2 + RCO3 #2.00E+16 @ 13542.0;
ROOH + CO #3.40E-13 @ -800;
.5*H02 + CO + R02 #1.8600E-12 @ -530;
H02 + CO + RC03 #2.8000E-12 @ -530;
MCO3 + HO2 + CO + RCO3 #0 . 016667E-03 /R15;
MCO3 + CO + RCO3 #1.7200E-11;
HN03 + MC03 + CO + RCO3 #1.4200E-12 @ 1860;
.19*HCHO + 0.31*ALD2 + 0.170*RCHO + 0.34*ACET + 0.44*MEK
+ 0.070*RO2N + 0.930*RO2R + 0.600*R2O2 + 1.600*RO2 #1.05E-11 @ 353;
{91} ALK7 H
+
{92} ALKN H
93
94
95
96
97
98
99
{100
101
102
103
104
RO2N H
R02N H
R02N H
RO2N H
R2O2 H
R202 H
R202 H
R2O2 H
ETHE H
ETHE H
ETHE H
ETHE H
'105} PRPE H
|l06} PRPE H
^107} PRPE H
108
109
110
PRPE H
TBUT H
TBUT H
111
112
113
TBUT H
TBUT H
TOLU H
H
h OH = 0
0.180*R02N
h OH
h NO
h H02
h R02
h RCO3 =
h NO
h H02
h R02
h RCO3 =
h OH
h 03
h 0
h NO3
h OH
h 03 = 0
h O = 0
h N03
h OH
h O3
h 0
h N03
h OH = 0
h 0.840*RO2
.02*HCHO + 0.03*ALD2 + 0.25*RCHO + 0.36*ACET + 0.88*MEK
+ 0.820*R02R + 0.840*R202 + 1.840*RO2 #1.62E-11 @ 288;
N02 + 0.155*MEK + 1.05*RCHO + 0.48*ALD2 + 0.16*HCHO
+ 1.390*R2O2 + 1.390*RO2 #2.19E-11 @ 708;
ALKN #4.20E-12 @ -181;
ROOH + MEK #3.4000E-13 @ -800;
R02 + 0.5*H02 + MEK #1.0000E-15;
RCO3 + 0.5*HO2 + MEK #1.8600E-12 @ -530;
NO2 #4.2000E-12 @ -181;
ROOH #3.4000E-13 @ -800;
R02 #1.0000E-15;
RCO3 #1.8600E-12 @ -530;
RO2R + RO2 + 1.56*HCHO + 0.22*ALD2 #1.9600E-12 @ -438;
HCHO + 0.12*H02 + 0.42*CO #9.1400E-15 @ 2580;
HCHO + H02 + CO + R02R + RO2 #1.0400E-11 @ 792;
NO2 + 2*HCHO + R2O2 + RO2 #5.4300E-12 @ 3041;
RO2R + HCHO + ALD2 + RO2 #4.8500E-12 @ -504;
.65*HCHO + 0.5*ALD2 + 0.285*CO + 0.06*OH + 0.165*HO2
+ 0.135*R02R + 0.135*R02 #1.32E-14 @ 2105.0;
.6*ACET + 0.4*HCHO + 0.2*ALD2 + 0.2*HO2
+ 0.6*RO2R + 0.4*CO + 0.6*RO2 #1.18E-11 @ 324;
N02 + HCHO + ALD2 + R2O2 + RO2 #5.00E-12 @ 1935;
R02R + 2*ALD2 + RO2 #1.0100E-11 @-549;
ALD2 + 0.15*CO + 0.27*RO2R + 0.12*OH + 0.21*HO2
+ 0.270*RO2 + 0.300*HCHO #9.08E-15 @ 1137;
MEK + 0.4*H02 #2.2600E-11 @ -10;
N02 + 2*ALD2 + R2O2 + RO2 #1.0000E-11 @ 975;
.16*CRES + 0.16*HO2 + 0.84*RO2R + 0.4*DIAL
+ 0.144*MGLY + 0.114*GLYX #2.1E-12 @ -322.0;
11
-------
'114} DIAL H
|ll5} DIAL
'll6} XYLE H
'117} TMBZ H
h OH
h OH
h OH
PC03 + RC03
HO2 + CO + MCO3 +
= 0.17*CRES + 0.17*HO2 + 0
+ 0.65*DIAL + 0.316*MGLY
= 0.17*CRES + 0.17*H02 + 0
RCO3
#3.0000E-11;
#0.016667E-03 /R16;
83*RO2R + 0.83*RO2
+ 0.095*GLYX
#1.66E
-11
@
-116;
83*R02R + 0.83*R02
+ 0.49*DIAL + 0.86*MGLY
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
135
136
137
CRES H
CRES H
BZO H
BZO H
BZO
PHEN H
PHEN H
NPHE H
BZN2 H
BZN2 H
BZN2
RO2P H
RO2P H
R02P H
R02P H
NRHC
ACRO H
ACRO H
ACRO H
!138} BDIE H
|l39} BDIE H
< (MECH)
h OH
h N03
h N02
h HO2
h OH
h N03
h NO3
h NO2
h H02
h NO
h HO2
h R02
h RC03
h OH
h 03
h NO3
h OH
h 03
= 0.2*MGLY + 0.15*RO2P + 0
HN03 + BZO
NPHE
PHEN
PHEN
= 0.2*GLYX + 0.15*R02P + 0
HN03 + BZO
HNO3 + BZN2
=
NPHE
NPHE
NPHE
ROOH
= 0.5*H02 + R02
= 0.5*H02 + RC03
NRHC
=
=
=
= ACRO
= ACRO
85*RO2R + RO2
#2.
#1.
#3.
#1.
85*R02R + R02
#1.
#3.
#1.
#4.
#3.
#1.
#1.
#1.
#1.
#2.
#1.
#1.
#3.
#6.2E-
11;
#4 .2000E
1000E-
3000E-
4000E-
OE-03;
#2.
#3.
#3.
3000E-
4000E-
OOOOE-
2000E-
4000E-
OOOOE-
8600E-
OOOOE+
9900E-
8000E-
2000E-
4800E-
3000E-
11;
11
13
-11
@ -
@ -
6300E-
6000E-
6000E-
11
13
03;
12
13
15;
12
00;
11;
18;
15;
11
14
@ -
@ -
@ -
@ -
@ -
;
300;
800;
11;
12;
12;
300;
800;
181.0;
800;
530;
@ 448;
@ -
2500;
12
-------
Section 4
VOC Species—Reactivity Parameters
For the model to predict city-specific secondary formation of formaldehyde, acetaldehyde,
and acrolein, a distribution of the VOCs emitted into the atmosphere in the selected urban areas was
required; data collected in the mid-1980s were used as the basis for establishing the distribution. This
study also used data found in the EPA report by Seila and others (1989). The distribution of
hydrocarbons has probably not changed substantially in the past 10 years, with the exception of a
moderate increase in the level of oxygenated hydrocarbons due to use of reformulated fuels. In the
study by Seila et al. (1989), identifications were made for 34 paraffins (ethane to w-decane), 14
olefins (ethene to oc-pinene, including 1,3-butadiene and isoprene), 14 aromatic hydrocarbons
(benzene to l,2-dimethyl-3-ethylbenzene), and acetylene. Beyond these specific compounds, those
investigators adopted techniques to classify the remaining unidentified peaks in terms of their
hydrocarbon class (i.e., paraffin, olefin, or aromatic). Since flame ionization detection, which has
a constant carbon response for all hydrocarbons, was employed for quantification, concentrations
could be assigned to compounds for which there was no unique identity. Thus, concentrations for
unidentified chromatographic peaks could be attributed to the appropriate class of paraffins, olefins,
or aromatics. The report also provided summations for the total sum of paraffins, olefins, aromatics,
acetylene, and unidentified hydrocarbons. From the report, data were obtained for 6 of the 10 cities.
For these cities, the number of chromatographic measurements during the 6-9 a.m. time frame are
given as follows: Atlanta (21 measurements), Boston (8), Chicago (10), Denver (13), Houston (36),
and Washington, DC (32).
Data for Los Angeles were obtained from an analysis done by the California Air Resources
Board during the 1997 Southern California Air Quality Study (Lonneman, personal communication).
For the remaining three study areas (Pittsburgh, Phoenix, and Seattle), there are no known sources
of city-specific VOC speciation available, and, therefore, cities of similar characteristics, for which
data are available, were used. The following substitutions were made: Akron for Pittsburgh, Salt
Lake City for Phoenix, and Boston for Seattle.
Given the information available for these study areas, it was possible to classify virtually the
entire carbon mass of VOCs according to the SAPRC97 mechanism classification scheme. This
13
-------
scheme combines all organic compounds into 14 groups, given below, which account for the total
VOC mass.
ALK4 - low molecular weight alkanes
ALK7 - high molecular weight alkanes
ETHE - ethene
PRPE - propene
BDIE - butadiene
TBUT - high molecular weight alkenes
TOLU - low molecular weight reactive aromatic hydrocarbons
XYLE - intermediate molecular weight aromatic hydrocarbons
TMBZ - high molecular weight aromatic hydrocarbons
HCHO - formaldehyde
ALD2 - acetaldehyde
RCHO - high molecular weight carbonyl compounds
ACRO - acrolein
NRHC - nonreactive hydrocarbons
The groups were generated according to the chemical characteristics of similar types of
compounds. The main criteria for grouping were reaction rate and the general identity of the
chemical products. For example, ALK4 represents C4 and C5 alkanes, two of which are w-butane and
isopentane. These compounds have rate constants of approximately 3 x 10"12 cnr'molec"1 s"1. The
products from the reactions, while not identical, give compounds having similar OH reactivity. On
the other hand, ALK7 compounds generally have higher OH rate constants and potentially give
different types of products than those formed from ALK4 compounds. For example, w-hexane can
give hydroxy products that result from isomerization reactions.
The procedure for classifying the hydrocarbons into the 14 groups is as follows: Except for
Seattle, each city has a unique distribution. For the six cities with hydrocarbon data from the study
by Seila et al. (1989), an average distribution of the hydrocarbons was calculated from the tabular
hydrocarbon data for each city. Each individual chromatogram making up the average was then
compared with the average distribution for that city. The chromatogram showing the least deviation
in the components was then used as the basis for determining the distribution for that city. (This
14
-------
approach was taken because there was no electronic version of the data available.) The following
combinations were then used to place the hydrocarbons into their appropriate groups:
ALK4 - The low molecular weight hydrocarbons were taken to be the sum of the
concentrations for isobutane, w-butane, isopentane, and w-pentane.
ALK7 - The high molecular weight alkanes were taken as the sum of the paraffin
concentration (provided in the report) minus the concentrations for ethane, propane,
isobutane, w-butane, isopentane, and w-pentane. To this value was added one-half of the
unidentified hydrocarbon concentration.
ETHE - The ethene concentration was used.
PRPE - The propene concentration was used.
BDIE - The 1,3-butadiene concentration was used.
TBUT - The high molecular weight alkenes (>C3) were taken as the sum of the olefm
concentration minus the concentrations for ethene, propene, and 1,3-butadiene. Isoprene was
included in this classification.
TOLU - The low molecular weight aromatic hydrocarbons were taken to be the sum of the
toluene and ethylbenzene concentrations. Ethylbenzene was included in this classification
rather than with XYLE since its OH rate constant is closer to toluene than the xylene
isomers.
XYLE - The intermediate weight aromatic hydrocarbons include (m+/>)-xylene and o-xylene
concentrations.
TMBZ - The high molecular weight aromatic hydrocarbons were taken as the total aromatic
hydrocarbon concentration minus the concentrations for benzene, toluene, ethylbenzene, (m
+/>)-xylene, and o-xylene. To this value was added one-half of the unidentified hydrocarbon
concentration.
HCHO - The formaldehyde concentration was taken to represent 2% of the total
hydrocarbons by carbon fraction.
ALD2 - The acetaldehyde concentration was taken to represent 2% of the total hydrocarbons
by carbon fraction.
RCHO - The high molecular weight carbonyl concentration (>C2) was taken to be 0.9% of
the total hydrocarbons by carbon fraction.
ACRO - The acrolein concentration was taken to represent 0.1% of the total hydrocarbons
15
-------
by carbon fraction.
NRHC - The nonreactive hydrocarbons were taken to be the sum of the concentrations for
acetylene, ethane, propane, and benzene. These compounds all have OH rate constants on
the order of 1 x 10"12 cm3 molec"1 s"1 or less.
To account for the 5% carbonyl concentration, the hydrocarbon distribution was scaled by
a factor of 0.95. Finally the individual reactivity components were summed to a precision of 0.01
to ensure that the total distribution gave unity. Adjustments were not required to account for round-
off error. The resultant reactivity factors on a city-by-city basis are given in Table 4-1.
The Los Angeles data were calculated differently because hydrocarbon data were available
from a different database that contained a finer speciation. Similar procedures as above were used
to combine the individual compounds into the 14 groupings of VOCs. The Los Angeles
chromatograms also had a 15% oxygenate component (e.g., alcohols, ethers, esters). Since these are
slowly reacting compounds, they were divided into the ALK4 and ALK7 classifications. The Los
Angeles data set also provided concentrations for carbonyl compounds. The total aldehyde
component was found to be 5.1%, a value consistent with our previous assumption. For consistency
with the other cities, the aldehyde component for Los Angeles was distributed similarly to that
described above.
16
-------
Table 4-1. Reactivity Factors Used for the 10 Cities Under Consideration. All values are in percent.
Parameter
ALK4
ALK7
ETHE
PRPE
BDIE
TBUT
TOLU
XYLE
TMBZ
HCHO
ALD2
RCHO
ACRO
NRHC
Atlanta
21.74
21.26
3.28
1.13
0.11
9.51
8.33
5.19
14.21
2
2
0.9
0.1
10.24
Boston
19.7
20.9
4.35
1.31
0.23
10.62
9.04
5.44
11.89
2
2
0.9
0.1
11.52
Chicago
17.14
27.2
2.28
1.46
0.24
12
8.51
5.29
14.14
2
2
0.9
0.1
6.74
Denver
24.61
24.69
4.85
1.05
0.15
6.28
7.86
4.76
9.48
2
2
0.9
0.1
11.27
Houston
27.94
20.73
3.56
1.91
0.25
7.21
5.11
3.33
10.18
2
2
0.9
0.1
14.78
Los Angeles
16.68
31.99
2.72
4.86
0.27
5.8
5.3
3.86
8.87
2
2
0.9
0.1
14.59
Phoenix
17.14
28.71
3.46
1.35
0.3
7.46
8.04
5.62
9.26
2
2
0.9
0.1
13.43
Pittsburgh
18.54
21.73
3.8
1.1
0.24
11.67
5.74
4.36
11.12
2
2
0.9
0.1
16.7
Seattle
19.7
20.9
4.35
1.31
0.23
10.62
9.04
5.44
11.89
2
2
0.9
0.1
11.52
Washington
19.1
21.64
4.19
1.67
0.35
8.26
10.34
10.01
8.96
2
2
0.9
0.1
10.48
-------
Section 5
Emissions Data and Boundary Conditions
Hourly emissions data (in units of kg km"2 h"1) were provided as inputs to the OZIPR model
for VOCs, NOX, CO, formaldehyde, acetaldehyde, acrolein, and 1,3-butadiene. The total VOC mass
is partitioned into species according to the fractions given in the reactivity table, which is part of the
OZIPR input file (see Section 7 or Appendix A for details). The emissions data consist of two parts:
anthropogenic and biogenic emissions. For the present study, the anthropogenic emissions came
from emissions inventories, while the biogenics came from running the BEIS-2 (Biogenic Emissions
Inventory System) model. Both types of emissions data were supplied by DynTel Corporation.
The anthropogenic emissions data consisted of 24-hour values for each of four seasons for
each pollutant and county. The units were kilograms per hour for total VOCs and moles per hour for
other pollutants (CO, NO, NO2, formaldehyde, acetaldehyde, acrolein, and 1,3-butadiene). The
pollutant NOX in OZIPR represents the sum of NO and NO2. The data were converted to emission
densities (kg km"2 h"1) by dividing by the area of each county (see Table 2-1). The biogenic emissions
were in the form of hourly emissions for each day of the year, for each of four species—isoprene,
monoterpenes, other VOCs, and NO. Seasonal averages were calculated for each species. The first
three species were added to the anthropogenic VOC emissions to obtain the total VOC emissions,
while the NO was added to the anthropogenic NOX emissions to obtain the total NOxemissions. All
emissions data were provided as county-wide average rates; in the cases where there were two or
more counties of the same type (urban or rural) in a given study area, multicounty averages
(weighted by area) were used.
The emissions data are input to OZIPR via the MASS option, which is described in Section
7 and also in Appendix A. The OZIPR model expects a series of hourly emission rates; for the
current study the first hour is for 8-9 a.m. and the twelfth (and last) is for 7-8 p.m.
The value provided for each hour represents the average emission rate for that hour across all the
days in the given season. The units required for the MASS option are kg km"2 h"1.
In addition to the hourly emissions, the initial conditions must be specified. This consists of
both the concentrations inside the box (the volume of air being modeled) and the concentrations in
18
-------
the aloft air. The aloft air is the air mass above the box being modeled; when the mixing height
increases, some of the aloft air is mixed into the box. In the present set of runs, four species are
initialized aloft: CO, ozone, VOC, and NOX. Inside the box, nine species are specified—the four
aloft species, the three target HAPs, 1-3 butadiene, and nitrous oxide (HONO). The initial
concentrations of the target HAPs were set to fixed fractions of the total VOCs. The initial
concentrations (both in the box and aloft) for total VOCs and NOX are based on concentrations
extracted from runs of the Urban Airshed Model (the lowest level was used for the surface
concentrations, and an upper level for the aloft concentrations). This model was run only for the
summer; it was assumed that the VOC and NOX concentrations are similar in other seasons. Urban
surface-level CO was varied seasonally, based on data for 8 a.m. measurements extracted from the
AIRS database. There were no CO monitors in any of the rural counties; initial (8 a.m.)
concentrations were set to 0.2 ppm. Aloft CO was also set to a level of 0.2 ppm. Surface-level
HONO was set to a value of 0.001 ppm. Aloft ozone was seasonally varied, based on surface-level
measurements from the AIRS database. Inside the box, initial ozone levels were set to zero, as ozone
levels are depleted throughout the night by surface contact and do not have much effect on the early
morning chemistry.
