EPA-R4-73-030f
July 1973
ENVIRONMENTAL MONITORING SERIES
ill
ill
mm
111
III
-------
EPA-R4-73-030f
URBAN AIR SHED PHOTOCHEMICAL
SIMULATION MODEL STUDY
VOLUME II - USES'S GUIDE AND DESCRIPTION
OF COMPUTER PROGRAMS
by
S. D. Reynolds
Systems Applications, Inc.
9418 Wilshire Boulevard
Beverly Hills, California 90212
Contract No. 68-02-0339
Program Element No. 1A1009
EPA Project Officer: Herbert Viebrock
Meteorology Laboratory
National Environmental Research Center
Research Triangle Park, North Carolina 27711
Prepared for
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
July 1973
-------
This report has been reviewed by the Environmental Protection Agency and
approved for publication. Approval does not signify that the contents
necessarily reflect the views and policies of the Agency, nor does
mention of trade names or commercial products constitute endorsement
or recommendation for use.
-------
TABLE OP CONTENTS
Page
INTRODUCTION 1
I. SYSTEM OVERVIEW 3
II. DESCRIPTION OP THE PROGRAMS 10
A. The Emissions Data Preparation Program 10
1. Program Objective and Capabilities 10
2. Program Structure 11
3. Input Data Requirements ..... 13
4. Program Output 14
B. The Meteorological Data Preparation Program 14
1. Program Objective and Capabilities 14
2. Program Structure 16
3. Input Data Requirements 17
4. Program Output 17
C. The Atmospheric Pollution Simulation Program .... 17
1. Program Objective and Capabilities 17
2. Program Structure 18
3. Input Data Requirements 28
4. Program Output 29
D. The Data Plotting Program 29
1. Program Objective and Capabilities 29
2. Program Structure 31
3. Input Data Requirements 38
4. Program Output 38
III. OPERATING PROCEDURES 39
A. General Information and Instructions 39
B. Program Operating Characteristics 40
1. tte Emissions Data Preparation Program 40
2. The Meteorological Data Preparation Program ... 50
-------
Page
3. The Atmospheric Pollution Simulation
Program 50
4. The Data Plotting Program 66
IV. LISTINGS, SYMBOL GLOSSARIES, AND SAMPLE OUTPUT. . . 71
A. The Emissions Data Preparation Program .... 72
1. Program Listing 72
2. Symbol Glossary 80
3. Sample Output 86
B. The Meteorological Data Preparation Program . . 90
1. Program Listing 90
2. Symbol Glossary 95
3. Sample Output 97
C. The Atmospheric Pollution Simulation Program . 100
1. Program Listing 100
2. Symbol Glossary 135
3. Sample Output 149
D. The Data Plotting Program 166
1. Program Listing 166
2. Symbol Glossary 179
3. Sample Output 185
REFERENCES 189
11
-------
INTRODUCTION
The simulation model described in Volume I of this report is
a dynamic model that was developed to predict ground level concentrations
of inert and photochemically reactive atmospheric contaminants and their
variations in space and time. Since the governing equations that comprise
the model are coupled, nonlinear, second-order partial differential equations,
a numerical, rather than analytical, solution is required. In addition,
a comprehensive emissions and meteorological data base is needed to
exercise the model. Clearly, a model of this computational complexity
must be implemented on a digital computer. It is the purpose of the
User's Guide to describe both the overall and specific operational
characteristics of the computer programs which embody the model.
To realize the full utility of the model, the computer programs have
been structured for general use, and not for exclusive application to a
particular airshed. Thus, the inputs to the programs are parameters
which specify the characteristics of the region to be modeled. For any
specific application, minor modification of the codes may be required to
account for individual airshed characteristics which are not treated in
the general formulation of the model or codes. Since the simulation study
of Los Angeles, reported in Volume I, is the first extensive application
of the model, the codes included in this Guide reflect some of the
characteristics which are unique to the Los Angeles airshed. In Section III,
we discuss the alterations that we have made in the general codes in
order to perform the present study.
As a part of this modeling effort, we have developed an airshed
simulation package consisting of four computer programs. Three of the
programs are subsidiary to the main program, which is used to predict
ground level pollutant concentrations. We note that there are a large
number and variety of inputs to an airshed model. As a result, we have
written two specialized programs which perform specific tasks involving
input data. One program carries out operations dealing with meteorological
input data, while the other processes pollutant emissions inputs. The
third subsidiary program prepares plots of the results predicted by the
main airshed simulation program. We have found that structuring the system
in this manner greatly simplifies its use.
The User's Guide is segmented into four sections. In Section I,
we present an overview of the airshed simulation package. The main
objective is to explain the basic computational structure of the package,
indicating the function of all programs. To illustrate the manner in which
a typical airshed simulation is performed, we describe the step-by-step
application of each program, discussing required inputs, calculations.
performed, and use of the program output.
Section II is devoted to a detailed description of the individual
programs, focusing on objectives, structure and capabilities. Since
complete disclosures of all algorithms are included in Volume I of this
report and in Roth et al. (1971), this discussion will deal only super-
ficially with the technical content of the model. When appropriate, we refer
the user to related sections in either or both of the reports cited for an
expanded discussion of the topic.
-------
The successful implementation of the airshed simulation package
requires detailed knowledge of individual program operating procedures.
In Section III, we present a complete set of instructions for the use
of all programs. Computing system information is given, including
program size, computing time, and hardware requirements. Since a large
volume of data, provided as punched cards, is input to the programs, we
have formulated a set of tables depicting the data-deck setup for every
program. The tables include a detailed description of the input
parameters, their units, and the format of each card.
Finally, listings of all programs are provided in Section IV. In
addition, we have compiled for each program a symbol glossary describing
important variables and their definitions. The glossary, in conjunction
with the comments contained in the program, is intended for use as an aid
in identifying the function of any particular section of the code. The
symbol glossary is located immediately after the listing of the program.
Following the glossary, we include several typical examples of program output.
In a practical sense, the computer codes may be viewed as FORTRAN
representations of the algorithms and ideas presented in Volume I of this
report and its earlier companion report. Thus, in order to fully appreciate
the technical content of the programs, the user should be thoroughly
familiar with Volume I and its Appendices, and with Roth et al. (1971)
and its associated Appendices. Of course, anyone primarily interested
in the technical aspects of the model will find the discussions in the
reports cited of greater benefit and interest than the general descriptions
included in the User's Guide.
For those who are interested in obtaining general information
regarding the computational structure or capabilities of the programs,
the discussions in Sections I and II of this Guide should prove sufficient.
Individuals responsible for actually operating any or all of the programs
should read the entire User's Guide, Sections II, III, and IV being of
particular interest.
-------
I. SYSTEM OVERVIEW
In this section, we describe the overall operating structure of the
airshed simulation package. Before presenting this discussion, however,
we briefly outline the major components of the model.
a. Emissions. An emissions inventory must be prepared for
all chemical species of interest. This involves calculating
the total mass of pollutant emitted from automobiles, aircraft, and
fixed sources into each ground level grid cell.
b. Meteorology. Meteorological inputs of wind speed, wind direction,
and mixing depth are specified at the center of all grid cells.
c. Chemical kinetic mechanism. A chemical kinetic mechanism
is required if any of the species of interest react in the
atmosphere. The mechanism is used to determine the rate at
which pollutant concentrations change due to chemical reaction.
The nonlinear, coupled partial differential equations expressing the
conservation of mass of each pollutant comprise the governing equations
of the model. Individual reaction rate expressions (from the kinetic
mechanism) are incorporated into the equations, and emissions and
meteorological data are inputs to the model. The solution is carried out by
numerically integrating the governing equations on a three dimensional grid
overlaying the modeling region to obtain the temporal variation of pollutant
concentrations at each cell on the grid. The ground level grid layout used
for the Los Angeles simulation study is shown in Figure 1.
The model developed in this study is embodied in four computer
programs. The most important of these is the Atmospheric Pollution
Simulation Program (JffiSP), which is used for predicting concentrations
of air contaminants at the grid cells comprising the region to be
modeled. We have also developed two specialized data preparation programs
to process the large volume of emissions and meteorological input data.
Digitized wind and mixing depth maps are input to the Meteorological Data
Preparation Program (MDPP). This information is processed and placed in the
Meteorological Data File (MDF). Emissions from automobiles, aircraft, and
fixed sources are all combined in the Emissions Data Preparation Program
(EDPP) to produce an array of total pollutant fluxes into each ground level
grid cell. These fluxes are then placed in the Emissions Data File (EDF). The
two data files serve as the primary mechanism for input of meteorological
and source data to the APSP.
Validation of the model involves comparison of simulated concentration
predictions with actual field measurements. As an aid in making these
comparisons, we developed a Data Plotting Program (DPP). This program
is used to prepare plots of both predicted and measured concentrations as
a function of time. A Calcomp Plotter is employed to produce the plots,
although any other plotter may be substituted after making the appropriate
software alterations.
-------
25
I i
2S
'. FIGURE 1
THE MODELING REGION
-------
FIGURE 2
SYSTEM FLOW DIAGRAM
/
Eromis.
Inventory
EDPP
en
Meteor.
Data
»-
MUFF
I.C. -B.C.
I
Hourly-
Averaged
Cone.
Maps
Instan-
taneous
Cone.
Maps
Station
Data
APSP
^
Station
Predictions
DPP
APSP - Atmospheric Pollution Simulation Program
DPP - Data Plotting Program
EDPP - Emission Data Preparation Program .
EDF - Emissions Data File
I.C.-B.C. - Initial and Boundary Conditions, and Other Program
Operating Parameters
MDPP - Meteorological Data Preparation Program
MDF - Meteorological Data File
-------
In Figure 2, we summarize the overall structure.of the airshed
simulation package, indicating the flow of information to and from each
program. To operate the system, wind and mixing depth maps are prepared
and transferred to punched cards. This information is input to the MDPP,
which is used to create the MDF. Next, a complete emissions inventory
is assembled. Once this inventory is digitized, the EDF is established by the
EDPP.
After the two input data files are created, initial and boundary
conditions are determined and placed oh punched.cards. In addition,
operating parameters for the APSP are also specified on punched cards
and included in the data-deck with the initial and boundary conditions.
The APSP is then used to perform the airshed simulation. Pollutant
concentrations are predicted at each grid cell, and using these values, an
estimate is obtained for the predicted contaminant level at each air quality
monitoring station. The results of the simulation are presented in three
forms:
1) "instantaneous"* ground level concentration maps for all species
printed at regular time intervals, such as once an hour - see
Figure 3.
2) printed hourly-averaged ground level concentration maps for
all species - see Figure 4.
3) punched cards containing hourly-averaged pollutant concentrations
predicted at each air quality monitoring station.
The punched cards from the APSP, along with cards containing the actual
pollutant concentrations measured at the stations, serve as the input data
to the DPP. This program is used to prepare the final plots, which illustrate
the model's performance.
At this point it may not be clear why all four programs are not
integrated into one large program. This may be accomplished, for example,
by writing each program as a subroutine and creating a small driving
program. The reasons for structuring the system in its present form are
based mainly on computing and user-oriented considerations. Once a
particular data file is created, it may be employed many times while
changes are made in the other data file, thus eliminating unnecessary
data processing. Consider the following two cases. One emissions inventory
may, for example, be used in conjunction with various meteorological
conditions, which is in effect the procedure adopted when model validation
runs are made for several different days. Similarly, one set of
meteorological conditions may be combined with different emissions
inventories to test the effectiveness of emission control strategies.
While the two cases discussed above are illustrative of the
computational savings inherent in the present package structure, user-
oriented considerations are even more important. To successfully perform
an airshed simulation, all inputs to the model must be specified correctly.
*An "instantaneous" concentration is a concentration predicted at a
particular time during the course of the numerical integration.
-------
FIGURE 3. TYPICAL COMPUTER DISPLAY OF THE INSTANTANEOUS GROUND LEVEL CONCENTRATIONS
1
I
j
25
23
22
21
20
19
18
17
16
IS
14
13
12
11
10
9
8
1 7
6
i
5
4
3
2
1
3.5 3.8 3.7 3.8 4.0 6.7 14.3
3.0 3.0 3.0 3.0 4.3 lOil 17.9
RESEDA
2.1 2.5 2.6 2.6-4.7 8.5 12.7
1.7 2.8 3.1 2.9 3.2 6.0 10.5
3.3 4.1 4.1 3.9 3.1 4.9 8.4
SANTA MONICA MTNS
3.2 3.7 3.6 3.5 3.2 3.9 5.6
3.2 3.4 3.3 3.3 3.1 3.0 3.6
. •..- NEST LA
4.7 3.5 2.8 2.1
5.1 3.7 2.0
4.7 1.7
2.0
5.0
5.3
': '• : : ' 5.3
's.i
03
8
29.5
25.5
18.9
16.5
14.2
8.6
4.4
1.9
1.4
GROUND LEVEL CONCENTRATIONS
-------
FIGURE 4. TYPICAL COMPUTER DISPLAY OF THE HOURLY-AVERAGED GROUND LEVEL CONCENTRATIONS
CD
AVERAGE GROUND LEVEL CONCENTRATIONS CPPHM) Cf 03 BETWEEN THE HOURS Of 1000. AND 1100. PST
1 2 3 4 5 * 7 8 9 10 11 12 13 14 IS 16 17 18 19 20 21 22 23 24 25
25 44433349 11 84
24 4 3 333 4 8 20 22 16 8
— j
I
!
j
I
SAN GABRIEL MTNS
23 3
22 2
18 3
17 3
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
3 33 3 6 12 23 30 29 20 10 6
RESEDA
2 22 3 5 9 15 23 33 34 21 10
' . ' BURBANK
SANTA MONICA NTNS
. OOWNTOMN LA
3 333 2 3 3 3 3 6 13 14
WEST LA
43 3 2 2 2 3 6 13 13
.779
11 9 10
PASADENA
16 11 11
17 11 9
11 10 7
10 10 8
:.. COCEHCE
4 4321127 13 13 10 87
.••••- 3 1 1 1 2 6 11 12
LENNOX
31113589
• 1113456
4113334
4223223
LONG
4333212
'••••• -4 4 3 3 2 1 2
PALOS VERD6S
' '• * '
i i ' '.
PACIFIC OCEAN
975
743
532
3 2 1
3 2 1
BEACH
221
32 1
13 14
7 7
EL MONTE
4 3
5 4
5 6
5 7
4 7
n
in
A
3
4
6
a
10
WHITT1ER
3 7 10
3 6
2 5
9
7
10
7
4
3
3
5
7
-------
Considering the number and types of parameters involved, it is convenient
to treat the input of emissions and meteorological data separately.