The OZIPR model is more sensitive to hourly emissions than to initial concentrations. In test
runs, even relatively large changes in initial concentrations (a factor of 2, for example) result in fairly
small changes (10% or less) in both ozone and HAP concentrations late in the day. However, the
OZIPR model is quite sensitive to meteorological inputs.
19
-------
Section 6
Meteorological Data
The OZIPR model requires hourly temperature and humidity data, along with daily average
pressure and twice-daily mixing heights. Temperature, relative humidity, and atmospheric pressure
data were taken from values reported by the National Weather Service (NWS) at airport monitoring
sites for each city. Hourly data from the years 1992-1995, inclusive, were used. Station identification
numbers are supplied in Table 6-1.
Mixing height data were supplied by EC/R, Inc. The data were generated from measurements
obtained from daily morning and afternoon weather balloon soundings; at all stations, observations
were made at midnight and noon, Greenwich mean time. As with the other meteorological data, the
years 1992-1995 were used. The mixing height stations are indicated in Table 6-2.
Table 6-1. NWS Stations Used for OZIPR Meteorological Input
City (Airport) Station Number
Atlanta 13874
Boston 14739
Chicago 94846
Denver (Stapleton) 23062
Houston 12960
Los Angeles 23174
Phoenix 23183
Pittsburgh 94823
Seattle 24233
Washington, DC (National) 13743
20
-------
Table 6-2. Mixing Height Stations Used for OZIPR Meteorological Input
City
Atlanta
Atlanta
Boston
Chicago
Denver
Houston
Los Angeles
Phoenix
Pittsburgh
Seattle
Washington, DC
Station
Athens, GA
Peachtree City, GA
Chatham, MA
Peoria, IL
Denver, CO
Lake Charles, LA
San Diego, CA
Tucson, AZ
Pittsburgh, PA
Quillayute, WA
Sterling, VA
Station Number
13873 (1992-1994)
53819(1995)
14684
14842
23062
03937
03190
23160
94823
94240
93734
6.1 Approach to the Utilization of Meteorological Data with OZIPR
The desired outputs from this study were annual and seasonal estimates of secondary
production of the HAPs of interest. The limited time resolution (at best, seasonal) of the input
emissions data precluded running OZIPR on a daily basis. For these reasons, it was decided to run
the model on a seasonal basis; however, even within a given season meteorological conditions can
vary substantially. These shifting weather conditions, in turn, can affect the secondary production
of HAPs. Since estimates were to be applied only over long time periods, the focus was on common
or typical meteorological conditions.
The seasons were defined as
• Winter - December, January, and February
• Spring - March, April, and May)
• Summer - June, July, and August
• Autumn - September, October, and November
21
-------
To accommodate the variability within seasons while at the same time avoiding excessive
calculations, it was decided to use at most three typical days per season. To identify the more
common types of days, cluster analysis was carried out on a seasonal basis on the meteorological
variables. To be counted as a typical day, a cluster had to contain at least 30 days (out of
approximately 365 days [4 years with three months each]). The OZIPR model was then run for each
of the typical days in each season, and afterwards seasonal averages of secondary HAP production
were calculated by averaging the results for the typical days weighted by their frequency of
occurrence (i.e., the number of days in each cluster). The averaging could also be done by dividing
each season into more, smaller clusters, but this would require more time and effort and would be
of limited value since other inputs (particularly emissions) were not available on a daily basis.
6.2 Typical Day Determination
The following algorithm describes how typical days were determined for each season to
supply the meteorological input to OZIPR. Before clustering days together, the hourly meteorological
data need to be expressed on a daily basis. To this end, the arithmetic mean for temperature,
humidity, and pressure for each day for the hours of 8 a.m. through 8 p.m., inclusive, were
calculated. Only averages computed from at least 9 hours were retained. Also, the difference (p.m.
minus a.m.), the ratio (p.m./a.m.), and the average of the mixing heights were calculated for each
day; both mixing heights had to be available for a given day to enter the typical day calculations. Any
mixing height originally furnished as less than 100 meters was reset to 100 meters before being used.
Cluster analysis was executed for each city-season combination for three separate sets of
variables: temperature, humidity, and average mixing height; temperature, humidity, and mixing
height difference; and temperature, humidity, and mixing height ratio. The clustering procedure
measured the distance between clusters as the average of the Euclidean distance between all possible
pairs of points. Clustering was intended to find groups of days that resembled each other, but were
distinct from days in other clusters.
Generally, the triplet containing the ratio of the afternoon to morning mixing heights yielded
better clustering results in the sense that more clusters contained more than 30 days and the cluster
sizes were more evenly distributed. In a few cases, the cluster results from either the mixing height
averages or differences were used. In some cases, the bulk of the data naturally divided into only two
clusters; for these, the "extra" cluster (needed to make three) consisted of a few outlying points
representing relatively rare types of weather.
Once the clusters were determined, data for each typical day had to be constructed for input
to the model. The model required hourly temperature and humidity numbers, a single daily
atmospheric pressure value, and one morning and one afternoon mixing height. To prepare this input
22
-------
data, the median temperature, humidity, and pressure were determined for each hour within each
season and cluster. The medians of the averages, differences, and ratios of the mixing heights, as
well as the medians of the mixing heights themselves, were also determined for each cluster within
the seasons. Subsequently, the mean of the median pressures for each cluster within the seasons was
calculated. These summary statistics then constituted the descriptors of the typical days, and the
median hourly temperatures and humidity levels, the mean of the hourly median atmospheric
pressures, and the median morning and afternoon mixing heights were supplied to the model as
input. All statistical calculations were done via the SAS programming language (SAS, 1990).
Tables 6-3 and 6-4 describe the typical conditions for the clusters obtained for each city and
season. The temperature and humidity values in Table 6-4 are from the middle of the day. (OZIPR
is a photochemical model, so midday temperature and humidity values are of particular interest.)
Table 6-3. Typical Conditions for Each City and Season
A. Atlanta
Season
Autumn
Spring
Summer
Winter
Days
in Cluster Comments on Midday Conditions
213 Upper 70s, moderate humidity, significant mixing
89 Mid to upper 60s, high humidity, no mixing (ht. ~ 500 m)
37 Mid 60s, low humidity, significant mixing
225 Low 70s, low humidity, significant mixing
76 Low 70s, moderate humidity, significant mixing
52 Low 60s, high humidity, little mixing (ht. -500 m)
204 Mid 80s, moderate humidity, significant mixing
120 Low 90s, moderate to low humidity, very much mixing (max. ht. -2200 m)
31 Mid 70s, high humidity, mixing height contraction
137 Upper 40s to low 50s, low humidity, some mixing (max. ht. -1000 m)
116 Upper 40s to low 50s, moderate humidity, slight mixing between -400 and
-700 m
90 Upper 40s to low 50s, high humidity, no mixing (-400 m)
23
-------
Table 6-3. Continued
B. Boston
Season
Autumn
Spring
Summer
Winter
Days
in Cluster
210
94
123
120
102
134
126
94
109
90
54
Comments on Midday Conditions
Upper 50s to low 60s, moderate to low humidity, some mixing (max. ht.
~1000m)
Low 60s, high humidity, no mixing (-400 m)
Low 50s, moderate humidity, some mixing (max. ht. -1000 m)
Upper 40s, high humidity, no mixing (-300 m)
Low 50s, low humidity, noticeable mixing from high morning level (-1000 to
-1800m)
Upper 70s to low 80s, low humidity, significant mixing
Upper 70s to low 80s, moderate humidity, noticeable mixing
Low 70s, high humidity, no mixing (-500 m)
Upper 30s, moderate humidity, little mixing
Upper 20s, low humidity, no mixing (-1000 m)
Low 40s, high humidity, no mixing (-300 m)
C. Chicago
Season
Autumn
Spring
Summer
Winter
Days
in Cluster Comments on Midday Conditions
178 Low 60s, moderate humidity, significant mixing
141 Low 50s, moderate to high humidity, no mixing (-600 m)
250 Mid 50s, moderate humidity, significant mixing
92 Mid 40s, high humidity, no mixing (-700 m)
205 Upper 70s, moderate humidity, significant mixing
151 Upper 70s, low to moderate humidity, much mixing (-100 to-1700 m)
258 Low 30s, moderate to high humidity, no mixing (-500 m)
66 Mid 20s, moderate humidity, little mixing
D. Denver
Season
Autumn
Spring
Summer
Winter
Days
in Cluster Comments on Midday Conditions
124 Upper 50s, low humidity, much mixing (-100 to -1700 m)
81 Upper 70s, very low humidity, very much mixing (-100 to -3300 m)
55 Low 40s, moderate humidity, little mixing
132 Upper 60s, very low humidity, very much mixing (-100 to -3300 m)
97 Upper 50s, low humidity, much mixing (-500 to -2500 m)
41 Mid 40s, moderate humidity, little mixing
158 Upper 70s, low humidity, much mixing (-300 to -2300 m)
96 Upper 80s, very low humidity, very much mixing (-100 to -4000 m)
203 Upper 40s, low humidity, much mixing (-100 to -1400 m)
100 Low 30s, moderate humidity, little mixing
24
-------
Table 6-3. Continued
E. Houston
Season
Autumn
Spring
Summer
Winter
Days
in Cluster
174
107
65
209
126
298
34
202
124
Comments on Midday Conditions
Mid 80s, moderate humidity, significant mixing
Lower 70s, low to moderate humidity, noticeable mixing
Lower 70s, high humidity, little mixing
Mid 70s, moderate humidity, some mixing
Mid 70s, moderate humidity, noticeable mixing
Low 90s, moderate humidity, noticeable mixing (
Upper 70s to low 80s, high humidity, no mixing
Mid 60s, moderate to high humidity, little mixing
Upper 50s, moderate humidity, some mixing
-600 to ~1400m)
F. Los Angeles
Season
Autumn
Spring
Summer
Winter
Days
in Cluster Comments on Midday Conditions
296 Low 70s, moderate humidity, no mixing (-500 m)
63 Low 70s, low humidity, significant mixing
281 Mid 60s, moderate humidity, no mixing (-700 m)
85 Mid 60s, moderate humidity, little mixing (-1400 m)
304 Low to mid 70s, moderate humidity, no mixing (-400 m)
51 Low 70s, moderate humidity, no mixing (-1000 m)
152 Low 60s, moderate humidity, noticeable mixing
131 Low 60s, moderate to high humidity, no mixing (-600 m)
69 Upper 60s, low humidity, significant mixing
G. Phoenix
Season
Autumn
Spring
Summer
Winter
Days
in Cluster Comments on Midday Conditions
283 Upper 80s, low humidity, very much mixing (-400 to -3000 m)
75 Low 70s, moderate humidity, some mixing
230 Upper 70s, low humidity, much mixing (-1000 to -3100 m)
89 Upper 80s, very low humidity, very much mixing (-400 to -4100 m)
44 Mid 60s, moderate humidity, little mixing (-2000 m)
229 -100, very low humidity, very much mixing (-1200 to -4400 m)
114 Mid 90s, low humidity, significant mixing (-2200 to -3400 m)
226 Mid 60s, low humidity, significant mixing
83 Upper 50s, moderate humidity, some mixing (-1100 to -1600 m)
49 Upper 50s, moderate to high humidity, no mixing (-1400 m)
25
-------
Table 6-3. Continued
H. Pittsburgh
Season
Autumn
Spring
Summer
Winter
Days
in Cluster
126
107
114
90
60
125
70
66
147
87
Comments on Midday Conditions
Lower 60s, moderate humidity, significant mixing
Upper 50s, moderate humidity, some mixing
Mid 60s, low humidity, significant mixing
Mid 50s, moderate humidity, noticeable mixing
Mid 40s, high humidity, little mixing
Low 80s, moderate humidity, significant mixing
Low 80s, low to moderate humidity, very much mixing i
Mid 70s, moderate humidity, some mixing
-100 to 2000 m)
Mid 30s, moderate to high humidity, mixed layer contraction (-700 to -500 m)
Low 30s, moderate humidity, little mixing
I. Seattle
Season
Autumn
Spring
Summer
Winter
Days
in Cluster Comments on Midday Conditions
227 Mid 50s, high humidity, no mixing
118 Mid 60s, moderate humidity, some mixing
238 Mid 50s, moderate humidity, some mixing
122 Mid 60s, moderate humidity, significant mixing
203 Upper 60s, moderate humidity, noticeable mixing (-700 to -1400 m)
133 Mid 60s, moderate humidity, noticeable mixing (-1200 to -1900 m)
277 Mid 40s, high humidity, mixed layer contraction (-600 to -300 m)
49 Mid 40s, moderate humidity, some mixing
J. Washington, DC
Days
in Cluster
Season
Autumn
Spring
Summer
Winter
241
101
182
98
75
151
115
84
203
67
62
Comments on Midday Conditions
Mid 60s, moderate humidity, noticeable mixing
Upper 60s, high humidity, no mixing
Low 60s, low to moderate humidity, significant mixing
Upper 60s, moderate humidity, noticeable mixing
Mid 50s, high humidity, no mixing
Mid 80s, moderate humidity, significant mixing
Mid 80s, moderate humidity, significant mixing
Upper 70s, moderate to high humidity, little mixing
Low 40s, moderate humidity, little mixing
Mid 30s, low humidity, no mixing (-1300 m)
Low 40s, high humidity, no mixing (-500 m)
26
-------
Table 6-4. Results of Cluster Analysis for Each City and Season
A. Atlanta
Season
Autumn
Spring
Summer
Winter
No. Days
213
89
37
225
76
52
204
120
31
137
116
90
Temperature (°F)
71
67
66
72
71
62
84
90
75
49
49
50
Humidity (%)
55
89
32
40
61
88
62
46
89
39
64
94
Mixing Height
Difference (m)
250-1300
500-500
100-1250
150-1800
500-1450
500-600
550-1650
200-2250
900-600
250-950
400-700
400-400
B. Boston
Season
Autumn
Spring
Summer
Winter
No. Days
210
94
123
120
102
134
126
94
109
90
54
Temperature (°F)
59
62
52
48
53
79
80
71
38
27
41
Humidity (%)
45
82
55
83
33
40
56
80
55
37
85
Mixing Height
Difference (m)
600-1050
450-400
500-1000
350-300
1000-1800
550-1650
550-1400
500-450
700-850
950-1000
300-300
27
-------
Table 6-4. Continued
C. Chicago
Season
Autumn
Spring
Summer
Winter
No. Days
178
141
250
92
205
151
258
66
Temperature (°F)
60
51
56
46
78
79
32
25
Humidity (%)
50
75
46
83
66
46
75
49
Mixing Height
Difference (m)
200-1150
600-650
300-1500
700-700
400-1300
150-1700
450-500
400-700
D. Denver
Season
Autumn
Spring
Summer
Winter
No. Days
124
81
55
132
97
41
158
96
203
100
Temperature (°F)
59
79
40
68
57
44
78
88
48
30
Humidity (%)
33
18
70
20
38
72
37
19
29
62
Mixing Height
Difference (m)
150-1650
100-3300
700-800
100-3350
550-2450
900-800
350-2350
100-4000
100-1500
300-500
28
-------
Table 6-4. Continued
E. Houston
Season No. Days
Autumn
Spring
Summer
Winter
174
107
65
209
126
298
34
202
124
Temperature (°F)
84
72
70
77
74
91
79
63
58
Humidity (%)
58
40
83
69
48
55
87
77
44
Mixing Height
Difference (m)
250-1250
200-1050
450-700
600-1100
400-1150
650-1450
850-850
350-600
350-900
F. Los Angeles
Season No. Days
Autumn
Spring
Summer
Winter
296
63
281
85
304
51
152
131
69
Temperature (°F)
71
72
66
63
73
72
62
60
68
Humidity (%)
66
24
70
65
68
65
52
77
21
Mixing Height
Difference (m)
450-500
100-1300
700-750
1350-1500
450-450
950-950
250-1050
700-600
100-1250
29
-------
Table 6-4. Continued
G. Phoenix
Season
Autumn
Spring
Summer
Winter
No. Days
283
75
230
89
44
229
114
226
83
49
Temperature (°F)
87
71
78
88
66
100
95
64
59
59
Humidity (%)
22
44
25
17
48
18
31
30
52
76
Mixing Height
Difference (m)
450-3000
900-1600
1000-3100
400-4150
2050-2250
1250-4400
2200-3450
300-1700
1100-1650
1400-1400
H. Pittsburgh
Season
Autumn
Spring
Summer
Winter
No. Days
126
107
114
90
60
125
70
66
147
87
Temperature (°F)
62
59
63
54
45
80
81
74
36
30
Humidity (%)
48
69
39
60
85
56
41
74
77
54
Mixing Height
Difference (m)
200-1400
700-1000
250-1900
750-1450
750-650
450-1800
100-2000
650-1200
700-550
750-950
30
-------
Table 6-4. Continued
I. Seattle
Season No. Days
Autumn
Spring
Summer
Winter
J. Washington,
227
118
238
122
203
133
277
49
DC
Season No. Days
Autumn
Spring
Summer
Winter
241
101
182
98
75
151
115
84
203
67
62
Temperature (°F)
54
66
54
63
69
66
46
45
Temperature (°F)
63
67
62
67
54
86
84
79
42
35
40
Humidity (%)
78
54
69
48
56
56
80
44
Humidity (%)
49
81
40
58
90
59
46
74
56
35
94
Mixing Height
Difference (m)
800-700
300-900
1100-1300
400-1750
700-1350
1250-1950
650-300
250-550
Mixing Height
Difference (m)
650-1400
700-750
800-1900
550-1350
700-600
500-1800
750-1950
850-1150
750-950
1250-1350
550-450
Note that the tables indicate that no single variable distinguishes one typical day from
another. Furthermore, the interpretation of the clusters is not standard across seasons or across study
areas.
For input into the OZIPR model, the morning and afternoon mixing heights are provided as
-------
the variables MHINIT and MHFINAL under the DILUTION option. The pressure is converted to
atmospheres and entered as the variable PRESSURE[AT]. For temperature and humidity, a series
of hourly values are required—a total of 13 hourly values are needed, from 8 a.m. to 8 p.m.,
inclusive. (The model treats these as instantaneous measurements at the top of each hour, not hourly
averages.) For the input files used in this study, the temperatures were converted to degrees Celsius.
A total of 120 meteorological files were prepared (3 per season, 4 seasons, 10 study areas). In
addition to establishing the typical days via clustering and preparing the meteorological data for
input, SAS programming was used to automatically generate all 120 files for use in OZIPR.
32
-------
Section 7
Running the OZIPR Model
The OZIPR model is a photochemical box model that simulates the production and removal
of certain chemical species over the course of a day. To run the model, parameter settings must be
defined and scenario-specific data must be supplied. Each scenario to be modeled with OZIPR
requires its own input file. A line-by-line description of an example input file is provided in
Appendix A, along with a discussion of file-naming conventions and batch processing.
Once the OZIPR input files are ready, the model can be run. Each run requires one DOS
command (OZIPR is not a Windows program). A set of batch files were prepared, one per study area,
to run all 48 scenarios for that study area and to combine the output tables into a single file. Figure
7-1 shows a complete OZIPR input file.
Figure 7-1. Complete OZIPR input file for summertime urban Houston, day type #1, with
primary emissions of carbonyls.
! The mechanism for OZIPR model is updated with SAPRC97 and two species added:
1,3-butadiene (BDIE) and acrolein (ARCO).
MECH [CM] >
CNUM=
ALK4 =4.5,
ALK7 = 7.0,
ETHE =2.0,
PRPE =3.0,
TBUT =4.0,
TOLU = 7.0,
XYLE = 8.0,
TMBZ = 9.0,
RCHO =3.0,
BDIE =4.0,
HCHO = 1.0,
ALD2 =2.0,
ACRO =3.0,
NRHC = 1.0;
REACTIONS =
{1} N02
{2} O
{3} O H
{4} O n
N02
N02
NO
03
NO
N03
#0.016667 /LI;
#1.0500E+03 @ -1282.0;
#6.5000E-12 @ -120;
#1.1100E-13 @ -894.0;
33
-------
5}
6}
7}
8}
9}
10}
11}
12}
13}
14}
15}
16}
17}
18}
19}
20}
21}
22}
23}
24}
25}
26}
27}
28}
29}
30}
31}
32}
33}
34}
35}
36}
37}
37a
38}
39}
40}
41}
42}
43}
44}
45}
46}
47}
48}
49}
50}
51}
52}
53}
54}
55}
56}
57}
58}
59}
60}
61}
62}
63}
64}
65}
66}
67}
68}
69}
70}
NO + O3
NO2 + O3
NO + NO3
NO + NO
NO2 + NO3
N2O5
N2O5 + H2O
NO2 + NO3
NO3
NO3
O3
O3
O1D + H2O
O1D
NO + OH
HONO
NO2 + H2O
NO2 + OH
HNO3 + OH
CO + OH
OH
O3 + OH
NO + HO2
NO2 + HO2
HNO4
HNO4 + OH
O3 + HO2
HO2 + HO2
HO2 + HO2 +
NO3 + HO2
NO3 + HO2 +
H2O2
H2O2 + OH
HO2 + OH
RO2 + NO
RCO3 + NO
RCO3 + NO2
RO2 + HO2
RCO3 + HO2
RO2 + RO2
RO2 + RCO3
RCO3 + RCO3
RO2R + NO
RO2R + HO2
RO2R + RO2
RO2R + RCO3
ROOH
HCHO
HCHO
HCHO + OH
HCHO + NO3
HCHO + HO2
ALD2
ALD2 + OH
ALD2 + NO3
MCO3 + NO
MCO3 + NO2
MCO3 + HO2
MCO3 + RO2
MCO3 + RCO3
PAN
RCHO
RCHO + OH
RCHO + NO3
PCO3 + NO
PCO3 + NO2
PCO3 + HO2
NO2
NO3
2*NO2
2*NO2
N2O5
NO2 + NO3
2*HNO3
NO + NO2
NO
NO2 + O
O
O1D
2*OH
O
HONO
NO + OH
HONO + -1* NO2 + HNO3
HNO3
NO3
HO2
HCHO + RO2R + RO2
HO2
NO2 + OH
HNO4
NO2 + HO2
NO2
OH
H2O2
H2O = H2O2
HNO3
H2O = HNO3
= 2*OH
HO2
=
NO
NO
NO2
HO2
HO2
=
=
=
NO2 + HO2
ROOH
= 0.5*HO2 + RO2
= 0.5*HO2 + RCO3
HO2 + OH
2*HO2 + CO
CO
HO2 + CO
HNO3 + HO2 + CO
RO2R + RO2
CO + HCHO + HO2 + RO2R +
MCO3 + RCO3
HNO3 + MCO3 + RCO3
NO2 + HCHO + RO2R + RO2
PAN
ROOH + HCHO
= 0.5*HO2 + HCHO + RO2
HO2 + HCHO + RCO3
MCO3 + NO2 + RCO3
ALD2 + HO2 + CO + RO2R
RCO3 + PCO3
HNO3 + PCO3 + RCO3
NO2 + ALD2 + RO2R + RO2
PPN
ROOH + ALD2
#2.0000E-12 @ 1400.0;
#1.4000E-13 @ 2500.0;
#1.7000E-11 @ -150.0;
#8.1500E-20 @ -528.0;
#4.6200E-13 @ -273.0;
#1.3300E+15 @ 11379;
#1 .OOOOE-21;
#2.5000E-14 @ 1228.0;
#0.016667 /Rl;
#0.016667 /R2;
#0 .016667E-03 /R3 ;
#0 .016667E-03 /R4 ;
#2 .2000E-10;
#7 .2000E + 08;
#4.0300E-13 @ -833.0;
#0.016667 /R5;
#4 .OOOOE-24;
#9.5700E-13 @ -737.0;
#6.4500E-15 @ -818;
#2 .400E-13;
#28.3*2 @ 1280.0;
#1.6002E-12 @ 941.0;
#3.7000E-12 @ -242.0;
#1.0200E-13 @ -773.0;
#4.3500E+13 @ 10103;
#1.3000E-12 @ -380;
#1.1000E-14 @ 500.0;
# 2.200E-13 @ -619.0;
# 3.100E-34 @ -2818.0;
# 2.200E-13 @ -619.0;
# 3.100E-34 @ -2818.0;
# 0.016667E-03 /R6 ;
# 3.3000E-12 @ 200.0;
# 4.6000E-11 @ -230;
# 4.2000E-12 @ -181.0;
# 4.2000E-12 @ -180.0;
# 2.8000E-12 @ -180.0;
# 3.4000E-13 @ -800;
# 3.4000E-13 @ -800;
# l.OOOOE-15;
# 1.8600E-12 @ -530;
# 2.8000E-12 @ -530;
# 4.2000E-12 @ -181.0;
# 3.4000E-13 @ -800;
# l.OOOOE-15;
# 1.8600E-12 @ -530;
# 0.016667E-03 /R7;
# 0.016667E-03 /R8 ;
# 0.016667E-03 /R9;
# 1.125000E-12*2 .0 @ -648;
#2.8000E-12 @ 2516;
#1. OOOOE-14;
RO2 #0. 016667E-03 /RIO;
#5.55000E-12 @ -311;
#1.40000E-12 @ 1860;
#4.2000E-12 @ -180;
#2.8000E-12 @ -180;
#3.4000E-13 @ -800;
#1.8600E-12 @ -530;
#2.8000E-12 @ -530;
#2.0000E+16 @ 13542;
+ RO2 #0 .016667E-03 /Rll
#8.5000E-12 @ -252;
#1.4000E-12 @ 1860;
#4.2000E-12 @ -180;
#8 .4000E-12;
#3.4000E-13 @ -800;
34
-------
71}
72}
73}
74}
75}
76}
77}
78}
79}
80}
81}
82}
83}
84}
85}
86}
87}
88}
89}
90}
PCO3 -
PCO3 -
PPN
ACET
ACET -
MEK
MEK +
GLYX
GLYX -
GLYX -
GCO3 -
GCO3 -
GPAN
GCO3 -
GCO3 -
GCO3 -
MGLY
MGLY -
MGLY -
ALK4 -
h RO2
h RCO3
h OH
OH
h OH
h NO3
h NO2
h NO
h HO2
h RO2
h RCO3
h OH
h NO3
h OH
= 0.5*HO2 + ALD2 + RO2 #1.8600E-12 ® -530;
HO2 + ALD2 + RCO3 #2.8000E-12 ® -530;
PCO3 + NO2 + RCO3 #1.6000E+17 ® 14073
MCO3 + HCHO + RCO3 + RO2R + RO2 #0 . 016667E-03 /R12
RO2R + RO2 + MCO3 +RCO3 + HCHO #4.81E-13 * 2.0 ® 230
MCO3 + ALD2 + RCO3 + RO2R + RO2 #0 . 016667E- 03 /R13
= 1.5*R2O2 + 1.5*RO2 + 0.5*MCO3 + 0.50*ALD2
+ 0.5*HCHO + 0.5*PCO3 + RCO3 #2 . 9200E-13*2 . 0® -414
= 0.13*HCHO + 1.87*CO #0 . 016667E- 03 /R14 ;
= 0.6*HO2 + 1.2*CO + 0.4*GCO3 + 0.4*RCO3 #1 . 1400E-11;
= HNO3 + 0.6*HO2 + 1.2* CO + 0.4*GCO3
+ RCO3 #1.4000E-12 ® 1860;
= GPAN #2.8000E-12 ® -180.0
= NO2 + HO2 + CO #4.2000E-12 ® -180.0
= GCO3 + NO2 + RCO3 #2.00E+16 ® 13542.0;
= ROOH + CO #3.40E-13 ® -800;
= 0.5*HO2 + CO + RO2 #1.8600E-12 ® -530;
= HO2 + CO + RCO3 #2.8000E-12 ® -530;
= MCO3 + HO2 + CO + RCO3 #0 . 016667E- 03 /R15
= MCO3 + CO + RCO3 #1 . 7200E-11 ;
= HNO3 + MCO3 + CO + RCO3 #1.4200E-12 ® 1860;
= 0.19*HCHO + 0.31*ALD2 + 0.170*RCHO + 0.34*ACET + 0.44*MEK
+ 0.070*RO2N + 0.930*RO2R + 0.600*R2O2 + 1.600*RO2 #1.05E-11 ® 353;
{91}
ALK7 -
h OH
= 0.02*HCHO + 0.03*ALD2 + 0.25*RCHO + 0.36*ACET + 0.88*MEK
+ 0.180*RO2N + 0.820*RO2R + 0.840*R2O2 + 1.840*RO2 #1.62E-11 ® 288;
{92}
93}
94}
95}
96}
97}
98}
99}
LOO}
101}
102}
103}
104}
105}
106}
107}
108}
109}
110}
111}
112}
113}
114}
115}
116}
117}
118}
119}
120}
121}
122}
123}
124}
125}
126}
ALKN -
RO2N -
RO2N -
RO2N -
RO2N -
R2O2 -
R2O2 -
R2O2 -
R2O2 -
ETHE -
ETHE -
ETHE -
ETHE -
PRPE -
PRPE -
PRPE -
PRPE -
TBUT -
TBUT -
TBUT -
TBUT -
TOLU -
DIAL -
DIAL
XYLE -
TMBZ -
CRES -
CRES -
BZO -
BZO -
BZO
PHEN -
PHEN -
NPHE -
BZN2 -
h OH
h NO
h HO2
h RO2
h RCO3
h NO
h HO2
h RO2
h RCO3
h OH
h O3
h O
h NO3
h OH
h O3
h O
h NO3
h OH
h O3
h O
h NO3
h OH
h 0. 840
h OH
h OH
h OH
h OH
h NO3
h NO2
h HO2
h OH
h NO3
h NO3
h NO2
= NO2 + 0.155*MEK + 1.05*RCHO + 0.48*ALD2 + 0.16*HCHO
+ 1.390*R2O2 + 1.390*RO2 #2.19E-11 ® 708;
ALKN #4.20E-12 ® -181;
ROOH + MEK #3.4000E-13 ® -800;
RO2 + 0.5*HO2 + MEK #1. OOOOE-15;
RCO3 + 0.5*HO2 + MEK #1.8600E-12 ® -530;
NO2 #4.2000E-12 ® -181;
ROOH #3.4000E-13 ® -800;
RO2 #1. OOOOE-15;
RCO3 #1.8600E-12 ® -530;
RO2R + RO2 + 1.56*HCHO + 0.22*ALD2 #1.9600E-12 ® -438;
HCHO + 0.12*HO2 + 0.42*CO #9.1400E-15 ® 2580;
HCHO + HO2 + CO + RO2R + RO2 #1.0400E-11 ® 792;
NO2 + 2*HCHO + R2O2 + RO2 #5.4300E-12 ® 3041;
RO2R + HCHO + ALD2 + RO2 #4.8500E-12 ® -504;
= 0.65*HCHO + 0.5*ALD2 + 0.285*CO + 0.06*OH + 0.165*HO2
+ 0.135*RO2R + 0.135*RO2 #1.32E-14 ® 2105.0;
= 0.6* ACET + 0.4*HCHO + 0.2*ALD2 + 0.2*HO2
+ 0.6*RO2R + 0.4*CO + 0.6*RO2 #1.18E-11 ® 324;
NO2 + HCHO + ALD2 + R2O2 + RO2 #5.00E-12 ® 1935;
RO2R + 2*ALD2 + RO2 #1.0100E-11 ®-549;
ALD2 + 0.15*CO + 0.27*RO2R + 0.12*OH + 0.21*HO2
+ 0.270*RO2 + 0.300*HCHO #9.08E-15 ® 1137;
MEK + 0.4*HO2 #2.2600E-11 ® -10;
NO2 + 2*ALD2 + R2O2 + RO2 #1.0000E-11 ® 975;
= 0.16*CRES + 0.16*HO2 + 0.84*RO2R + 0.4*DIAL
*RO2 + 0.144*MGLY + 0.114*GLYX #2.1E-12 ® -322.0;
PCO3 + RCO3 #3.0000E-11;
HO2 + CO + MCO3 + RCO3 #0 . 016667E- 03 /R16;
= 0.17*CRES + 0.17*HO2 + 0.83*RO2R + 0.83*RO2
+ 0.65*DIAL + 0.316*MGLY + 0.095*GLYX #1.66E-11 ® -116;
= 0.17*CRES + 0.17*HO2 + 0.83*RO2R + 0.83*RO2
+ 0.49*DIAL + 0.86*MGLY #6.2E-11;
= 0.2*MGLY + 0.15*RO2P + 0.85*RO2R + RO2 #4 . 2000E-11 ;
HNO3 + BZO #2.1000E-11;
NPHE #1.3000E-11 ® -300;
PHEN #3.4000E-13 ® -800;
PHEN #1.0E-03;
= 0.2*GLYX + 0.15*RO2P + 0.85*RO2R + RO2 #2 . 6300E-11 ;
HNO3 + BZO #3.6000E-12;
HNO3 + BZN2 #3.6000E-12;
#1.3000E-11 ® -300;
35
-------
{127} BZN2 + HO2 = NPHE
{128} BZN2 = NPHE
{129} RO2P + NO = NPHE
{130} RO2P + HO2 = ROOH
{131} RO2P + RO2 = 0.5*HO2 +
{132} RO2P + RCO3 = 0.5*HO2 +
{133} NRHC = NRHC
{135} ACRO + OH
{136} ACRO + O3
{137} ACRO + NO3
{138} BDIE + OH = ACRO
{139} BDIE + O3 = ACRO
< (MECH)"
ZENITH>
LI = 0.560,
0. 188,
Rl : 2.2600,
2 .8300,
R2 : 20.4000,
25 .4000,
R3 : 55.2000,
64 .5000,
R4 : 4.5600,
1 .0200,
R5 : 0.1960,
0 .1900,
R6 : 0.9080,
0 .6250,
R7 : 0.9080,
0 .6250,
R8 : 3.6700,
2 .0700,
R9 : 4.6600,
3 .4700,
RIO: 0.5870,
0 .2580,
Rll: 1.1900,
0 .7260,
R12 : 0.1340,
0 .0705,
R13 : 0.1910,
0 .1010,
R14 : 7.7900,
8 .9700,
R15: 17.2000,
19 .8000,
R16: 63.8000,
52 .5000,
< (ZENI)
TITLE > Houston
PLACE >
0 .550,
0 .0724,
2 .2600,
3 .3400,
20 .5000,
29 .7000,
55 .3000,
74 .3000,
4 .4100,
0 .3500,
0 .1960,
0 .1850,
0 .9030,
0 .5120,
0 .9030,
0 .5120,
3 .6200,
1 .3300,
4 .6400,
2 .6500,
0 .5790,
0 .1630,
1 .1800,
0 .5530,
0 .1320,
0 .0498,
0 .1890,
0 .0711,
7 .8100,
9 .8800,
17 .2000,
21 .7000,
63 .6000,
47 .0000,
- Summer
0 .548,
0 .00436
2 .2800,
4 .0600,
20 .7000,
35 .5000,
55 .6000,
89 .0000,
4 .0500,
0 .0810,
0 .1960,
0 .1810,
0 .8820,
0 .4310,
0 .8820,
0 .4310,
3 .4500,
0 .7100,
4 .4900,
1 .7700,
0 .5510,
0 .1010,
1 .1400,
0 .4270,
0 .1270,
0 .0355,
0 .1820,
0 .0507,
7 .8700,
10 .9000,
17 .4000,
23 .7000,
62 .9000,
42 .9000,
- Day#l
#3.4000E-13 @ -800;
#1. OOOOE-03;
#4.2000E-12 @ -181.0
#3.4000E-13 @ -800;
RO2 #1.0000E-15;
RCO3 #1.8600E-12 @ -530;
#1. OOOOE+00;
#1. 9900E-11;
#2 . 8000E-18;
#1.2000E-15;
#1.4800E-11 @ 448;
#3.3000E-14 @ -2500;
0.520, 0.479, 0.417, 0.322,
'
2.3200, 2.3900, 2.5600,
2 .7900,
21.1000, 21.8000, 23.1000,
25 .8000,
56.1000, 57.1000, 59.9000,
66 .0000,
3.5000, 2.7300, 1.8800,
0 .0230,
0.1960, 0.1940, 0.1930,
0 .1880,
0.8450, 0.7910, 0.7180,
0 .3830,
0.8450, 0.7910, 0.7180,
0 .3830,
3.3000, 3.0200, 2.6000,
0 .6500,
4.4200, 4.2300, 3.8900,
1 .7700,
0.5010, 0.4360, 0.3530,
0 .0598,
1.0800, 0.9940, 0.8750,
0 .3310,
0.1180, 0.1060, 0.0897,
0 .0249,
0.1690, 0.1510, 0.1280,
0 .0356,
7.9700, 8.1700, 8.4600,
9 .7300,
17.6000, 18.0000, 18.7000,
21 .4000,
61.5000, 59.5000, 56.6000,
42 .7000,
- Urban - Carbonyls <
CITY = Houston;
LAT = 29.8;
YEAR = 1996;
< (PLACE)
LON
MONTH =
95.2;
TZONE = 6;
7; DAY = 15;
TIME > 0800, 2000 <
BOUNDARY >
REAC =
ALK4,
ALK7,
ETHE,
PRPE,
0.2794,
0.2073,
0. 0356,
0. 0191,
0.2794,
0.2073,
0. 0356,
0. 0191,
0.2794,
0.2073,
0. 0356,
0. 0191,
36
-------
TBUT,
TOLU,
XYLE,
TMBZ,
RCHO,
BDIE,
HCHO,
ALD2,
ACRO,
NRHC,
0. 0721,
0. 0511,
0. 0333,
0. 1018,
0. 0090,
0. 0000,
0. 0000,
0. 0000,
0. 0000,
0. 1913,
0. 0721,
0. 0511,
0. 0333,
0. 1018,
0. 0090,
0. 0025,
0. 0200,
0. 0200,
0. 0010,
0. 1478,
0. 0721,
0. 0511,
0. 0333,
0. 1018,
0. 0090,
0. 0025,
0. 0200,
0. 0200,
0. 0010,
0. 1478;
I FRACTION NO2 = 0.20;
TRANSPORT =
VOCALOFT
NOXALOFT
O3ALOFT
COALOFT
INIT =
HCHO = 0 .