This enables the user to concentrate fully on the task of establishing
the emissions and meteorological data files. Thus, the EOF and the MDF
are created and checked for accuracy and correctness before an airshed
simulation is attempted. The contents of the EOF and the MDF are listed
in the program output of the EDPP and the MDPP respectively.
This concludes our discussion of the general system considerations.
In the next section we examine each of the four programs in sufficient
detail to give the user a basic understanding of individual program
structure and capabilities.
-------
II. DESCRIPTION OF THE PROGRAMS
A. The Emissions Data Preparation Program
1. Program Objective and Capability
Pollutant emissions from automobiles, aircraft, and distributed
fixed sources are combined in the Emissions Data Preparation Program to
produce a total flux of each pollutant species into every ground
level grid cell. The fluxes are stored in the Emissions Data File for
future use by the APSP. Calculation of the fluxes requires an emissions
model and inventory. The emissions model forms the basis for the algorithms
in the code, while the emissions inventory serves as the input data to the
program. To fully appreciate the scope of a comprehensive emissions model
and inventory for a large urban area, the user is encouraged to read
Appendix A of Roth et al. (1971), and Appendix A in Volume I of this
report.
Since most source emissions rates undergo diurnal variations,
fluxes are calculated at a sufficient number of times throughout the day to
represent the characteristic temporal changes. In the present study, we
calculate emissions at five minute intervals between 0500* and 1000 hours,
and hourly thereafter. For convenience, all times used in this study are
Pacific Standard Time (PST). The emissions inventory, however, is based
on "local time". We note that when daylight savings time is in effect,
local time and standard time differ by one hour. Therefore, we include
the appropriate logic in the EDPP to make a one hour shift in the emissions
for use on those days when local time and standard time do not correspond.
An important point must be made regarding the treatment of sources.
Emissions from large point sources, such as power plants, are treated
separately in the APSP. Thus, two fixed source inventories are required.
The first contains all distributed, ground-based fixed source emissions,
such as those from petroleum marketing operations. These sources are
processed by the EDPP. The second fixed source inventory includes the
remaining point sources, which only emit pollutants from a relatively few
locations. Typically, these sources inject pollutants into the atmosphere
from high stacks, which makes it inappropriate to treat them with the other
ground-based sources. The point source inventory is input to the APSP on
punched cards, while the ground-based emissions are input to the APSP using
the EOF.
We wish to note that the EDPP is presently structured to make
computations of emissions between 0500 and 1655 hours. However, the program
can easily be modified to compute emissions to 2355 hours. Further
alterations to the code will be required to compute emissions prior to
0500 hours, or after 2400 hours (i.e., into the next day). In Section III,
we include instructions for extending the computation of emissions from
1655 to 2355 hours.
*0500 hours corresponds to 5 A.M.
10
-------
The EDPP is primarily used to create the EOF. This file contains
the information needed by the APSP to calculate ground level emissions fluxes.
In the following discussion we explain the structure of the EOF. The
discussion is illustrated in Figures 5 and 6.
First, emissions fluxes are calculated for all species at the
starting time, 0500 hours, and written on the EOF. Then emissions fluxes
are calculated at 0505 hours. If we assume that the flux varies linearly
between 0500 and 0505 hours, the time rate of change of the emissions flux
is given by
<0505> - Q, • .(0500)
- for 0500 + :t (t-osoo)
The two times, 0500 and 0505, are written on the EOF followed by values of
the time rate of change of the emissions fluxes.
Next, we compute the emissions fluxes at 0510 hours and again assume
linear time behavior. We calculate the time rate of change of the fluxes,
as before, applicable between 0505 and 0510 hours. The two times, 0505
and 0510 are written on the EOF followed by values of the time rate of change
of the emissions fluxes. To obtain the values of the emissions fluxes
at any time t, we simply reconstruct the connected line segment curves
(see Figures 5 and 6) using the initial fluxes, the time intervals, and the
slopes of the lines (i.e., dQ^ j_ j/dt) . This is essentially the way in
which emissions fluxes are calculated by the APSP from the information
contained in the EOF.
2. Program Structure
The EDPP consists of a MAIN program and one subprogram, TIME.
MAIN
Structurally, the MAIN program is divided into four sections. The
first section is devoted to program initialization, which involves the input
of the emissions inventory. In the second section, emissions are calculated
every five minutes between 0500 and 0955 hours. The time rate of change
of the emissions fluxes (ppm-ft/min2) are computed and written in the EOF.
The third section is very similar to the second, except emissions fluxes
are calculated hourly from 1000 to 1655 hours. Since the temporal distribution
11
-------
FIGURE 5
ASSUMED TEMPORAL VARIATION OF THE EMISSIONS FLUX AT A
PARTICULAR LOCATION BETWEEN 0500 AND 0955 HOURS
at
Oi
W
§
•H
W
M
•g
W
0500 0505 0510
TIME
0515
FIGURE 6
ASSUMED TEMPORAL VARIATION OF THE EMISSIONS FLUX AT A
PARTICULAR LOCATION AFTER 1000 HOURS
Cx
i
(0
c
o
•••i
01
m
1000 1100 1200 1300
TIME
12
-------
of automobile emissions is constant over hourly periods but discontinuous
from one hour to the next, the computation for each hour takes place in two
steps. Emissions rates are assumed constant over the first 55 minutes of the
hour. Then from the 55 minute point to the beginning of the next hour, the
emissions vary linearly in time between the two hourly values (see Figure 6).
In the last section of the EDPP, emissions fluxes are calculated
using the newly created EDF. The fluxes applicable at the beginning of
each hour are printed on the program output to provide a partial record of
the contents of the EDF.
TIME (Tl, T2, M)
Three types of arithemacic operations involving times
in minutes and times based on a 2400 hour clock are performed in function
subprogram TIME.
If M=l, add T2 to Tl and return the sum, where Tl is 2400 hour time, and
T2 has the units of minutes. For example, if Tl=0530 and T2=40, the result
is 0610 (add 40 minutes to 5:30 A.M. to obtain 6:10 A.M. = 0610 hours).
If M=2, subtract T2 from Tl, where Tl and T2 are 2400 hour times. The
result is expressed in minutes. For example, if Tl=0610 and T2=0530, the
result is 40 minutes.
If M=3, the operation is analogous to that with M=l, except a
subtraction is performed. If Tl=0610 and T2=40, the result is 0530
(subtract 40 minutes from 6:10 A.M. to obtain 5:30 A.M.=0530 hours).
3. Input Data Requirements
The following outline lists the general types of information
that must be provided as inputs to the EDPP. See Table 2 for a complete
summary of the input data required by the program.
a. Automotive Emissions
• daily miles driven on all streets (excluding freeways)
in each grid square
• daily miles driven on freeways in each grid square
• hot and cold start emissions factors
• fraction of vehicles cold started
• correction factor (0) for the nonuniform distribution
of trip starts
• freeway driving speeds in the fast and slow directions
• emissions - driving speed correlation factors
• ratio of freeway vehicle miles driven in the slow
direction to the number driven in the fast direction
• temporal distributions of freeway and non-freeway.
traffic activity
b. Aircraft Emissions (Ground Operations)
• number of daily aircraft flights at each airport
• temporal distribution of aircraft activity
• emissions/aircraft engine
• number of aircraft engines/aircraft
13
-------
c. Fixed Source Emissions (Ground-Based)
• total distributed fixed source emissions from each grid
square
4. Program Output
The output of the EDPP consists of the EOF and the printed
listing of hourly values of the emissions fluxes. The units of the fluxes
displayed on the printed output are ppm-ft/minute. Examples of the printed
output are included in Section IV.
B. The Meteorological Data Preparation Program
1. Program Objective and Capabilities
Digitized wind speed, wind direction, and mixing depth maps are
converted in the MDPP to a format suitable for input to the APSP. The
meteorological data is then placed in the Meteorological Data File, where
it can be accessed at a later time. Thus, the operation performed by
this program is basically that of transferring information from cards to
tape or disc.
Hourly-averaged wind speed and wind direction maps must be
prepared for each hour to be simulated. The time averaging is assumed to
extend from the half hour to the half hour. Thus, wind speeds applicable
at 0500 hours, for example, are obtained from the 0430 - 0530 wind speed
map. Mixing depth maps are also prepared hourly, applicable at the
beginning of each hour (i.e., at 0500 PST, 0600 PST, and etc.). Mixing
depths are assumed to vary linearly in time between the values prescribed
at two consecutive hours. In Figure 7, we represent the assumed temporal
changes of the wind speed, wind direction, and the mixing depth.
In the following discussion, we describe the structure and content
of the MDF. Initial values of the mixing depth at 0500 hours are read from
the data-deck and placed directly in the MDF. Next, wind speeds, wind
directions, and the time rate of change of the mixing depths are determined,
applicable between 0500 and 0530 hours. The wind data are obtained from
the data-deck, as are the values of the mixing depths at 0600 hours. The
time rate of change of the mixing depths are calculated in the following
manner. Assuming linear temporal behavior, the rate of change for the
mixing depth at grid square (i,j) is given by:
AHi>:)(0600) -
dt 60 minutes
for 0500
-------
FIGURE 7
§
u
2
•H
T3
tJ Oi
0) O
T3 -H
C .X
•H -H
» B
>M
O
TREATMENT OF THE TEMPORAL CHANGES
IN WIND SPEED, WIND DIRECTION, AND MIXING DEPTH
AT A PARTICULAR LOCATION
Or'
•fr
•e-
0500
0600
0700
Time
O Mixing depth data point
D Wind speed or wind direction data point
'""— Assumed temporal variation of mixing depth
Assumed temporal variation of the wind speed or
wind direction
15
-------
The time interval, designated by 0500 and 0530, is written on the
MDF followed by values of the wind speed (feet/minute), wind direction
(radians), and time rate of change of the mixing depth (feet/minute).
We summarize below the order in which the first hour of meteorological
data is arranged in the MDF.
Mixing depths at 0500 PST
Time interval, designated by 0500 and 0530,
over which the following data are applicable
Wind speeds
Wind directions
Time rate of change of the mixing depths
Next time interval, designated by 0530 and 0600,
over which the following data are applicable
Wind speeds
Wind directions
Time rate of change of the mixing depths
Succeeding blocks of meteorological data are arranged as follows:
Time interval, designated by t1 and t2 ,
over which the following data are applicable
Wind speeds
Wind directions
Time rate of change of the mixing depths
2. Program Structure
The MDPP is composed of a MAIN program and one subprogram, TIME.
MAIN
The MAIN program is used to perform two basic functions. First,
meteorological input data supplied on punched cards is converted into a
format suitable for input to the APSP and written on tape or disc. Wind
speeds are converted from miles/hour to feet/minute, and wind directions
are converted from degrees to radians. After all meteorological data is
written in the MDF, hourly printed maps of wind speed, wind direction,
and mixing depth are produced. These maps provide a complete summary
of all meteorological data contained in the MDF.
16
-------
TIME (Tl, T2, M)
See the description given in the discussion of the EDPP.
3. Input Data Requirements
The following information must be supplied on punched cards for
each hour
• Wind speeds
• Wind directions
• Mixing depths
A complete description of the order and format of the input data
is given in Table 3.
4. Program Output
The primary output of the MDPP is the MDF. In addition, printed
hourly maps of wind speed (m.p.h. x 10), wind direction (degrees), and
mixing depth (feet) are also produced. Typical examples of the printed
output are included in Section IV.
C. The Atmospheric Pollution Simulation Program
1. Program Objective and Capabilities
The actual airshed simulation is performed by the APSP.
Meteorological inputs consisting of wind speed, wind direction, and
mixing depth are obtained from the data in the MDF, while ground level
emission inputs are calculated from information in the EOF. The simulation
involves the numerical integration of the governing model equations by
the finite-difference techniques described in Appendix D of Volume I
of this report. The primary objective of the APSP is the prediction of
ground level pollutant concentrations of both inert and photochemical
species on a grid such as that illustrated in Figure 1.
Several useful features have been incorporated in the APSP.
At present, the program is capable of simulating up to six chemical species.
Each species is designated by a number, and we have adopted the following
convention for this study:
Species Number Species Name
1 Reactive Hydrocarbon
2 Nitric Oxide
3 Ozone
4 Nitrogen Dioxide
5 Carbon Monoxide
6 Unreactive Hydrocarbon
This numbering convention is consistent with that used in the EDPP. A
full photochemical simulation involves all six species, but it is possible
to operate the program simulating fewer species. The following list contains
all the possible combinations of species that can be simulated simultaneously:
17
-------
a. species 1, 2, 3, 4, 5, and 6
b. species 1, 2, 3, 4, and 5
c. species 5 and 6
d. species 5 or 6
The user specifies which species are to be simulated as part of the input
data to the APSP. During photochemical validation studies, we use this
feature as follows. First, CO is simulated to test the meteorological
and automotive source inputs. If the results indicate that these parameters
are specified correctly, then a full photochemical simulation is carried
out involving all six chemical species. We also note that other inert
species can be substituted for CO and/or unreactive hydrocarbon simply by
changing a few input parameters in the data-deck of the program.
Another important feature designed into the program is the
capability of performing the numerical integration on an irregularly shaped
grid. We note in Figure 1, that many grid squares are located over
the Pacific Ocean and the San Gabriel Mountains. It is computationally
advantageous to eliminate these low population areas since a 30% reduction
in the total computing can be realized. To generalize this procedure,
the shape of the modeling region is specified as part of the input
data to the program, in Figure 1, we illustrate that portion of the
original 25x25 ground level grid layout actually used in the Los Angeles
simulation study.
The chemical kinetic mechanism is an important component of this
simulation model. The mechanism is also subject to change as further studies
of atmospheric chemistry are carried out. Since rate constants are
treated as input data to the program, they can easily be changed at any
time. The mechanism currently employed in the APSP is described in
Appendix B of Volume I of this report. To alter the formulation of the
mechanism in the program, minor changes in the code are required. In
Section III, we include a detailed discussion of these changes.
Before closing this section, we wish to mention that the coding of
the APSP has been formulated very carefully. We note that a major portion
of the total computing time is spent solving the governing model equations.
Therefore, computational efficiency is stressed in the coding of the
numerical integration procedure. In other sections of the code, general
utility is considered more important. In the next section, we discuss in
more detail the structural and computational aspects of the APSP.
2. Program Structure
The APSP consists of a MAIN program and several subroutines.
In Figure 8 we show a schematic diagram of the structure of the APSP.