ALD2 = 0 .
ACRO = 0 .
BDIE = 0.
HONO = 0 .
DEPO [12] =
NO2 = 0 .
O3 =0.
HNO3 = 3 .
H2O2 = 1.
PAN = 0.
< (BOUNDARY)
= 0.1000,
= 0.0003,
= 0.0200,
= 0.2000;
00600,
00600,
00030,
00075,
001;
48, 0 .54, 0
60,0.70,0
00,3 .30, 3
80,1.90,2
48, 0 .54, 0
MET > {HOUSTON, SUMMER
DILUTION=
MHINIT =
MHFINAL =
671,
1457;
.60,0.60,
.80,0.80,
. 50,3 .50,
. 00,2 .00,
.60,0.60,
1 }
0.60,0.60,0.
0.80,0.80,0.
3 . 50,3 .50, 3 .
2 . 00,2 .00, 2 .
0.60,0.60,0.
.60,0.60
.80,0.80
. 50,3 .50
. 00,2 .00
.60,0.60
TEMPERATURE [13 , C] =
27.8, 29.
33.9, 33 .
PRESSURE [AT]
RH[13] =
82.0, 72 .
50.0, 51.
< (MET)
4, 30.6,
3, 32.8,
= 1.00;
0, 63.0,
0, 52 . 5,
31.7, 32 .
32.2, 31.
58.0, 54 .
54 . 0, 59.
.8, 33.3,
. 1, 30.0, 28 .
.5, 52 . 0,
. 0, 65.5, 70.
. 9;
. 5;
0.54,0 .48,0.36,0 .24,
0.70,0.60,0.50,0.30,
3.30,3 .20,3.00,2 .60,
1.90,1 .80,1.70,1 .70,
0. 54,0.48,0.36,0.24;
MASS [12] >
VOC = 0.3000,
9.65,10.65,11.44,12.26,12.95,13.33,13.79,
4.85, 4.95, 4 .92,
38.4, 39.4, 39.1,
.0499, .0505, .0503,
NOX = 0.0300,
4.61, 4.57, 4.69,
CO = 0.5000,
31.2, 35.6, 36.9,
HCHO[30] = 0.0000001,
.0466, .0478, .0487,
ALD2 [44] = 0.0000001,
.1594, .1725, .1789, .1870, .1918, .1902,
ACRO[56] = 0.0000001,
.0074, .0074, .0074, .0074, .0074, .0074,
BDIE[54] = 0.0000001,
.0262, .0259, .0259,
(MASS)
.0259, .0259, .0259,
5.00,
40.0,
.0510,
.1953,
.0074,
.0259,
14 .25,
5.24,
42.6,
.0529,
.2089,
.0074,
.0259,
14 .13,13 .68,10 .29, 7.33,
5.35, 5.29, 4.62, 2.25,
43.8, 43.5, 35.9, 16.5,
.0538, .0535, .0468, .0230,
.2161, .2139, .1797, .0972,
.0074, .0074, .0074, .0074,
.0260, .0259, .0259, .0259;
CALCULATE >
VOC = 0.3000;
NOX = 0.0300;
CO = 0.5000;
PRINT[CONG] =
NAMES[5] = CO,O3,HCHO,ALD2,ACRO,
= CO,O3,HCHO,ALD2,ACRO;
<
END.
AVG [5]
(CALCULATE)
37
-------
7.1 Estimation of Secondary HAP Production
The concentration of HAPs can be conceptually divided into two types: primary and
secondary. Primary HAPs are the result of the direct emission into the air of the species in question,
that is, the concentration that would result if there were no atmospheric chemical reactions that either
form or remove the species. Secondary HAPs are the net results of chemical reactions in the
atmosphere.
The method used to determine the relative amount of primary to secondary production for
each target HAP was to conduct two OZIPR model runs for each scenario: one with emissions of the
target HAP and the other without these emissions. For the latter run, all of the HAP production is
secondary. This can then be compared to the HAP concentrations from the first run which represent
the sum of both primary and secondary production.
If the target HAP concentrations were principally due to primary emissions, then this
technique of eliminating the primary emissions would result in changes in the chemical equilibrium,
particularly in the concentration of radicals that interact with the target HAPs as well as with many
other compounds. This is not the case, however, since secondary formation generally dominates
primary emissions, and test runs have shown that the radical concentrations are not greatly affected
by removing primary target HAP emissions.
To eliminate primary target HAP emissions from an OZIPR run, several changes must be
made to the input file. The first change is to remove the terms representing the target HAPs from the
MASS option (that is, remove the HCHO, ALD2, and ACRO terms). Also, the second and third
columns of data under the REAC option are modified to reflect changes in the aloft air
concentrations. The assumption is made that in the absence of primary emissions, the aloft air would
contain only 80% of the normal concentration of each carbonyl species. This is based on an assumed
20% primary to 80% secondary breakdown. The effect of the aloft concentrations on the surface
concentrations seen later in the day is generally not that large, so uncertainties in the initial aloft
concentrations should not result in large uncertainties in the final results. The final input file change
needed is to set the initial carbonyl concentrations in the surface air (i.e., in the box) at the start of
the run. These concentrations would not be zero, even without emissions, since there would be
secondary production of carbonyls left in the atmosphere from the previous day. Similarly for the
aloft concentrations, the initial surface concentrations are set to 80% of the values of the run, which
includes primary emissions. This is done under the INIT option of the boundary command. In test
runs, it was found that the concentrations of the target HAPs late in the day depend much more on
the chemistry, the weather, and the hourly emissions than they do on the initial concentrations.
Therefore, like the aloft concentrations, moderate uncertainties in the initial concentrations result in
small uncertainties in the final HAP concentrations.
38
-------
7.2 OZIPR Model Output
Each run of the OZIPR model (each scenario) produces two output files. The larger of the
two has the filename extension .OUT, and it contains a log of the model run, detailing inputs,
parameter settings, and certain values calculated internally by the model. For the purposes of this
study, the second output file (with filename extension .TXT) is more important. This file contains
a table of the concentrations of selected chemical species at the top of each hour; in particular, the
three target HAPs. The concentrations are all in parts per million volume (ppmV). The times are in
local standard (not daylight saving); standard time was used since all four seasons were being
modeled. For presentation in the output tables in Section 9 of this report, the concentrations were
converted to units of micrograms per cubic meter (|_ig m"3), using the ideal gas law applied to the
model results with the appropriate temperature and pressure for each hour.
The OZIPR model was not run for the nighttime hours since it is a photochemical model and
relatively little chemistry happens at night. At night, two effects will generally serve to reduce
atmospheric concentrations: (1) horizontal transport away from the emission sources and
(2) deposition on the ground. The OZIPR model does not account for dilution due to possible
spreading of the pollutant "cloud." Also, the model, as implemented here, does not reduce mixing
height after the afternoon value has been reached. (OZIPR would allow varying nighttime mixing
heights if one supplied them on an hourly basis as input.) The effect of stabilizing the nighttime air
is that the upper levels become decoupled from, and no longer influence, the ground-level
concentration. Also, in the relatively thin ground-level layer that is still mixed, surface deposition
becomes much more important than in the daytime when the mixing box is much taller. Since these
effects are not well modeled in OZIPR, a simple interpolation was used to estimate nighttime values.
This was done by linearly interpolating in time between the 8p.m. and 8 a.m. concentrations for each
HAP of interest, assuming that the evening value was higher. In the relatively few cases where the
morning concentrations were higher than the evening ones, the nighttime values were set equal to
the evening value, under the assumption that the larger morning value was due to early morning
emissions just prior to the start of the model run at 8 a.m. These interpolated nighttime values were
used in the construction of seasonal and annual average concentrations.
39
-------
Section 8
Results and Discussion
The OZIPR model was run for up to 48 scenarios for each of the 10 study areas; all runs were
for the 8 a.m. to 8 p.m. time period. Within each of the four seasons, two or three typical days were
used, and runs were made separately for both urban and rural areas. To estimate the relative
importance of secondary production, each of these scenarios was run both with and without primary
emissions of the target HAPs. The runs without target HAPs emissions provide estimates of
secondary formation alone. The runs with emissions yield total concentrations. The ratio between
these two gives the percentage of secondary production.
8.1 Summary Tables of OZIPR Model Results
Results for the secondary formation of the target HAPs for the 10 study areas for both urban
and rural scenarios are presented in Tables 8-1, 8-2, and 8-3, corresponding to formaldehyde,
acetaldehyde, and acrolein, respectively. The values are instantaneous concentrations at the top of
each hour (indicated in military time), spatially averaged through the model study area and
seasonally averaged across typical days. Seasons were defined as indicated in Sections 2 and 6 of
this report.
All concentrations have been converted from parts per million (ppm), the OZIPR output
units, to micrograms per cubic meter (|_ig m"3). Using the ideal gas law, conversions were made with
the temperature and barometric pressure pertaining to each hour. For 25°C and a pressure of
1 atmosphere, the multiplicative factors for ppb units are 1.23, 1.80, and 2.29 for formaldehyde,
acetaldehyde, and acrolein, respectively (i.e., 1 ppb of acetaldehyde gives 1.80 |_ig m"3). Each
concentration in these tables represents a seasonal average, with each typical day contributing to the
weighted average according to its frequency of occurrence as listed in Table 6-3. As an example,
for autumn in Atlanta (Table 6-3 A) the statistical weights were 213/339 for the first typical day,
89/339 for the second, and 37/339 for the third.
40
-------
Table 8-1. Secondary Production of Formaldehyde in |ig m3 as estimated by the OZIPR
Model
SEASON
AUTUMN
SPRING
SUMMER
WINTER
0800
3.9
3.9
3.8
4.1
0900
4.8
4.8
5.5
4.4
1000 1100
5.4 5.7
5.0 4.9
6.1 6.3
5.2 6.1
1200
5.7
4.9
6.5
6.5
r^lTV— ATI AMTA ADCA-
SEASON
AUTUMN
SPRING
SUMMER
WINTER
SEASON
AUTUMN
SPRING
SUMMER
WINTER
SEASON
AUTUMN
SPRING
SUMMER
WINTER
0800
3.9
3.9
3.8
4.1
0800
0.8
0.8
0.8
0.9
0800
1.2
1.2
1.2
1.3
0900
4.1
4.2
4.8
1000 1100 1200
5.2 6.4
5.2 6.1
6.2 7.3
4.1 4.5 5.6
PITY— RO^TOM
0900 1000 1100
2.0 2.5 2.8
2.5
2.6
1.2
r\~\
0900
2.0
2.5
2.7
1.5
2.7 2.8
3.0 3.3
2.0 2.6
rV— ROQTOM
1000 1100
2.6 3.2
3.1 3.6
3.5 4.2
2.0 2.7
7.4
6.6
7.9
1300 1400
5.8 5.9
4.9 5.0
6.4 6.3
6.5 6.2
-I IRRAM
1300 1400
8.1 8.4
6.8 6.8
7.9 7.8
6.9 7.8 8.3
ARPA — PI IPAI
1200 1300 1400
3.0 3.2 3.3
2.9
3.4
2.9
3.0 3.0
3.5 3.5
3.1 3.2
1500
5.9
5.0
6.1
6.0
1500
8.6
6.8
7.6
8.6
1500
3.4
3.0
3.5
3.3
1600
6.1
5.1
5.9
6.0
1600
9.0
7.0
7.5
8.9
1600
3.5
3.1
3.6
3.3
1700
6.3
5.2
5.8
6.1
1700
9.3
7.2
7.5
9.2
1700
3.6
3.1
3.8
3.3
1800
6.4
5.4
5.8
6.1
1800
9.6
7.5
7.7
9.4
1800
3.6
3.2
3.9
3.3
1900
6.5
5.5
5.9
6.2
1900
9.7
7.7
8.0
9.5
1900
3.7
3.2
4.0
3.4
2000
6.5
5.6
6.0
6.2
2000
9.8
7.8
8.1
9.6
2000
3.7
3.3
4.1
3.4
ARPA — I IRRAM
1200 1300 1400
3.7
4.0
5.0
3.3
4.1 4.5
4.3 4.7
5.7 6.1
3.8 4.1
1500
4.9
4.9
6.5
4.4
1600
5.2
5.2
6.9
4.5
1700
5.4
5.5
7.3
4.5
1800
5.5
5.7
7.7
4.5
1900
5.5
5.8
7.9
4.6
2000
5.6
5.8
8.0
4.6
41
-------
Table 8-1. Continued
SEASON
AUTUMN
SPRING
SUMMER
WINTER
SEASON
AUTUMN
SPRING
SUMMER
WINTER
0800
1.0
1.0
1.0
1.1
0800
2.4
2.4
2.3
2.6
0900
1.5
1.8
1.8
1000 1100
1.8 2.0
1.8 1.9
1.9 2.0
1.3 2.0 2.6
PITY— PHI PAl^n
0900 1000 1100
2.3 3.2 4.5
2.7
2.7
2.7
3.4 4.1
3.5 4.5
3.2 4.5
r^iTV— npMWPD
SEASON
AUTUMN
SPRING
SUMMER
WINTER
SEASON
AUTUMN
SPRING
SUMMER
WINTER
0800
3.4
3.4
3.3
3.6
0800
3.4
3.4
3.3
3.6
0900 1000 1100
3.0
3.4
3.2
3.1 3.0
3.2 3.0
2.9 2.9
3.1 3.1 3.1
PITY— DP MX/PR
0900 1000 1100
3.0 3.7 4.6
3.5
3.4
3.0
4.2 4.8
4.1 4.9
2.9 3.5
1200
2.2
1.9
2.1
3.0
ARPA-
1200
5.6
4.8
5.5
6.4
1300 1400
2.3 2.3
1.9 1.8
2.1 2.1
3.1 3.2
•I IRRAM
1300 1400
6.6 7.3
5.3 5.7
6.2 6.6
8.2 9.6
1500
2.4
1.8
2.1
3.3
1500
8.0
6.0
6.9
10.5
1600
2.5
1.8
2.1
3.3
1600
8.6
6.4
7.3
11.0
1700
2.5
1.8
2.2
3.3
1700
9.0
6.8
7.8
11.1
1800
2.5
1.8
2.3
3.3
1800
9.2
7.1
8.3
11.1
1900
2.5
1.9
2.3
3.4
1900
9.3
7.4
8.6
11.2
2000
2.6
1.9
2.3
3.4
2000
9.5
7.5
8.9
11.2
ARPA — Rl IRAI
1200 1300 1400
2.9
2.8
2.9
2.9 2.8
2.7 2.6
3.0 3.0
3.1 3.0 2.9
ARPA— I IRRAM
1200 1300 1400
5.4 5.8 6.1
5.2
5.6
4.4
r^lTV— U/~ll IOTV>M ADPA
SEASON
AUTUMN
SPRING
SUMMER
WINTER
SEASON
AUTUMN
SPRING
SUMMER
WINTER
0800
4.0
4.0
3.9
4.2
0800
6.0
6.0
5.8
6.2
0900
4.5
5.7
6.3
1000 1100 1200
4.4 4.5
5.9 5.9
7.1 7.3
4.6
5.9
7.3
4.8 5.3 5.3 5.2
PITY— HOI I^TDM ARPA
0900 1000 1100 1200
6.0 6.8 7.8 8.6
8.2
9.1
6.5
9.3 9.9
11.2 11.9
7.4 8.4
10.1
11.5
9.2
5.3 5.4
5.9 6.0
5.2 5.8
-Rl IRAI
1300 1400
4.7 4.8
5.8 5.8
7.0 6.7
5.1 5.1
-I IRRAM
1300 1400
8.9 9.0
9.7 9.2
10.6 9.9
9.7 10.0
1500
2.9
2.6
3.0
2.9
1500
6.4
5.4
6.0
6.2
1500
4.9
5.7
6.4
5.2
1500
9.2
8.9
9.4
10.4
1600
2.9
2.6
3.1
2.9
1600
6.7
5.5
6.1
6.5
1600
5.0
5.6
6.1
5.3
1600
9.5
8.9
9.2
10.8
1700
3.0
2.6
3.1
3.0
1700
6.8
5.6
6.2
6.6
1700
5.1
5.6
6.0
5.4
1700
9.8
9.1
9.3
11.1
1800
3.0
2.7
3.1
3.0
1800
6.9
5.7
6.4
6.6
1800
5.2
5.7
6.0
5.5
1800
10.0
9.4
9.5
11.3
1900
3.0
2.7
3.2
3.0
1900
7.0
5.8
6.5
6.7
1900
5.3
5.8
6.1
5.5
1900
10.2
9.6
9.8
11.5
2000
3.1
2.7
3.2
3.1
2000
7.1
5.9
6.6
6.7
2000
5.3
5.9
6.2
5.6
2000
10.4
9.8
10.0
11.6
Table 8-1. Continued
42
-------
SEASON
AUTUMN
SPRING
SUMMER
WINTER
SEASON
AUTUMN
SPRING
SUMMER
WINTER
SEASON
AUTUMN
SPRING
SUMMER
WINTER
SEASON
AUTUMN
SPRING
SUMMER
WINTER
SEASON
AUTUMN
SPRING
SUMMER
WINTER
SEASON
AUTUMN
SPRING
SUMMER
WINTER
0800
8.4
8.4
8.3
8.6
0800
8.4
8.4
8.3
8.6
0800
3.8
3.9
3.7
4.0
0800
3.8
3.9
3.7
4.0
0800
4.0
4.0
3.8
4.2
0800
4.0
4.0
3.8
4.2
0900
11.5
14.7
15.3
8.0
PITV-
0900
10.3
13.2
14.0
1000 1100 1200
14.0 15.0 14.8
17.2 17.1 15.5
18.2 18.0 16.4
1300
14.0
14.0
15.4
1400
13.4
13.2
14.5
9.1 10.3 11.0 11.2 11.1
-I OQ AMf^PI PQ ARPA — I IRRAM
1000 1100 1200 1300 1400
13.3 15.1 15.9 16.2 16.3
16.4 17.2 17.0
17.8 19.2 19.2
7.4 8.3 9.9 11.2
PITY— PHOFMIY ARPA-
0900 1000 1100 1200
3.9 3.9 4.0 4.1
4.6
5.5
4.5 4.4 4.4
6.2 6.3 6.1
4.0 4.1 4.2 4.2
PITV— PMOPMIY ARPA-
0900 1000 1100 1200
3.8 4.2 5.0 5.6
4.5
5.3
3.8
PITY
0900
4.7
5.7
5.6
4.5
PITY
0900
4.3
5.1
5.1
4.3
5.1 5.5 5.7
6.4 7.1 7.1
16.3
18.5
15.3
17.5
11.9 12.2
-Rl IRAI
1300 1400
4.2 4.3
4.5
5.8
4.5
5.4
4.2 4.2
-I IRRAM
1300 1400
5.8 5.8
5.7
6.6
5.4
6.0
3.9 4.4 4.9 5.3 5.5
— PITT^RI IRf^H ARPA — Rl IRAI
1000 1100 1200 1300 1400
5.4 5.7 5.8 5.7 5.6
6.2 6.0 5.6
5.9 5.9 6.0
5.2
6.0
4.9
5.9
5.6 7.1 8.1 8.5 8.6
— PITTQRI IRf^M ARPA — I IRRAM
1000 1100 1200 1300 1400
5.1 6.0 6.6 7.0 7.2
6.1 6.6 6.7
6.0 6.7 7.0
5.0 6.3 7.6
6.6
7.0
8.4
6.4
6.8
8.9
1500
13.2
12.5
13.7
11.1
1500
16.5
14.5
16.5
12.6
1500
4.4
4.4
5.0
4.3
1500
5.8
5.2
5.4
5.6
1500
5.7
4.8
5.7
8.5
1500
7.3
6.2
6.7
9.1
1600
13.2
12.1
13.2
11.2
1600
16.8
14.1
15.9
12.9
1600
4.5
4.4
4.7
4.5
1600
5.8
5.1
5.0
5.7
1600
5.8
4.7
5.7
8.4
1600
7.5
6.2
6.7
9.3
1700
13.4
12.1
13.0
11.3
1700
17.2
14.2
15.9
13.2
1700
4.6
4.4
4.5
4.6
1700
5.9
5.0
4.7
5.8
1700
5.9
4.8
5.6
8.5
1700
7.7
6.3
6.7
9.5
1800
13.6
12.3
13.2
11.4
1800
17.6
14.4
16.3
13.4
1800
4.7
4.4
4.4
4.6
1800
6.0
5.1
4.6
5.8
1800
5.9
4.8
5.6
8.5
1800
7.8
6.4
6.9
9.6
1900
13.8
12.5
13.5
11.5
1900
17.8
14.7
16.7
13.5
1900
4.7
4.5
4.4
4.7
1900
6.0
5.1
4.6
5.9
1900
6.0
4.9
5.7
8.6
1900
8.0
6.6
7.1
9.8
2000
13.9
12.7
13.7
11.5
2000
18.0
14.9
17.1
13.6
2000
4.8
4.6
4.4
4.7
2000
6.0
5.2
4.7
5.9
2000
6.0
4.9
5.8
8.6
2000
8.1
6.7
7.2
9.8
Table 8-1. Continued
43
-------
SEASON
AUTUMN
SPRING
SUMMER
WINTER
SEASON
AUTUMN
SPRING
SUMMER
WINTER
SEASON
AUTUMN
SPRING
SUMMER
WINTER
SEASON
AUTUMN
SPRING
SUMMER
WINTER
0800
0.8
0.8
0.8
0.8
0800
1.2
1.2
1.2
1.2
0800
1.2
1.2
1.2
1.3
0800
3.1
3.1
3.0
3.2
0900
1.6
2.1
2.4
1000 1100
2.2 2.5
2.4 2.4
2.6 2.6
0.9 1.3 1.9
PITV— QPATTI P
0900 1000 1100
1.7 2.3 2.8
2.1
2.4
1.3
PITY
0900
1.9
2.3
2.4
1.5
PITY
0900
3.9
4.5
4.7
3.4
2.5 2.8
2.8 3.0
1200
2.7
2.4
2.6
2.4
ARPA-
1200
3.3
3.0
3.2
1300
2.8
2.4
2.6
1400
2.9
2.3
2.5
2.7 2.9
-I IRRAM
1300 1400
3.6 3.9
3.1
3.4
3.2
3.6
1.5 2.1 2.6 3.1 3.6
— WA^HIMl^TOM ARFA — Rl IRAI
1000 1100 1200 1300 1400
2.4 2.7 2.9 3.0 3.1
2.6 2.7
2.8 3.1
2.8
3.4
2.8
3.5
2.8
3.3
2.0 2.5 2.9 3.0 3.1
— WAQMIMf^TOM ARPA — I IRRAM
1000 1100 1200 1300 1400
5.2 6.4 7.5 8.4 9.2
5.6 6.4
5.9 7.1
4.2 5.7
7.0
8.2
7.2
7.6
9.0
8.5
8.1
9.4
9.5
1500
3.0
2.2
2.5
3.0
1500
4.2
3.2
3.7
3.9
1500
3.1
2.7
3.2
3.1
1500
10.1
8.5
9.5
10.2
1600
3.1
2.2
2.5
3.1
1600
4.4
3.3
3.8
4.1
1600
3.2
2.7
3.2
3.1
1600
10.8
9.0
9.9
10.7
1700
3.1
2.2
2.6
3.1
1700
4.5
3.4
3.9
4.1
1700
3.2
2.7
3.1
3.2
1700
11.3
9.6
10.4
10.9
1800
3.2
2.3
2.6
3.1
1800
4.6
3.4
4.1
4.1
1800
3.3
2.7
3.2
3.2
1800
11.6
10.1
10.9
11.0
1900 2000
3.2 3.2
2.3 2.3
2.7 2.8
3.1 3.1
1900 2000
4.6 4.7
3.5 3.6
4.2 4.3
4.2 4.2
1900 2000
3.3 3.3
2.8 2.8
3.2 3.2
3.2 3.2
1900 2000
11.8 11.9
10.4 10.6
11.3 11.6
11.1 11.3
44
-------
Table 8-2. Secondary Production of Acetaldehyde (|ig m3) Estimated by the OZIPR Model
SEASON
AUTUMN
SPRING
SUMMER
WINTER
SEASON
AUTUMN
SPRING
SUMMER
WINTER
SEASON
AUTUMN
SPRING
SUMMER
WINTER
SEASON
AUTUMN
SPRING
SUMMER
WINTER
0800
5.8
5.8
5.6
6.0
0800
5.8
5.8
5.6
6.0
0800
1.2
1.2
1.2
1.3
0800
1.8
1.8
1.7
1.9
0900
11.4
10.8
11.8
1000 1100
11.0 9.6
9.4 8.2
9.5 8.1
1200
8.7
7.7
7.2
9.0 12.8 13.2 12.0
PITY— ATI AMTA ARPA-
0900 1000 1100 1200
9.4 12.1 13.0 12.6
9.8
11.2
7.1
r\~\
0900
3.7
3.8
3.5
11.2 10.7
11.8 10.2
10.6 14.4
rV— ROQTOM
1000 1100
3.7 3.6
3.5 3.3
3.1 2.6
3.1 4.3 4.6
PITY— RO^TOM
0900 1000 1100
4.9 6.1 6.9
5.8
5.6
3.4
6.6 7.3
6.1 6.4
5.5 7.2
r* ITV— r* u i r* A r^r>
SEASON
AUTUMN
SPRING
SUMMER
WINTER
SEASON
AUTUMN
SPRING
SUMMER
WINTER
0800
1.5
1.5
1.4
1.6
0800
3.6
3.6
3.4
3.8
0900 1000 1100
3.6
3.5
3.1
3.3 3.1
2.8 2.2
2.1 1.4
3.5 5.2 5.6
PITY— PHI PAl^n
0900 1000 1100
7.0 10.5 13.5
8.3
8.3
5.6
10.1 11.2
9.4 9.3
10.1 16.4
9.4
8.1
16.5
1300 1400
8.2 7.7
7.3 6.8
6.4 5.7
10.7 9.8
-I IRRAM
1300 1400
11.6 10.7
8.1 7.1
6.5 5.7
17.2 17.3
1500
7.4
6.4
5.3
9.4
1500
10.3
6.6
5.3
17.4
1600
7.3
6.2
5.0
9.3
1600
10.3
6.4
5.3
17.9
1700
7.3
6.0
4.8
9.3
1700
10.8
6.5
5.4
18.5
1800
7.5
6.0
4.9
9.4
1800
11.6
6.9
5.8
19.2
1900
7.6
6.2
5.0
9.5
1900
12.6
7.4
6.3
20.0
2000
7.7
6.3
5.1
9.6
2000
13.3
8.0
6.8
20.6
ARPA — Rl IRAI
1200 1300 1400
3.6
3.1
2.2
3.5 3.5
3.0 2.9
2.1 2.1
4.7 4.8 4.9
ARPA — I IRRAM
1200 1300 1400
7.7 8.3 9.0
7.9
6.6
8.4
ARPA-
1200
2.9
1.8
1.0
5.6
ARPA-
1200
15.7
12.0
8.8
22.5
8.5 9.1
6.5 6.5
9.4 10.1
-Rl IRAI
1300 1400
2.7 2.6
1.6 1.4
0.9 0.9
5.5 5.5
-I IRRAM
1300 1400
17.4 18.8
12.7 13.2
8.0 7.4
27.3 30.9
1500
3.7
3.0
2.2
5.0
1500
9.6
9.7
6.6
10.7
1500
2.6
1.3
0.9
5.6
1500
20.5
13.8
7.1
33.5
1600
3.9
3.1
2.5
5.1
1600
10.3
10.5
7.0
11.0
1600
2.7
1.4
1.0
5.7
1600
22.2
14.9
7.5
34.8
1700
4.1
3.4
2.7
5.2
1700
10.8
11.3
7.6
11.1
1700
2.8
1.4
1.1
5.8
1700
23.5
16.1
8.3
35.1
1800
4.4
3.6
3.0
5.3
1800
11.3
11.9
8.4
11.1
1800
3.0
1.5
1.2
5.9
1800
24.7
17.3
9.4
35.2
1900
4.6
3.9
3.3
5.4
1900
11.6
12.4
9.1
11.1
1900
3.2
1.6
1.3
6.0
1900
25.6
18.6
10.6
35.3
2000
4.9
4.1
3.5
5.4
2000
11.9
12.8
10.0
11.2
2000
3.3
1.7
1.5
6.1
2000
26.2
19.8
11.9
35.4
Table 8-2. Continued
45
-------
SEASON
AUTUMN
SPRING
SUMMER
WINTER
SEASON
AUTUMN
SPRING
SUMMER
WINTER
SEASON
AUTUMN
SPRING
SUMMER
WINTER
SEASON
AUTUMN
SPRING
SUMMER
WINTER
SEASON
AUTUMN
SPRING
SUMMER
WINTER
SEASON
AUTUMN
SPRING
SUMMER
WINTER
Table 8-2.