MAIN
Two tasks are performed by the MAIN program:
(1) the control of model initialization
(2) the numerical integration of the governing airshed equations
18
-------
FIGURE 8
STRUCTURE OF THE APSP
MAIN
BLOCTD
«£>
DIFCOF
METEOR
PTSOUR
SIMDIG
SOURCE
AVCONC
MAPPER
PRINT1
TIME
-------
Model initialization consists of the following procedures that
must be performed before the numerical integration can be initiated:
input of program operating parameters specified on punched cards
open the EOF and initialize the ground level pollutant flux
array (subroutine SOURCE)
open the MDF and initialize all meteorological variables
(subroutine METEOR)
establish the initial concentration distribution (subroutine
ICCONC)
After all parameters in the model have been initialized, the airshed
simulation may proceed. Predicted concentrations are obtained by integrating
the governing equations in discrete steps over time. The operations involved
in advancing from one instant in time to the next, over a discrete time
interval, are as follows:
• establish the size of the integration time step
update meteorological variables (subroutine METEOR) and
the distribution of elevated point source emissions
(subroutine PTSOUR)
• update ground level emissions fluxes (subroutine SOURCE)
increment the time-integrated ground level concentration
array and the time-integrated station prediction array
(entry point CALIVC in subroutine AVCONC)
update boundary concentrations on the horizontal boundaries
(subroutine BCCONC)
• perform Steps I and II of the numerical integration procedure -
integration in the x and y directions
• perform Step III of the numerical integration procedure -
integration in the z direction; the following calculations
are involved:
1. compute turbulent diffusivities (subroutine DIFCOF)
2. solve the finite-difference equations for inert species
(subroutine SIMDIG)
3. solve the finite-difference equations for reactive
species (subroutine BLCKTD)
increment the time-integrated ground level concentration array
and the time-integrated station prediction array (entry point
CALIVC in subroutine AVCONC)
• print hourly-averaged ground level concentrations and station
predictions if one hour has elapsed since the last hourly-averaged
concentrations were printed (subroutines AVCONC and MAPPER)
20
-------
When the simulation has been completed, the user may specify
that the entire concentration field be dumped onto a save disc/tape
area. These concentrations may be used at a later time as initial
conditions in a program restart. The restart procedure is discussed
further in Section III.
AVCONC
All tasks related to the calculation and printing of the hourly-
averaged predicted concentrations are performed in subroutine AVCONC.
Hourly-averaged concentrations are calculated at every ground level
cell on the grid, and at each air quality monitoring station. We de-
fine the hourly-averaged concentration as follows:
t0+60
c(t)dt
j
— - O /I \
c ~ 60 minutes UJ
where
c(t) is the concentration predicted at time t ,
c is the hourly-averaged concentration, and
t is the beginning of the time interval over
which the concentrations are averaged.
The integration is approximated numerically by the trapezoidal rule.
Since monitoring stations are not, in general, located at the center
of a ground level grid cell, we interpolate for the predicted con-
centration at the station using the concentrations predicted in the
four nearest ground level grid cells. The interpolation procedure
is illustrated in Figure 9.
There are two entry points in subroutine AVCONC, and we now
discuss the calculations performed in each.
1. Entry Point AVINIT
When AVINIT is entered from MAIN, the arrays which contain
the numerical approximations of the integrals are all set to zero.
This is an initialization procedure.
21
-------
FIGURE 9
INTERPOLATION PROCEDURE FOR CALCULATING
THE PREDICTED CONCENTRATION AT A MONITORING STATION
J+l
AX-
AY,
\j
|*-AX,
9
AY
1+1
9 - monitoring station location
"station
[clfJU-€>
where 5 =
AX
AY
22
-------
2. Entry Point CALIVC
The numerical evaluation of the integration defined in
Equation (1) is performed in CALIVC. The trapezoidal rule is em-
ployed and may be written as follows:
n-1
where
c(t) dt " TZ-f 1°^- -i) + c(t.)](t. - t.)
-° * X
o
t = t +60 minutes
n o
and ^ designates a discrete point in time at which the governing
equations of the model are solved.
Before the governing equations are integrated from t^ to
ti+i , CALIVC is entered, and the following values of A are cal-
culated for each ground level grid cell and monitoring station.
A-Ic(ti)(tu.i-V (
Arrays that will eventually contain the complete expression on the
right hand side of Equation (2) for all ground level cells and stations
are incremented by the appropriate value of A given in Equation (3).
After the governing equations have been integrated from ti to
fci+l ' CALIVC is again entered from MAIN, and the following values of B
are calculated for each ground level cell and station.
The arrays that will contain the expression in Equation (2) are this
time incremented by the appropriate values of B given in Equation (4).
If A and B are summed, we obtain:
23
-------
which is one of the terms in Equation (2). When the governing equations
have been integrated to tn , the complete expression on the right hand
side of Equation (2) is available for all ground level grid cells and
monitoring stations. This completes our discussion of CALIVC.
AVCONC is called from MAIN once each hour, and the expression
for each ground level grid cell and monitoring station given in Equation
(2) is divided by 60 minutes to yield the hourly-averaged concentrations.
Then the following output is produced by this subroutine;
printed hourly-averaged ground level concentration
maps for each species (see Figure 4 and Section IV)
. printed summary of the hourly-averaged concentrations
predicted at the monitoring stations (see Section IV)
. punched cards containing the hourly-averaged con-
centrations predicted at the monitoring stations
AVINIT
An entry point in subroutine AVCONC (see AVCONC).
BCCONC
In Steps I and II of the numerical integration procedure (in-
tegration in the x and y directions), pollutant concentrations are
required at points of horizontal inflow to the model. During the simu-
lation, subroutine BCCONC is called from MAIN to update the values
of the boundary concentrations. These concentrations are inputs to
the model and included in the data-deck of the APSP. A concentration
specified at a particular point on the boundary is assumed to apply at
all levels between the ground and the inversion base.
To account for temporal changes, the boundary concentrations
are allowed to vary in time in a stepwise manner. The user defines
a time interval (designated by two times) and a set of boundary con-
centrations that are to apply during that time interval. For example,
to specify a set of boundary concentrations that are to apply be-
tween 0500 and 0600 hours, the time interval is designated by the
two times, 0500 and 0600.
24
-------
The operational characteristics of BCCONC are best illustrated
by an example. Suppose we wish to perform a two hour simulation from
0500 to 0700 hours, and also suppose that the boundary concentrations
are to change hourly. Thus, we specify a set of boundary concentra-
tions that apply from 0500 to 0600 hours, and another set that apply
from 0600 to 0700 hours. The cards in the data-deck are arranged in
the following order (see Card No. 18-19 in Table 4):
time interval, designated by 0500 and 0600, and
boundary concentrations that apply between 0500
and 0600 hours
time interval, designated by 0600 and 0700, and
boundary concentrations that apply between 0600
and 0700 hours.
The first time BCCONC is called from MAIN, the time interval and
boundary concentrations are determined from the first set of data
cards. Thus, a set of boundary concentrations are established and
applied between 0500 and 0600 hours. Each time BCCONC is called to
update the boundary concentrations, the current simulation time is
checked against the time interval. In the present example, when the
simulation time finally becomes greater than 0600 hours, the naxt
time interval and boundary concentrations are read from the data-deck.
These concentrations that apply from 0600 to 0700 hours are then used
for the remainder of the simulation.
When photochemical simulations are performed, ozone concen-
trations at the boundary are determined by assuming that the rate
of reaction of ozone is equal to zero. Upon making this assumption
and utilizing the reaction rate expression given in Appendix B of
Volume I, we obtain the following algebraic relation for the boundary
concentration of ozone:
k2[o]
where
k. [NO.]
[O] = ,. A 2
and [NO] , [NO_] , and [HC] are the boundary concentrations of NO
N02 , and reactive hydrocarbon respectively.
25
-------
BLCKTD (N,M)
Subroutine BLCKTD is used to solve a block-tridiagonal system
of N-M linear equations. Each block is an M x M matrix, and all off-
diagonal matrices are assumed to be diagonal. A discussion of the
algorithm is presented in Appendix D of Volume I. The solution of
a system of equations in this form is required as part of the
numerical integration procedure.
CALIVC
An entry point in subroutine AVCONC (see AVCONC).
DIFCOF (I,J,KZ)
In Step III of the numerical integration procedure, vertical
turbulent diffusivities must be determined to compute the vertical
diffusion of pollutants. Vertical diffusivities are specified in
the array KZ applicable in grid cells above grid square (I,J) .
KZ(K) is the value of the diffusivity at the interface between grid
cells (I,J,K-1) and (I,J,K) . See Appendix C of Roth et al. (1971)
for a discussion of the algorithm.
ICCONC
The initial concentration distribution is established in
subroutine ICCONC. Two modes of initial concentration input are
possible. In the first mode, the initial ground level concentrations
are read from cards included in the data-deck. The values in the
ground cells are extended vertically to the inversion to complete the
definition of the initial concentration field. The second mode of in-
put arises when a simulation begins at a point in time at which a pre-
vious simulation terminated. Recall from our discussion of the MAIN
program that at the end of a simulation, the entire concentration field
may be "dumped" onto a save disc area. Thus, the second mode of
initial concentration input simply consists of a READ statement to
read these "dumped" values.
Initial ozone concentrations are determined by making the
assumption that the rate of reaction of ozone is equal to zero at
the starting time of the simulation. Using the reaction rate expression
for ozone given in Appendix B of Volume I, we obtain
[Oj = 2
JUo]
'3J k3[NO]
where k, [N0_]
[o] - * 2
k2
26
-------
and [0,] / [NO] , [NC>2] / and [HC] are the initial concentrations of
0^ , NO , NC>2 , and reactive hydrocarbon respectively.
MAPPER (I, IHOLD, ISTART, ISTOP, APRINT)
This subroutine is used during the printing of the ground
level concentration maps to produce "blanks" in those locations on the
map which are outside the computational grid. For example, cells
over the Pacific Ocean are not included in the current simulation
study of Los Angeles. Note that in Figures 3 and 4, no concentations
are printed in areas of the grid over the Pacific Ocean. The blanks
printed in this area are a direct result of the use of subroutine
MAPPER.
The subroutine operates as follows. Before a particular
row of ground level predictions are printed in subroutines AVCONC
or PRINT1, the concentrations in that row are stored in the array
IHOLD. Next, the values in IHOLD which represent predictions within
the computational grid are determined. These values are stored in
IHOLD between IHOLD (ISTART) and IHOLD (ISTOP). This information is
used in subroutine MAPPER to produce the array APRINT, which is re-
turned to the calling program. This alphameric array is then printed
(128 printing columns) in the calling program, and only numbers within
the computational grid appear on the printed map. If 1=0, integers
are produced on the printed output. To obtain numbers with one decimal
digit, we set 1=1. We note that this subroutine is written
in Assembly Language.
METEOR
The processing of all meteorological data is done in subroutine
METEOR. During model initialization, the MDF is opened, and initial
values of the meteorological variables are calculated. As the simula-
tion progresses, the meteorological variables are updated as necessary
by reading the file. See Appendix C of Roth et al. (1971) for a further
discussion of the meteorological data. Subroutine PTSOUR is called
from METEOR every time the wind changes speed or direction to estab-
lish the spatial distribution of emissions from elevated point sources.
PRINT1
Instantaneous concentrations are printed on the program output
by subroutine PRINT1. Vertical concentration profiles above monitoring
stations and ground level concentration maps are printed at regular
27
-------
time intervals. We have found that an interval of 60 minutes be-
tween printouts gives a good indication of hourly trends and does
not result in endless pages of unused computer output. A typical
example of a ground level concentration map printed by this sub-
routine is illustrated in Figure 3. Other examples are included
in Section IV. Subroutine MAPPER is used in the preparation of
the ground level maps.
PTSOUR
Emissions from elevated point sources are distributed among
several downwind grid cells in subroutine PTSOUR. The algorithm
employed is described fully in Appendix A of Volume I. This sub-
routine is called from METEOR each time the wind changes speed or
direction since the downwind distribution of the emissions is
directly affected by a change in either of these parameters.
SIMDIG(N)
Subroutine SIMDIG is employed in the solution of a tridiagonal
system of N linear equations. The computational algorithm is given
in Appendix D of Volume I.
SOURCE
Ground level pollutant fluxes are calculated in subroutine
SOURCE using the data contained in the EOF. During model initializa-
tion/ the EDF is opened, and initial values of the fluxes are computed.
SOURCE is called from MAIN throughout the course of a simulation to
update the ground level fluxes.
TIME (Tl, 12, M)
This subprogram is discussed as part of the EDPP.
3. Input Data Requirements
The following outline lists the types of input information
needed to run the APSP. The detailed listing of all parameters is
given in Table 4.
28
-------
grid characteristics, including number of
vertical strata, horizontal grid spacing,
and shape of the region
starting and stopping time of the simula-
tion
time interval between instantaneous con-
centration printouts
location of monitoring stations
rate, location, and temporal variation of
major elevated point source emissions
integration parameters, including minimum
and maximum, time step size, relative error
tolerance to be satisfied in the iterative
solution of the nonlinear finite-difference
equations, maximum number of iterations to
be allowed in the iterative solution, and
the initial size of the integration time
step
rate constants, temporal variation of the
photolysis rate constants, and values of the
generalized coefficients
initial conditions
concentrations alofjt
boundary concentrations at points of horizontal
inflow
4. Program Output
Three types of output are provided by the APSP
. printed hourly-averaged ground level concentration
maps and a printed summary of the hourly-averaged
concentrations predicted at each monitoring station
printed instantaneous ground level concentration
maps and a printed summary of the instantaneous
vertical concentration distribution above each
monitoring station
punched cards for input to the DPP containing the
hourly-averaged concentrations predicted at
each monitoring station
See Section IV for examples of the printed output from the APSP.
D. The Data Plotting Program
1. Program Objective and Capabilities
The evaluation procedure adopted for the simulation study reported
in Volume I requires that model predictions be compared with actual
29
-------
measurements of pollutant concentrations. These comparisons are
best made by displaying the temporal variation of the measured
and predicted concentrations together on one plot. In this manner,
similarities and differences in the two concentrations at any par-
ticular time are readily apparent. Thus, the primary objective of
the DPP is the automatic preparation of plots of predicted and measured
concentrations at each air quality monitoring station.