0800
5.0
5.0
4.8
5.2
0800
5.0
5.0
4.8
5.2
0800
5.9
5.9
5.7
6.1
0800
8.8
8.8
8.6
9.1
0800
6.1
6.1
6.0
6.3
0800
6.1
6.1
6.0
6.3
0900
5.3
6.1
5.5
1000 1100
5.2 5.2
5.5 5.4
4.7 4.5
1200 1300
5.3 5.4
5.3 5.3
4.4 4.3
1400
5.4
5.2
4.2
5.4 5.4 5.4 5.4 5.4 5.4
PITV— DPMVPR ARPA— I IRRAM
0900 1000 1100 1200 1300 1400
5.2 6.7 7.7 8.0 8.1 8.1
6.5
6.2
7.1 7.1
6.4 6.0
4.6 5.2 7.0
PITY— HOI I^TOM
0900 1000 1100
9.3 8.4 8.0
12.1
12.4
11.0 10.0
11.2 9.7
10.5 11.4 10.7
PITV— MO! IQTOM
0900 1000 1100
12.7 13.0 12.2
17.4
18.3
12.6
PITY-
0900
20.8
27.2
28.1
11.5
PITV-
0900
17.9
24.3
25.9
9.3
16.9 14.3
16.7 12.8
6.8 6.4
5.4 4.7
6.1
4.2
8.9 10.1 10.9
ARPA — PI IRAI
1200 1300 1400
7.9 7.7 7.5
9.1 8.4
8.5 7.5
7.7
6.8
10.2 9.9 9.6
ARPA — I IRRAM
1200 1300 1400
10.8 9.4 8.5
11.3 8.7
9.1 7.1
7.3
6.4
16.1 17.4 17.4 16.9 16.3
-I O^ AMl^PI P1^ APPA — PI IRAI
1000 1100 1200 1300 1400
24.2 22.7 19.6 17.0 15.5
28.9 24.9
28.6 23.3
20.1 17.7
19.0 16.7
16.1
15.0
16.1 18.0 17.7 16.7 15.7
-I OQ AMf^PI PQ ARPA — I IRRAM
1000 1100 1200 1300 1400
24.5 26.1 25.6 24.6 23.7
30.0 29.6
31.3 30.6
14.4 18.2
27.3 24.5
27.8 24.4
19.8 20.2
21.8
21.0
20.2
1500
5.5
5.1
4.0
5.5
1500
8.4
6.0
4.0
11.5
1500
7.4
7.2
6.3
9.5
1500
8.2
6.8
6.2
16.2
1500
15.0
15.0
13.8
15.3
1500
23.4
19.8
18.5
20.4
1600
5.6
5.2
4.0
5.6
1600
8.7
6.1
4.0
12.0
1600
7.4
6.9
6.1
9.5
1600
8.4
6.8
6.2
16.5
1600
14.9
14.4
13.3
15.2
1600
23.8
18.9
17.5
20.8
1700
5.6
5.2
3.9
5.8
1700
9.0
6.4
4.1
12.3
1700
7.5
6.9
6.1
9.6
1700
8.8
7.2
6.6
17.0
1700
15.1
14.3
13.2
15.3
1700
24.6
18.9
17.7
21.2
1800
5.7
5.2
3.9
5.9
1800
9.4
6.6
4.4
12.4
1800
7.7
7.0
6.2
9.7
1800
9.6
7.7
7.1
17.8
1800
15.4
14.5
13.6
15.4
1800
25.7
19.4
18.5
21.9
1900
5.8
5.3
4.0
5.9
1900
9.7
6.9
4.6
12.4
1900
7.9
7.1
6.4
9.8
1900
10.4
8.4
7.7
18.5
1900
15.6
14.7
14.0
15.5
1900
26.5
19.9
19.4
22.4
2000
5.8
5.3
4.0
6.0
2000
10.0
7.2
4.9
12.5
2000
8.0
7.3
6.7
9.9
2000
10.9
8.9
8.1
19.1
2000
15.8
14.8
14.5
15.6
2000
27.2
20.3
20.3
22.9
Continued
riTY-PHDPMiy ARPA-RI IPAI
46
-------
SEASON
0800 0900 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 2000
AUTUMN
SPRING
SUMMER
WINTER
SEASON
AUTUMN
SPRING
SUMMER
WINTER
SEASON
AUTUMN
SPRING
SUMMER
WINTER
SEASON
AUTUMN
SPRING
SUMMER
WINTER
SEASON
AUTUMN
SPRING
SUMMER
WINTER
SEASON
AUTUMN
SPRING
SUMMER
WINTER
5.6
5.7
5.4
5.9
0800
5.6
5.7
5.4
5.9
0800
5.8
5.9
5.6
6.1
0800
5.8
5.9
5.6
6.1
0800
1.2
1.2
1.2
1.2
0800
1.8
1.8
1.8
1.8
8.3 7.4
9.8 8.7
10.8 9.4
7.1 6.9
8.0 7.5
7.9 6.8
8.1 8.9 8.6 8.4
PITY— PHOFMIY ARPA-
0900 1000 1100 1200
7.4 7.8 7.4 6.3
9.3 9.0
10.4 9.5
7.8 6.3
7.3 5.2
6.7
7.0
5.8
6.3
6.5
5.1
8.4 8.2
-I IRRAM
1300 1400
5.3 4.7
5.0
4.0
4.1
3.4
6.6 7.8 8.8 9.0 8.6 7.9
PITY— PITTQRI IRf^M ARPA — PI IRAI
0900 1000 1100 1200 1300 1400
12.7 13.0 11.8 10.5 9.6 9.0
14.8 13.2
13.4 10.7
11.0 9.6
9.0 7.9
8.8
7.0
8.2
6.1
9.5 15.8 19.0 19.1 18.0 16.8
PITY— PITT^RI IRf^H ARPA — I IRRAM
0900 1000 1100 1200 1300 1400
11.0 13.6 14.2 13.6 12.5 11.6
13.8 14.8
13.2 12.8
13.9 12.3
10.6 8.4
10.8
6.8
9.6
5.9
7.8 13.0 18.3 21.4 22.8 23.5
PITY— QPATTI P ARPA — PI IRAI
0900 1000 1100 1200 1300 1400
3.4 3.5 3.4 3.2 3.1 3.0
3.4 2.9
3.4 2.7
2.5 2.1
2.2 1.6
1.8 3.2 4.1 4.4
PITY— ^PATTI P ARPA-
0900 1000 1100 1200
4.1 5.4 6.0 6.5
4.8 5.0
5.1 5.1
2.2 3.8
5.0 4.8
4.8 4.3
5.7 7.1
1.7
1.3
1.5
1.1
4.5 4.6
-I IRRAM
1300 1400
6.8 7.1
4.5
3.9
8.3
4.3
3.4
9.3
6.1
6.0
4.6
8.2
1500
4.2
3.6
3.1
7.4
1500
8.7
7.6
5.5
15.9
1500
11.3
8.8
5.4
24.0
1500
3.0
1.4
1.1
4.7
1500
7.4
4.2
3.2
10.1
6.0
5.7
4.2
8.2
1600
4.0
3.4
2.8
7.2
1600
8.6
7.2
5.0
15.7
1600
11.4
8.4
5.2
24.5
1600
3.1
1.4
1.2
4.8
1600
7.8
4.2
3.1
10.6
5.9
5.5
4.0
8.3
1700
4.0
3.2
2.7
7.1
1700
8.7
7.0
4.7
15.8
1700
11.7
8.4
5.1
25.2
1700
3.2
1.4
1.3
4.9
1700
8.1
4.4
3.2
10.8
6.0
5.4
3.9
8.4
1800
4.0
3.2
2.7
7.2
1800
8.8
7.0
4.6
15.9
1800
12.4
8.7
5.4
25.8
1800
3.3
1.5
1.5
5.0
1800
8.6
4.6
3.5
10.9
6.0
5.4
3.9
8.4
1900
4.1
3.3
2.7
7.3
1900
9.0
7.0
4.7
16.1
1900
13.0
9.2
5.7
26.4
1900
3.5
1.5
1.7
5.0
1900
9.0
4.9
3.8
11.0
6.1
5.5
4.0
8.5
2000
4.0
3.3
2.7
7.3
2000
9.1
7.1
4.8
16.2
2000
13.6
9.7
6.2
26.8
2000
3.7
1.6
1.9
5.1
2000
9.5
5.2
4.2
11.0
Table 8-2. Continued
CITY=WASHINGTONAREA=RURAL
SEASON 0800 0900 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 2000
47
-------
AUTUMN
SPRING
SUMMER
WINTER
1.8
1.8
1.7
1.9
3.8
4.2
4.0
2.8
4.1
3.9
3.6
4.1
3.9
3.4
2.9
4.7
3.6
3.0
2.2
4.8
3.3
2.6
1.7
4.7
3.1
2.3
1.4
4.6
3.0
2.0
1.3
4.5
3.0
1.9
1.3
4.5
3.1
1.9
1.3
4.6
3.2
2.0
1.4
4.6
3.2
2.0
1.5
4.6
3.3
2.1
1.5
4.6
CITY=WASHINGTON AREA=URBAN
SEASON 0800 0900 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 2000
AUTUMN
SPRING
SUMMER
WINTER
4.5
4.5
4.3
4.7
9.1
10.3
10.2
6.4
11.8
11.8
11.0
10.2
13.5 14.7
12.4 12.8
10.6 9.9
14.4 17.6
15.5
13.1
9.1
20.1
16.3
13.4
8.3
21.9
17.4
13.7
7.8
23.4
18.6
14.6
8.1
24.5
19.8
15.7
8.7
25.0
21.2
16.9
9.8
25.1
22.4 23.2
18.1 19.3
10.7 11.7
25.2 25.4
48
-------
Table 8-3. Secondary Production of Acrolein (|ig m3) Estimated by the OZIPR Model
SEASON
AUTUMN
SPRING
SUMMER
WINTER
SEASON
AUTUMN
SPRING
SUMMER
WINTER
SEASON
AUTUMN
SPRING
SUMMER
WINTER
SEASON
AUTUMN
SPRING
SUMMER
WINTER
0800
0.4
0.4
0.4
0.4
0800
0.4
0.4
0.4
0.4
0800
0.1
0.1
0.1
0.1
0800
0.1
0.1
0.1
0.1
0900
0.5
0.4
0.4
1000 1100
0.3 0.2
0.2 0.2
0.2 0.1
1200
0.2
0.1
0.1
0.8 0.6 0.4 0.3
PITY— ATI AMTA ARPA-
0900 1000 1100 1200
0.6 0.4 0.3 0.2
0.5
0.5
0.9
r\~\
0900
0.2
0.1
0.1
0.3 0.2
0.3 0.2
0.7 0.6
rV— ROQTOM
1000 1100
0.1 0.1
0.1 0.1
0.1 0.0
0.3 0.2 0.2
PITY— RO^TOM
0900 1000 1100
0.4 0.3 0.3
0.3
0.3
0.5
0.3 0.2
0.2 0.1
0.4 0.4
r* ITV— r* u i r* A r^r>
SEASON
AUTUMN
SPRING
SUMMER
WINTER
SEASON
AUTUMN
SPRING
SUMMER
WINTER
0800
0.1
0.1
0.1
0.1
0800
0.2
0.2
0.2
0.2
0900 1000 1100
0.2
0.1
0.1
0.1 0.1
0.1 0.0
0.0 0.0
0.3 0.3 0.2
PITY— PHI PAl^n
0900 1000 1100
0.6 0.5 0.4
0.6
0.4
1.0
0.4 0.3
0.3 0.2
1.0 0.9
0.1
0.1
0.5
1300 1400
0.1 0.1
0.1 0.1
0.1 0.0
0.2 0.2
-I IRRAM
1300 1400
0.2 0.1
0.1 0.1
0.0 0.0
0.4 0.4
1500
0.1
0.1
0.0
0.2
1500
0.1
0.0
0.0
0.4
1600
0.1
0.1
0.0
0.2
1600
0.1
0.0
0.0
0.3
1700
0.1
0.0
0.0
0.2
1700
0.1
0.0
0.0
0.4
1800
0.1
0.0
0.0
0.2
1800
0.1
0.0
0.0
0.4
1900
0.1
0.0
0.0
0.2
1900
0.1
0.1
0.0
0.4
2000
0.1
0.0
0.0
0.2
2000
0.2
0.1
0.0
0.4
ARPA — Rl IRAI
1200 1300 1400
0.1
0.1
0.0
0.1 0.1
0.0 0.0
0.0 0.0
0.2 0.1 0.1
ARPA — I IRRAM
1200 1300 1400
0.2 0.2 0.2
0.2
0.1
0.4
ARPA-
1200
0.1
0.0
0.0
0.2
ARPA-
1200
0.4
0.3
0.1
0.9
0.2 0.2
0.1 0.1
0.4 0.4
-Rl IRAI
1300 1400
0.1 0.0
0.0 0.0
0.0 0.0
0.2 0.2
-I IRRAM
1300 1400
0.4 0.4
0.2 0.2
0.1 0.1
0.9 0.9
1500
0.1
0.0
0.0
0.1
1500
0.2
0.2
0.1
0.4
1500
0.0
0.0
0.0
0.2
1500
0.4
0.2
0.1
0.9
1600
0.1
0.0
0.0
0.1
1600
0.3
0.2
0.1
0.5
1600
0.0
0.0
0.0
0.2
1600
0.4
0.2
0.1
1.0
1700
0.1
0.0
0.0
0.1
1700
0.3
0.2
0.1
0.5
1700
0.0
0.0
0.0
0.2
1700
0.5
0.2
0.1
1.1
1800
0.1
0.0
0.0
0.1
1800
0.3
0.3
0.1
0.5
1800
0.0
0.0
0.0
0.2
1800
0.5
0.3
0.1
1.1
1900
0.1
0.0
0.0
0.2
1900
0.4
0.3
0.1
0.5
1900
0.0
0.0
0.0
0.2
1900
0.5
0.3
0.1
1.1
2000
0.1
0.0
0.0
0.2
2000
0.4
0.3
0.1
0.5
2000
0.1
0.0
0.0
0.2
2000
0.6
0.3
0.1
1.1
49
-------
Table 8-3. Continued
SEASON
AUTUMN
SPRING
SUMMER
WINTER
SEASON
AUTUMN
SPRING
SUMMER
WINTER
SEASON
AUTUMN
SPRING
SUMMER
WINTER
SEASON
AUTUMN
SPRING
SUMMER
WINTER
SEASON
AUTUMN
SPRING
SUMMER
WINTER
SEASON
AUTUMN
SPRING
SUMMER
WINTER
0800
0.3
0.3
0.3
0.3
0800
0.3
0.3
0.3
0.3
0800
0.4
0.4
0.4
0.4
0800
0.6
0.6
0.5
0.6
0800
0.2
0.2
0.2
0.2
0800
0.2
0.2
0.2
0.2
0900
0.3
0.3
0.3
1000 1100
0.3 0.2
0.2 0.2
0.2 0.1
1200 1300
0.2 0.2
0.2 0.2
0.1 0.1
1400
0.2
0.2
0.1
0.6 0.3 0.3 0.2 0.2 0.2
PITY— DPMVPR ARPA— 1 IRRAM
0900 1000 1100 1200 1300 1400
0.4 0.3 0.2 0.2 0.2 0.2
0.4
0.3
0.7
0900
0.7
0.9
0.8
0.3 0.2
0.2 0.1
0.5 0.4
/—HOI IQTOM
1000 1100
0.5 0.4
0.6 0.5
0.5 0.4
1.3 0.9 0.6
PITY— HOI I^TOM
0900 1000 1100
1.2 0.7 0.5
1.5
1.4
2.1
PITV-
0900
2.2
2.4
2.2
2.3
PITY-
0900
2.4
2.6
2.5
2.4
0.9 0.6
0.8 0.4
1.5 1.1
0.2 0.1
0.1 0.0
0.4 0.3
0.1
0.0
0.3
1500
0.2
0.2
0.1
0.2
1500
0.2
0.1
0.0
0.3
1600
0.2
0.1
0.1
0.2
1600
0.2
0.1
0.0
0.3
1700
0.2
0.1
0.1
0.2
1700
0.2
0.1
0.0
0.4
1800
0.2
0.1
0.1
0.2
1800
0.2
0.1
0.0
0.4
1900
0.2
0.1
0.1
0.2
1900
0.2
0.1
0.0
0.4
2000
0.2
0.1
0.1
0.2
2000
0.2
0.1
0.0
0.4
ARPA — Rl IRAI
1200 1300 1400
0.3 0.3
0.3 0.3
0.2 0.2
0.2
0.2
0.1
0.6 0.5 0.4
ARPA — I IRRAM
1200 1300 1400
0.3 0.2 0.1
0.3 0.2
0.2 0.1
0.9 0.7
0.1
0.0
0.6
1500
0.2
0.2
0.1
0.4
1500
0.1
0.1
0.0
0.6
1600
0.2
0.1
0.1
0.4
1600
0.1
0.1
0.0
0.6
1700
0.2
0.1
0.0
0.4
1700
0.1
0.1
0.0
0.6
1800
0.2
0.1
0.0
0.4
1800
0.1
0.1
0.0
0.6
1900
0.2
0.1
0.0
0.4
1900
0.2
0.1
0.1
0.7
2000
0.2
0.1
0.0
0.4
2000
0.2
0.1
0.1
0.7
-I OQ AMf^PI PQ ARPA — Rl IRAI
1000 1100 1200 1300 1400
1.6 1.1
1.7 1.1
1.5 0.8
0.7 0.5
0.7 0.5
0.5 0.3
0.4
0.4
0.2
1.6 1.2 0.9 0.7 0.6
-I O1^ AMf^PI P1^ ARPA — 1 IRRAM
1000 1100 1200 1300 1400
1.8 1.4 1.1 0.9 0.8
2.0 1.6
1.9 1.4
1.8 1.4
1.2 0.8
1.0 0.6
1.2 1.0
0.6
0.4
0.9
1500
0.3
0.3
0.1
0.5
1500
0.7
0.5
0.3
0.8
1600
0.3
0.2
0.1
0.5
1600
0.7
0.4
0.2
0.8
1700
0.3
0.2
0.1
0.5
1700
0.7
0.4
0.2
0.8
1800
0.3
0.2
0.1
0.5
1800
0.7
0.4
0.2
0.8
1900
0.3
0.2
0.1
0.5
1900
0.7
0.4
0.2
0.9
2000
0.3
0.2
0.1
0.5
2000
0.7
0.4
0.3
0.9
Table 8-3. Continued
50
-------
SEASON
AUTUMN
SPRING
SUMMER
WINTER
SEASON
AUTUMN
SPRING
SUMMER
WINTER
SEASON
AUTUMN
SPRING
SUMMER
WINTER
SEASON
AUTUMN
SPRING
SUMMER
WINTER
SEASON
AUTUMN
SPRING
SUMMER
WINTER
SEASON
AUTUMN
SPRING
SUMMER
WINTER
Table 8-3.
0800
0.4
0.4
0.3
0.4
0800
0.4
0.4
0.3
0.4
0800
0.4
0.4
0.4
0.4
0800
0.4
0.4
0.4
0.4
0800
0.1
0.1
0.1
0.1
0800
0.1
0.1
0.1
0.1
0900 1000
0.8 0.5
0.8 0.6
0.9 0.5
1100
0.4
0.4
0.4
1200
0.3
0.4
0.2
1.3 0.9 0.7 0.5
PITY— PHPiPMlY ARPA-
0900 1000 1100 1200
0.9 0.5 0.3 0.2
0.9 0.6
0.9 0.5
0.4
0.3
0.2
0.1
1300 1400
0.3 0.2
0.3 0.2
0.2 0.1
0.5 0.5
-I IRRAM
1300 1400
0.1 0.1
0.1 0.1
0.0 0.0
1.4 1.0 0.7 0.5 0.4 0.4
PITY— PITT^RI IPf^H ARPA — Rl IRAI
0900 1000 1100 1200 1300 1400
0.8 0.6 0.4 0.3 0.2 0.2
0.8 0.5
0.6 0.3
0.3
0.2
0.3
0.1
0.2 0.2
0.1 0.1
1.5 1.3 1.1 0.9 0.7 0.6
PITY— PITTQRI IRf^M ARPA — I IRRAM
0900 1000 1100 1200 1300 1400
0.9 0.7 0.5 0.4 0.3 0.2
0.9 0.6
0.7 0.4
0.4
0.2
1.5 1.4 1.3
PITY— ^PATTI P
0900 1000 1100
0.2 0.1 0.1
0.1 0.1
0.1 0.1
0.1
0.0
0.3 0.2 0.2
PITY— QPATTI P
0900 1000 1100
0.4 0.3 0.3
0.3 0.2
0.3 0.2
0.5 0.4
0.2
0.1
0.4
0.3
0.1
0.3 0.2
0.1 0.1
1.1 1.0 0.9
ARPA — Rl IRAI
1200 1300 1400
0.1 0.1 0.1
0.0
0.0
0.2
ARPA-
1200
0.2
0.1
0.1
0.4
0.0 0.0
0.0 0.0
0.2 0.1
-I IRRAM
1300 1400
0.2 0.2
0.1 0.1
0.1 0.0
0.4 0.3
1500
0.2
0.2
0.1
0.4
1500
0.1
0.1
0.0
0.3
1500
0.2
0.2
0.0
0.6
1500
0.2
0.1
0.0
0.9
1500
0.1
0.0
0.0
0.1
1500
0.2
0.1
0.0
0.4
1600
0.2
0.2
0.1
0.4
1600
0.1
0.0
0.0
0.3
1600
0.2
0.1
0.0
0.6
1600
0.2
0.1
0.0
0.9
1600
0.1
0.0
0.0
0.2
1600
0.2
0.1
0.0
0.4
1700
0.2
0.1
0.1
0.4
1700
0.1
0.0
0.0
0.3
1700
0.2
0.1
0.0
0.6
1700
0.2
0.1
0.0
0.9
1700
0.1
0.0
0.0
0.2
1700
0.2
0.1
0.0
0.4
1800
0.2
0.1
0.1
0.4
1800
0.1
0.0
0.0
0.3
1800
0.2
0.1
0.0
0.6
1800
0.2
0.1
0.0
0.9
1800
0.1
0.0
0.0
0.2
1800
0.2
0.1
0.0
0.4
1900
0.2
0.1
0.1
0.4
1900
0.1
0.0
0.0
0.3
1900
0.2
0.1
0.0
0.6
1900
0.2
0.1
0.0
0.9
1900
0.1
0.0
0.0
0.2
1900
0.2
0.1
0.0
0.4
2000
0.2
0.1
0.1
0.4
2000
0.1
0.0
0.0
0.3
2000
0.2
0.1
0.0
0.6
2000
0.2
0.1
0.0
1.0
2000
0.1
0.0
0.0
0.2
2000
0.2
0.1
0.0
0.4
Continued
nTY-WASHIMrvrnM ARPA-RI IRAI
51
-------
SEASON 0800 0900 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 2000
AUTUMN
SPRING
SUMMER
WINTER
0.1
0.1
0.1
0.1
0.4
0.4
0.3
0.6
0.3
0.2
0.2
0.5
0.2
0.2
0.1
0.4
0.2
0.1
0.0
0.4
0.1
0.1
0.0
0.3
0.1
0.1
0.0
0.3
0.1
0.1
0.0
0.3
0.1
0.0
0.0
0.3
0.1
0.0
0.0
0.3
0.1
0.0
0.0
0.3
0.1
0.0
0.0
0.3
0.1
0.0
0.0
0.3
CITY=WASHINGTON AREA=URBAN
SEASON 0800 0900 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 2000
AUTUMN
SPRING
SUMMER
WINTER
0.3
0.3
0.3
0.3
1.2
1.1
0.9
1.6
1.0
0.8
0.6
1.5
0.8
0.6
0.4
1.4
0.7
0.5
0.3
1.3
0.6
0.4
0.2
1.2
0.6
0.4
0.1
1.2
0.5
0.4
0.1
1.2
0.6
0.4
0.1
1.3
0.6
0.4
0.1
1.3
0.7
0.4
0.1
1.4
0.7
0.5
0.2
1.4
0.8
0.5
0.2
1.4
8.2 Discussion of OZIPR Output Tables
8.2.1 Variation by Pollutant
The above tables for secondary production generally show that late in the day (when levels
tend to be highest) formaldehyde and acetaldehyde concentrations are comparable in summer, with
somewhat more acetaldehyde than formaldehyde in winter. For most hours in most scenarios, the
formaldehyde concentration was between 3 and 10 pg m"3, with extreme values ranging from 0.8 to
19.2 |_ig m"3. For acetaldehyde, the typical values ranged from 3 to 20 |_ig m"3, with extreme values
varying from 1.1 to 35.4 |_ig m"3. The levels of acrolein were much lower than the other two HAPs
at all times. For acrolein, the typical levels were 0.1 to 0.4 |_ig m"3, with extreme levels ranging from
zero (i.e., below 0.05 |_ig m"3) to 2.6 |_ig m"3. Concentrations of acrolein above 1 |_ig m"3 are generally
seen only early in the day, as photochemical destruction of acrolein tends to outpace formation.
8.2.2 Variation by Hour of Day
For most scenarios, both formaldehyde and acetaldehyde concentrations tend to increase over
the course of the day, with the daily maximum occurring near (or at) the end of the simulation. Some
scenarios also show a morning peak around 10 or 11 a.m., creating a bimodal profile. It should be
noted that the 8 a.m. values in the above tables are determined by the initial conditions provided to
the OZIPR model, which are somewhat uncertain because the chemistry of the immediately
preceding hours (before 8 a.m.) was not modeled. Because the reaction rates that dominate
production and removal of the target HAPs are fairly fast, values later in the day tend to reflect the
chemical transformation of recent emissions, with relatively little impact from the initial conditions.
For both formaldehyde and acetaldehyde, the largest concentrations occur late in the day when the
initial conditions have almost no influence. On the other hand, acrolein levels, while always small,
52
-------
tend to peak in the morning around 9 a.m. Therefore, the estimation of the daily maximum
concentration for acrolein presumably would benefit from an improved characterization of the early
morning chemistry.
8.2.3 Variation by Season
The formaldehyde concentrations in Table 8-1 show relatively little variation by season. Even
the final hour of the simulations (8 p.m.), which shows more variation than other hours, is generally
within 20% of the annual mean in all seasons. Acetaldehyde, particularly in urban areas, shows much
more seasonal variation, with autumn and winter values sometimes at two or three times the levels
seen in spring and summer. The ratio of acetaldehyde to formaldehyde is generally close to unity in
the summer, but several scenarios show substantially larger ratios in winter. Observed data (see the
discussion later in this section) tend to support ratios near unity, but most observations are made in
the summertime.
8.2.4 Urban/Rural Variation
In every study area except Phoenix, the urban area showed higher concentrations of all three
pollutants than the rural area. The size of the differences between the urban and rural results depend
on the particular choice of counties in each study area. For cases including Washington, DC, and
Chicago, the urban area is almost entirely built up, with a universally very high road and traffic
density. The rural areas for these cities (see Table 2-1 for the counties) are largely farms or forest,
with very different emissions. For several western cities, particularly Phoenix, the nominally "urban"
county is extremely large, and in fact is mostly nonurbanized. In such cases, the box model as used
by OZIPR effectively averages urban and rural areas within the same county, which reduces the
contrast with truly rural areas. Note that for all three pollutants, the very high concentrations
occurred in urban areas. The only exception to this was for Los Angeles, where Ventura County was
used as the rural study area. Compared to the rest of the Los Angeles area, Ventura County is
relatively unpopulated, but with 700,000 people, it is comparable to some of the urban counties in
other study areas.
Another observation from these OZIPR runs is that there often appeared to be a time lag of
about an hour or two for the rural areas in the development of the peak for formaldehyde and
acetaldehyde. This may be a reflection of morning rush hour emissions in the urban areas and the
greater importance of biogenic emissions in the rural areas.
53
-------
8.3 Annual Averages of Target Compounds from Secondary Formation
The results discussed above were used to estimate the secondary contribution of
formaldehyde, acetaldehyde, and acrolein on an annual basis. The determination of an annual
average required estimates for the 12-hour nighttime period for which the model was not run. OZIPR
or any photochemical model is designed to be used during daylight hours when radical
concentrations are sufficiently high to render a reasonable conversion of precursors to products.
During periods when reactive conversions are minimized, advective transport plays a much bigger
role. Therefore, the OZIPR model was not run at night and the nighttime formaldehyde,
acetaldehyde, and acrolein concentrations were obtained by linear interpolation. This was done by
interpolating between the 8 p.m. value and using the 8 a.m. value as the concentration 12 hours later.
In the occasional instances where the initial value of the day was greater than the ending value, the
nighttime concentrations were simply set equal to the 8 p.m. level. This procedure was consistent
with the expectation of decreasing aldehyde concentrations at night.
Once nighttime values were obtained for each OZIPR scenario, annual estimates of secondary
production were determined as follows. First a daily value for each pollutant under each scenario was
computed by averaging across all hours. Seasonal means were constructed by weighting each daily
value according to the frequency of the typical day. (The weighting technique is illustrated in the
example calculation for Tables 8-1 through 8-3.) The seasonal means were then averaged to generate
the final annual estimate of secondary production. Similarly, using the OZIPR runs that included
HAPs emissions, annual total levels were derived for each pollutant. Annual primary levels were
found by subtraction. The resulting annual numbers are tabulated in Tables 8-4, 8-5, and 8-6 for
formaldehyde, acetaldehyde, and acrolein, respectively. Table 8-7 provides the percentage of
secondary production on an annual basis for both urban and rural areas for each city. (Note that the
percentages for acrolein are somewhat more uncertain than the others due to the low concentrations
combined with model round-off error.) Figures 8-1 through 8-3 display maps of annual average
secondary production estimates for the target HAPs.
54
-------
Table 8-4. Annual Averages for Formaldehyde Concentrations. All values are |ig m3.
City
Atlanta
Boston
Chicago
Denver
Houston
Los Angeles
Phoenix
Pittsburgh
Seattle
Washington, DC
Secondary
6.8
4.0
6.2
5.2
8.8
13.4
5.0
6.4
3.0
7.8
Table 8-5. Annual Averages
City
Atlanta
Boston
Chicago
Denver
Houston
Los Angeles
Phoenix
Pittsburgh
Seattle
Washington, DC
Secondary
9.8
7.5
14.9
7.0
10.9
18.0
5.0
11.4
5.2
13.4
Urban
Primary
0.7
0.5
0.8
0.8
0.9
1.1
0.5
0.7
0.3
1.0
Total
7.5
4.5
7.1
6.0
9.7
14.5
5.4
7.1
3.3
8.8
Secondary
5.3
2.6
2.0
3.0
5.2
11.8
4.4
5.6
2.2
2.5
for Acetaldehyde Concentrations. All
Urban
Primary
1.3
1.1
1.9
1.2
1.2
1.3
0.3
1.1
0.6
2.4
Total
11.1
8.6
16.8
8.3
12.0
19.3
5.3
12.5
5.9
15.8
Secondary
7.2
3.2
2.6
5.1
7.7
13.8
6.2
8.9
2.4
2.7
Rural
Primary
0.5
0.2
0.2
0.3
0.5
0.8
0.4
0.5
0.2
0.2
values are |ig
Rural
Primary
0.6
0.3
0.2
0.4
0.7
0.8
0.5
0.7
0.2
0.2
Total
5.8
2.8
2.2
3.3
5.7
12.7
4.8
6.1
2.3
2.7
m3.
Total
7.8
3.5
2.8
5.5
8.4
14.6
6.7
9.6
2.7
2.9
55
-------
Table 8-6. Annual Averages for Acrolein Concentrations. All values are |ig m"3.
City
Secondary
Atlanta
Boston
Chicago
Denver
Houston
Los Angeles
Phoenix
Pittsburgh
Seattle
Washington, DC
Table 8-7. Annual
City
0.2
0.2
0.4
0.2
0.4
0.7
0.2
0.4
0.2
0.6
Secondary
Formaldehyde
Atlanta
Boston
Chicago
Denver
Houston
Los Angeles
Phoenix
Pittsburgh
Seattle
Washington, DC
91%
89%
87%
87%
91%
92%
92%
90%
91%
89%
Urban
Primary
0.0
0.1
0.1
0.1
0.1
0.0
0.0
0.1
0.0
0.1
Production as
Urban
Acetaldehyde
88%
87%
89%
84%
91%
93%
94%
91%
88%
85%
Total Secondary
0.2
0.3
0.5
0.3
0.4
0.7
0.2
0.4
0.2
0.7
a Percentage
0.1
0.1
0.1
0.2
0.3
0.4
0.3
0.3
0.1
0.1
of Total
Acrolein Formaldehyde
82%
76%
78%
78%
88%
93%
94%
87%
83%
80%
91%
93%
91%
91%
91%
93%
92%
92%
96%
92%
Rural
Primary
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Rural
Acetaldehyde
92%
91%
93%
93%
92%
94%
92%
93%
89%
93%
Total
0.2
0.1
0.1
0.2
0.3
0.5
0.3
0.3
0.1
0.1
Acrolein
84%
76%
82%
89%
88%
94%
94%
88%
77%
89%
56
-------
Urban Formaldehyde
s
Formaldehyde (ug/m3)
2.0- 3.9
• 4.0- 6.9
• 7.0- 8.9
9.0 - 10.9
A 11.0-14.0
Figure 8-la. Annual averages for secondary formaldehyde concentrations in
urban areas.
Rural Formaldehyde
PHOE
Formaldehyde (ug/m3)
• 2.0- 3.9
• 4.0- 6.9
• 7.0- 8.9
% 9.0-10.9
A 11.0-14.0
Figure 8-lb. Annual averages for secondary formaldehyde concentrations in
rural areas.
57
-------
Urban Acetaldehyde
SEA
LOS ANGEL
Acetaldehyde (ug/m3)
• 2.0- 5.9
• 6.0- 8.9
• 9.0-11.9
12.0 -14.9
ft 15.0-18.0
Figure 8-2a. Annual averages for secondary acetaldehyde concentrations in
urban areas.
Rural Acetaldehyde
LOS ANGEL
Acetaldehyde (ug/m3)
• 2.0- 5.9
• 6.0- 8.9
• 9.0-11.9
12.0 - 14.9
A 15.0-18.0
Figure 8-2b. Annual averages for secondary acetaldehyde concentrations in rural
areas.
58
-------
Urban Acrolein
PHOENIX (0.2)
——Lj
Acrolein (ug/m3)
• 0.1 - 0.2
• 0.2 - 0.3
• 0.3 - 0.5
O 0.5-0.6
A 0.6-0.7
Figure 8-3a. Annual averages for secondary acrolein concentrations in urban areas.
Rural Acrolein
s
(0.1)
I
ITTSBURGH (0.3)
Acrolein (ug/m3)
0.1 - 0.2
• 0.2 - 0.3
• 0.3 - 0.5
£ 0.5-0.6
A 0.6-0.7
Figure 8-3b. Annual averages for secondary acrolein concentrations in rural areas.
59
-------
The most striking feature of the annual estimates is that in all cases the secondary production
accounts for the vast maj ority of the total for each pollutant. For urban formaldehyde, the percentage
due to secondary formation ranged from 87% to 92%, while in rural areas the percentages were even
higher, ranging from 91% to 96%. For acetaldehyde, the urban range was 84% to 94%, and the rural
range was 89% to 94%. For acrolein, the percentages ranged from 76% to 94% in both urban and
rural areas. These results are notable for their uniformity across the diverse set of study areas
selected. Table 8-7 also indicates that the percentage of secondary formation in rural areas is
generally higher than in urban areas.
For formaldehyde, in every case the rural estimate is lower than the corresponding urban
estimate for primary, secondary, and total levels. The Los Angeles study area has distinctly higher
formaldehyde concentrations than the other nine areas. These results also apply to acetaldehyde,
except that the rural estimates for Phoenix exceed the urban estimates. These results also apply to
acrolein. However, as mentioned before, the levels are quite low, so these differences are very small
and possibly subject to round-off error.
Depending on the city, rural areas were estimated to have between 31% and 89% of the total
formaldehyde present in urban areas. Rural acetaldehyde fell between 17% and 77% of the urban
levels, except in Phoenix, where rural values exceeded urban ones.
8.4 Comparison of Target Compounds with Other Models or Measurements
The OZIPR model results were compared to monitored values from the Photochemical
Assessment Monitoring Stations (PAMS) and numbers from the Regional Acid Deposition Model
(RADM). The available PAMS measurements for formaldehyde, acetaldehyde, and acrolein from
1995 to 1997 are summarized in Table 8-8. Measurements of the carbonyl species in the PAMS
network are typically taken using silica gel or C18 cartridges coated with 2,4-dinitrophenyl-hydrazine
(DNPH). Measurements of formaldehyde using the silica gel cartridge can suffer from a negative
artifact due to ozone. While the use of potassium iodide scrubbers is recommended for both types
of cartridges (Kleindienst et al., 1998), there is generally no consistent manner of conducting these
measurements, and the possibility of systematic errors exists. Table 8-8 also contains the urban levels
estimated by OZIPR.
60
-------
Table 8-8. Comparison Values for Carbonyl Compounds (|j,g m3) from the Photochemical
Assessment Monitoring Stations (PAMS) Collected During the Photochemical Pollutant Seasons
for 1995-1997
City
Atlanta
Atlanta
Chicago
Chicago
Denver
Denver
Denver
Houston
Houston
Los Angeles
Los Angeles
Washington, DC
Washington, DC
Pollutant
Acetaldehyde
Formaldehyde
Acetaldehyde
Formaldehyde
Acetaldehyde
Acrolein
Formaldehyde
Acetaldehyde
Formaldehyde
Acetaldehyde
Formaldehyde
Acetaldehyde
Formaldehyde
No.
Observations
200
212
277
277
7
7
7
1362
1364
604
590
1461
1461
PAMS
Mean
3.1
3.3
2.6
8.0
37.9
0.3
5.2
4.0
6.4
12.4
6.3
3.1
7.4
Standard
Deviation
3.7
2.8
1.6
7.5
28.5
0.5
0.9
3.7
4.5
37.5
3.8
1.7
4.0
OZIPR
Urban
11.1
7.5
16.8
7.1
8.3
0.3
6.0
12.0
9.7
19.3
14.5
15.8
8.8
For the six cities in Table 8-8, one finds that the differences between the OZIPR and PAMS
mean values range from -0.9 to 8.2 |_ig m"3 for formaldehyde. Except Denver, with a difference of
-29.6 |_ig m"3, the differences for acetaldehyde varied from 6.9 to 14.2 |_ig m"3. The lone case of
acrolein showed no difference.