Available measurements of particular interest are those of
total hydrocarbon (as parts per million carbon), NO , N02 , CO , and
oxidant. Before the air quality measurements can be plotted, however,
they must be corrected to account for interferences caused by the
presence of other pollutants. These interferences are discussed in
Appendix E of Roth et al. (1971). The following corrections are made
to the NO and oxidant measurements:
N0plotted mi'K N0^asured
OX . .. , = OX , - 0.15 NO? + SO,
plotted measured ^measured Measured
A useful feature of the DPP is that the user may specify, as
part of the program input data, which species are to be plotted at
each station. If measurements are not available for a particular
species at a station, the DPP may be used to plot only the predicted
results at that station. In fact, the user may wish to completely
suppress the plotting since no comparisons can be made. As an example,
total hydrocarbon is measured at six stations in the Los Angeles
area. Thus, for evaluation purposes, we only prepare plots of the
hydrocarbon concentrations at these six stations.
In Appendix C of Volume I, we describe a technique to account
for the effects of near-by sources on the field measurements. These
effects are especially important to include when a monitoring station
is located adjacent to a heavily-traveled street. The local concen-
tration elevations are calculated using the algorithms in the Appendix
cited and included on punched cards in the data-deck of the DPP. These
concentration elevations are then added to the model predictions to
obtain an "improved prediction" at the station. We wish to note that,
as of this writing, local corrections may be calculated only for
inert species, such as CO, due to the nonlinear effects of chemical
reactions.
30
-------
2. Program Structure
The DPP consists of a MAIN program and three subroutines,
each of which performs a specific task in the drawing of the plots.
The creation of a plot may be described by the following four step
process:
1. draw and label the axes; print the
title block
2. plot the predicted concentrations
3. plot the station measurements
4. plot the locally corrected predicted concentrations
We now discuss the function of the MAIN program and each subroutine.
The structure of the DPP is illustrated in Figure 10.
MAIN
In the first section of code in MAIN, all input cards are read,
and the NO and oxidant measurements are corrected for interferences.
Having accomplished this, the program is ready to make all plots
specified by the user. The plots are drawn in the following order:
1. CO plots for each station
2. total hydrocarbon plots for each station
3. NO, N0_, and O. plots for each station
We describe the operations performed in the MAIN program by
discussing the manner in which a CO plot is made. First, the maximum
CO concentration to be plotted is found, and the scale of the con-
centration axis is calculated accordingly. Then the data points are
examined to be certain that none will be plotted in the title block
located in the upper right hand corner of the plot (See Figure 11).
If a data point does fall in the title block, then the concentration
axis is scaled in such a manner that no data points will be plotted
in this area.
31
-------
FIGURE 10
THE STRUCTURE OF THE DPP
MAIN
LABEL
PLOTXY
CD ASH
32
-------
FIGURE 11
EXAMPLE OF PLOTS FOR CO AND TOTAL HYDROCARBON
15
10
OL_
Q_
O
O
O
LONG BERCH
9/29/69
E MEflSURED CQ
PREDICTED CO
WITH LOCflL
CORRECTION
a
J I
8 9 10 11
TIME (PST)
12 13
15
COMMERCE
9/29/69
HEflSURED HC
- PREDICTED HC
o 10
Q_
Q_
CJ
•z.
O
to
J L
J L
_L
8 9 10 11
TIME (PST)
12 13
33
-------
After determining the scale of the concentration axis, sub-
routine LABEL is called to draw and label the two axes representing
concentration and time. Then subroutine PLOTXY is called to plot
the model predictions. This plot consists of solid line segments
connecting the discrete hourly-averaged predictions. After the pre-
dictions have been plotted, subroutine PLOTXY is called again, this
time to produce a symbol plot representing the measured concentrations.
We note that the solid lines have no other meaning than to correct
the predicted concentration data points and to provide a contrast
to the discrete symbol plot.
If local corrections are required, the predictions at the
station and the corresponding local corrections are transferred to
subroutine CDASH. An array is defined by CDASH consisting of model
predictions augmented by the local corrections. This array is then
plotted by PLOTXY as a dashed line. In Figure 11, we illustrate a
typical CO plot drawn by the DPP.
The basic operations described above are repeated for each
succeeding plot. All CO and total hydrocarbon plots are drawn, dis-
playing two sets of results on each page. The format is somewhat
different, however, for the display of the NO , N02 , and 03
results. In this case, the predictions and measurements for all
three species at a particular station are presented on the same page.
A typical plot involving these three species is given in Figure 12.
CDASH (NOHOUR, T, C, B, NOCORR, TCORR, CCORR)
Before the local corrections to the model predictions are
plotted, the MAIN program supplies subroutine CDASH with the number
of data points, NOHOUR, and three arrays T, C, and B, containing
the times, model predictions, and local corrections respectively at the
particular station. An array of times, TCORR, and an array of model
predicted concentrations plus local corrections, CCORR, are calculated
in subroutine CDASH. The number of data points in these arrays is
equal to NOCORR. The MAIN program passes NOCORR, TCORR, and CCORR
to subroutine PLOTXY, which then produces the dashed line plot
representing the locally corrected model predictions.
34
-------
FIGURE 12
EXAMPLE OF PLOTS FOR NO, N02 and 03
50
Q- 30
Q_
20
10
BURBRNK
9/29/69
E MERSURED NO
PREDICTED NO
A MERSURED N02
PREDICTED N02
I I
35
30
~ 25
z:
3:
S: 20
o
•z.
o
o
15
10
5
7 8 9 10 11 12 13
TIME (PST)
BURBRNK
9/29/69
MERSURED 03
PREDICTED 03
CD
8 9 10 11
TIME (PST)
12 13 m
35
-------
LABEL (X,Y,N, XDATA, NY, YMAX, UNIT, DATE, STN, NONAME, NAME, CORECT)
Subroutine LABEL is used to 1) draw and label the axes,
and 2) print the title block for each plot. The axes intersect at
the point (X,Y) on the plotting area, where X and Y are measured
in inches. N tick marks are drawn on the X axis, representing
the N data points that will eventually be plotted. XDATA is an
array of hours (or abscissa), and the subroutine uses this array to
write the hours under the X-axis. NY is the number of increments
into which the Y-axis is to be subdivided, and YMAX is the largest
value to be represented on the axis. The Y-axis is labeled to reflect
that it represents concentration, and the units (PPM, PPTM, or PPHM)
are also written according to the alphameric information contained in
the variable UNIT. For example, if the concentration units are to be
pphm (parts per hundred million), then UNIT = 'PPHM1. DATE and STN
are alphameric designations of the date and monitoring station name
respectively, which appear in the title block. NONAME is the number of
species to be plotted on the plot, and NAME is an alphameric array
containing the abbreviation of each species' name. For example, an
abbreviation for hydrocarbon is 'HC1. CORECT is a logical variable
passed by MAIN to inform LABEL whether or not local corrections will
be plotted on the current plot. If CORECT = .TRUE., then the local
corrections will appear, and LABEL will make a comment to that effect
in the title block.
PLOTXY (N,X,Y,XMN, XMX, YMN, YMX, IP, ISYS, SUPRES)
Subroutine PLOTXY is a general data plotting routine which
is capable of making point, solid line, or dashed line plots. The
program operates in a conventional manner, in which the user desig-
nates N points of an array of abscissa and ordinates, X and Y, to
be plotted. The values of the abscissa on the left and right hand
borders of the plotting area are specified by XMN and XMX. Similarly,
the values of the ordinate on the bottom and top of the plotting
area are YMN and YMX. The pen is moved from one data point to the
next, where the location of the pen is determined by linear inter-
polation using the particular values of X and Y to be plotted,
and the values of the abscissa and ordinate along the plotting area
borders.
36
-------
The type of plot produced by PLOTXY is controlled by the
parameter IP, where
!0 v symbol plot
1 > produces a ) solid line plot
2 ' 'dashed line plot
The value of ISYS determines which symbol is plotted when IP = 0.
A symbol will not be plotted for any value of Y which is
equal to SUPRES. This feature is used to denote a missing data
point, and hence, no symbol should be plotted. For the simulation
study reported in Volume I, we set all missing values of the measured
concentrations equal to minus one. To correspond with this con-
vention, we also set SUPRES equal to minus one.
The subroutine normally uses a 15 x 10 inch plotting area,
but the user can alter this specification by including a BLOCK DATA
subprogram as follows:
BLOCK DATA
COMMON/COMPXY/
DATA ITEST, XLNGTH, YLNGTH/1,''','''/
END
where XLNGTH and YLNGTH are the altered X-length and Y-length of the
plotting area, in inches, respectively. XLNGTH must be specified
at least 8-1/2 inches, and YLNGTH must be specified at least 10 inches.
37
-------
3. Input Data Requirements
The DPP requires the following types of input data:
number of carbon atoms per molecule of
reactive and unreactive hydrocarbon
air quality measurements
. model predictions
corrections due to local effects
The complete data-deck setup for the DPP is given in Table 5.
4. Program Output
The only output from the DPP are the plots, such as those
illustrated in Figures 11 and 12.
38
-------
III. OPERATING PROCEDURES
This section of the User's Guide is divided into two parts. In
the first part, we give general instructions regarding the use of all
programs. Included is a four or five step procedure outlining the manner
in which each program is to be operated. The second part of this section
is devoted to a presentation of the specific operating characteristics of each
program. The following topics are discussed:
limitations in the use of the programs
instructions for making important changes in the codes
• instructions for assembling the input data-deck
computing system information
Before discussing these specific details, however, we first consider the
general operating procedures for the use of airshed simulation package.
A. General Information and Instructions
The main tasks involved in exercising the airshed simulation package
are those of establishing the meteorological and emissions input data files,
and then running the APSP. All programs, except the APSP, operate solely
from punched card input data; the APSP requires inputs from both punched
cards and the two data files. Consequently, we work primarily with the
input data-deck to each program and are not required to make substantial
coding changes in any of the programs. There are some model parameters
that have been specified in the programs, however, and in the next section
we discuss which parameters these are and where they are located in the
code.
Assembling the input data-deck for each program is one of the most
important aspects of operating the model. The data-decks for the EDPP,
MDPP, and APSP consist of hundreds of cards, all of which must be specified
correctly. As an aid in defining the contents and format of the data-decks,
we have prepared, for all programs, tables listing the parameters appearing
on each input card. The units of the parameter and the card format are
included in the table.
Since the number of input data cards may vary from one simulation
to the next, • generalized means of denoting the card order is employed in the
data-deck tables. To illustrate this convention, we refer the reader to
Table 4. We find that Card No. 1 actually consists of two physical cards
on which the array NOTE is stored. Card No. 2, however, is only one physical
card on which several input parameters are specified. The number of cards
required to input arrays can be determined from the format information
given in the table. For example, the format of Card No. 1 is 2(20A4).
The "two" outside the parentheses denotes the number of physical cards.
The next topic we wish to discuss deals with making coding changes
in the programs. In Section III-B, we discuss several alterations that
39
-------
the user may wish to make in the codes. It is conceivable, however, that
a change may be required that we have not discussed. If this is the case,
the user should first determine which subroutine contains the algorithm
of interest. The discussion in Section II will be helpful in this respect.
The comments in the program can then be used to further pinpoint the location
of the algorithm, and with the aid of the appropriate symbol glossary,
minor code alterations can be made. Major alterations in any code will,
of course, require that the user understand the basic computational structure
of the program.
Another important consideration involves the use of subscripts in the
codes. This is especially important since several multidimensional arrays
are employed in the programs. We note that any grid cell may be addressed
by the integer triple (I,J,K). The numbering convention for I and J is
illustrated in Figure 1. K designates the vertical level in which the
cell is located, where K = 1 corresponds to the ground level cells. Those
variables which are independent of height are designated by the ordered pair
(I,J). For example, since the wind direction is invariant with height above
grid square (I,J), it is denoted as WINDAN(I,J). The pollutant concentration
array, CONC(L,I,J,K), denotes the concentration of species L in grid cell
(I,J,K), where
1 - Reactive Hydrocarbon
2 - NO
L - 3 - °3
4 - NO-
5 - CO
6 - Unreactive Hydrocarbon
Having discussed several topics related to the operation of the
programs, we now present a set of instructions for the use of the airshed
simulation package. These instructions are included in Table 1. Since
the instructions are formulated in a general manner, they apply to all
simulations performed by the programs given in this Guide. The user will
find information pertaining to computing time, program size, pages of printed
output, and hardware requirements for each program in the next section.
B. Program Operating Characteristics
1. The Emissions Data Preparation Program
The following parameters are specified in the code of the EDPP:
Parameter Card ID Description
XMOL EDPP 0330 molecular weights of auto emissions species
WTFS EDPP 0370 molecular weights of fixed source emissions
species
DX EDPP 0410 grid spacing in x-direction (feet)
DY EDPP 0420 grid spacing in y-direction (feet)
40
-------
TABLE 1
OPERATING INSTRUCTIONS
A, EDPP - use this program to create the EOF
1. Make any required coding changes
2. Prepare the data-deck (use Table 2)
3. Specify the JCL
4. Run the program
B. MDPP - use this program to create the MDF
1. Make any required coding changes
2. Prepare the data-deck (use Table 3)
3. Specify the JCL
4. Run the program
C. APSP - use this program to perform the airshed simulation
1. Create the EOF and the MDF (see A and B above)
2. Make any required coding changes
3. Prepare the data-deck (use Table 4)
4. Specify the JCL
5. Run the program
D. DPP - use this program to plot the results obtained from the APSP
1. Obtain model predictions from the APSP (see C above)
2. Make any required coding changes
3. Prepare the data-deck (use Table 5)
4. Specify the JCL
5. Run the program
41
-------
Parameter Card ID Description
FAIRNO EDPP 0470 mole fraction of NO in aircraft NOX emissions
FCARHC EDPP 0480 mole fraction of reactive hydrocarbon in
auto hydrocarbon emissions
FCARNO EDPP 0490 mole fraction of NO in auto NOX emissions
FFSNO EDPP 0500 mole fraction of NO in fixed source NOX
emissions
HCEVAP EDPP 0590 molecular weight of evaporative hydrocarbon
emissions from autos
These parameters must be checked for correctness before the EDPP is used.
We list below two alterations that may be required in the code.
The temporal distribution for traffic activity is delayed
by one hour in the Downtown Los Angeles area - grid squares
(11,17), (12,17) and (12,16). See cards EDPP 1890, EDPP 1900,
EDPP 1910 and EDPP 3060, EDPP 3070, EDPP 3080.
Emissions are not calculated after 1655 hours. To compute
emissions to a later time, the upper bound on the range of
the DO loop at card EDPP 2940 should be increased, but specified no
larger than 19.
We wish to note the following limitations on the use of the EDPP.
These limitations can be overcome, however, with further program development.