Note that the PAMS values are not annual averages, and Denver only had seven
measurements. Most of the PAMS observations represented in Table 8-8 are summertime
measurements. Additionally, the PAMS data are collected at specific sites and may not represent
areal averages. Also recall that the time periods for the PAMS data and the OZIPR modeled results
do not match exactly. Despite these limitations, the data suggest that results from the model are
comparable to those values measured in corresponding ambient environments. This conclusion is
strongest for formaldehyde. Note from Table 8-2 that acetaldehyde is elevated in the cooler seasons,
and, therefore, the annual averages from OZIPR may be expected to be larger than the PAMS
measurements.
61
-------
Comparisons of the results of the OZIPR model were also made with a sample output from
the RADM model. RADM is a three-dimensional model that predicts (among other quantities)
formaldehyde and acetaldehyde levels by combining both dispersion and chemistry. Since the model
is highly complex and expensive to run, comparisons were made for an archived model run. RADM
output values for the period July 11-15, 1995, are given in Table 8-9.
Table 8-9. Comparison Values for Carbonyl Compounds (|ig m"3) from the RADM Model
Run for July 11-15,1995, for Cities in This Study
City
Atlanta
Atlanta
Boston
Boston
Chicago
Chicago
Houston
Houston
Pittsburgh
Pittsburgh
Washington
Washington, DC
No.
Pollutant Observations
Acetaldehyde
Formaldehyde
Acetaldehyde
Formaldehyde
Acetaldehyde
Formaldehyde
Acetaldehyde
Formaldehyde
Acetaldehyde
Formaldehyde
Acetaldehyde
Formaldehyde
120
120
120
120
120
120
120
120
120
120
120
120
Mean
Value
27.4
14.1
7.5
7.3
9.5
7.3
20.9
12.1
10.2
8.6
12.6
10.5
Standard
Deviation
22.9
6.5
3.9
3.5
6.5
2.6
18.6
6.6
8.2
4.4
10.7
4.9
OZIPR
Urban
11.1
7.5
8.6
4.5
16.8
7.1
12.0
9.7
12.5
7.1
15.8
8.8
Generally, the OZIPR results are slightly lower than the RADM results, though there are
exceptions. It should be noted that the RADM results are for a single ozone episode, whereas the
OZIPR values are annual averages. The data from both OZIPR and RADM show a molar ratio of
formaldehyde to acetaldehyde that is much closer to unity than seen in the PAMS data. The
consistency between the values generated from OZIPR and RADM is thus quite good. This may
result from a common set of assumptions, particularly regarding the photochemical mechanism.
There were four cities (Atlanta, Chicago, Houston, and Washington, DC) with both PAMS
and RADM data. For Atlanta and Houston, the OZIPR estimates for both formaldehyde and
62
-------
acetaldehyde were between the PAMS and RADM values. For Chicago, the OZIPR estimate of
acetaldehyde exceeded both of the others, but the formaldehyde was very slightly below both the
PAMS and RADM data. For Washington, DC, the acetaldehyde was above both (but close to the
RADM value), while formaldehyde was between the PAMS and RADM values.
The general conclusion from the PAMS and RADM comparisons is that the simplified
approach taken here is adequate for obtaining reasonable estimates for the target HAPs considered
in this study. In the opinion of the authors, concentrations from this approach are probably within
a factor of two of the ambient concentrations and considerably better than an order of magnitude.
This level of accuracy may be sufficient to estimate risk factors.
8.5 Limitations of Current Study
The use of a photochemical box model, such as OZIPR, to determine area-wide
concentrations of toxic compounds involves a number of assumptions that will qualify the results.
The initial concept in formulating the approach here was to use the results from a box model
representing secondary formation to augment the results from a dispersion model. Ideally, the
dispersion model should be intrinsically combined with the photochemical model since the physical
system is highly coupled. Examples of such models include RADM and EPA's Urban Airshed
Model (UAM). The present approach has rendered reasonable results because, for the HAPs
considered here, a high percentage of the compound formation is due to secondary chemistry. Thus,
the resultant concentrations of HAPs are due more to the area emissions of hydrocarbon precursors
rather than to point source emissions of the compounds themselves.
The chemical mechanism itself has some additional limitations. First, little experimental
information is available on the products from secondary reactions of first-generation reaction
products. This will lead to some uncertainty in carbonyl formation from these processes. Second, the
mechanism contains no chemical reactions for a wide number of oxygenated hydrocarbons, such as
alcohols, ethers, and esters. (This might be significant in cities that use reformulated fuels.) In the
present study, the oxygenates represent a minor fraction of the hydrocarbons and were distributed
among the reactivity classes given in Section 4. Finally, for three of the cities (Pittsburgh, Seattle,
and Phoenix), no reactivity data were available and were assumed based on other cities similar in
geography and population. Note that the limitations described here would apply equally to a three-
dimensional model.
Box models in general have some additional limitations. Perhaps the most significant of these
is that the quantities being modeled are distributed homogeneously throughout the box. In this study,
the finest spatial scale for emissions data was the county level. For a large county such as Los
63
-------
Angeles County, which has an area of 10,515 km2, most of the emissions are concentrated over a
much smaller land area; localized areas of high concentration were not identified.
Nighttime conditions can be difficult to represent in a photochemical model since the basic
assumption that photochemistry is dominant is violated. Finally, some physical parameters such as
wet deposition and cloud cover are not considered in the OZIPR model; these effects may lead to
overestimates of HAPs when these phenomena are present.
8.6 Study Summary
In this study, emissions and meteorological data were obtained as input to a one-dimensional
photochemical model to estimate the contributions to formaldehyde, acetaldehyde, and acrolein from
secondary processes. These data were incorporated into the model without any use of adjustable
parameters or other devices to tune the output concentrations. Within the constraints of the approach,
the model gave reasonable concentrations for the aldehydes of interest.
The following specific results were found:
1. In ambient air, photochemical production, on average, accounts for approximately
90% of the formaldehyde and acetaldehyde and approximately 85% of the acrolein
concentrations.
2. For the cities with both PAMS and RADM data, the OZIPR results generally were
between the two. The simplified approach taken here is adequate for obtaining
reasonable estimates for the target HAPs considered.
3. The estimates for the rural scenarios were lower than for the corresponding urban
scenarios.
4. The simplified approach used in this study (to be used to augment the results of a
dispersion model) is a viable alternative to more complex and expensive
methodology, such as the RADM and UAM models.
64
-------
Section 9
Practical Use of the Results from This Study
The results from this report are designed to be of practical use to organizations that may be
required to scrutinize ambient standards for formaldehyde, acetaldehyde, or acrolein. The
information presented here can be used at any of three levels of application depending on the type
of analysis desired. The results presented can also be extended for cities not explicitly modeled. (1)
The most detailed information on secondary carbonyl formation for a specific metropolitan area can
be obtained from running this model using city-specific parameters. (2) Tables 8-1 through 8-3 can
be used in conjunction with a dispersion model to provide seasonal and time-specific adjustments.
Table 8-4 through 8-6 and Figures 8-1 through 8-3 can be used to make annual adjustments for
secondary formation to apply to a dispersion model's results. (3) When the cities examined in this
study are too dissimilar from the area of interest and running the model is not feasible, the results
found here can still be used to adjust predictions of aldehyde concentrations for secondary
production.
9.1 Use of the OZIPR Model to Estimate Secondary HAP Production
To obtain information for a specific city regarding secondary formation of formaldehyde,
acetaldehyde, or acrolein, OZIPR may be run with input tailored for that given city. The computer
code for the model is available from EPA. Sample input files are given elsewhere in this report. (See
Section 7 and Appendix A.) This section will describe the general principles involved in using the
model and input data requirements. This will allow the user to determine whether there is sufficient
information to run the model. The user may be able to obtain most of the data needed with little
trouble. Mixing height and emissions inputs may be the most problematic to acquire. For example,
mixing heights are collected from a relatively sparse network; a user may need to decide whether or
not to utilize data from the closest sounding site or otherwise estimate mixing heights for input to
the model. As another example, certain VOC species may not be available explicitly in many
emissions data sets; the user may need to devise a means for supplying these. As there are myriad
possibilities for scenarios that may be of interest, no attempt is made here to specify how a user
should address the specific situation being modeled.
The user-supplied input file requires geographic, temporal, meteorological, and emissions
65
-------
data, including the VOC composition (e.g., see Section 4). The user must also supply boundary
conditions. Geographical information includes the longitude and latitude of the city being modeled;
temporal information includes the date of the simulation and the starting and stopping times. The
meteorological information includes the temperature, relative humidity, and barometric pressure. The
model accepts mixing heights on an hourly or twice daily basis. If two values per run are supplied,
the user may supply specific times or use the model defaults.
Required emissions data include the area-wide emissions of volatile organic compounds
(VOCs), nitrogen oxides, and specific HAPs (if total HAPs levels are desired). The model also
allows for including 1,3-butadiene, which is a hydrocarbon precursor for acrolein. As run in this
study, emission rates must be provided in units of kg km"2 h"1. For example, if total mass emission
rate (kg/hr) of a chemical as a function of time of day are known for a county, these values must be
divided by the area of the county to get the appropriate units.
Depending on the user's needs, the OZIPR model may be run for one or more specific days
or one or more "representative" days (as was done for this report). One input file is necessary for
each modeled day. Individual model scenarios run in a few seconds on a modern personal computer.
Output values from the model are given in parts per million by volume (ppm). The values from
secondary formation can then be used in conjunction with concentration estimates from dispersion
models, as described below.
9.2 Use of Secondary Formation Values with Dispersion Models
The tables in Section 8 were developed to allow adjustments for secondary production to be
applied to dispersion models. Here the term dispersion model is used to indicate a model that
predicts concentration as a function of location, and possibly time, utilizing emissions from point
sources as input, without considering photochemical reactions.
Concentration values output from the OZIPR model have no spatial component. The
concentrations of all compounds are homogeneously distributed within the box. As explained earlier,
the model was run with and without primary emissions. The run that includes primary emissions
gives the total HAPs concentrations. The HAPs components due to secondary formation are obtained
directly from the run without primary emissions. The difference between the two runs gives the
component due to primary emissions.
The following is an example of how the entries in Tables 8-1 through 8-3 can be used.
Consider the entries for urban Atlanta for formaldehyde in Table 8-1. All values in the table give pg
m"3 secondary contributions as a function of season and hour of day. These represent instantaneous
values from the model. Thus, if one has output from a dispersion model for Atlanta for a typical
66
-------
autumn day at 12:00 noon, a value of 7.4 |_ig m"3 would have to be added to every grid point to
represent the secondary contribution. Advancing the clock by one hour to 1:00 p.m., a new value of
8.1 |_ig m"3 would now be used to represent the secondary contribution. For other times, a linear
interpolation of the tabular data could be used.
As in the case above, the formaldehyde, acetaldehyde, and acrolein values in Tables 8-4
through 8-6 are designed to be additive adjustments. These tables are to be applied on an annual time
scale. Thus, they would be used with output from a dispersion model that did not have a finer time
resolution. Again taking urban Atlanta as an example, Table 8-4 would be used for formaldehyde,
and a value of 6.8 \ig m"3 would be added to the dispersion estimates in a spatially uniform manner.
If one is interested in a location not appearing in this study and running the OZIPR model
is not feasible, then the area of interest should be matched as closely as possible to one of those
modeled here. The cities examined in this study cover a variety of urban areas. Users will need to
make the decision of suitability on a case-by-case basis. The most obvious considerations are factors
such as population size, industrial base, geography, land use, and climate. Note that both urban and
rural scenarios are included in this report. Generally, secondary formation in metropolitan areas
should be estimated using the tabulated urban values and outlying areas should use the rural values.
However, for smaller and medium-size urban areas, the rural values reported here may be more
appropriate.
9.3 Estimation of Secondary Formation without Using Tables
If none of the cities in this report typifies the area of interest, the secondary component of the
HAPs concentrations may be estimated from values for the primary component. This can be seen by
examining the percentages in Table 8-7. Note that for formaldehyde and acetaldehyde the percentage
of secondary production is near 90% for all of the urban areas (and slightly higher in the rural areas).
For acrolein, both the urban and rural areas averaged near 85% secondary, with somewhat more
variation from city to city than for the other two HAPs. In the absence of better information, these
average percentages could be used to estimate the effect of secondary formation in any study area.
It should be noted that the percentages in Table 8-7 represent the ratio of spatially averaged
secondary production to spatially averaged total production for each HAP. These averages are over
the entire study area (one or more counties). There is good reason to believe that the spatial
distribution of secondary HAPs does not match the spatial distribution for primary HAPs. Primary
HAPs will be concentrated around the large point sources in the study area. The distribution of
secondary HAPs will differ for two reasons: (1) the pattern of sources for the chemical precursors
that transform into the target HAPs may not match the source pattern for direct emission of the
67
-------
HAPs, and (2) the reactions take some time, which means that the precursors may move some
distance from the emission location before being transformed into HAPs.
As a result, it is not correct to simply amplify the distribution of primary HAPs as determined
by a dispersion model by a multiplicative factor to account for secondary formation. That would lead
to very high estimates of total HAPs concentrations at the localized "hot spots" seen in the primary
distribution, with little evidence that such levels are accurate.
A better approach is to first calculate a spatial average for the primary HAPs concentrations,
then to multiply this by a factor of 9.0 (90% secondary / 10% primary), or 5.7 (85% / 15%), as
appropriate, to find the average secondary contribution. The resultant secondary contribution may
then be added in a spatially uniform manner to the distribution of primary HAPs.
The method described here will preserve the character of the spatial distribution seen for the
primary HAPs (in particular, the location of any hot spots in the study area), since the secondary
component affects all locations equally. However, the method is more conservative than a simple
multiplicative scaling of the primary distribution. There is no reason to expect the secondary HAPs
concentrations at primary hot spots to be either higher or lower than elsewhere. In the absence of
knowledge of the secondary spatial distribution, a uniform distribution is a reasonable choice.
68
-------
References
Atkinson, R. 1990. Gas-phase tropospheric chemistry of organic compounds: A review. Atmospheric
Environment 24 A: 1-41.
Carlier, P., H. Hannachi, G. Mouvier. 1986. The chemistry of carbonyl compounds in the
atmosphere—A review. Atmospheric Environment 20:2079-2099.
Carter, W.P.L. 1990. A detailed mechanism for the gas-phase atmospheric reactions of organic
compounds. Atmospheric Environment 24A:481-518.
Carter, W.P.L. and F.W. Lurmann. 1991. Evaluation of a detailed gas-phase reaction mechanism
using environmental chamber data. Atmospheric Environment 25A:2771-2806.
Gery M.W. and Grouse, R.R. 1990. User's Guide for Executing OZIPR, Office of Research and
Development, U.S. Environmental Protection Agency, Research Triangle Park, NC.
Grosjean, D., R.D. Swanson, C. Ellis. 1983. Carbonyls in Los Angeles air: Contribution of direct
emissions and photochemistry. Sci. Total Environ., 29:65-85.
Harley, R.A. and G.R. Cass. 1994. Modeling the concentrations of gas-phase toxic organic air
pollutants: Direct emissions and atmospheric formation. Environ. Sci. TechnoL, 28:88-98.
Kleindienst, T.E., E.W. Corse, F.T. Blanchard, and W.A. Lonneman. 1998. Evaluation of the
performance of DNPH-coated silica gel and C18 cartridges in the measurement of formaldehyde in
the presence and absence of ozone. Environ. Sci. TechnoL, 32:124-130.
SAS. 1990. SAS/STATUser 's Guide, Version 6, Fourth Edition, Volume /, SAS Institute, Cary, NC.
Seila, R.L., W.A. Lonneman, and S.A. Meeks. 1989. Determination of C2 to C12 hydrocarbons in 39
U.S. cities from 1984 through 1986. U.S. EPA/600/3-89/058 U.S. EPA, Office of Research and
Development, U.S. Government Printing Office, Washington, D.C.
69
-------
TECHNICAL REPORT DATA
(Please read Instructions on reverse before completing)
1. REPORT NO.
EPA-454/R-99-054
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
A Simplified Approach for Estimating Secondary
Production of Hazardous Air Pollutants (HAPs) Using the
OZIPR Model
Replaces Section 2.4 in Dispersion Modeling of Toxic
Pollutants in Urban Areas, EPA-454/R-99-021
5. REPORT DATE
December 1999
PERFORMING ORGANIZATION CODE
AUTHOR(S)
PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
10. PROGRAM ELEMENT NO.
ManTech Environmental Technology
2 Triangle Drive
Research Triangle Park, NC 27411
11. CONTRACT/GRANT NO.
EPA Contract No. 68D50049
12. SPONSORING AGENCY NAME AND ADDRESS
13. TYPE OF REPORT AND PERIOD COVERED
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Emissions, Monitoring & Analysis Division
Research Triangle Park, NC 27711
Final Report
15. SUPPLEMENTARY NOTES
EPA Work Assignment Manager: Thomas McCurdy
16. ABSTRACT
Hazardous air pollutants are found in the atmosphere as a result of primary emissions
or from the transformation of organic compounds emitted into the atmosphere. Several
very complex models exist that can include both dispersion and atmospheric chemistry to
yield HAPs concentration estimates. However, these models are very expensive to
execute, often requiring the use of supercomputers. This report describes a simplified
approach for estimating total HAPs concentrations by estimating secondary HAP with a
stand-alone model, run in a personal computing environment, that incorporated only non
dispersive processes, such as photochemistry. The results from this model would then
be coupled to those from a relatively simple dispersion model such as the EPA's
Industrial Source Complex (ISC) dispersion model which uses primary emissions.
17. KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b. IDENTIFIERS/OPEN ENDED TERMS
Air Pollution
Air Quality
Dispersion Models
Meteorology
Air Toxics
Urban Area Modeling
18. DISTRIBUTION STATEMENT
Release Unlimited
SECURITY CLASS (Report)
Unclassified
20. SECURITY CLASS (Pacje)
Unclassified
EPA Form 2220-1 (Rev. 4-77)
PREVIOUS EDITION IS OBSOLETE
70
------- |