• The calculation of emissions after 2355 hours should not
be attempted.
• Emissions cannot be computed prior to 0500 hours.
Computing System Information
Hardware:
1 - disc area/tape drive FT04F001
1 - card reader FT05F001
1 - printer FT06F001
Program Size: 130K bytes
CPU Time (IBM 370/155): 1 minute
Printed Output: 144 pages
In Table 2 which follows, we give a detailed specification of the
data-deck setup for the EDPP.
42
-------
TABLE 2
DATA-DECK SETUP FOR THE EMISSIONS DATA PREPARATION PROGRAM
Card No.
1
Column No.
1-80
Variable
YCS(I)
Format:
3(8F10.0)
Units
1-30
EC(L)
3F10.0
grams/mile
31
PDT
LI
1-30
EH(L)
3F10.0
grains /mile
1-30
XA(L)
3F10.0
Note
Fraction of cars "cold-
started" during hourly per-
iod I (Is! corresponds to
the period midnight to
1 A.M.); input sequence:
(YCS(I),1 1,24)
Cold-start emissions from
automobiles for species
L, where
!1 CO
2 HC
3 NOX
input sequence:
(EC(L),L=1,3)
If daylight savings time is
in effect, PDT=.TRUE.; if
standard time is in effect,
PDT=.FALSE.
Hot start emissions from
automobiles for species
L; input sequence*
(EH(L),L=1,3)
The parameter "a" in the
automotive speed-emissions
correlation for species L,
input sequence:
(XA(L),L=1,3)
-------
TABLE 2 (Cont'd)
DATA-DECK SETUP FOR THE EMISSIONS DATA PREPARATION PROGRAM
Card No.
Column No.
1-30
Variable
XB(L)
Units
1-80
TSURF(I) 3(8F10.0) minute-1
1-80
TFREE(I) 3{8F10.0) minute'1
1-75
SURMLG(I,J) 2B(25F3.0) 103 miles
Note
The parameter "b" in the
automotive speed-emissions
correlation for species L;
input sequence:
Temporal distribution of
non-freeway traffic;
fraction of total daily
miles driven in each grid
square each minute during
hourly period I; on input,
this array is to be
specified 10** times larger
than its true value; input
sequence :
(TSURF(I) ,I«1,24)
Temporal distribution of
freeway traffic! fraction
of total daily miles driven
on freeways in each grid
square each minute during
hourly period I; on input,
this array is to be speci-
fied 10** times larger than
its true value; input
sequence :
(TFREE(I) ,1=1,24)
Total daily n OB- free way
mileage driven in grid
square (I,J); input
sequence :
((SURMLG(I,J),I=1,25) ,
J=l,25)
-------
TABLE 2 (Cont'd)
DATA-DECK SETUP FOR THE EMISSIONS DATA PREPARATION PROGRAM
Card No.
Column No.
1-75
Variable
Format
Units
FWYMLG(I,J) 25{25F3.0) 103 miles
10
1-72
BETA(M,L,K) 15(12F6.3)
11
1-65/1-60
VS1(I,J)
25(13F5.0
/12F5.0)
mph
12
1-65/1-60
VF1(I,J)
25U3F5.0
/12F5.0)
mph
Note
Total daily freeway
mileage driven in grid
square (I,J); input
sequence:
((FWYMLG(I,J),1=1,25),
J=l,25)
The "beta" correction
curve for species M accoun-
ting for the nonuniform
distribution of vehicle
starts, where
CO
HC
-I!
NO
and L designates the hour
(L=l,5), where L=l,2,...,5
corresponds to 5 A.M., 6
6 A.M., ...» 9 A.M.
K designates each 5 minutes
of hour L, (K=l,12); input
sequence:
(((BETA(M,L,K),K=1,12),
L=1,5),M=1,3)
Average automobile speed
in the slow freeway dir-
ection in grid square (I,J)
at 5 A.M. local time;
input sequence:
((VS1(I,J),I=1,25),J=1,25)
Average automobile speed in
the fast freeway direction
in grid square (I,J) at
5 A.M. local time; input
sequence]
((VF1(I,J),1=1,25),J=1,25)
-------
TABLE 2 (Cont'd)
DATA-DECK SETUP FOR THE EMISSIONS DATA PREPARATION PROGRAM
Card No.
13
Column No.
1-65/1-60
Variable
Format
FXSOUR(I,J,L) 150U3F5.0
/12F5.0)
14
1-5
NOARPT
15
Note
Fixed source emissions for
species L in grid square
(I,J); input sequence:
(((FXSOUR(I,J,L),1=1,25),
J=l,25),L=1,6)
where
1
2
3
4
5
6
reactive HC
NO
03
NO2
CO
unreactive 1
Number of airports
If NOARPT=0, then do NOT include Card No. 15-23.
15
1-70
AIREMS(M,L,I)
18(7F10.1)
Ib./min
Emissions rate of species
L from class I aircraft
ie operating mode M,
where
1 long range jet
transport
2 medium range jet
transport
3 business jet
transport
4 turboprop trans-
port
5 piston engine
transport
6 piston engine
utility
7 turbine engine
helicopter
-------
16
*»
TABLE 2 (Cont'd)
DATA-DECK SETUP FOR THE EMISSIONS DATA PREPARATION PROGRAM
Card No. Column No.
Variable
Format
Units
1-70
TMODE(M,D 3(7F10.0)
minute
M =< 2
(3
Note
taxi
landing
take-off
L - see Card No. 13
input sequence:
(((AIREMS(M,L,I),1=1,7),
L=l,6),M=1,3)
The amount of time class
I aircraft spend in opera-
ting mode M; see Card No.
15 for description of I and
M; input sequence:
((TMODE(M,I) ,1=1,7) ,M=1,3)
Include a complete set of Card No. 17-23 for each airport (J=1,NOARPT)
17
1-4
AIRPRT
A4
17
18
5-6
1-30
NOORT(J)
IXORT(J,I)
12
615
Alphameric designation of
the airport initials
(such as 'LAX','BUR', and
etc.)
Number of grid squares
which receive emissions
from airport J
Column numbers of the grid
squares that receive emis-
sions from airport J; input
sequence:
(IXORT(J,I) ,I=1,NCORT(J))
-------
TABLE 2 (Cont'd)
DATA-DECK SETUP FOR THE EMISSIONS DATA PREPARATION PROGRAM
Card No.
19
Column No.
1-30
Variable Format
IYORT(J,I) 615
20
1-60
21
1-78
PCT(J,I)
CO
22
1-42
ACFT(J,I)
23
1-42
EMIX(J,I)
Units
DIST(J,I) 2(12F5.3)
hour-1
13F6.2
7F6.2
aircraft/
day
7F6.2
engines/
aircraft
Note
Row numbers of the grid
squares that receive
emissions from airport J;
input sequence:
(IYORT(J,I),I=1,NCORT(J))
Fraction of total daily
aircraft operations occur-
ring at airport J during
hour I (Is! corresponds to
the period midnight -
1 A.M.); input sequence:
(DIST(J,I),1=1,24)
Fraction of emissions
from airport J that are
apportioned to grid square
(IXORT(J,I),IYORT(J,I));
input sequence:
(PCT(J,I),I-1,NCORT(J))
Number of daily flights
of class I aircraft at
airport J) input sequence:
(ACFT(J,I),I=1,7)
See Card No. 15 for a
detailed description of I.
Number of engines on class
I aircraft operating at
airport J; input sequence:
(EMIX(J,I),1=1,7)
-------
TABLE 2 (Cont'd)
DATA-DECK SETUP FOR THE EMISSIONS DATA PREPARATION PROGRAM
Card No.
24
Column No.
1-65/1-60
Variable
Format
25U3F5.0/
12F5.0)
Units
25
1-65/1-60
VS1(I,J)
25(13F5.0
/12F5.0
mph
vo
26
1-65/1-60
VF1(I,J)
25(13F5.0
/12F5.0)
mph
Note
Ratio of the number of
miles traveled in the
slow direction to the num-
ber traveled in the fast
direction in grid square
(I,J) between 5 and 6 A.M.
local time) input sequence:
((X(ItJ), 1-1,25) ,J-1,25)
Average speed traveled by
automobiles in the slow
direction on freeways in
grid square (I,J) at
6 A.M. local time i input
sequence :
((VS(I,J) ,1-1,25) ,0*1,25)
Average speed traveled by
automobiles in the fast
direction on freeways
in grid square (I,J) at
6 A.M. local time; input
sequence t
((VF(I,J), 1-1,25) ,J=1,25)
Include four additional sets of Card No. 24-26 to cover the time period
6 A.M. - 10 A.M.
-------
2. The Meteorological Data Preparation Program
The MDPP is a general program, and the only limitations on its
use ares
• mixing depths must be specified on the hour
• wind speed and wind direction are constant over hourly
periods measured from the half hour to the half hour
• the initial mixing depths are to be applicable at the
same time the airshed simulation is to start. That is,
if a simulation is to start at 0500 hours, then the first
mixing depth data should be applicable at 0500 hours.
This is a limitation imposed by the APSP.
Computing System Information
Hardwares
1 - disc area/tape drive FT04F001
1 - card reader FT05F001
1 - printer FT06F001
Program Size: 60K bytes
CPU Time (IBM 370/155) «
time required to process 10 hours of meteorological data - 25 seconds
Printed Output:
number of pages of printed output from 10 hours of meteorological
input data
10 hours x 3 pages/hour = 30 pages
The complete data-deck setup for the MDPP is described in
Table 3.
3. The Atmospheric Pollution Simulation Program
We list below four operating parameters which are specified in the
code of the APSP. They are all located in the MAIN program.
Parameter Card ID Description
KXY APS 00850 horizontal turbulent diffusivity
NOSPEC APS 00860 maximum number of chemical species
NORATE APS 00870 number of chemical reactions
DTAVEC APS 00910 t^ period (in minutes) over which the
predicted concentrations are time averaged
50
-------
TABLE 3
DATA-DECK SETUP FOR THE METEOROLOGICAL DATA PREPARATION PROGRAM
ui
Card No.
1
Column No.
1-5
6-10
Variable
NX
NY
Format
15
15
Units
1
1
1
2
3
4
5
6
11-20 TSTART F10.0
21-30 TSTOP F10.0
31-38 DATE (I) 2A4
1-65/1-60 H(I,J,1) 25(13F5.0/
12F5.O
1-65/1-60 W(I,J) 25(13F5.1/
12F5.1)
1-65/1-60 A(I,J) 25(13F5.1/
12F5.1)
1-65/1-60 H(I,J,2) 25U3F5.1/
12F5.1)
1-65/1-60 W(I,J) 25U3F5.1/
12F5.1)
2400 time
2400 time
—
feet
mph
degrees
feet
mph
Note
Number of grid squares in
the x-direction
Number of grid squares in
the y-direction
Starting time
Terminating time
Alphameric designation of
the date; input sequence:
(DATE(I),1-1,2)
Mixing depths applicable
at TSTART; input sequence:
((H(I,J,1),1=1,NX),J=1,NY)
Wind speed applicable at
TSTART; input sequence:
((W(I,J),I=1,25),J=1,25)
Wind direction measured
clockwise from north,*
applicable at TSTART; input
sequence:
((A(I,J),1=1,25),J=1,25)
Mixing depths applicable at
the next hour; input
sequence:
((H(I,J,2),I=1,NX),J»1,NY)
Wind speeds applicable at
the next hour; input
sequence:
((W(I,J),I=1,25),J=1,25)
The direction in which the wind is flowing.
-------
Card No.
7
Column No. Variable
1-65/1-60 A(I,J)
Format
25U3F5.1/
12F5.1)
Units
degrees
TABLE 3 (Cont'd)
DATA-DECK SETUP FOR THE METEOROLOGICAL DATA PREPARATION PROGRAM
Note
Wind directions applicable
at the next hour; input
sequence:
<(A(I,J),I=1,25),J=1,25)
Include a complete set of Card No. 5-7 for each hour of meteorological data.
Note: Mixing depths are assigned on the hour (0500, 0600, and etc.) , and wind speeds
and wind directions are applicable from the half hour to the half hour (0430-0530,
0530-0630, and etc.). TSTART must be assigned an even hour, such as 0500 hours.
Ul
to
-------
An important change the user may wish to make in the APSP involves
the formulation of the chemical kinetic mechanism. Rate constants and
certain stoichiometric coefficients are input via punched cards, which
allows these parameters to be altered without changing the code. We
summarize below those sections of the code which are directly affected
by the formulation of the kinetic mechanism.
Routine Card ID Description
MAIN APS 06560-06630 Expressions relating the concentrations
of the "steady state" species to the
concentrations of HC, NO, N02, and O3
MAIN APS 00920-00950 RATE(L) - rate of reaction of species L
APS 06710-06890
3(rate expression for species I)
R(I,J; = - g^concentration of species J)
MAIN APS 05520-05530 Initial and boundary concentrations of ozone
BCCONC A1"8 11490-11540 are determined by assuming that the rate
ICCONC APS 13200-13250 of reaction of ozone, both initially and
at the boundary, is equal to zero. Thus,
we can write an algebraic expression
relating the initial and boundary concentrations
of ozone to the initial and boundary
concentrations of HC, NO, and NO2.
MMN APS 04980-05010 Calculations involving the temporal
APS 02790-02810 variation of photolysis rate constants.
ICCONC APS 13150
In order to establish the boundary concentrations aloft correctly,
the first set of mixing depths in the MDF must be applicable at the starting
time of the airshed simulation.
During the evaluation phase of model development, we found it advantageous
to perform a 10 hour simulation in two 5 hour segments. Thus, a simulation
for 0500-1500 hours is carried out in two sequential steps: 0500-1000 and
1000-1500 hours. To do this, we have made provisions for restarting the
program using the results obtained at the end of a previous simulation as
initial conditions. The parameter PUNCH, specified on Card No. 2 of the
data-deck (see Table 4), determines whether or not the concentration array,
CONC(L,I,J,K), is "dumped" onto a save disc area or tape at the time the
program terminates. A program restart is controlled by the parameter RESTRT
(see Card No. 2 in Table 4). During the restart procedure the initial
concentrations on the grid are established by reading the file containing
the last concentrations predicted in the previous run. We note that the
restart option should only be used when it is of interest to closely monitor
the outcome of a particular simulation.
One of the important features of the APSP is that the user can
easily modify the computational grid - both horizontally and vertically.
The number of vertical cells (or strata) employed must be at least three,
53
-------
and up to six can be specified in the present code. Only available storage
restrictions limit the present code to a maximum of six levels. As far
as the horizontal extent of the grid is concerned, the user defines,
as part of the program input, the shape of the modeling area. In Figure 1,
we illustrate that portion of the original 25x25 grid actually used during
the simulation studies of the Los Angeles airshed.
To obtain the shape of the computational grid, we eliminate a
series of consecutive grid squares starting from the left and/or right
hand borders. Consider the region illustrated in Figure 1. In row 1,
20 consecutive grid squares, starting from the left hand border, are
eliminated from the original 25x25 grid. As another example, 14 consecutive
grid squares, starting from the right hand border, are eliminated from row 25.
The region designation is made by specifying two arrays, ILEFT(J) and IRIGHT(J),
containing the column numbers of the left and right-most cells in row j.
The following values of ILEFT and I RIGHT are obtained from Figure 1:
ILEFT(l) = 21 IRIGHT(l) = 25
ILEFT(2) = 20 IRIGHT(2) = 25
ILEFT(25) = 1 IRIGHT(25) = 11
Instantaneous and time-averaged ground level concentrations can be
predicted at up to 16 user-selected sites. The x and y-coordinates of each
site are input to the program on punched cards, where the center of grid
square (I,J) has an x-coordinate of I and a y-coordinate of J. In addition,
the name of each station is printed on the ground level concentration maps
near its location on the grid (see Figures 3 and 4). The user can also
specify up to 8 landmarks. The name of each landmark is printed on the ground
level concentration maps as an aid in identifying the location of a
particular area of the grid. Four landmarks are illustrated in Figures 3
and 4, including the Pacific Ocean, the Santa Monica Mountains, the San
Gabriel Mountains, and Palos Verdes.
Computing System Information
Hardware:
5 - disc areas/tape drives
a. EOF (required) FT02F001
b. MDF (required) FT04F001
c. Scratch-Area (required) FT08F001
d. Save area where concentrations can be written at
the end of a simulation for use as initial conditions
in a program restart (optional) FT03F001
e. Save area where concentrations are located to
be used as initial conditions in a.program restart
(optional). FT09F001
1 - card reader FT05F001
1 - printer FT06F001
1 - card punch FT07F001
54
-------
Program Size: 304K bytes
CPU time (IBM 370/155):
10 hour CO simulation - 4 minutes
10 hour photochemical simulation involving RHC, URHC, NO, NO2, 03,
and CO - 73 minutes
Printed Output:
10 hour CO simulation with instantaneous concentration values
printed once an hour.
• hourly-averaged ground level concentration maps
1 species x 10 hours x 1 page/species hour = 10 pages
• hourly-averaged predictions at monitoring stations
10 hours x 1 page/hour = 10 pages
instantaneous ground level concentration maps
1 species x 10 hours x 1 page/species hour = 10 pages
instantaneous concentration profiles above each
station (15 stations)
10 hours x 2 pages/hour = 20 pages
• miscellaneous, including grid specifications, rate
constants, and initial conditions = 4 pages
Total Printed Output = 54 pages
10 hour photochemical simulation (6 species) with instantaneous
concentration values printed once an hour.
• hourly-averaged ground level concentration maps
6 species x 10 hours x 1 page/species hour = 60 pages
• hourly-averaged predictions at monitoring stations
10 hours x 1 page/hour = 10 pages
instantaneous ground-level concentration maps
6 species x 10 hours x 1 page/species hour - 60 pages
• instantaneous concentration profiles above each
station (15 stations)
10 hours x 3 pages/hour = 30 pages
• miscellaneous, including grid specifications,
rate constants, and initial conditions = 10 pages
Total Printed Output =170 pages
In Table 4 which follows, we describe in detail the data-deck
setup for the APSP.
55
-------
TABLE 4
DATA-DECK SETUP FOR THE ATMOSPHERIC POLLUTION SIMULATION PROGRAM
Card No.
Column No.
Variable
Format
Units
Note
1-80
NOTE(I)
PUNCH
2(20A4)
LI
Two cards of user-
supplied alphameric infor-
mation to be printed on
program output; input
sequence:
(NOTE(I),1=1,40)
If PUNCH=.TRUE., then the
entire concentration array
(CONG) will be dumped onto
a peripheral storage area
at the end of the simula-
tion. These concentrations
can be used for a program
restart.
in
RESTRT
LI
If RESTRT=.TRUE., then the
program restarts using the
results obtained in a
previous simulation for
initial conditions
6-10
NOX
15
Number of grid points in
x-direction
11-15
16-20
NOY
NOZ
15
15
Number of grid points in
y-direction
Number of grid points in
z-direction
21-30
DT
F10.0
minute
Initial time step size
for the numerical inte-
gration
-------
TABLE 4 (Cont'd)
DATA-DECK SETUP FOR THE ATMOSPHERIC POLLUTION SIMULATION PROGRAM
Card No.
2
2
2
2
2
Column No.
31-40
41-50
51-60
61-70
71-80
Variable
DELTAX
DELTA*
PRTDT
TSTART
TSTOP
Format
F10.0
F10.0
F10.0
F10.0
F10.0
Units
feet
feet
minute
2400 time
2400 time
1-6
7-30
ISPEC(L)
UNIT(L)
6L1
6A4
Note
Grid spacing in x-direction
Grid spacing in y-direction
Time interval between
instantaneous concentration
print-outs
Simulation starting time
Simulation stopping time
If ISPEC(L)=.TRUE., then
species L will be simulated
1 reactive hydrocarbon
2 nitric oxide
3 ozone
4 nitrogen dioxide
5 carbon monoxide
6 unreactive hydro-
carbon
Species 1-5 must all be
simulated when performing
photochemical calculations;
species 5' and/or 6 can be
simulated independently;
input sequence:
(ISPEC(L),L=1,6)
Alphameric designation of
the units to be used on
input/output for the con-
centration of species L.
For the above species
-------
TABLE 4 (Cont'd)
DATA-DECK SETUP FOR THE ATMOSPHERIC POLLUTION SIMULATION PROGRAM
Card No.
Column No.
Variable
Format
Units
Note
31-60
FACTOR(L)
6F5.1
oo
61-64
1-80
TIMEZN
REGION(I)
A4
20A4
UNIT(L)
L=1,'PPHM'
L=2,'PPHM'
L=3,'PPHM'
L=4,'PPHM'
L=5,'PPM '
L=6,'PPTM'
input sequence:
(UNIT(L),L=1,6)
Conversion factor; must
be consistant with UNIT(L)
If UNIT(L)
then FACTOR(L)
{'PPHM1 }
•PPTM'> ,
'PPM ')
(100. |
• h: •
input sequence:
(FACTOR(L),L=1,6)
Alphameric designation of
the time zone (for Los
Angeles, use 'PST ')
Alphameric designation of
the geographical location
to be modeled (for example,
'LOS ANGELES AND VICINITV') ;
input sequence:
(REGION(I),1=1,20) .
-------
TABLE 4 (Cont'd)
DATA-DECK SETUP FOR THE ATMOSPHERIC POLLUTION SIMULATION PROGRAM
Card No.
Column No.
1-75
Variable
ILEFT(J)
Format
2513
Units
1-75
IRIGHT(J)
2513
in
ID
1-3
NOSTN
13
4-6
7-9
NOLAND
NOPTS
13
13
Note
ILEFT defines the left
(western) border of the
model. ILEFT(J) is the
column number of the left-
most cell in row J;
input sequence:
{ILEFT(J),J=1,NOY)
IRIGHT defines the right
(eastern) border of the
model. IRIGHT(J) is the
column number of the right-
most cell in row J; input
sequence:
(IRIGHT(J),J=1,NOY)
Number of stations for
which vertical concentra-
tion distributions and
hourly-averaged ground
level concentrations will
be calculated and printed
Number of landmarks that
will be printed on the
program output
Number of point sources
If NOSTN = 0, do NOT include Card No. 8.
If NOSTN>0, include one Card No. 8 for each station (1=1,NOSTN).
Card No. 8 until YSTN(1)> YSTN(2)>...> YSTN(NOSTN).
Sort all
-------
Card No.
8
8
TABLE 4 (Cont'd)
DATA-DECK SETUP FOR THE ATMOSPHERIC POLLUTION SIMULATION PROGRAM
Column No.
1-20
21-30
31-40
Variable
STNAME(I,J)
XSTN(I)
YSTN(I)
Format
5A4
F10.0
F10.0
Units
Note
Alphameric designation of
station I (for example
'DOWNTOWN LA1); input
sequence:
(STNAME(I,J),J=1,5)
x-coordinate of station I.
y-coordinate of station I.
If NOLAND
, do NOT include Card No. 9.
If NOLAND>0, include one Card No. 9 for each landmark (1 = 1, NOLAND).
Sort all Card No. 9 until YLAND(l) >YLAND(2)>.. .>YLAND(NOLAND) .
9
9
1-20
21-30
31-40
LANDMK(I,J)
5A4
XLAND(I)
YLAND(I)
F1O.O
F1O.O
Alphameric designation of
landmark I (for example
•PACIFIC OCEAN')j input
sequence t
(LANDMK(I,J),J=1,5)
x-coordinate of landmark I
y-coordinate of landmark I
If NOPTS = 0, do NOT include Card No. 10-11.
If NOPTS>0, include a set of Card No. 10-11 for each point source (I = 1, NOPTS),
10
10
1-5
6-10
XPTS(I)
YPTS(I)
F5.0
F5.0
x-coordinate of point
source I
y-coordinate of point
source I
-------
Card No.
10
10
11
12
12
12
12
TABLE 4 (Cont'd)
DATA-DECK SETUP FOR THE ATMOSPHERIC POLLUTION SIMULATION PROGRAM
Column No. Variable Format Units
HEMIS(I) F10.0 feet
11-20
21-80
1-80
1-5
6-10
11-15
16-20
PTEMIS(L,I)
6F10.0 pound-mole
/hour
TEMPTS(I,J) 3(8F10.0)
MAXITR
RERROR
DTMIN
DTMAX
15
F5.0
F5.0
F5.0
Minute
Minute
Note
Height at which point
source I emits pollutants
Nominal emissions rate
of species L from point
source I; input sequence:
(PTEMIS(L,I),L=1,6)
Temporal distribution for
point source I; fraction
of nominal emissions rate,
PTEMIS(L,I), during hourly
period J. PTEMIS(L,D*
TEMPTS(I,J) = emissions
rate of species L from
point source I during
hourly period J; input
sequence:
(TEMPTS(I,J),J=l,24)
Maximum number of itera-
tions allowed in the
Newton solution of the non-
linear difference equations
Relative error tolerance
that must be satisfied
before the Newton iteration
is considered finished
Minimum allowable integra-
tion time step. If the
program must take a smaller
step, it terminates.
Maximum allowable inte-
gration time step
-------
ov
to
13
14
TABLE 4 (Cont'd)
DATA-DECK SETUP FOR THE ATMOSPHERIC POLLUTION SIMULATION PROGRAM
Card No.
12
12
12
12
12
Column No.
21-25
26-30
31-35
36-40
41-45
Variable
ALPHA
BETA
GAMMA
EPS
HTOP
Format
F5.0
F5.O
F5.0
F5.O
F5.O
1-80
1-80
RKO(I)
FACTRK(J)
(8F10.0)
3(8F10.0)
Units
feet
ppm-min
Note
Chemical kinetics parameter
Chemical kinetics parameter
Chemical kinetics parameter
Chemical kinetics parameter
Height above which the
ambient concentration above
the inversion base is
CALOFT (see Card No. 16)
Nominal chemical reaction
rate constant for reaction
I; input sequence:
(RKO(I),I=1,NORATE)
NORATE = number of chemical
reactions (this parameter
is defined in the MAIN
program)
Light intensity curve;
fraction of nominal photo-
lysis rate constant (k^,k_)
at hour J
(J=l corresponds to mid-
night)
k =RKO(1)*FACTRK(J) at hour
J; input sequence:
(FACTRK(J),J=1,24)
-------
TABLE 4 (Cont'd)
DATA-DECK SETUP FOR THE ATMOSPHERIC POLLUTION SIMULATION PROGRAM
Card No.
15
Column No.
1-24
Variable
NAME(L)
Format
6A4
Units
16
1-75
CALQFT(L,I,J)
(25F3.0)
(see note)
see note
u>
Note
Alphameric designation of
the name of species L
NAME(L)
L=2,
' HC '
1 NO '
L=3, ' 03 '
L=4, ' NO21
>L=5, ' CO '
L=6, 'URHC'
input sequence:
(NAME(L),L=1,6)
Ambient concentration aloft
of species L above grid
square (I,J). These con-
centrations apply when the mixing
depth above grid square (I,J)
is greater than HTOP (see
Card No. 12).
The concentration units
are the same as those
designated by UNIT(L)(see
Card No. 3). The input
sequence is governed by the
following logic:
DO 3 L=1,NOSPEC
IF(.NOT.ISPEC(L)) GO TO 3
DO 1 J=1,NOY
1 READ(5,2)(CALOFT(L,I,J),
11=1,NOX)
2 FORMAT(25F3.0)
3 CONTINUE
Note that input cards are
included only for those
species simulated.
-------
TABLE 4 (Cont'd)
DATA-DECK SETUP FOR THE ATMOSPHERIC POLLUTION SIMULATION PROGRAM
Card No. Column No. Variable Format Units Note
17 1-75 CONCI(L,I,J) 25F3.0 see note Initial ground level con-
(see note) centration of species L in
grid cell (I,J,1). The
concentration units are the
same as those designated
by UNIT(L) (see Card No. 3X
The input sequence is
governed by the same logic
as that used for the input
of CALOFT(L,I,J) (see Card
No. 16). No cards are in-
cluded for those species
not being simulated.
Include a sufficient number of sets of Card No. 18-19 to cover the entire
time period to be simulated (i«e., TSTART to TSTOP-rsee Card No. 2). The
sets of Card No. 18-19 should be arranged sequentially, according to the time
interval designated by TDATA and TNEXT. For example, consider the following
sequence:
Card No. 18 - TDATA = 0500, TNEXT = 0600.
Card No. 19 - Boundary conditions that apply between 0500 and 0600.
Set #1 {
Card No. 19 - Boundary conditions that apply between 0600 and 0700
0 ^ „_ i Card No. 18 - TDATA = 0600, TNEXT = 0700.
Set #2 •{
18 1-10 TDATA F10.0 2400 time Beginning of the time in-
terval over which the
following boundary condi-
tions hold.
18 11-20 TNEXT FlO.O 2400 time End of the time interval
over which the following
boundary conditions hold.
-------
TABLE 4 (Cont'd)
DATA-DECK SETUP FOR THE ATMOSPHERIC POLLUTION SIMULATION PROGRAM
Card No.
19
Column No.
1-75
Variable
BC(L,I,J)
Format
25F3.0
(see note)
Note
Boundary concentrations for
species L at points of hor~
izontal inflow to the
model. I denotes the row
(east or west border) or
column (north or south bor-
der) number of the boundary
cell on border j.
1
2
3
4
refers to the
m
top (north)
bottom (south)
right (east )
left (west )
border
The concentration units are
the same as those designa-
ted by UNIT(L) (see Card
No. 3). The input sequence
is governed by the follow-
ing logic:
DO 3 L=1,NOSPEC
IF(.NOT.ISPEC(L))GO TO 3
. ISTOP=HOX
DO 1 J=l,4
IF(J.GT.2) ISTOP=NOY
1 READ(5,2)(BC(L,I,J),I=1,
1 ISTOP)
2 FORMAT(25F3.0)
3 CONTINUE
Note that input cards
are included only for
those species simulated.
-------
4. The Data Plotting Program
We summarize below the changes that may be required in the
code of the DPP.
• The NO measurements at stations 5 and 9 (El Monte
and Commerce) are not corrected. To make corrections
at these stations, remove card DPP 1920.
• Oxidant measurements are not corrected at stations
5 and 9 (El Monte and Commerce). To make oxidant
corrections at these stations, remove card
DPP 2000.
• Oxidant measurements are not corrected at station
number 13 (Long Beach). To correct oxidant measurements
at this station, remove card DPP 1990.
The following limitations are placed on the use of the DPP:
• Only local corrections to CO predictions will be plotted
• No more than 12 hours of data can be plotted
Computing System Information
Hardware:
1 - card reader FT05F001
1 - printer FT06F001
1 - Calcomp Plotter
Program size: 68K bytes
CPU Time (IBM 370/155)i 30 seconds
Printed Output: 1 page
Software:
The following Calcomp routines are used in the DPP. All
coordinates and heights are measured in inches.
SYSPLT(X,Y.IPN)
This routine moves the pen from its present location to (X,Y)
with the pen up (IPN=3) or down (IPN»2).
SYSSYM(X,Y,H,SYMBOL,N,PEG)
This routine plots the alphameric information contained
in SYMBOL at the point (X,Y). The height of the plotting
is H, and the plot is produced at an angle of DEC with
respect to the x-axis. The number of symbols to be plotted
is N.
66
-------
SYSEND(LAB,FLAG)
This routine is used to terminate a plot with or without
plotting the page number at the lower-right corner of
the paper.
= 0, the pen will return to the current
origin without plotting a page number.
LAB > 0, the page number is plotted. Then the
pen is moved to the next origin.
< 0, the pen is moved to the next origin without
plotting the page number.
yt 0, the statements "ONE SHEET PLOTTED" and
"ENDING PLOT LABELED..." will be printed
FLAG on user's output.
=0, the above printing will be suppressed.
The data-deck setup for the DPP is described in Table 5.
67
-------
oo
Card No.
TABLE 5
DATA-DECK SETUP FOR THE DATA PLOTTING PROGRAM
Column No.
1-5
31-40
41-48
Variable
NOSTN
CARUHC
DATE(I)
Format
15
1
1
1
6-10
11-20
21-30
NOHOUR
TSTART
CARRHC
15
F10.0
F10.0
F10.0
2A4
Units
2400 time
Note
Number of monitoring
stations
Number of hours simulated
Simulation starting time
Number of carbon atoms
per molecule of reactive
hydrocarbon
Number of carbon atoms
per molecule of unreactive
hydrocarbon
Alphameric designation of
the date - for example,
'9/29/69'; input sequence:
(DATE(I),1=1,2)
Include a Card No. 2 for each monitoring station (K=1,NOSTN); be certain
station K corresponds to station number K in the Atmospheric Pollution
Simulation Program.-
1-20
21-26
STNAME(L,K)
5A4
PLSPEC (K,L) 6L1
20 columns of alphameric
information designating the
name or location of moni-
toring station K; input
sequence:
(STNAME (I,,K) ,L=1,5)
If PLSPEC(K,L)=.TRUE.,
then plot species L at
monitoring station K;
input sequence:
(PLSPEC(K,L),L=1,6)
-------
TABLE 5 (Cont'd)
DATA-DECK SETUP FOR THE DATA PLOTTING PROGRAM
Card No.
Column No.
Variable
Format
Units
Note
Include a Card No. 3 for each monitoring station reporting and each chemical
species measured. The cards may be in any order; columns 7-9 should be left
blank on all cards EXCEPT the last one, on which a "T" must be punched
somewhere in columns 7-9.
7-9
LAST
L3
3
3
10-11
12-13
ISTN
ISPEC
12
12
vo
33-80
DATA(I)
24F2.O
If LAST=.TRUE., then the
present card is the last
Card No. 3 in the data-
deck
Monitoring station number
Chemical species number
1 SC>2 (pphm)
2 NO (pphm)
3 °3 (pphm)
4 NO 2 (pphm)
5 CO (ppm )
6 total HC (ppmC)
ISPEC=
Measured hourly-averaged
pollutant concentrations
starting at midnight (see
ISPEC above for units) ;
input sequence :
(DATA(I) ,1=1,24)
Include a Card No. 4 for each station and each chemical species for which a
local correction is to be made; columns 1-3 should be left blank on all cards
EXCEPT the last one, on which a "T" must be punched somewhere in columns 1-3.
1-3
LAST
L3
If LAST=.TRUE., then the
present card is the last
Card No. 4 in the data-deck
-------
TABLE 5 (Cont'd)
DATA-DECK SETUP FOR THE DATA PLOTTING PROGRAM
Card No.
4
4
Column No.
4-6
7-9
Variable
ISTN
ISPEC
Format
13
13
Units
10-45
STCORR(ISTN,ISPEC,I) 12F3.0
Note
Monitoring station number
Chemical species number
1 reactive hydrocarbon (pphm)
2 NO (pphm)
ISPEC=
3 0-
4 NO,
(pphm)
(pphm)
5 CO (ppm )
6 unreactive hydrocarb on(pptm)
Local correction to be
applied to the model pre-
diction at station ISTN
for species ISPEC at hour
I; 1=1 corresponds to the
hourly average starting at
TSTART (see Card No. 1);
see ISPEC above for the con-
centration units; input
sequence:
(STCORR(ISTN,ISPEC,I),1=1,
NOHOUR)
Include a Card No. 5 for each chemical species and each hour simulated.
The cards may be in any order. These cards are punched by the
Atmospheric Pollution Simulation Program.
5
5
5-10
11-16
17-80
ISPEC
Tl
16 — Chemical species number
F6.0 2400 time Hour the model predictions
were averaged over
16F4.0 -- Concentration of species
ISPEC predicted at monitor-
ing site I over hour Tl;
input sequence:
,I=1,NOSTN)
-------
IV. LISTINGS, SYMBOL GLOSSARIES, AND SAMPLE OUTPUT
In this section we include the listing, symbol glossary,
and sample output for each computer program.
71
-------
C*<**$ *****************************************************************
C*
C*
C*
C*
C*
C*
C*
C*
rC*
C*
C*
C*
C*
C*
C*
C*
C*
C*
C*
THE EMISSIONS DATA PREPARATION PROGRAM
WRITTEN BYJ STEVEN D, REYNOLDS
FOR
SYSTEMS APPLICATIONS, INC,
9418 WILSHIRF BLVO,
BEVERLY HILLS, CALIFORNIA 90212
JCL INFORMATION
FT04F001 EMISSIONS DATA FILE
PT05F001 CARD READER
FT06F001 PRINTER
C* THE EMISSIONS DATA FILE IS DEFINED AS FOLLOWS (USING A 9
C* TRACK
C*
C*FT04P001
C*
C*
, 1600 BPI TAPE) 1
DO UNI T» ( T APE U,, DEFER), LABEL* (1,BLP),D I SP« (OLD, KEEP),
VQLiSERBEDPOOl,DCBs(RECFHeVS,LRECLc2504,BLKSIZE32508)
**EOPP0010
*tDPP0020
*EOPP0030
*EDPP0040
*EDPPOObO
*EOPP0060
*£OPPU070
*EOPP0080
*£OPP0090
*EOPPOIOO
*EOPP0110
*tDPPQ120
*FDPP0130
*EDPP0150
*EDPP0160
*£f)PPO!70
AEDPP0180
*EOPPt)200
*EDPPQ210
*EOPP0220
*EOPP0240
C
C
C
C
DIMENSION EC(3),EH(3),XAC3),TSURF(24),TPREE(24),8URMLG(2b,25),F
1LG(25,25),BETA(6,5,12),V3(23,25),VF(25,25),V81(25,25),VF1(25,25
1 SMILES(25,25),FMILES(25,25),E(3,25,25),X(25,25),91(6,25,25),
1 Q2(6,25,25),QSLOPE(25,25),XMOL(6),XB(3)
DIMENSION PXSOUR(25,25,6),FACTFS(6),AIRFLX(6,60),YCS(24)
DATA XMOL/47,8,30,,48,,46.,28,,21.1 /
COMMON/AIRCFT/NOARPT,NCORT(15),EMIX(15,7),ACFT(15,7),PCT(15,4),
1 Dl8T(i5,24),IXORT(15,4),IYORT(15,fl),TMODE(3,7),AIREMS(3,6,7)
REAL NAME(6),WTF8(6)
DATA WTP8/60,,SO,,48,,46tf28,,18./
DATA NAME/4H HC ,4H NO ,«H OS ,4H N02,4H CO ,4HURHC/
LOGICAL KFLAG,PDT
KFLAC«,FALSE,
DX110560,
OY-10S60,
*** DEFINE MOLE FRACTIONS OF NO AND REACTIVE HYDROCARBON IN
TOTAL NOX AND HYDROCARBON EMISSIONS
WYMEOPP0260
), fcDPPO?90
EOPP0300
EOPP0310
EDPP0320
FAXRNQpO,99
FCARHCiO.674
PCARNO«0,99
PFlNOtOf8
C
C
C
C
*** COMPUTE THE MASS FRACTION OF NO AND REACTIVE HYDROCARBON
IN TOTAL NOX AND HYDROCARBON EMISSIONS
XCARHC>FCARHC*XMOL(i)'(FCARHC*XMOL(l)
XAZRNQWFAXRNO*XMOL(2)/(FAIRNO*XHOL(2)
XCARNO»FCARNO*XMOL(2)/(FCARNO*XMOL(2)
XP8NO«FP8NO*WTF8(2)/(FFSNO*WTFS(2) *
HCEVAP-67,6
DENCARP46./30, * (1,»XCARNO)/XCARNO
U,»FCARHC)*XMQL(6))
(1,»FAIRNO)*XMOL(4))
U,»FCARNO)*XMOL<4))
•FFSNO)*WTFS(4))
EDPP0.340
EDPP0350
EOPP0360
EOPP03/0
EOPP0380
EOPP0390
EOPP0400
EDPP0410
EDPP0420
EDPP0430
EOPP0440
EDPP0450
EDPP0460
EOPP0470
EDPP0480
EDPP0490
EOPP0500
EOPP0510
EDPP0520
EDPPOS30
EOPP0540
EDPPOS50
EDPPObbO
EDPP0570
EDPP0580
EDPPOS90
EDPP06UO
72
-------
c
c
c
900
901
c
c
c
902
699
c
c
c
100
c
c
c
101
c
c
c
«00
300
C
C
C
10
11
C
C
C
+ 1,-XAlRNQ
»DENFSsq6,/30, * (1 ,-XFSNQ)/XFSNO
FACTEv/aO, 343*8. 68E5/(60, *HCEVAP*DX*DY)
*** INPUT CARD MO, 1*7
READ(5,900)
READ(5,901>
READC5.900)
READ(5,900)
READ(5,900)
READ(5,900)
RFAD(5,9QO)
VC8
EC,POT
EH
XA
XB
TSURF
TFREE
FQRMATC8P10.0)
FORMAT(3F10, 0,1.1)
*** ESTABLISH STARTING TIME
TDATAaaSS,
TNEXTsSOO,
IPC.NQT.PQT) 6C TO 902
TOATA8355.
TNEXTs
-------
READ(5,t7) AIRPRT,NCORT(J)
15
16
19
17
12
13
C
C
C
C
105
C
C
C
55
36
C
C
C
C
C
I
C
C
C
C
C
C"
C
C
C'
MBNCORT(J)
REAOC5,16)
READ(5,16)
REAI3(5,19)
READ(5,15)
READ(5,15)
READ(5,15J
(IXORT(J,I),IP1,M)
(IYORT(J,I),X«1,M)
(PCTU,I),Ift,M)
(ACFTCJ,n.I»l,7)
FQRMAT(6I5)
FORMATU2F5.3)
FORMAT(Aa,I2)
CONTINUE
CONTINUE
*** CONVERT FIXED SOURCE EMISSIONS INTO
THE UNITS PPM»PEET/M1NUTE
A SURFACE KLUX WITH
DO 14 L«l,6
FACTFs(L)»1000,*e,680E5/(WTFS(L)*60,*DX*DY)
FACTFS(2)«PACTFS(2)/DENFS
FACTFS<4)«FACTFSU)*C1,0-XFSNO)/CXF8NO*DENFS)
DO 105 Lei,6
DO 105 I»l,25
DO 105 J?l,25
FXSOUR(I,J,L)»FX80UR(X,J,L)*FACTFS(L)
*** CALCULATE TOTAL POLLUTANT FLUX EMITTED FRUM EACH AIRPORT
IF(NOARPT,LE,0) GO TO 50
DO .16 LSI,6
DO 35 J«l,NQARPT
FLUX*0,
DD 37 Ill,7
FLUX«FLUX + ACFT(J,I)*EMIX(J,I)*(AIREMS(l,L,n*TMODE(l,I) * AX
l(2,L,n*TMODEC2,I) * AIREMS(3,L, I )*TMOOE(3,1))
AIRFLX(L,J)»FLUX*t,OE»3*39alO*l,OE6/(60,*XMGL(L)*OX*DY)
IF(L
-------
c
c
c
c
00 a50 J«!,25
VS(1,J)»V81U,J)
VF(I,J)«VF1(I,J)
LOiLL
*** SHIFT THE TEMPORAL DISTRIBUTION OF TRAFFIC ACTIVITY IN
DOWNTOWN LOS ANGELES AREA BACK BY ONE HOUR
THE
450
IFU.EQ.ll, AND, J, £0,17)
IF(I,EQ,12tAND.J,EQ,17)
IF(I,EQ.12,ANO(J,EQfU> LO»LL«i
SMlLES(I,J)sSURMLGa,J)*TSURFCLO+5)
FHESt J)»FWVMC(I»J)*TFREE(LO*S)
C
C
C
C
C
C
c
c
c
c
*** INPUT CARD NO, 24*26
READ (5,300X(X (I, J), I«1, 2b), J= 1, ?5)
READ e5,300)((VSl(I,JM?l»25),J*l,25), ((VF1 (I, J), Is! ,25), J*l, ?S )
*** DO LOOP FOR EACH FIVE MINUTES IN THE HOUR
DO 111 K»l,12
XK9FLOATCK.D/12,
*** CALCULATE TOTAL EMISSIONS FROM AUTOMOBILES IN EACH GRIO
SQUARE
DO 120 1*1,25
00 120 Jll,25
VS9aVS(I,J)+ *8,680E5/(DX*OY)
*** COMPLETE THE CONVERSION OF THE AUTOMOTIVE FLUX TO
FEET/MINUTE AND COMBINE WITH THE FIXED SOURCE FLUX
THE
i
PPM-
EOPP1810
EDPPlSiO
EDPP1860
EOPP1870
EOPP1880
EDPP1B90
EDPP1900
EDPP1910
EOPP1920
tOPP1930
£!>PP19UO
tDPPt960
F. DPP 1970
tOPP2030
EOPP2040
EOPP2060
tDPP2070
EOPP2100
et)PP21iO
EDPP2140.
EOPP21SO
EDPP2160
EDPP21«0
EOPP2190
EDPP2200
EDPP222Q
EDPP2240
EOPP2250
EDPP2260
EOPP2270
EOPP2280
EDPP2290
EOPP2300
EDPP2310
EDPP2320
EDPP2330
EOPP2340
EOPP2350
EDPP2360
DO 331 I«l,25 EDPP2370
DO 331 J«l,25 EOPP2380
Q2(1,I,4)BEC2»I»J)*XCARHC/XMOLC1)*FXSOUR(I»J»1)*LM*LN*FACTEV*SURMLEDPP2390
1G(1,J) EOPP2400
75
-------
331
C
C
C
02(2,
02(3,
Q2(4,
1 FXSO
02(5,
Q2(6,
,J)*E(3,I,J)/(XMOL(2)*DENCAR) t FXSOUR ( I , J,2) *LM
, J)»0,
,J)«E(3,I,J)*(1.*XCARNO)/(XMOL(U)*XCARNO*DENCAR) +
R(I,J,4)*LH
,J)»EU,I,J)/XHOL(b) * FXSOUR(I,J,5>)*L,M
,J)s£(2,I,J)*(l,*XCARHC)/XMnL(6) * FXSOUR ( I , J,6) *LM
*** ADD EMISSIONS FLUXES FROM AIRPORTS
336
41
40
55
C
C
C
66
C
C
C
86
C
C
C
C
90
91
92
554
111
&OPP2410
e.OPP«?420
EOPP2430
topp^aao
fcOPP2aSO
1
C
C
C
C
C
C
IF(NOARPT.LE,0) CO TO 55
00 40 J«t,NOARPT
M*NCORT(J)
00 41 II»1,M
DO 336 L«1.6
Q2(L»IXORT(J,II),lYORT(J,n))sQ2(L,IXORT(J,II),IYORT(J,II))
J, II)*OIST(J,LL+5)*AIRFLX(L»J)
CONTINUE
CONTINUE
CONTINUE
IF(KFLAG) 60 TO 8»
EOPP24BO
fei)PP2a<)0
6i)PP2bOO
tDPPiJIblO
ef)PP2520
PCT
***
THIE INITIAL EMISSIONS FLUXES ON THE FILE
KPLASs.TRUE,
00 86 L»l,6
I*
25)
CONTINUE
CO TO 92
*** DEFINE AND KRITF. THE TIME INTERVAL ON THE FILfc
TDATAaTlME(TDATA,5,ii)
TNEXTaTIME(TNEXT,5til)
) TDATA,TNEXT
*** CALCULATE THE TIME RATE OF CHANGE OF THE EMISSIONS FLUXFS
AND WRITE THEM ON THE FILE
DO 91 L«li6
DO 90 J « 1,25
DO 90 lat*25
Q8LOPE(I,J)»(Q2(L»I»J)«»Ql(L,I,J))/bt
WRlTt(a) QSLOPE
DO 554 L9l*6
DO 554 I«l,a5
DO 554 Jil,25
Qia,I,J)«Q2(l,I,J)
CONTINUE
CONTINUE
*** IN THE NEXT SECTION OF THE CODE, EMISSIONS ARE CALCULATED
HOURLY
DO 3 LL*6,12
*** COMPUTE AUTOMOTIVE EMISSIONS FLUXES
DO 2 M«l,3
DO 2 Ifl,25
DO 2 Jal,25
(LDPP2550
tOPP2b60
EPPP25/0
EOPP2610
EOPP2630
EOPP2640
EOPP2670
S-.DPP2680
EOPP2690
F.OPP2700
EDPP2710
fef)PP27iO
EDPP2740
60PP2760
EOPP2770
fcOPP27»0
EDPP2790
EDPP2900
t:i)PP2810
fcOPP2820
EDPH2840
tOPP2860
EOPP2870
EOPP2880
EDRP2890
EDPP2900
EDPP2910
EOPP2920
EDPP2930
EOPP2940
FOPP2950
EDPP2960
EDPP2970
EDPP2960
EDPP2990
EDPP3000
76
-------
c
c
c
c
2
C
c
c
631
C
C
c
641
640
655
C
C
C
c
c
c
c
c
c
c
690
691
C*
C
C
C "
C
LO*LL
*** SHIFT TEMPORAL DISTRIBUTION OF AUTOMOBILE TRAFFIC ACTIVITY
IN THE DOWNTOWN LOS ANGELES AREA BACK BY ONE HOUR
IF(I,EQ,ll,ANDtJtEQ,17) LO=LL»1
IF(I,EQ,12,AND,J,EQ,17) LO»LL»1
IF(I,EQ.12,AND, J.EQ.16) LQeLL-1
SMltES(I,J)sSURMLGCI,J)*T3URFUO + 5)
FMlLES(I,J)sFWYMLGCI,J)*TFREEUOt5)
E(M,I,J)s SMlUES(IrJ)*(YC3(LL*b)*EC(M)+(l,«YC3
1 LL+5) )*EH(M) ) + FMHESU, J)*XA(M)*(60,**XB(M) )
E(M,I,J)sE(M,I,J)*8,6eQE5/(DX*DY)
*** ADD FIXED SOURCE EMISSIONS FLUXES TO AUTOMOTIVE FLUXES
DO 6M Isl,25
DO 631 Jsl,2b
Q2(lf I,J)«E(2,IiJ)*XC.ARHC/XMOL(l)+FXSOU«
-------
c
c
c
c
c
692
C
C
C
C
»TOATAsTNEXT
TNEXTsTIME(TDATA,55,,l)
WRITE(4) TDATA,TNEXT
*** SET THE TIME RATE OF CHANCE OF THE EMISSIONS FLUXtS TO
ZERO SINCE EMISSIONS ARE ASSUMED TO BE CONSTANT OVtR THfc
FIRST 55 MINUTES OF THE HOUR
00 692 1st,25
OQ 692 Jsl,25
QSLOPE(I,J)»0,
*** 4RITE THt
THt FILE
TIME RATE OF CHANGE OF THE EMISSIONS FLUXES ON
693
DO 693 Lai,6
WRITEC4) QSLOPE
DO 664 1*1,6
00 66a 1*1,25
DO 664 J*l,25
Qia,I,J>*Q2(L,I,J>
CONTINUE
C
C
c
c
*** IN THE
FLUXES
NEXT SECTION
ARE PRINTED
OF CODE, HOURLY MAPS OF EMISSIONS
2000 REWIND 4
TWRITE=500,
00 1000 L»l»6
1000 READ(a) ((Q1(L,I,J),I*1,25),J*1,25)
1011 REAO(4,END«1010) TPAT*,TNfcXT
IF(TDATA,LT,T*RITE) GO TO 1001
TWRITEsTWRITE + 100,
00 1002 1*1,6
WRITE(6,1001) NAME(L),TDATA,(I,I«1,13)
1003 FORMAT(lHt,47X,'SOURCE STRENGTHS OF ',(
U3I9//)
DO 1004 JJ*1,25
1004
1005
1007
1006
1002
1001
AT «,F6tl,' PST'///SX,
1006
1010
WRITE (6, 1005) J, (Qi(L, I, J), 1=1, 13)
FORMAT(/1X,I2,4X,1P13E9,1)
WRITE(6,1007) (I,I«14,25)
FORMAT (///////5X,l2I9//)
DO 1006 JJ«1,25
J326-JJ
WRITE (6, 1005) J,(Ql(L,I,J),I«H,25)
CONTINUE
OT«TIME(TNEXT,TDATA,2)
00 1008 L"li6
REAOC4) QSLOPt
DO 1008 I>1,25
DO 1008 J*l,25
QlU,I,J)*Qi(L,I,J)
60 TO 1011
STOP
END
DT*QSLQPE(I,J)
EOPP3610
E!5PPi620
EQPP5630
fc()PP3640
fcf)PP36bO
fcOPP3670
E DPP 3680
fcDPP"4700
t.OPPi/10
E'-)PP37i»0
EOPP37i»0
tOPP37ao
F.OPP37SO
EOPP37/0
EOPP3780
tOPP3800
KOPP3H10
EOPP3820
EDPP3630
EDPP3840
EOPPJ860
KDPP3H70
tDPPJBBO
EOPP3900
fcDPP3910
KOPP3920
EOPP3930
fcDPP4940
fcOPP39liO
fcnPPJ960
EOPP3970
EOPP3980
EOPP3990
fcfiPPaOOO
fcf)PP4010
EOPP4020
EDPPaO'iO
EDPP4040
tOPP40bO
EDPP4060
EOPP4070
EOPP4080
EDPP4090
EDPP4100
EOPP4110
EDPP4120
EDPP4130
EOPP4140
EOPP4150
EDPP4160
78
-------
c
c
c
c
c
c
c
c
c
c
c
*** ADD A TIME IN MINUTES TO A 2400 HOUR TIME AND RETURN A 2400
HOUR TIME
»REAL FUNCTION TIME(Tl,T2,M)
GO TO (1,2,3),*
EOPP419Q
EOPPa200
EDPP4210
EDPP4220
UOO = T1/100
T3«T1 -100,0*1100 + T?
I1008I100 * INT(T3/60)
TIMEsilOOMOO.O * T3 • 60,0*INT(T3/60)
RETURN
EOPP4280
*** SUBTRACT TWO 2400 HOUR TIMES AMD RFTURN THE RESULT IN MlNUTf;SfcOPP4290
EOPP4500
I100»T1/100
TIMC*T1 • 1100*40,0
I100ST2/100
TIME«TIM£ • T2 + 1100*40,0
RETURN
Ef)PP«240
EDPPM
-------
oo
o
Variable
ACFT(J,I)
AIREMS (M,L,I)
AIRFLX(L,J)
AIRPRT
BETA(M,K,L)
DIST(J,I)
SYMBOL GLOSSARY
EMISSIONS DATA PREPARATION PROGRAM
Units
aircraft/day
Ib/min
Dimension
15x7
3x6x7
ppm-ft/min
6x60
Scalar
6x5x12
15x24
DT
DX
DY
E(M,I,J)
minute
feet
feet
gm/min
Scalar
Scalar
Scalar
3x25x25
Description
Number of flights per day of class I
aircraft at airport J
Aircraft emissions from class I
aircraft operating in mode M
1 reactive hydrocarbon
2 NO
3 03
4 NO2
5 CO
6 unreacting hydrocarbon
Total flux of species L from airport J
Name of the airport
The "6" correction factor that accounts
for the nonuniform distribution of trip
starts. M denotes the chemical species
M =
H
CO
HC
NO,,
Fraction of daily aircraft flights at
airport J assigned to hourly period I
(1*1 corresponds to the time period
midnight - 1 A.M.)
Time interval
Grid spacing in x-direction
Grid spacing in y-direction
Total emissions of species M from automobiles
driven in grid square (I,J)
M
-------
SYMBOL GLOSSARY (Cont'd)
EMISSIONS DATA PREPARATION PROGRAM
oo
Variable
EC (I)
EH (I)
EMIX(J,I)
FAIRHC
FAIRNO
FCARHC
Units
grams/mile
grains/mile
Dimension
engines/
aircraft
15x7
Scalar
Scalar
Scalar
Description
Automotive cold-start emissions rates
{1 CO
2 HC
3 NOX
Automotive hot-start emissions rates
!1 CO
2 HC
3 NOX
Number of engines on class I aircraft
operating at airport J
Mole fraction of reactive hydrocarbon in
total hydrocarbon emissions from aircraft
Mole fraction of NO in total NOX emissions
from aircraft
Mole fraction of reactive hydrocarbon
in total hydrocarbon emissions from
automobiles
FCARNO
Scalar
Mole fraction of NO in total NOX emissions
from automobiles
FFSNO
Scalar
Mole fraction of NO in total NOX emissions
from fixed sources
FMILES(I,J)
miles/min
25x25
Number of miles driven each minute on
freeways in grid square (I,J)
-------
00
to
Variable
FWYMLG(I,J)
FXSOUR(I,J,L)
HCEVAP
IXORT(J,I)
IYORT(J,I)
NAME(L)
SYMBOL GLOSSARY (Cont'd)
EMISSIONS DATA PREPARATION PROGRAM
Units
mile/day
Kgm/hr
Dimension
25x25
25x25x6
gm/gm mole
Scalar
15x4
15x4
Scalar
Scalar
Description
Total freeway mileage driven each day
in grid square (I,J)
Fixed source emissions assigned to
grid square (I,J)
1 reactive hydrocarbon
2 NO
3 03
4 N02
5 CO
6 unreactive hydrocarbon
Molecular weight of evaporative hydrocarbon
x-coordinates of the grid squares
receiving emissions from airport J
y-coordinates of the grid squares
receiving emissions from airport J
Controls temporal distribution of fixed
sources
0 before 0600 (local time)
1 after 0600
LM
Controls the temporal distribution of
evaporative hydrocarbon emissions from
automobiles
j 0 before 0700 (local time)
1 1 after 0700
LN =
Alphameric designation of the name of
species L
------- |